From 750a82a4ff615190a2793c0cfae9f4c1f5c75433 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Thu, 15 Feb 2024 13:23:33 +0100
Subject: [PATCH 001/443] Update doc

---
 .../mergetree-family/mergetree.md             |   2 +
 docs/en/operations/storing-data.md            | 146 ++++++++++++++++--
 2 files changed, 134 insertions(+), 14 deletions(-)

diff --git a/docs/en/engines/table-engines/mergetree-family/mergetree.md b/docs/en/engines/table-engines/mergetree-family/mergetree.md
index f185c11bab3..e1eef8db9ab 100644
--- a/docs/en/engines/table-engines/mergetree-family/mergetree.md
+++ b/docs/en/engines/table-engines/mergetree-family/mergetree.md
@@ -1106,6 +1106,8 @@ Configuration markup:
 </storage_configuration>
 ```
 
+Also see [configuring external storage options](/docs/en/operations/storing-data.md/#configuring-external-storage).
+
 :::note cache configuration
 ClickHouse versions 22.3 through 22.7 use a different cache configuration, see [using local cache](/docs/en/operations/storing-data.md/#using-local-cache) if you are using one of those versions.
 :::
diff --git a/docs/en/operations/storing-data.md b/docs/en/operations/storing-data.md
index 003277c8d4f..7a7edfb1a90 100644
--- a/docs/en/operations/storing-data.md
+++ b/docs/en/operations/storing-data.md
@@ -11,45 +11,163 @@ To work with data stored on `Amazon S3` disks use [S3](/docs/en/engines/table-en
 
 To load data from a web server with static files use a disk with type [web](#storing-data-on-webserver).
 
-## Configuring HDFS {#configuring-hdfs}
+## Configuring external storage {#configuring-external-storage}
 
-[MergeTree](/docs/en/engines/table-engines/mergetree-family/mergetree.md) and [Log](/docs/en/engines/table-engines/log-family/log.md) family table engines can store data to HDFS using a disk with type `HDFS`.
+[MergeTree](/docs/en/engines/table-engines/mergetree-family/mergetree.md) and [Log](/docs/en/engines/table-engines/log-family/log.md) family table engines can store data to `S3`, `AzureBlobStorage`, `HDFS` using a disk with types `s3`, `azure_blob_storage`, `hdfs` accordingly.
 
 Configuration markup:
 
+Let's take a loop at different storage configuration options on the example of `S3` storage.
+Firstly, define configuration in server configuration file. In order to configure `S3` storage the following configuration can be used:
+
 ``` xml
 <clickhouse>
     <storage_configuration>
         <disks>
-            <hdfs>
-                <type>hdfs</type>
-                <endpoint>hdfs://hdfs1:9000/clickhouse/</endpoint>
-            </hdfs>
+            <s3>
+                <type>s3</type>
+                <endpoint>https://s3.eu-west-1.amazonaws.com/clickhouse-eu-west-1.clickhouse.com/data/</endpoint>
+                <use_invironment_credentials>1</use_invironment_credentials>
+            </s3>
         </disks>
         <policies>
-            <hdfs>
+            <s3>
                 <volumes>
                     <main>
-                        <disk>hdfs</disk>
+                        <disk>s3</disk>
                     </main>
                 </volumes>
-            </hdfs>
+            </s3>
         </policies>
     </storage_configuration>
+</clickhouse>
+```
 
+Starting with 24.1 clickhouse version, a different type of configuration is supported in addition to the older one:
+
+``` xml
+<clickhouse>
+    <storage_configuration>
+        <disks>
+            <s3>
+                <type>object_storage</type>
+                <object_storage_type>s3</object_storage_type>
+                <metadata_type>local</metadata_type>
+                <endpoint>https://s3.eu-west-1.amazonaws.com/clickhouse-eu-west-1.clickhouse.com/data/</endpoint>
+                <use_invironment_credentials>1</use_invironment_credentials>
+            </s3>
+        </disks>
+        <policies>
+            <s3>
+                <volumes>
+                    <main>
+                        <disk>s3</disk>
+                    </main>
+                </volumes>
+            </s3>
+        </policies>
+    </storage_configuration>
+</clickhouse>
+```
+
+In order to make a specific kind of storage a default option for all `MergeTree` tables add the following section to configuration file:
+
+``` xml
+<clickhouse>
     <merge_tree>
-        <min_bytes_for_wide_part>0</min_bytes_for_wide_part>
+        <storage_policy>s3</storage_policy>
     </merge_tree>
 </clickhouse>
 ```
 
-Required parameters:
+If you want to configure a specific storage policy only to specific table, you can define it in settings while creating the table:
 
-- `endpoint` — HDFS endpoint URL in `path` format. Endpoint URL should contain a root path to store data.
+``` sql
+CREATE TABLE test (a Int32, b String)
+ENGINE = MergeTree() ORDER BY a
+SETTINGS storage_policy = 's3';
+```
 
-Optional parameters:
+You can also use `disk` instead of `storage_policy`. In this case it is not requires to have `storage_policy` section in configuration file, only `disk` section would be enough.
 
-- `min_bytes_for_seek` — The minimal number of bytes to use seek operation instead of sequential read. Default value: `1 Mb`.
+``` sql
+CREATE TABLE test (a Int32, b String)
+ENGINE = MergeTree() ORDER BY a
+SETTINGS disk = 's3';
+```
+
+There is also a possibility to specify storage configuration without a preconfigured disk in configuration file:
+
+``` sql
+CREATE TABLE test (a Int32, b String)
+ENGINE = MergeTree() ORDER BY a
+SETTINGS disk = disk(name = 's3_disk', type = 's3', endpoint = 'https://s3.eu-west-1.amazonaws.com/clickhouse-eu-west-1.clickhouse.com/data/', use_environment_credentials = 1);
+```
+
+Adding cache is also possible:
+
+``` sql
+CREATE TABLE test (a Int32, b String)
+ENGINE = MergeTree() ORDER BY a
+SETTINGS disk = disk(name = 'cached_s3_disk', type = 'cache', max_size = '10Gi', path = '/s3_cache', disk = disk(name = 's3_disk', type = 's3', endpoint = 'https://s3.eu-west-1.amazonaws.com/clickhouse-eu-west-1.clickhouse.com/data/', use_environment_credentials = 1));
+```
+
+A combination of config file disk configuration and sql-defined configuration is also possible:
+
+``` sql
+CREATE TABLE test (a Int32, b String)
+ENGINE = MergeTree() ORDER BY a
+SETTINGS disk = disk(name = 'cached_s3_disk', type = 'cache', max_size = '10Gi', path = '/s3_cache', disk = 's3');
+```
+
+Here `s3` is a disk name from server configuration file, while `cache` disk is defined via sql.
+
+Let's take a closer look at configuration parameters.
+
+All disk configuration require `type` section, equal to one of `s3`, `azure_blob_storage`, `hdfs`, `local`, `cache`, `web`. Then goes configuration of a specific storage type.
+Starting from 24.1 clickhouse version, you can you a new configuration option. For it you are required to specify `type` as `object_storage`, `object_storage_type` as one of `s3`, `azure_blob_storage`, `hdfs`, `local`, `cache`, `web`, and optionally you can specify `metadata_type`, which is `local` by default, but it can also be set to `plain`, `web`.
+
+E.g. first configuration option:
+``` xml
+<s3>
+    <type>s3</type>
+    <endpoint>https://s3.eu-west-1.amazonaws.com/clickhouse-eu-west-1.clickhouse.com/data/</endpoint>
+    <use_invironment_credentials>1</use_invironment_credentials>
+</s3>
+```
+
+and second (from `24.1`):
+``` xml
+<s3>
+    <type>object_storage</type>
+    <object_storage_type>s3</object_storage_type>
+    <metadata_type>local</metadata_type>
+    <endpoint>https://s3.eu-west-1.amazonaws.com/clickhouse-eu-west-1.clickhouse.com/data/</endpoint>
+    <use_invironment_credentials>1</use_invironment_credentials>
+</s3>
+```
+
+Configuration like
+``` xml
+<s3_plain>
+    <type>s3_plain</type>
+    <endpoint>https://s3.eu-west-1.amazonaws.com/clickhouse-eu-west-1.clickhouse.com/data/</endpoint>
+    <use_invironment_credentials>1</use_invironment_credentials>
+</s3_plain>
+```
+
+is equal to
+``` xml
+<s3_plain>
+    <type>object_storage</type>
+    <object_storage_type>s3</object_storage_type>
+    <metadata_type>plain</metadata_type>
+    <endpoint>https://s3.eu-west-1.amazonaws.com/clickhouse-eu-west-1.clickhouse.com/data/</endpoint>
+    <use_invironment_credentials>1</use_invironment_credentials>
+</s3_plain>
+```
+
+For details configuration options of each storage see [MergeTree](/docs/en/engines/table-engines/mergetree-family/mergetree.md).
 
 ## Using Virtual File System for Data Encryption {#encrypted-virtual-file-system}
 

From 9bcd4daabe56e29132fc5098420afb4dcba9001d Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Thu, 15 Feb 2024 16:19:31 +0100
Subject: [PATCH 002/443] Better

---
 .../mergetree-family/mergetree.md             | 294 +------------
 docs/en/operations/storing-data.md            | 411 +++++++++++++++---
 2 files changed, 346 insertions(+), 359 deletions(-)

diff --git a/docs/en/engines/table-engines/mergetree-family/mergetree.md b/docs/en/engines/table-engines/mergetree-family/mergetree.md
index e1eef8db9ab..0fff13c906f 100644
--- a/docs/en/engines/table-engines/mergetree-family/mergetree.md
+++ b/docs/en/engines/table-engines/mergetree-family/mergetree.md
@@ -987,49 +987,6 @@ ORDER BY (postcode1, postcode2, addr1, addr2)
   # highlight-end
 ```
 
-### Nested Dynamic Storage
-
-This example query builds on the above dynamic disk configuration and shows how to
-use a local disk to cache data from a table stored at a URL. Neither the cache disk
-nor the web storage is configured in the ClickHouse configuration files; both are
-configured in the CREATE/ATTACH query settings.
-
-In the settings highlighted below notice that the disk of `type=web` is nested within
-the disk of `type=cache`.
-
-```sql
-ATTACH TABLE uk_price_paid UUID 'cf712b4f-2ca8-435c-ac23-c4393efe52f7'
-(
-    price UInt32,
-    date Date,
-    postcode1 LowCardinality(String),
-    postcode2 LowCardinality(String),
-    type Enum8('other' = 0, 'terraced' = 1, 'semi-detached' = 2, 'detached' = 3, 'flat' = 4),
-    is_new UInt8,
-    duration Enum8('unknown' = 0, 'freehold' = 1, 'leasehold' = 2),
-    addr1 String,
-    addr2 String,
-    street LowCardinality(String),
-    locality LowCardinality(String),
-    town LowCardinality(String),
-    district LowCardinality(String),
-    county LowCardinality(String)
-)
-ENGINE = MergeTree
-ORDER BY (postcode1, postcode2, addr1, addr2)
-  # highlight-start
-  SETTINGS disk = disk(
-    type=cache,
-    max_size='1Gi',
-    path='/var/lib/clickhouse/custom_disk_cache/',
-    disk=disk(
-      type=web,
-      endpoint='https://raw.githubusercontent.com/ClickHouse/web-tables-demo/main/web/'
-      )
-  );
-  # highlight-end
-```
-
 ### Details {#details}
 
 In the case of `MergeTree` tables, data is getting to disk in different ways:
@@ -1058,19 +1015,17 @@ During this time, they are not moved to other volumes or disks. Therefore, until
 
 User can assign new big parts to different disks of a [JBOD](https://en.wikipedia.org/wiki/Non-RAID_drive_architectures) volume in a balanced way using the [min_bytes_to_rebalance_partition_over_jbod](/docs/en/operations/settings/merge-tree-settings.md/#min-bytes-to-rebalance-partition-over-jbod) setting.
 
-## Using S3 for Data Storage {#table_engine-mergetree-s3}
+## Using External Storage for Data Storage {#table_engine-mergetree-s3}
 
-:::note
-Google Cloud Storage (GCS) is also supported using the type `s3`. See [GCS backed MergeTree](/docs/en/integrations/gcs).
-:::
+[MergeTree](/docs/en/engines/table-engines/mergetree-family/mergetree.md) family table engines can store data to `S3`, `AzureBlobStorage`, `HDFS` using a disk with types `s3`, `azure_blob_storage`, `hdfs` accordingly. See [configuring external storage options](/docs/en/operations/storing-data.md/#configuring-external-storage) for more details.
 
-`MergeTree` family table engines can store data to [S3](https://aws.amazon.com/s3/) using a disk with type `s3`.
+Example for [S3](https://aws.amazon.com/s3/) as external storage using a disk with type `s3`.
 
 Configuration markup:
 ``` xml
 <storage_configuration>
     ...
-    <disks>
+e   <disks>
         <s3>
             <type>s3</type>
             <support_batch_delete>true</support_batch_delete>
@@ -1112,247 +1067,6 @@ Also see [configuring external storage options](/docs/en/operations/storing-data
 ClickHouse versions 22.3 through 22.7 use a different cache configuration, see [using local cache](/docs/en/operations/storing-data.md/#using-local-cache) if you are using one of those versions.
 :::
 
-### Configuring the S3 disk
-
-Required parameters:
-
-- `endpoint` — S3 endpoint URL in `path` or `virtual hosted` [styles](https://docs.aws.amazon.com/AmazonS3/latest/dev/VirtualHosting.html). Endpoint URL should contain a bucket and root path to store data.
-- `access_key_id` — S3 access key id.
-- `secret_access_key` — S3 secret access key.
-
-Optional parameters:
-
-- `region` — S3 region name.
-- `support_batch_delete` — This controls the check to see if batch deletes are supported. Set this to `false` when using Google Cloud Storage (GCS) as GCS does not support batch deletes and preventing the checks will prevent error messages in the logs.
-- `use_environment_credentials` — Reads AWS credentials from the Environment variables AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY and AWS_SESSION_TOKEN if they exist. Default value is `false`.
-- `use_insecure_imds_request` — If set to `true`, S3 client will use insecure IMDS request while obtaining credentials from Amazon EC2 metadata. Default value is `false`.
-- `expiration_window_seconds` — Grace period for checking if expiration-based credentials have expired. Optional, default value is `120`.
-- `proxy` — Proxy configuration for S3 endpoint. Each `uri` element inside `proxy` block should contain a proxy URL.
-- `connect_timeout_ms` — Socket connect timeout in milliseconds. Default value is `10 seconds`.
-- `request_timeout_ms` — Request timeout in milliseconds. Default value is `5 seconds`.
-- `retry_attempts` — Number of retry attempts in case of failed request. Default value is `10`.
-- `single_read_retries` — Number of retry attempts in case of connection drop during read. Default value is `4`.
-- `min_bytes_for_seek` — Minimal number of bytes to use seek operation instead of sequential read. Default value is `1 Mb`.
-- `metadata_path` — Path on local FS to store metadata files for S3. Default value is `/var/lib/clickhouse/disks/<disk_name>/`.
-- `skip_access_check` — If true, disk access checks will not be performed on disk start-up. Default value is `false`.
-- `header` —  Adds specified HTTP header to a request to given endpoint. Optional, can be specified multiple times.
-- `server_side_encryption_customer_key_base64` — If specified, required headers for accessing S3 objects with SSE-C encryption will be set.
-- `server_side_encryption_kms_key_id` - If specified, required headers for accessing S3 objects with [SSE-KMS encryption](https://docs.aws.amazon.com/AmazonS3/latest/userguide/UsingKMSEncryption.html) will be set. If an empty string is specified, the AWS managed S3 key will be used. Optional.
-- `server_side_encryption_kms_encryption_context` - If specified alongside `server_side_encryption_kms_key_id`, the given encryption context header for SSE-KMS will be set. Optional.
-- `server_side_encryption_kms_bucket_key_enabled` - If specified alongside `server_side_encryption_kms_key_id`, the header to enable S3 bucket keys for SSE-KMS will be set. Optional, can be `true` or `false`, defaults to nothing (matches the bucket-level setting).
-- `s3_max_put_rps` — Maximum PUT requests per second rate before throttling. Default value is `0` (unlimited).
-- `s3_max_put_burst` — Max number of requests that can be issued simultaneously before hitting request per second limit. By default (`0` value) equals to `s3_max_put_rps`.
-- `s3_max_get_rps` — Maximum GET requests per second rate before throttling. Default value is `0` (unlimited).
-- `s3_max_get_burst` — Max number of requests that can be issued simultaneously before hitting request per second limit. By default (`0` value) equals to `s3_max_get_rps`.
-- `read_resource` — Resource name to be used for [scheduling](/docs/en/operations/workload-scheduling.md) of read requests to this disk. Default value is empty string (IO scheduling is not enabled for this disk).
-- `write_resource` — Resource name to be used for [scheduling](/docs/en/operations/workload-scheduling.md) of write requests to this disk. Default value is empty string (IO scheduling is not enabled for this disk).
-- `key_template` — Define the format with which the object keys are generated. By default, Clickhouse takes `root path` from `endpoint` option and adds random generated suffix. That suffix is a dir with 3 random symbols and a file name with 29 random symbols. With that option you have a full control how to the object keys are generated. Some usage scenarios require having random symbols in the prefix or in the middle of object key. For example: `[a-z]{3}-prefix-random/constant-part/random-middle-[a-z]{3}/random-suffix-[a-z]{29}`. The value is parsed with [`re2`](https://github.com/google/re2/wiki/Syntax). Only some subset of the syntax is supported. Check if your preferred format is supported before using that option. Disk isn't initialized if clickhouse is unable to generate a key by the value of `key_template`. It requires enabled feature flag [storage_metadata_write_full_object_key](/docs/en/operations/settings/settings#storage_metadata_write_full_object_key). It forbids declaring the `root path` in `endpoint` option. It requires definition of the option `key_compatibility_prefix`.
-- `key_compatibility_prefix` — That option is required when option `key_template` is in use. In order to be able to read the objects keys which were stored in the metadata files with the metadata version lower that `VERSION_FULL_OBJECT_KEY`, the previous `root path` from the `endpoint` option should be set here.
-
-### Configuring the cache
-
-This is the cache configuration from above:
-```xml
-        <s3_cache>
-            <type>cache</type>
-            <disk>s3</disk>
-            <path>/var/lib/clickhouse/disks/s3_cache/</path>
-            <max_size>10Gi</max_size>
-        </s3_cache>
-```
-
-These parameters define the cache layer:
-- `type` — If a disk is of type `cache` it caches mark and index files in memory.
-- `disk` — The name of the disk that will be cached.
-
-Cache parameters:
-- `path` — The path where metadata for the cache is stored.
-- `max_size` — The size (amount of disk space) that the cache can grow to.
-
-:::tip
-There are several other cache parameters that you can use to tune your storage, see [using local cache](/docs/en/operations/storing-data.md/#using-local-cache) for the details.
-:::
-
-S3 disk can be configured as `main` or `cold` storage:
-``` xml
-<storage_configuration>
-    ...
-    <disks>
-        <s3>
-            <type>s3</type>
-            <endpoint>https://clickhouse-public-datasets.s3.amazonaws.com/my-bucket/root-path/</endpoint>
-            <access_key_id>your_access_key_id</access_key_id>
-            <secret_access_key>your_secret_access_key</secret_access_key>
-        </s3>
-    </disks>
-    <policies>
-        <s3_main>
-            <volumes>
-                <main>
-                    <disk>s3</disk>
-                </main>
-            </volumes>
-        </s3_main>
-        <s3_cold>
-            <volumes>
-                <main>
-                    <disk>default</disk>
-                </main>
-                <external>
-                    <disk>s3</disk>
-                </external>
-            </volumes>
-            <move_factor>0.2</move_factor>
-        </s3_cold>
-    </policies>
-    ...
-</storage_configuration>
-```
-
-In case of `cold` option a data can be moved to S3 if local disk free size will be smaller than `move_factor * disk_size` or by TTL move rule.
-
-## Using Azure Blob Storage for Data Storage {#table_engine-mergetree-azure-blob-storage}
-
-`MergeTree` family table engines can store data to [Azure Blob Storage](https://azure.microsoft.com/en-us/services/storage/blobs/) using a disk with type `azure_blob_storage`.
-
-As of February 2022, this feature is still a fresh addition, so expect that some Azure Blob Storage functionalities might be unimplemented.
-
-Configuration markup:
-``` xml
-<storage_configuration>
-    ...
-    <disks>
-        <blob_storage_disk>
-            <type>azure_blob_storage</type>
-            <storage_account_url>http://account.blob.core.windows.net</storage_account_url>
-            <container_name>container</container_name>
-            <account_name>account</account_name>
-            <account_key>pass123</account_key>
-            <metadata_path>/var/lib/clickhouse/disks/blob_storage_disk/</metadata_path>
-            <cache_path>/var/lib/clickhouse/disks/blob_storage_disk/cache/</cache_path>
-            <skip_access_check>false</skip_access_check>
-        </blob_storage_disk>
-    </disks>
-    ...
-</storage_configuration>
-```
-
-Connection parameters:
-* `storage_account_url` - **Required**, Azure Blob Storage account URL, like `http://account.blob.core.windows.net` or `http://azurite1:10000/devstoreaccount1`.
-* `container_name` - Target container name, defaults to `default-container`.
-* `container_already_exists` - If set to `false`, a new container `container_name` is created in the storage account, if set to `true`, disk connects to the container directly, and if left unset, disk connects to the account, checks if the container `container_name` exists, and creates it if it doesn't exist yet.
-
-Authentication parameters (the disk will try all available methods **and** Managed Identity Credential):
-* `connection_string` - For authentication using a connection string.
-* `account_name` and `account_key` - For authentication using Shared Key.
-
-Limit parameters (mainly for internal usage):
-* `s3_max_single_part_upload_size` - Limits the size of a single block upload to Blob Storage.
-* `min_bytes_for_seek` - Limits the size of a seekable region.
-* `max_single_read_retries` - Limits the number of attempts to read a chunk of data from Blob Storage.
-* `max_single_download_retries` - Limits the number of attempts to download a readable buffer from Blob Storage.
-* `thread_pool_size` - Limits the number of threads with which `IDiskRemote` is instantiated.
-* `s3_max_inflight_parts_for_one_file` - Limits the number of put requests that can be run concurrently for one object.
-
-Other parameters:
-* `metadata_path` - Path on local FS to store metadata files for Blob Storage. Default value is `/var/lib/clickhouse/disks/<disk_name>/`.
-* `skip_access_check` - If true, disk access checks will not be performed on disk start-up. Default value is `false`.
-* `read_resource` — Resource name to be used for [scheduling](/docs/en/operations/workload-scheduling.md) of read requests to this disk. Default value is empty string (IO scheduling is not enabled for this disk).
-* `write_resource` — Resource name to be used for [scheduling](/docs/en/operations/workload-scheduling.md) of write requests to this disk. Default value is empty string (IO scheduling is not enabled for this disk).
-
-Examples of working configurations can be found in integration tests directory (see e.g. [test_merge_tree_azure_blob_storage](https://github.com/ClickHouse/ClickHouse/blob/master/tests/integration/test_merge_tree_azure_blob_storage/configs/config.d/storage_conf.xml) or [test_azure_blob_storage_zero_copy_replication](https://github.com/ClickHouse/ClickHouse/blob/master/tests/integration/test_azure_blob_storage_zero_copy_replication/configs/config.d/storage_conf.xml)).
-
-:::note Zero-copy replication is not ready for production
-Zero-copy replication is disabled by default in ClickHouse version 22.8 and higher.  This feature is not recommended for production use.
-:::
-
-## HDFS storage {#hdfs-storage}
-
-In this sample configuration:
-- the disk is of type `hdfs`
-- the data is hosted at `hdfs://hdfs1:9000/clickhouse/`
-
-```xml
-<clickhouse>
-    <storage_configuration>
-        <disks>
-            <hdfs>
-                <type>hdfs</type>
-                <endpoint>hdfs://hdfs1:9000/clickhouse/</endpoint>
-                <skip_access_check>true</skip_access_check>
-            </hdfs>
-            <hdd>
-                <type>local</type>
-                <path>/</path>
-            </hdd>
-        </disks>
-        <policies>
-            <hdfs>
-                <volumes>
-                    <main>
-                        <disk>hdfs</disk>
-                    </main>
-                    <external>
-                        <disk>hdd</disk>
-                    </external>
-                </volumes>
-            </hdfs>
-        </policies>
-    </storage_configuration>
-</clickhouse>
-```
-
-## Web storage (read-only) {#web-storage}
-
-Web storage can be used for read-only purposes. An example use is for hosting sample
-data, or for migrating data.
-
-:::tip
-Storage can also be configured temporarily within a query, if a web dataset is not expected
-to be used routinely, see [dynamic storage](#dynamic-storage) and skip editing the
-configuration file.
-:::
-
-In this sample configuration:
-- the disk is of type `web`
-- the data is hosted at `http://nginx:80/test1/`
-- a cache on local storage is used
-
-```xml
-<clickhouse>
-    <storage_configuration>
-        <disks>
-            <web>
-                <type>web</type>
-                <endpoint>http://nginx:80/test1/</endpoint>
-            </web>
-            <cached_web>
-                <type>cache</type>
-                <disk>web</disk>
-                <path>cached_web_cache/</path>
-                <max_size>100000000</max_size>
-            </cached_web>
-        </disks>
-        <policies>
-            <web>
-                <volumes>
-                    <main>
-                        <disk>web</disk>
-                    </main>
-                </volumes>
-            </web>
-            <cached_web>
-                <volumes>
-                    <main>
-                        <disk>cached_web</disk>
-                    </main>
-                </volumes>
-            </cached_web>
-        </policies>
-    </storage_configuration>
-</clickhouse>
-```
-
 ## Virtual Columns {#virtual-columns}
 
 - `_part` — Name of a part.
diff --git a/docs/en/operations/storing-data.md b/docs/en/operations/storing-data.md
index 7a7edfb1a90..baf4e1999a7 100644
--- a/docs/en/operations/storing-data.md
+++ b/docs/en/operations/storing-data.md
@@ -5,21 +5,68 @@ sidebar_label: "External Disks for Storing Data"
 title: "External Disks for Storing Data"
 ---
 
-Data, processed in ClickHouse, is usually stored in the local file system — on the same machine with the ClickHouse server. That requires large-capacity disks, which can be expensive enough. To avoid that you can store the data remotely — on [Amazon S3](https://aws.amazon.com/s3/) disks or in the Hadoop Distributed File System ([HDFS](https://hadoop.apache.org/docs/current/hadoop-project-dist/hadoop-hdfs/HdfsDesign.html)).
+Data, processed in ClickHouse, is usually stored in the local file system — on the same machine with the ClickHouse server. That requires large-capacity disks, which can be expensive enough. To avoid that you can store the data remotely. Various storages are supported:
+1. [Amazon S3](https://aws.amazon.com/s3/) object storage.
+2. The Hadoop Distributed File System ([HDFS](https://hadoop.apache.org/docs/current/hadoop-project-dist/hadoop-hdfs/HdfsDesign.html))
+3. [Azure Blob Storage](https://azure.microsoft.com/en-us/products/storage/blobs).
 
-To work with data stored on `Amazon S3` disks use [S3](/docs/en/engines/table-engines/integrations/s3.md) table engine, and to work with data in the Hadoop Distributed File System — [HDFS](/docs/en/engines/table-engines/integrations/hdfs.md) table engine.
-
-To load data from a web server with static files use a disk with type [web](#storing-data-on-webserver).
+Note: to work with data stored on `Amazon S3` disks use [S3](/docs/en/engines/table-engines/integrations/s3.md) table engine, to work with data in the Hadoop Distributed File System — [HDFS](/docs/en/engines/table-engines/integrations/hdfs.md) table engine, and to work with data stored in Azure Blob Storage use [AzureBlobStorage](/docs/en/engines/table-engines/integrations/AzureBlobStorage.md) table engine. They are different from external storage described on this page as they allow to read data stored in some general file format (like Parquet), while on this page we are describing storage configuration for ClickHouse `MergeTree` famility or `Log` family tables.
 
 ## Configuring external storage {#configuring-external-storage}
 
 [MergeTree](/docs/en/engines/table-engines/mergetree-family/mergetree.md) and [Log](/docs/en/engines/table-engines/log-family/log.md) family table engines can store data to `S3`, `AzureBlobStorage`, `HDFS` using a disk with types `s3`, `azure_blob_storage`, `hdfs` accordingly.
 
-Configuration markup:
+Disk configuration requires:
+1. `type` section, equal to one of `s3`, `azure_blob_storage`, `hdfs`, `local_blob_storage`, `web`.
+2. Configuration of a specific external storage type.
 
-Let's take a loop at different storage configuration options on the example of `S3` storage.
-Firstly, define configuration in server configuration file. In order to configure `S3` storage the following configuration can be used:
+Starting from 24.1 clickhouse version, it is possible to use a new configuration option.
+It requires to specify:
+1. `type` equal to `object_storage`
+2. `object_storage_type`, equal to one of `s3`, `azure_blob_storage`, `hdfs`, `local_blob_storage`, `web`.
+Optionally, `metadata_type` can be specified (it is equal to `local` by default), but it can also be set to `plain`, `web`.
 
+E.g. configuration option
+``` xml
+<s3>
+    <type>s3</type>
+    <endpoint>https://s3.eu-west-1.amazonaws.com/clickhouse-eu-west-1.clickhouse.com/data/</endpoint>
+    <use_invironment_credentials>1</use_invironment_credentials>
+</s3>
+```
+
+is equal to configuration (from `24.1`):
+``` xml
+<s3>
+    <type>object_storage</type>
+    <object_storage_type>s3</object_storage_type>
+    <metadata_type>local</metadata_type>
+    <endpoint>https://s3.eu-west-1.amazonaws.com/clickhouse-eu-west-1.clickhouse.com/data/</endpoint>
+    <use_invironment_credentials>1</use_invironment_credentials>
+</s3>
+```
+
+Configuration
+``` xml
+<s3_plain>
+    <type>s3_plain</type>
+    <endpoint>https://s3.eu-west-1.amazonaws.com/clickhouse-eu-west-1.clickhouse.com/data/</endpoint>
+    <use_invironment_credentials>1</use_invironment_credentials>
+</s3_plain>
+```
+
+is equal to
+``` xml
+<s3_plain>
+    <type>object_storage</type>
+    <object_storage_type>s3</object_storage_type>
+    <metadata_type>plain</metadata_type>
+    <endpoint>https://s3.eu-west-1.amazonaws.com/clickhouse-eu-west-1.clickhouse.com/data/</endpoint>
+    <use_invironment_credentials>1</use_invironment_credentials>
+</s3_plain>
+```
+
+Example of full storage configuration will look like:
 ``` xml
 <clickhouse>
     <storage_configuration>
@@ -43,8 +90,7 @@ Firstly, define configuration in server configuration file. In order to configur
 </clickhouse>
 ```
 
-Starting with 24.1 clickhouse version, a different type of configuration is supported in addition to the older one:
-
+Starting with 24.1 clickhouse version, it can also look like:
 ``` xml
 <clickhouse>
     <storage_configuration>
@@ -71,7 +117,6 @@ Starting with 24.1 clickhouse version, a different type of configuration is supp
 ```
 
 In order to make a specific kind of storage a default option for all `MergeTree` tables add the following section to configuration file:
-
 ``` xml
 <clickhouse>
     <merge_tree>
@@ -96,80 +141,259 @@ ENGINE = MergeTree() ORDER BY a
 SETTINGS disk = 's3';
 ```
 
-There is also a possibility to specify storage configuration without a preconfigured disk in configuration file:
+## Dynamic Configuration {#dynamic-configuration}
 
-``` sql
-CREATE TABLE test (a Int32, b String)
-ENGINE = MergeTree() ORDER BY a
-SETTINGS disk = disk(name = 's3_disk', type = 's3', endpoint = 'https://s3.eu-west-1.amazonaws.com/clickhouse-eu-west-1.clickhouse.com/data/', use_environment_credentials = 1);
+There is also a possibility to specify storage configuration without a predefined disk in configuration in a configuration file, but can be configured in the CREATE/ATTACH query settings.
+
+The following example query builds on the above dynamic disk configuration and shows how to use a local disk to cache data from a table stored at a URL.
+
+```sql
+ATTACH TABLE uk_price_paid UUID 'cf712b4f-2ca8-435c-ac23-c4393efe52f7'
+(
+    price UInt32,
+    date Date,
+    postcode1 LowCardinality(String),
+    postcode2 LowCardinality(String),
+    type Enum8('other' = 0, 'terraced' = 1, 'semi-detached' = 2, 'detached' = 3, 'flat' = 4),
+    is_new UInt8,
+    duration Enum8('unknown' = 0, 'freehold' = 1, 'leasehold' = 2),
+    addr1 String,
+    addr2 String,
+    street LowCardinality(String),
+    locality LowCardinality(String),
+    town LowCardinality(String),
+    district LowCardinality(String),
+    county LowCardinality(String)
+)
+ENGINE = MergeTree
+ORDER BY (postcode1, postcode2, addr1, addr2)
+  # highlight-start
+  SETTINGS disk = disk(
+    type=web,
+    endpoint='https://raw.githubusercontent.com/ClickHouse/web-tables-demo/main/web/'
+  );
+  # highlight-end
 ```
 
-Adding cache is also possible:
+The example below adds cache to external storage.
 
-``` sql
-CREATE TABLE test (a Int32, b String)
-ENGINE = MergeTree() ORDER BY a
-SETTINGS disk = disk(name = 'cached_s3_disk', type = 'cache', max_size = '10Gi', path = '/s3_cache', disk = disk(name = 's3_disk', type = 's3', endpoint = 'https://s3.eu-west-1.amazonaws.com/clickhouse-eu-west-1.clickhouse.com/data/', use_environment_credentials = 1));
+```sql
+ATTACH TABLE uk_price_paid UUID 'cf712b4f-2ca8-435c-ac23-c4393efe52f7'
+(
+    price UInt32,
+    date Date,
+    postcode1 LowCardinality(String),
+    postcode2 LowCardinality(String),
+    type Enum8('other' = 0, 'terraced' = 1, 'semi-detached' = 2, 'detached' = 3, 'flat' = 4),
+    is_new UInt8,
+    duration Enum8('unknown' = 0, 'freehold' = 1, 'leasehold' = 2),
+    addr1 String,
+    addr2 String,
+    street LowCardinality(String),
+    locality LowCardinality(String),
+    town LowCardinality(String),
+    district LowCardinality(String),
+    county LowCardinality(String)
+)
+ENGINE = MergeTree
+ORDER BY (postcode1, postcode2, addr1, addr2)
+  # highlight-start
+  SETTINGS disk = disk(
+    type=cache,
+    max_size='1Gi',
+    path='/var/lib/clickhouse/custom_disk_cache/',
+    disk=disk(
+      type=web,
+      endpoint='https://raw.githubusercontent.com/ClickHouse/web-tables-demo/main/web/'
+      )
+  );
+  # highlight-end
 ```
 
-A combination of config file disk configuration and sql-defined configuration is also possible:
+In the settings highlighted below notice that the disk of `type=web` is nested within
+the disk of `type=cache`.
 
-``` sql
-CREATE TABLE test (a Int32, b String)
-ENGINE = MergeTree() ORDER BY a
-SETTINGS disk = disk(name = 'cached_s3_disk', type = 'cache', max_size = '10Gi', path = '/s3_cache', disk = 's3');
+A combination of config-based configuration and sql-defined configuration is also possible:
+
+```sql
+ATTACH TABLE uk_price_paid UUID 'cf712b4f-2ca8-435c-ac23-c4393efe52f7'
+(
+    price UInt32,
+    date Date,
+    postcode1 LowCardinality(String),
+    postcode2 LowCardinality(String),
+    type Enum8('other' = 0, 'terraced' = 1, 'semi-detached' = 2, 'detached' = 3, 'flat' = 4),
+    is_new UInt8,
+    duration Enum8('unknown' = 0, 'freehold' = 1, 'leasehold' = 2),
+    addr1 String,
+    addr2 String,
+    street LowCardinality(String),
+    locality LowCardinality(String),
+    town LowCardinality(String),
+    district LowCardinality(String),
+    county LowCardinality(String)
+)
+ENGINE = MergeTree
+ORDER BY (postcode1, postcode2, addr1, addr2)
+  # highlight-start
+  SETTINGS disk = disk(
+    type=cache,
+    max_size='1Gi',
+    path='/var/lib/clickhouse/custom_disk_cache/',
+    disk=disk(
+      type=web,
+      endpoint='https://raw.githubusercontent.com/ClickHouse/web-tables-demo/main/web/'
+      )
+  );
+  # highlight-end
 ```
 
-Here `s3` is a disk name from server configuration file, while `cache` disk is defined via sql.
+where `web` is a from a server configuration file:
 
-Let's take a closer look at configuration parameters.
-
-All disk configuration require `type` section, equal to one of `s3`, `azure_blob_storage`, `hdfs`, `local`, `cache`, `web`. Then goes configuration of a specific storage type.
-Starting from 24.1 clickhouse version, you can you a new configuration option. For it you are required to specify `type` as `object_storage`, `object_storage_type` as one of `s3`, `azure_blob_storage`, `hdfs`, `local`, `cache`, `web`, and optionally you can specify `metadata_type`, which is `local` by default, but it can also be set to `plain`, `web`.
-
-E.g. first configuration option:
 ``` xml
-<s3>
-    <type>s3</type>
-    <endpoint>https://s3.eu-west-1.amazonaws.com/clickhouse-eu-west-1.clickhouse.com/data/</endpoint>
-    <use_invironment_credentials>1</use_invironment_credentials>
-</s3>
+<storage_configuration>
+    <disks>
+        <web>
+            <type>web</type>
+            <endpoint>'https://raw.githubusercontent.com/ClickHouse/web-tables-demo/main/web/'</endpoint>
+        </web>
+    </disks>
+</storage_configuration>
 ```
 
-and second (from `24.1`):
+### Using S3 Storage {#s3-storage}
+
+Required parameters:
+
+- `endpoint` — S3 endpoint URL in `path` or `virtual hosted` [styles](https://docs.aws.amazon.com/AmazonS3/latest/dev/VirtualHosting.html). Endpoint URL should contain a bucket and root path to store data.
+- `access_key_id` — S3 access key id.
+- `secret_access_key` — S3 secret access key.
+
+Optional parameters:
+
+- `region` — S3 region name.
+- `support_batch_delete` — This controls the check to see if batch deletes are supported. Set this to `false` when using Google Cloud Storage (GCS) as GCS does not support batch deletes and preventing the checks will prevent error messages in the logs.
+- `use_environment_credentials` — Reads AWS credentials from the Environment variables AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY and AWS_SESSION_TOKEN if they exist. Default value is `false`.
+- `use_insecure_imds_request` — If set to `true`, S3 client will use insecure IMDS request while obtaining credentials from Amazon EC2 metadata. Default value is `false`.
+- `expiration_window_seconds` — Grace period for checking if expiration-based credentials have expired. Optional, default value is `120`.
+- `proxy` — Proxy configuration for S3 endpoint. Each `uri` element inside `proxy` block should contain a proxy URL.
+- `connect_timeout_ms` — Socket connect timeout in milliseconds. Default value is `10 seconds`.
+- `request_timeout_ms` — Request timeout in milliseconds. Default value is `5 seconds`.
+- `retry_attempts` — Number of retry attempts in case of failed request. Default value is `10`.
+- `single_read_retries` — Number of retry attempts in case of connection drop during read. Default value is `4`.
+- `min_bytes_for_seek` — Minimal number of bytes to use seek operation instead of sequential read. Default value is `1 Mb`.
+- `metadata_path` — Path on local FS to store metadata files for S3. Default value is `/var/lib/clickhouse/disks/<disk_name>/`.
+- `skip_access_check` — If true, disk access checks will not be performed on disk start-up. Default value is `false`.
+- `header` —  Adds specified HTTP header to a request to given endpoint. Optional, can be specified multiple times.
+- `server_side_encryption_customer_key_base64` — If specified, required headers for accessing S3 objects with SSE-C encryption will be set.
+- `server_side_encryption_kms_key_id` - If specified, required headers for accessing S3 objects with [SSE-KMS encryption](https://docs.aws.amazon.com/AmazonS3/latest/userguide/UsingKMSEncryption.html) will be set. If an empty string is specified, the AWS managed S3 key will be used. Optional.
+- `server_side_encryption_kms_encryption_context` - If specified alongside `server_side_encryption_kms_key_id`, the given encryption context header for SSE-KMS will be set. Optional.
+- `server_side_encryption_kms_bucket_key_enabled` - If specified alongside `server_side_encryption_kms_key_id`, the header to enable S3 bucket keys for SSE-KMS will be set. Optional, can be `true` or `false`, defaults to nothing (matches the bucket-level setting).
+- `s3_max_put_rps` — Maximum PUT requests per second rate before throttling. Default value is `0` (unlimited).
+- `s3_max_put_burst` — Max number of requests that can be issued simultaneously before hitting request per second limit. By default (`0` value) equals to `s3_max_put_rps`.
+- `s3_max_get_rps` — Maximum GET requests per second rate before throttling. Default value is `0` (unlimited).
+- `s3_max_get_burst` — Max number of requests that can be issued simultaneously before hitting request per second limit. By default (`0` value) equals to `s3_max_get_rps`.
+- `read_resource` — Resource name to be used for [scheduling](/docs/en/operations/workload-scheduling.md) of read requests to this disk. Default value is empty string (IO scheduling is not enabled for this disk).
+- `write_resource` — Resource name to be used for [scheduling](/docs/en/operations/workload-scheduling.md) of write requests to this disk. Default value is empty string (IO scheduling is not enabled for this disk).
+- `key_template` — Define the format with which the object keys are generated. By default, Clickhouse takes `root path` from `endpoint` option and adds random generated suffix. That suffix is a dir with 3 random symbols and a file name with 29 random symbols. With that option you have a full control how to the object keys are generated. Some usage scenarios require having random symbols in the prefix or in the middle of object key. For example: `[a-z]{3}-prefix-random/constant-part/random-middle-[a-z]{3}/random-suffix-[a-z]{29}`. The value is parsed with [`re2`](https://github.com/google/re2/wiki/Syntax). Only some subset of the syntax is supported. Check if your preferred format is supported before using that option. Disk isn't initialized if clickhouse is unable to generate a key by the value of `key_template`. It requires enabled feature flag [storage_metadata_write_full_object_key](/docs/en/operations/settings/settings#storage_metadata_write_full_object_key). It forbids declaring the `root path` in `endpoint` option. It requires definition of the option `key_compatibility_prefix`.
+- `key_compatibility_prefix` — That option is required when option `key_template` is in use. In order to be able to read the objects keys which were stored in the metadata files with the metadata version lower that `VERSION_FULL_OBJECT_KEY`, the previous `root path` from the `endpoint` option should be set here.
+
+:::note
+Google Cloud Storage (GCS) is also supported using the type `s3`. See [GCS backed MergeTree](/docs/en/integrations/gcs).
+:::
+
+### Using Azure Blob Storage {#azure-blob-storage}
+
+`MergeTree` family table engines can store data to [Azure Blob Storage](https://azure.microsoft.com/en-us/services/storage/blobs/) using a disk with type `azure_blob_storage`.
+
+As of February 2022, this feature is still a fresh addition, so expect that some Azure Blob Storage functionalities might be unimplemented.
+
+Configuration markup:
 ``` xml
-<s3>
-    <type>object_storage</type>
-    <object_storage_type>s3</object_storage_type>
-    <metadata_type>local</metadata_type>
-    <endpoint>https://s3.eu-west-1.amazonaws.com/clickhouse-eu-west-1.clickhouse.com/data/</endpoint>
-    <use_invironment_credentials>1</use_invironment_credentials>
-</s3>
+<storage_configuration>
+    ...
+    <disks>
+        <blob_storage_disk>
+            <type>azure_blob_storage</type>
+            <storage_account_url>http://account.blob.core.windows.net</storage_account_url>
+            <container_name>container</container_name>
+            <account_name>account</account_name>
+            <account_key>pass123</account_key>
+            <metadata_path>/var/lib/clickhouse/disks/blob_storage_disk/</metadata_path>
+            <cache_path>/var/lib/clickhouse/disks/blob_storage_disk/cache/</cache_path>
+            <skip_access_check>false</skip_access_check>
+        </blob_storage_disk>
+    </disks>
+    ...
+</storage_configuration>
 ```
 
-Configuration like
-``` xml
-<s3_plain>
-    <type>s3_plain</type>
-    <endpoint>https://s3.eu-west-1.amazonaws.com/clickhouse-eu-west-1.clickhouse.com/data/</endpoint>
-    <use_invironment_credentials>1</use_invironment_credentials>
-</s3_plain>
+Connection parameters:
+* `storage_account_url` - **Required**, Azure Blob Storage account URL, like `http://account.blob.core.windows.net` or `http://azurite1:10000/devstoreaccount1`.
+* `container_name` - Target container name, defaults to `default-container`.
+* `container_already_exists` - If set to `false`, a new container `container_name` is created in the storage account, if set to `true`, disk connects to the container directly, and if left unset, disk connects to the account, checks if the container `container_name` exists, and creates it if it doesn't exist yet.
+
+Authentication parameters (the disk will try all available methods **and** Managed Identity Credential):
+* `connection_string` - For authentication using a connection string.
+* `account_name` and `account_key` - For authentication using Shared Key.
+
+Limit parameters (mainly for internal usage):
+* `s3_max_single_part_upload_size` - Limits the size of a single block upload to Blob Storage.
+* `min_bytes_for_seek` - Limits the size of a seekable region.
+* `max_single_read_retries` - Limits the number of attempts to read a chunk of data from Blob Storage.
+* `max_single_download_retries` - Limits the number of attempts to download a readable buffer from Blob Storage.
+* `thread_pool_size` - Limits the number of threads with which `IDiskRemote` is instantiated.
+* `s3_max_inflight_parts_for_one_file` - Limits the number of put requests that can be run concurrently for one object.
+
+Other parameters:
+* `metadata_path` - Path on local FS to store metadata files for Blob Storage. Default value is `/var/lib/clickhouse/disks/<disk_name>/`.
+* `skip_access_check` - If true, disk access checks will not be performed on disk start-up. Default value is `false`.
+* `read_resource` — Resource name to be used for [scheduling](/docs/en/operations/workload-scheduling.md) of read requests to this disk. Default value is empty string (IO scheduling is not enabled for this disk).
+* `write_resource` — Resource name to be used for [scheduling](/docs/en/operations/workload-scheduling.md) of write requests to this disk. Default value is empty string (IO scheduling is not enabled for this disk).
+
+Examples of working configurations can be found in integration tests directory (see e.g. [test_merge_tree_azure_blob_storage](https://github.com/ClickHouse/ClickHouse/blob/master/tests/integration/test_merge_tree_azure_blob_storage/configs/config.d/storage_conf.xml) or [test_azure_blob_storage_zero_copy_replication](https://github.com/ClickHouse/ClickHouse/blob/master/tests/integration/test_azure_blob_storage_zero_copy_replication/configs/config.d/storage_conf.xml)).
+
+:::note Zero-copy replication is not ready for production
+Zero-copy replication is disabled by default in ClickHouse version 22.8 and higher.  This feature is not recommended for production use.
+:::
+
+## Using HDFS storage {#hdfs-storage}
+
+In this sample configuration:
+- the disk is of type `hdfs`
+- the data is hosted at `hdfs://hdfs1:9000/clickhouse/`
+
+```xml
+<clickhouse>
+    <storage_configuration>
+        <disks>
+            <hdfs>
+                <type>hdfs</type>
+                <endpoint>hdfs://hdfs1:9000/clickhouse/</endpoint>
+                <skip_access_check>true</skip_access_check>
+            </hdfs>
+            <hdd>
+                <type>local</type>
+                <path>/</path>
+            </hdd>
+        </disks>
+        <policies>
+            <hdfs>
+                <volumes>
+                    <main>
+                        <disk>hdfs</disk>
+                    </main>
+                    <external>
+                        <disk>hdd</disk>
+                    </external>
+                </volumes>
+            </hdfs>
+        </policies>
+    </storage_configuration>
+</clickhouse>
 ```
 
-is equal to
-``` xml
-<s3_plain>
-    <type>object_storage</type>
-    <object_storage_type>s3</object_storage_type>
-    <metadata_type>plain</metadata_type>
-    <endpoint>https://s3.eu-west-1.amazonaws.com/clickhouse-eu-west-1.clickhouse.com/data/</endpoint>
-    <use_invironment_credentials>1</use_invironment_credentials>
-</s3_plain>
-```
-
-For details configuration options of each storage see [MergeTree](/docs/en/engines/table-engines/mergetree-family/mergetree.md).
-
-## Using Virtual File System for Data Encryption {#encrypted-virtual-file-system}
+### Using Data Encryption {#encrypted-virtual-file-system}
 
 You can encrypt the data stored on [S3](/docs/en/engines/table-engines/mergetree-family/mergetree.md/#table_engine-mergetree-s3), or [HDFS](#configuring-hdfs) external disks, or on a local disk. To turn on the encryption mode, in the configuration file you must define a disk with the type `encrypted` and choose a disk on which the data will be saved. An `encrypted` disk ciphers all written files on the fly, and when you read files from an `encrypted` disk it deciphers them automatically. So you can work with an `encrypted` disk like with a normal one.
 
@@ -230,7 +454,7 @@ Example of disk configuration:
 </clickhouse>
 ```
 
-## Using local cache {#using-local-cache}
+### Using local cache {#using-local-cache}
 
 It is possible to configure local cache over disks in storage configuration starting from version 22.3.
 For versions 22.3 - 22.7 cache is supported only for `s3` disk type. For versions >= 22.8 cache is supported for any disk type: S3, Azure, Local, Encrypted, etc.
@@ -393,7 +617,56 @@ Cache profile events:
 
 - `CachedWriteBufferCacheWriteBytes`, `CachedWriteBufferCacheWriteMicroseconds`
 
-## Storing Data on Web Server {#storing-data-on-webserver}
+### Using static Web storage (read-only) {#web-storage}
+
+Web storage can be used for read-only purposes. An example use is for hosting sample
+data, or for migrating data.
+
+:::tip
+Storage can also be configured temporarily within a query, if a web dataset is not expected
+to be used routinely, see [dynamic storage](#dynamic-storage) and skip editing the
+configuration file.
+:::
+
+In this sample configuration:
+- the disk is of type `web`
+- the data is hosted at `http://nginx:80/test1/`
+- a cache on local storage is used
+
+```xml
+<clickhouse>
+    <storage_configuration>
+        <disks>
+            <web>
+                <type>web</type>
+                <endpoint>http://nginx:80/test1/</endpoint>
+            </web>
+            <cached_web>
+                <type>cache</type>
+                <disk>web</disk>
+                <path>cached_web_cache/</path>
+                <max_size>100000000</max_size>
+            </cached_web>
+        </disks>
+        <policies>
+            <web>
+                <volumes>
+                    <main>
+                        <disk>web</disk>
+                    </main>
+                </volumes>
+            </web>
+            <cached_web>
+                <volumes>
+                    <main>
+                        <disk>cached_web</disk>
+                    </main>
+                </volumes>
+            </cached_web>
+        </policies>
+    </storage_configuration>
+</clickhouse>
+```
 
 There is a tool `clickhouse-static-files-uploader`, which prepares a data directory for a given table (`SELECT data_paths FROM system.tables WHERE name = 'table_name'`). For each table you need, you get a directory of files. These files can be uploaded to, for example, a web server with static files. After this preparation, you can load this table into any ClickHouse server via `DiskWeb`.
 
@@ -595,7 +868,7 @@ If URL is not reachable on disk load when the server is starting up tables, then
 Use [http_max_single_read_retries](/docs/en/operations/settings/settings.md/#http-max-single-read-retries) setting to limit the maximum number of retries during a single HTTP read.
 
 
-## Zero-copy Replication (not ready for production) {#zero-copy}
+### Zero-copy Replication (not ready for production) {#zero-copy}
 
 Zero-copy replication is possible, but not recommended, with  `S3` and `HDFS` disks. Zero-copy replication means that if the data is stored remotely on several machines and needs to be synchronized, then only the metadata is replicated (paths to the data parts), but not the data itself.
 

From 5ae410e6339fe52e33b41bbc9c6c115ac6293f57 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Thu, 15 Feb 2024 18:33:38 +0100
Subject: [PATCH 003/443] A bit more explanation

---
 .../mergetree-family/mergetree.md             | 49 +------------------
 docs/en/operations/storing-data.md            | 44 ++++++++++++++++-
 2 files changed, 44 insertions(+), 49 deletions(-)

diff --git a/docs/en/engines/table-engines/mergetree-family/mergetree.md b/docs/en/engines/table-engines/mergetree-family/mergetree.md
index 0fff13c906f..f23b251f3a1 100644
--- a/docs/en/engines/table-engines/mergetree-family/mergetree.md
+++ b/docs/en/engines/table-engines/mergetree-family/mergetree.md
@@ -940,53 +940,6 @@ You could change storage policy after table creation with [ALTER TABLE ... MODIF
 
 The number of threads performing background moves of data parts can be changed by [background_move_pool_size](/docs/en/operations/server-configuration-parameters/settings.md/#background_move_pool_size) setting.
 
-### Dynamic Storage
-
-This example query shows how to attach a table stored at a URL and configure the
-remote storage within the query. The web storage is not configured in the ClickHouse
-configuration files; all the settings are in the CREATE/ATTACH query.
-
-:::note
-The example uses `type=web`, but any disk type can be configured as dynamic, even Local disk. Local disks require a path argument to be inside the server config parameter `custom_local_disks_base_directory`, which has no default, so set that also when using local disk.
-:::
-
-#### Example dynamic web storage
-
-:::tip
-A [demo dataset](https://github.com/ClickHouse/web-tables-demo) is hosted in GitHub.  To prepare your own tables for web storage see the tool [clickhouse-static-files-uploader](/docs/en/operations/storing-data.md/#storing-data-on-webserver)
-:::
-
-In this `ATTACH TABLE` query the `UUID` provided matches the directory name of the data, and the endpoint is the URL for the raw GitHub content.
-
-```sql
-# highlight-next-line
-ATTACH TABLE uk_price_paid UUID 'cf712b4f-2ca8-435c-ac23-c4393efe52f7'
-(
-    price UInt32,
-    date Date,
-    postcode1 LowCardinality(String),
-    postcode2 LowCardinality(String),
-    type Enum8('other' = 0, 'terraced' = 1, 'semi-detached' = 2, 'detached' = 3, 'flat' = 4),
-    is_new UInt8,
-    duration Enum8('unknown' = 0, 'freehold' = 1, 'leasehold' = 2),
-    addr1 String,
-    addr2 String,
-    street LowCardinality(String),
-    locality LowCardinality(String),
-    town LowCardinality(String),
-    district LowCardinality(String),
-    county LowCardinality(String)
-)
-ENGINE = MergeTree
-ORDER BY (postcode1, postcode2, addr1, addr2)
-  # highlight-start
-  SETTINGS disk = disk(
-      type=web,
-      endpoint='https://raw.githubusercontent.com/ClickHouse/web-tables-demo/main/web/'
-      );
-  # highlight-end
-```
-
 ### Details {#details}
 
 In the case of `MergeTree` tables, data is getting to disk in different ways:
@@ -1025,7 +978,7 @@ Configuration markup:
 ``` xml
 <storage_configuration>
     ...
-e   <disks>
+    <disks>
         <s3>
             <type>s3</type>
             <support_batch_delete>true</support_batch_delete>
diff --git a/docs/en/operations/storing-data.md b/docs/en/operations/storing-data.md
index baf4e1999a7..0f818b813bf 100644
--- a/docs/en/operations/storing-data.md
+++ b/docs/en/operations/storing-data.md
@@ -213,6 +213,10 @@ ORDER BY (postcode1, postcode2, addr1, addr2)
 In the settings highlighted below notice that the disk of `type=web` is nested within
 the disk of `type=cache`.
 
+:::note
+The example uses `type=web`, but any disk type can be configured as dynamic, even Local disk. Local disks require a path argument to be inside the server config parameter `custom_local_disks_base_directory`, which has no default, so set that also when using local disk.
+:::
+
 A combination of config-based configuration and sql-defined configuration is also possible:
 
 ```sql
@@ -302,6 +306,11 @@ Optional parameters:
 Google Cloud Storage (GCS) is also supported using the type `s3`. See [GCS backed MergeTree](/docs/en/integrations/gcs).
 :::
 
+### Using Plain Storage {#s3-storage}
+
+There is a disk type `s3_plain`, which provides a write-once storage. Unlike `s3` disk type, it stores data as is, e.g. instead of randomly-generated blob names, it uses normal file names as clickhouse stores files on local disk. So this disk type allows to keeper a static version of the table and can also be used to create backups on it.
+Configuration parameters are the same as for `s3` disk type.
+
 ### Using Azure Blob Storage {#azure-blob-storage}
 
 `MergeTree` family table engines can store data to [Azure Blob Storage](https://azure.microsoft.com/en-us/services/storage/blobs/) using a disk with type `azure_blob_storage`.
@@ -672,7 +681,40 @@ There is a tool `clickhouse-static-files-uploader`, which prepares a data direct
 
 This is a read-only disk. Its data is only read and never modified. A new table is loaded to this disk via `ATTACH TABLE` query (see example below). Local disk is not actually used, each `SELECT` query will result in a `http` request to fetch required data. All modification of the table data will result in an exception, i.e. the following types of queries are not allowed: [CREATE TABLE](/docs/en/sql-reference/statements/create/table.md), [ALTER TABLE](/docs/en/sql-reference/statements/alter/index.md), [RENAME TABLE](/docs/en/sql-reference/statements/rename.md/#misc_operations-rename_table), [DETACH TABLE](/docs/en/sql-reference/statements/detach.md) and [TRUNCATE TABLE](/docs/en/sql-reference/statements/truncate.md).
 
-Web server storage is supported only for the [MergeTree](/docs/en/engines/table-engines/mergetree-family/mergetree.md) and [Log](/docs/en/engines/table-engines/log-family/log.md) engine families. To access the data stored on a `web` disk, use the [storage_policy](/docs/en/engines/table-engines/mergetree-family/mergetree.md/#terms) setting when executing the query. For example, `ATTACH TABLE table_web UUID '{}' (id Int32) ENGINE = MergeTree() ORDER BY id SETTINGS storage_policy = 'web'`.
+:::tip
+A [demo dataset](https://github.com/ClickHouse/web-tables-demo) is hosted in GitHub.  To prepare your own tables for web storage see the tool [clickhouse-static-files-uploader](/docs/en/operations/storing-data.md/#storing-data-on-webserver)
+:::
+
+In this `ATTACH TABLE` query the `UUID` provided matches the directory name of the data, and the endpoint is the URL for the raw GitHub content.
+
+```sql
+# highlight-next-line
+ATTACH TABLE uk_price_paid UUID 'cf712b4f-2ca8-435c-ac23-c4393efe52f7'
+(
+    price UInt32,
+    date Date,
+    postcode1 LowCardinality(String),
+    postcode2 LowCardinality(String),
+    type Enum8('other' = 0, 'terraced' = 1, 'semi-detached' = 2, 'detached' = 3, 'flat' = 4),
+    is_new UInt8,
+    duration Enum8('unknown' = 0, 'freehold' = 1, 'leasehold' = 2),
+    addr1 String,
+    addr2 String,
+    street LowCardinality(String),
+    locality LowCardinality(String),
+    town LowCardinality(String),
+    district LowCardinality(String),
+    county LowCardinality(String)
+)
+ENGINE = MergeTree
+ORDER BY (postcode1, postcode2, addr1, addr2)
+  # highlight-start
+  SETTINGS disk = disk(
+      type=web,
+      endpoint='https://raw.githubusercontent.com/ClickHouse/web-tables-demo/main/web/'
+      );
+  # highlight-end
+```
 
 A ready test case. You need to add this configuration to config:
 

From 09e630e02be9ccd19681b34f33e24cea849ca9fd Mon Sep 17 00:00:00 2001
From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com>
Date: Thu, 15 Feb 2024 19:00:08 +0100
Subject: [PATCH 004/443] Update storing-data.md

---
 docs/en/operations/storing-data.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/en/operations/storing-data.md b/docs/en/operations/storing-data.md
index 0f818b813bf..60e33fe2849 100644
--- a/docs/en/operations/storing-data.md
+++ b/docs/en/operations/storing-data.md
@@ -10,7 +10,7 @@ Data, processed in ClickHouse, is usually stored in the local file system — on
 2. The Hadoop Distributed File System ([HDFS](https://hadoop.apache.org/docs/current/hadoop-project-dist/hadoop-hdfs/HdfsDesign.html))
 3. [Azure Blob Storage](https://azure.microsoft.com/en-us/products/storage/blobs).
 
-Note: to work with data stored on `Amazon S3` disks use [S3](/docs/en/engines/table-engines/integrations/s3.md) table engine, to work with data in the Hadoop Distributed File System — [HDFS](/docs/en/engines/table-engines/integrations/hdfs.md) table engine, and to work with data stored in Azure Blob Storage use [AzureBlobStorage](/docs/en/engines/table-engines/integrations/AzureBlobStorage.md) table engine. They are different from external storage described on this page as they allow to read data stored in some general file format (like Parquet), while on this page we are describing storage configuration for ClickHouse `MergeTree` famility or `Log` family tables.
+Note: to work with data stored on `Amazon S3` disks use [S3](/docs/en/engines/table-engines/integrations/s3.md) table engine, to work with data in the Hadoop Distributed File System — [HDFS](/docs/en/engines/table-engines/integrations/hdfs.md) table engine, and to work with data stored in Azure Blob Storage use [AzureBlobStorage](/docs/en/engines/table-engines/integrations/AzureBlobStorage.md) table engine. They are different from external storage described on this page as they allow to read data stored in some general file format (like Parquet), while on this page we are describing storage configuration for ClickHouse `MergeTree` family or `Log` family tables.
 
 ## Configuring external storage {#configuring-external-storage}
 

From 8c11f59ba82bd9ae3a322f7a9729c4a5a8644512 Mon Sep 17 00:00:00 2001
From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com>
Date: Mon, 19 Feb 2024 11:01:37 +0100
Subject: [PATCH 005/443] Fix bad link, update disk web description

---
 docs/en/operations/storing-data.md | 19 ++++++++-----------
 1 file changed, 8 insertions(+), 11 deletions(-)

diff --git a/docs/en/operations/storing-data.md b/docs/en/operations/storing-data.md
index 60e33fe2849..4b0345a3206 100644
--- a/docs/en/operations/storing-data.md
+++ b/docs/en/operations/storing-data.md
@@ -628,14 +628,9 @@ Cache profile events:
 
 ### Using static Web storage (read-only) {#web-storage}
 
-Web storage can be used for read-only purposes. An example use is for hosting sample
-data, or for migrating data.
-
-:::tip
-Storage can also be configured temporarily within a query, if a web dataset is not expected
-to be used routinely, see [dynamic storage](#dynamic-storage) and skip editing the
-configuration file.
-:::
+This is a read-only disk. Its data is only read and never modified. A new table is loaded to this disk via `ATTACH TABLE` query (see example below). Local disk is not actually used, each `SELECT` query will result in a `http` request to fetch required data. All modification of the table data will result in an exception, i.e. the following types of queries are not allowed: [CREATE TABLE](/docs/en/sql-reference/statements/create/table.md), [ALTER TABLE](/docs/en/sql-reference/statements/alter/index.md), [RENAME TABLE](/docs/en/sql-reference/statements/rename.md/#misc_operations-rename_table), [DETACH TABLE](/docs/en/sql-reference/statements/detach.md) and [TRUNCATE TABLE](/docs/en/sql-reference/statements/truncate.md).
+Web storage can be used for read-only purposes. An example use is for hosting sample data, or for migrating data.
+There is a tool `clickhouse-static-files-uploader`, which prepares a data directory for a given table (`SELECT data_paths FROM system.tables WHERE name = 'table_name'`). For each table you need, you get a directory of files. These files can be uploaded to, for example, a web server with static files. After this preparation, you can load this table into any ClickHouse server via `DiskWeb`.
 
 In this sample configuration:
 - the disk is of type `web`
@@ -677,9 +672,11 @@ In this sample configuration:
 </clickhouse>
 ```
 
-There is a tool `clickhouse-static-files-uploader`, which prepares a data directory for a given table (`SELECT data_paths FROM system.tables WHERE name = 'table_name'`). For each table you need, you get a directory of files. These files can be uploaded to, for example, a web server with static files. After this preparation, you can load this table into any ClickHouse server via `DiskWeb`.
-
-This is a read-only disk. Its data is only read and never modified. A new table is loaded to this disk via `ATTACH TABLE` query (see example below). Local disk is not actually used, each `SELECT` query will result in a `http` request to fetch required data. All modification of the table data will result in an exception, i.e. the following types of queries are not allowed: [CREATE TABLE](/docs/en/sql-reference/statements/create/table.md), [ALTER TABLE](/docs/en/sql-reference/statements/alter/index.md), [RENAME TABLE](/docs/en/sql-reference/statements/rename.md/#misc_operations-rename_table), [DETACH TABLE](/docs/en/sql-reference/statements/detach.md) and [TRUNCATE TABLE](/docs/en/sql-reference/statements/truncate.md).
+:::tip
+Storage can also be configured temporarily within a query, if a web dataset is not expected
+to be used routinely, see [dynamic configuration](#dynamic-configuration) and skip editing the
+configuration file.
+:::
 
 :::tip
 A [demo dataset](https://github.com/ClickHouse/web-tables-demo) is hosted in GitHub.  To prepare your own tables for web storage see the tool [clickhouse-static-files-uploader](/docs/en/operations/storing-data.md/#storing-data-on-webserver)

From 601b1dfaa14323db28f169b6b193d59ec75e8bfc Mon Sep 17 00:00:00 2001
From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com>
Date: Mon, 19 Feb 2024 12:21:52 +0100
Subject: [PATCH 006/443] Fix bad link

---
 docs/en/operations/storing-data.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/en/operations/storing-data.md b/docs/en/operations/storing-data.md
index 4b0345a3206..4f676904375 100644
--- a/docs/en/operations/storing-data.md
+++ b/docs/en/operations/storing-data.md
@@ -10,7 +10,7 @@ Data, processed in ClickHouse, is usually stored in the local file system — on
 2. The Hadoop Distributed File System ([HDFS](https://hadoop.apache.org/docs/current/hadoop-project-dist/hadoop-hdfs/HdfsDesign.html))
 3. [Azure Blob Storage](https://azure.microsoft.com/en-us/products/storage/blobs).
 
-Note: to work with data stored on `Amazon S3` disks use [S3](/docs/en/engines/table-engines/integrations/s3.md) table engine, to work with data in the Hadoop Distributed File System — [HDFS](/docs/en/engines/table-engines/integrations/hdfs.md) table engine, and to work with data stored in Azure Blob Storage use [AzureBlobStorage](/docs/en/engines/table-engines/integrations/AzureBlobStorage.md) table engine. They are different from external storage described on this page as they allow to read data stored in some general file format (like Parquet), while on this page we are describing storage configuration for ClickHouse `MergeTree` family or `Log` family tables.
+Note: to work with data stored on `Amazon S3` disks use [S3](/docs/en/engines/table-engines/integrations/s3.md) table engine, to work with data in the Hadoop Distributed File System — [HDFS](/docs/en/engines/table-engines/integrations/hdfs.md) table engine, and to work with data stored in Azure Blob Storage use [AzureBlobStorage](/docs/en/engines/table-engines/integrations/azureBlobStorage.md) table engine. They are different from external storage described on this page as they allow to read data stored in some general file format (like Parquet), while on this page we are describing storage configuration for ClickHouse `MergeTree` family or `Log` family tables.
 
 ## Configuring external storage {#configuring-external-storage}
 

From e98d09c93e6c54a2cc4eadab8614539c0a5eb0f8 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 21 Feb 2024 21:41:04 +0100
Subject: [PATCH 007/443] Do not load useless columns from the index in memory

---
 contrib/rapidjson                             |  2 +-
 src/Processors/QueryPlan/PartsSplitter.cpp    |  9 ++++---
 src/Storages/MergeTree/IMergeTreeDataPart.cpp | 21 +++++++++++++++
 .../MergeTree/MergeTreeDataSelectExecutor.cpp | 26 +++++++++++++++----
 src/Storages/MergeTree/MergeTreeSettings.h    |  2 +-
 5 files changed, 49 insertions(+), 11 deletions(-)

diff --git a/contrib/rapidjson b/contrib/rapidjson
index c4ef90ccdbc..a9bc56c9165 160000
--- a/contrib/rapidjson
+++ b/contrib/rapidjson
@@ -1 +1 @@
-Subproject commit c4ef90ccdbc21d5d5a628d08316bfd301e32d6fa
+Subproject commit a9bc56c9165f1dbbbcada64221bd3a59042c5b95
diff --git a/src/Processors/QueryPlan/PartsSplitter.cpp b/src/Processors/QueryPlan/PartsSplitter.cpp
index 0fc6ddd6408..fcb1d8dd92c 100644
--- a/src/Processors/QueryPlan/PartsSplitter.cpp
+++ b/src/Processors/QueryPlan/PartsSplitter.cpp
@@ -32,9 +32,9 @@ std::string toString(const Values & value)
 
 int compareValues(const Values & lhs, const Values & rhs)
 {
-    chassert(lhs.size() == rhs.size());
+    size_t size = std::min(lhs.size(), rhs.size());
 
-    for (size_t i = 0; i < lhs.size(); ++i)
+    for (size_t i = 0; i < size; ++i)
     {
         if (applyVisitor(FieldVisitorAccurateLess(), lhs[i], rhs[i]))
             return -1;
@@ -55,8 +55,9 @@ public:
     Values getValue(size_t part_idx, size_t mark) const
     {
         const auto & index = parts[part_idx].data_part->getIndex();
-        Values values(index.size());
-        for (size_t i = 0; i < values.size(); ++i)
+        size_t size = index.size();
+        Values values(size);
+        for (size_t i = 0; i < size; ++i)
         {
             index[i]->get(mark, values[i]);
             if (values[i].isNull())
diff --git a/src/Storages/MergeTree/IMergeTreeDataPart.cpp b/src/Storages/MergeTree/IMergeTreeDataPart.cpp
index 11ede661f78..629f3688874 100644
--- a/src/Storages/MergeTree/IMergeTreeDataPart.cpp
+++ b/src/Storages/MergeTree/IMergeTreeDataPart.cpp
@@ -869,6 +869,27 @@ void IMergeTreeDataPart::loadIndex() const
             for (size_t j = 0; j < key_size; ++j)
                 key_serializations[j]->deserializeBinary(*loaded_index[j], *index_file, {});
 
+        /// Cut useless suffix columns, if necessary.
+        Float64 ratio_to_drop_suffix_columns = storage.getSettings()->primary_key_ratio_of_unique_prefix_values_to_skip_suffix_columns;
+        if (key_size > 1 && ratio_to_drop_suffix_columns > 0 && ratio_to_drop_suffix_columns < 1)
+        {
+            chassert(marks_count > 0);
+            for (size_t j = 0; j < key_size - 1; ++j)
+            {
+                size_t num_changes = 0;
+                for (size_t i = 1; i < marks_count; ++i)
+                    if (0 != loaded_index[j]->compareAt(i, i - 1, *loaded_index[j], 0))
+                        ++num_changes;
+
+                if (static_cast<Float64>(num_changes) / marks_count >= ratio_to_drop_suffix_columns)
+                {
+                    key_size = j + 1;
+                    loaded_index.resize(key_size);
+                    break;
+                }
+            }
+        }
+
         for (size_t i = 0; i < key_size; ++i)
         {
             loaded_index[i]->shrinkToFit();
diff --git a/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp b/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp
index 1ba28713680..175419f20e0 100644
--- a/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp
+++ b/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp
@@ -1110,7 +1110,11 @@ MarkRanges MergeTreeDataSelectExecutor::markRangesFromPKRange(
     DataTypes key_types;
     for (size_t i : key_indices)
     {
-        index_columns->emplace_back(ColumnWithTypeAndName{index[i], primary_key.data_types[i], primary_key.column_names[i]});
+        if (i < index.size())
+            index_columns->emplace_back(index[i], primary_key.data_types[i], primary_key.column_names[i]);
+        else
+            index_columns->emplace_back(); /// The column of the primary key was not loaded in memory - we'll skip it.
+
         key_types.emplace_back(primary_key.data_types[i]);
     }
 
@@ -1119,7 +1123,6 @@ MarkRanges MergeTreeDataSelectExecutor::markRangesFromPKRange(
     std::function<void(size_t, size_t, FieldRef &)> create_field_ref;
     if (key_condition.hasMonotonicFunctionsChain())
     {
-
         create_field_ref = [index_columns](size_t row, size_t column, FieldRef & field)
         {
             field = {index_columns.get(), row, column};
@@ -1159,7 +1162,11 @@ MarkRanges MergeTreeDataSelectExecutor::markRangesFromPKRange(
             {
                 for (size_t i = 0; i < used_key_size; ++i)
                 {
-                    create_field_ref(range.begin, i, index_left[i]);
+                    if ((*index_columns)[i].column)
+                        create_field_ref(range.begin, i, index_left[i]);
+                    else
+                        index_left[i] = NEGATIVE_INFINITY;
+
                     index_right[i] = POSITIVE_INFINITY;
                 }
             }
@@ -1170,8 +1177,17 @@ MarkRanges MergeTreeDataSelectExecutor::markRangesFromPKRange(
 
                 for (size_t i = 0; i < used_key_size; ++i)
                 {
-                    create_field_ref(range.begin, i, index_left[i]);
-                    create_field_ref(range.end, i, index_right[i]);
+                    if ((*index_columns)[i].column)
+                    {
+                        create_field_ref(range.begin, i, index_left[i]);
+                        create_field_ref(range.end, i, index_right[i]);
+                    }
+                    else
+                    {
+                        /// If the PK column was not loaded in memory - exclude it from the analysis.
+                        index_left[i] = NEGATIVE_INFINITY;
+                        index_right[i] = POSITIVE_INFINITY;
+                    }
                 }
             }
             key_condition_maybe_true = key_condition.mayBeTrueInRange(used_key_size, index_left.data(), index_right.data(), key_types);
diff --git a/src/Storages/MergeTree/MergeTreeSettings.h b/src/Storages/MergeTree/MergeTreeSettings.h
index b64632b6139..1cff44142bc 100644
--- a/src/Storages/MergeTree/MergeTreeSettings.h
+++ b/src/Storages/MergeTree/MergeTreeSettings.h
@@ -202,7 +202,7 @@ struct Settings;
     M(UInt64, marks_compress_block_size, 65536, "Mark compress block size, the actual size of the block to compress.", 0) \
     M(UInt64, primary_key_compress_block_size, 65536, "Primary compress block size, the actual size of the block to compress.", 0) \
     M(Bool, primary_key_lazy_load, true, "Load primary key in memory on first use instead of on table initialization. This can save memory in the presence of a large number of tables.", 0) \
-    \
+    M(Float, primary_key_ratio_of_unique_prefix_values_to_skip_suffix_columns, 0.9f, "If the value of a column of the primary key in data part changes at least in this ratio of times, skip loading next columns in memory. This allows to save memory usage by not loading useless columns of the primary key.", 0) \
     /** Projection settings. */ \
     M(UInt64, max_projections, 25, "The maximum number of merge tree projections.", 0) \
 

From 31de27b149ab2922647ff7d9141871330cc9d743 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 21 Feb 2024 21:42:19 +0100
Subject: [PATCH 008/443] Do not load useless columns from the index in memory

---
 contrib/rapidjson | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/contrib/rapidjson b/contrib/rapidjson
index a9bc56c9165..c4ef90ccdbc 160000
--- a/contrib/rapidjson
+++ b/contrib/rapidjson
@@ -1 +1 @@
-Subproject commit a9bc56c9165f1dbbbcada64221bd3a59042c5b95
+Subproject commit c4ef90ccdbc21d5d5a628d08316bfd301e32d6fa

From d0d84a840151675ee3c7a108709e4b0b486af577 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 21 Feb 2024 23:03:01 +0100
Subject: [PATCH 009/443] Add a test

---
 .../02998_primary_key_skip_columns.reference  | 18 ++++++++++
 .../02998_primary_key_skip_columns.sql        | 33 +++++++++++++++++++
 2 files changed, 51 insertions(+)
 create mode 100644 tests/queries/0_stateless/02998_primary_key_skip_columns.reference
 create mode 100644 tests/queries/0_stateless/02998_primary_key_skip_columns.sql

diff --git a/tests/queries/0_stateless/02998_primary_key_skip_columns.reference b/tests/queries/0_stateless/02998_primary_key_skip_columns.reference
new file mode 100644
index 00000000000..9df0a2c097c
--- /dev/null
+++ b/tests/queries/0_stateless/02998_primary_key_skip_columns.reference
@@ -0,0 +1,18 @@
+100000
+14954
+798
+15908
+108
+120
+2334
+19
+Key size: 	2400000
+100000
+14954
+798
+15907
+108
+120
+2334
+19
+Key size: 	800008
diff --git a/tests/queries/0_stateless/02998_primary_key_skip_columns.sql b/tests/queries/0_stateless/02998_primary_key_skip_columns.sql
new file mode 100644
index 00000000000..801fa35fb52
--- /dev/null
+++ b/tests/queries/0_stateless/02998_primary_key_skip_columns.sql
@@ -0,0 +1,33 @@
+DROP TABLE IF EXISTS test;
+
+CREATE TABLE test (a UInt64, b UInt64, c UInt64) ENGINE = MergeTree ORDER BY (a, b, c) SETTINGS index_granularity = 1, primary_key_ratio_of_unique_prefix_values_to_skip_suffix_columns = 1;
+INSERT INTO test SELECT sipHash64(number, 1), sipHash64(number, 2), sipHash64(number, 3) FROM numbers(100000);
+
+SELECT count() FROM test;
+SELECT count() FROM test WHERE a > 1849813033528774208 AND a < 4594276315503201760;
+SELECT count() FROM test WHERE b > 7898976344263989848 AND b < 8040320939819153137;
+SELECT count() FROM test WHERE c > 13239894303140990071 AND c < 16179795840886947236;
+SELECT count() FROM test WHERE a > 1849813033528774208 AND a < 4594276315503201760 AND b > 7898976344263989848 AND b < 8040320939819153137;
+SELECT count() FROM test WHERE b > 7898976344263989848 AND b < 8040320939819153137 AND c > 13239894303140990071 AND c < 16179795840886947236;
+SELECT count() FROM test WHERE a > 1849813033528774208 AND a < 4594276315503201760 AND c > 13239894303140990071 AND c < 16179795840886947236;
+SELECT count() FROM test WHERE a > 1849813033528774208 AND a < 4594276315503201760 AND b > 7898976344263989848 AND b < 8040320939819153137 AND c > 13239894303140990071 AND c < 16179795840886947236;
+
+SELECT 'Key size: ', round(sum(primary_key_bytes_in_memory), -5) FROM system.parts WHERE database = currentDatabase() AND table = 'test';
+
+ALTER TABLE test MODIFY SETTING primary_key_ratio_of_unique_prefix_values_to_skip_suffix_columns = 0.9;
+
+DETACH TABLE test;
+ATTACH TABLE test;
+
+SELECT count() FROM test;
+SELECT count() FROM test WHERE a > 1849813033528774208 AND a < 4594276315503201760;
+SELECT count() FROM test WHERE b > 7898976344263989848 AND b < 8040320939819153137;
+SELECT count() FROM test WHERE c > 13239894303140990071 AND c < 16179795840886947236;
+SELECT count() FROM test WHERE a > 1849813033528774208 AND a < 4594276315503201760 AND b > 7898976344263989848 AND b < 8040320939819153137;
+SELECT count() FROM test WHERE b > 7898976344263989848 AND b < 8040320939819153137 AND c > 13239894303140990071 AND c < 16179795840886947236;
+SELECT count() FROM test WHERE a > 1849813033528774208 AND a < 4594276315503201760 AND c > 13239894303140990071 AND c < 16179795840886947236;
+SELECT count() FROM test WHERE a > 1849813033528774208 AND a < 4594276315503201760 AND b > 7898976344263989848 AND b < 8040320939819153137 AND c > 13239894303140990071 AND c < 16179795840886947236;
+
+SELECT 'Key size: ', round(sum(primary_key_bytes_in_memory), 5) FROM system.parts WHERE database = currentDatabase() AND table = 'test';
+
+DROP TABLE test;

From 395ad35c93a185291a16449b1ac4d1dcecb1a127 Mon Sep 17 00:00:00 2001
From: Shanfeng Pang <shanfengp@myscale.com>
Date: Thu, 22 Feb 2024 10:41:16 +0800
Subject: [PATCH 010/443] fix LRUResource Cache bug

---
 src/Common/LRUResourceCache.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Common/LRUResourceCache.h b/src/Common/LRUResourceCache.h
index 4ccaa272346..60b4053bff5 100644
--- a/src/Common/LRUResourceCache.h
+++ b/src/Common/LRUResourceCache.h
@@ -221,7 +221,7 @@ private:
         {
             std::lock_guard lock(mutex);
             auto it = cells.find(key);
-            if (it != cells.end() && !it->second.expired)
+            if (it != cells.end())
             {
                 if (!it->second.expired)
                 {

From 999cf88ab79cf71bc82e7be3140496697a661416 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Thu, 22 Feb 2024 09:54:39 +0100
Subject: [PATCH 011/443] Improve test

---
 tests/queries/0_stateless/02998_primary_key_skip_columns.sql | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/queries/0_stateless/02998_primary_key_skip_columns.sql b/tests/queries/0_stateless/02998_primary_key_skip_columns.sql
index 801fa35fb52..b2dadcc5e7c 100644
--- a/tests/queries/0_stateless/02998_primary_key_skip_columns.sql
+++ b/tests/queries/0_stateless/02998_primary_key_skip_columns.sql
@@ -28,6 +28,6 @@ SELECT count() FROM test WHERE b > 7898976344263989848 AND b < 80403209398191531
 SELECT count() FROM test WHERE a > 1849813033528774208 AND a < 4594276315503201760 AND c > 13239894303140990071 AND c < 16179795840886947236;
 SELECT count() FROM test WHERE a > 1849813033528774208 AND a < 4594276315503201760 AND b > 7898976344263989848 AND b < 8040320939819153137 AND c > 13239894303140990071 AND c < 16179795840886947236;
 
-SELECT 'Key size: ', round(sum(primary_key_bytes_in_memory), 5) FROM system.parts WHERE database = currentDatabase() AND table = 'test';
+SELECT 'Key size: ', round(sum(primary_key_bytes_in_memory), -5) FROM system.parts WHERE database = currentDatabase() AND table = 'test';
 
 DROP TABLE test;

From 9b10aebecc690e36ec3591ba7115991b00920289 Mon Sep 17 00:00:00 2001
From: Shanfeng Pang <shanfengp@myscale.com>
Date: Thu, 22 Feb 2024 17:24:59 +0800
Subject: [PATCH 012/443] add unit-test for bug fix

---
 src/Common/tests/gtest_lru_resource_cache.cpp | 27 +++++++++++++++++++
 1 file changed, 27 insertions(+)

diff --git a/src/Common/tests/gtest_lru_resource_cache.cpp b/src/Common/tests/gtest_lru_resource_cache.cpp
index bc037824ff8..94490d1e86d 100644
--- a/src/Common/tests/gtest_lru_resource_cache.cpp
+++ b/src/Common/tests/gtest_lru_resource_cache.cpp
@@ -45,6 +45,33 @@ struct MyWeight
     size_t operator()(const int & x) const { return static_cast<size_t>(x); }
 };
 
+TEST(LRUResourceCache, remove2)
+{
+    using MyCache = DB::LRUResourceCache<int, int, MyWeight>;
+    auto mcache = MyCache(10, 10);
+    for (int i = 1; i < 5; ++i)
+    {
+        auto load_int = [&] { return std::make_shared<int>(i); };
+        mcache.getOrSet(i, load_int);
+    }
+
+    auto n = mcache.size();
+    ASSERT_EQ(n, 4);
+    auto w = mcache.weight();
+    ASSERT_EQ(w, 10);
+    auto holder4 = mcache.get(4);
+    ASSERT_TRUE(holder4 != nullptr);
+    mcache.tryRemove(4);
+    auto holder_reget_4 = mcache.get(4);
+    ASSERT_TRUE(holder_reget_4 == nullptr);
+    mcache.getOrSet(4, [&]() { return std::make_shared<int>(4); });
+    holder4.reset();
+    auto holder1 = mcache.getOrSet(1, [&]() { return std::make_shared<int>(1); });
+    ASSERT_TRUE(holder1 != nullptr);
+    auto holder7 = mcache.getOrSet(7, [&] { return std::make_shared<int>(7); });
+    ASSERT_TRUE(holder7 != nullptr);
+}
+
 TEST(LRUResourceCache, evictOnWweight)
 {
     using MyCache = DB::LRUResourceCache<int, int, MyWeight>;

From 3ef159853c78e438e6088d60a64bcff2bbb77b17 Mon Sep 17 00:00:00 2001
From: Michael Kolupaev <michael.kolupaev@clickhouse.com>
Date: Thu, 22 Feb 2024 14:34:18 +0000
Subject: [PATCH 013/443] Do something to the test

---
 .../0_stateless/02998_primary_key_skip_columns.reference        | 2 +-
 tests/queries/0_stateless/02998_primary_key_skip_columns.sql    | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/queries/0_stateless/02998_primary_key_skip_columns.reference b/tests/queries/0_stateless/02998_primary_key_skip_columns.reference
index 9df0a2c097c..ec44acbd16b 100644
--- a/tests/queries/0_stateless/02998_primary_key_skip_columns.reference
+++ b/tests/queries/0_stateless/02998_primary_key_skip_columns.reference
@@ -15,4 +15,4 @@ Key size: 	2400000
 120
 2334
 19
-Key size: 	800008
+Key size ok: 	1	1
diff --git a/tests/queries/0_stateless/02998_primary_key_skip_columns.sql b/tests/queries/0_stateless/02998_primary_key_skip_columns.sql
index b2dadcc5e7c..27672d7854e 100644
--- a/tests/queries/0_stateless/02998_primary_key_skip_columns.sql
+++ b/tests/queries/0_stateless/02998_primary_key_skip_columns.sql
@@ -28,6 +28,6 @@ SELECT count() FROM test WHERE b > 7898976344263989848 AND b < 80403209398191531
 SELECT count() FROM test WHERE a > 1849813033528774208 AND a < 4594276315503201760 AND c > 13239894303140990071 AND c < 16179795840886947236;
 SELECT count() FROM test WHERE a > 1849813033528774208 AND a < 4594276315503201760 AND b > 7898976344263989848 AND b < 8040320939819153137 AND c > 13239894303140990071 AND c < 16179795840886947236;
 
-SELECT 'Key size: ', round(sum(primary_key_bytes_in_memory), -5) FROM system.parts WHERE database = currentDatabase() AND table = 'test';
+SELECT 'Key size ok: ', (sum(primary_key_bytes_in_memory) as s) >= 800000, s < 1200000 FROM system.parts WHERE database = currentDatabase() AND table = 'test';
 
 DROP TABLE test;

From f1e95fb78bae190bb87e93704cf5f88c70cdccf4 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Thu, 22 Feb 2024 15:38:44 +0100
Subject: [PATCH 014/443] Add a way to force read-through cache for merges

---
 src/Core/Settings.h                           |   1 +
 src/Disks/IO/ReadBufferFromRemoteFSGather.cpp |  10 +-
 .../Cached/CachedObjectStorage.cpp            |  14 --
 .../Cached/CachedObjectStorage.h              |   2 -
 src/IO/ReadSettings.h                         |   2 +-
 src/Interpreters/Cache/FileSegment.cpp        |   3 +-
 src/Interpreters/Context.cpp                  |   1 +
 .../MergeTree/MergeTreeSequentialSource.cpp   |   2 +-
 .../integration/test_filesystem_cache/test.py |  79 ++++++++
 .../users.d/cache_on_write_operations.xml     |   7 +
 .../force_read_through_cache_on_merge.xml     |   7 +
 ...system_cache_on_write_operations.reference | 170 ++++++++++++++++++
 ...41_filesystem_cache_on_write_operations.sh |  81 +++++----
 13 files changed, 317 insertions(+), 62 deletions(-)
 create mode 100644 tests/integration/test_filesystem_cache/users.d/cache_on_write_operations.xml
 create mode 100644 tests/integration/test_filesystem_cache/users.d/force_read_through_cache_on_merge.xml

diff --git a/src/Core/Settings.h b/src/Core/Settings.h
index 433195af9c3..db060bf712d 100644
--- a/src/Core/Settings.h
+++ b/src/Core/Settings.h
@@ -769,6 +769,7 @@ class IColumn;
     M(Bool, enable_filesystem_cache_on_write_operations, false, "Write into cache on write operations. To actually work this setting requires be added to disk config too", 0) \
     M(Bool, enable_filesystem_cache_log, false, "Allows to record the filesystem caching log for each query", 0) \
     M(Bool, read_from_filesystem_cache_if_exists_otherwise_bypass_cache, false, "Allow to use the filesystem cache in passive mode - benefit from the existing cache entries, but don't put more entries into the cache. If you set this setting for heavy ad-hoc queries and leave it disabled for short real-time queries, this will allows to avoid cache threshing by too heavy queries and to improve the overall system efficiency.", 0) \
+    M(Bool, force_read_through_cache_for_merges, false, "Force read-through cache for merges", 0) \
     M(Bool, skip_download_if_exceeds_query_cache, true, "Skip download from remote filesystem if exceeds query cache size", 0) \
     M(UInt64, filesystem_cache_max_download_size, (128UL * 1024 * 1024 * 1024), "Max remote filesystem cache size that can be downloaded by a single query", 0) \
     M(Bool, throw_on_error_from_cache_on_write_operations, false, "Ignore error from cache when caching on write operations (INSERT, merges)", 0) \
diff --git a/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp b/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp
index 0b3ecca3587..1da39c7011c 100644
--- a/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp
+++ b/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp
@@ -16,12 +16,10 @@ using namespace DB;
 
 namespace
 {
-bool withCache(const ReadSettings & settings)
-{
-    return settings.remote_fs_cache && settings.enable_filesystem_cache
-        && (!CurrentThread::getQueryId().empty() || settings.read_from_filesystem_cache_if_exists_otherwise_bypass_cache
-            || !settings.avoid_readthrough_cache_outside_query_context);
-}
+    bool withCache(const ReadSettings & settings)
+    {
+        return settings.remote_fs_cache && settings.enable_filesystem_cache;
+    }
 }
 
 namespace DB
diff --git a/src/Disks/ObjectStorages/Cached/CachedObjectStorage.cpp b/src/Disks/ObjectStorages/Cached/CachedObjectStorage.cpp
index 1444f4c9c76..e3ab772e3b5 100644
--- a/src/Disks/ObjectStorages/Cached/CachedObjectStorage.cpp
+++ b/src/Disks/ObjectStorages/Cached/CachedObjectStorage.cpp
@@ -43,10 +43,6 @@ ReadSettings CachedObjectStorage::patchSettings(const ReadSettings & read_settin
 {
     ReadSettings modified_settings{read_settings};
     modified_settings.remote_fs_cache = cache;
-
-    if (!canUseReadThroughCache(read_settings))
-        modified_settings.read_from_filesystem_cache_if_exists_otherwise_bypass_cache = true;
-
     return object_storage->patchSettings(modified_settings);
 }
 
@@ -206,14 +202,4 @@ String CachedObjectStorage::getObjectsNamespace() const
     return object_storage->getObjectsNamespace();
 }
 
-bool CachedObjectStorage::canUseReadThroughCache(const ReadSettings & settings)
-{
-    if (!settings.avoid_readthrough_cache_outside_query_context)
-        return true;
-
-    return CurrentThread::isInitialized()
-        && CurrentThread::get().getQueryContext()
-        && !CurrentThread::getQueryId().empty();
-}
-
 }
diff --git a/src/Disks/ObjectStorages/Cached/CachedObjectStorage.h b/src/Disks/ObjectStorages/Cached/CachedObjectStorage.h
index 437baead7be..961c2709efc 100644
--- a/src/Disks/ObjectStorages/Cached/CachedObjectStorage.h
+++ b/src/Disks/ObjectStorages/Cached/CachedObjectStorage.h
@@ -119,8 +119,6 @@ public:
 
     const FileCacheSettings & getCacheSettings() const { return cache_settings; }
 
-    static bool canUseReadThroughCache(const ReadSettings & settings);
-
 #if USE_AZURE_BLOB_STORAGE
     std::shared_ptr<const Azure::Storage::Blobs::BlobContainerClient> getAzureBlobStorageClient() override
     {
diff --git a/src/IO/ReadSettings.h b/src/IO/ReadSettings.h
index c397689d6ad..2c79735317d 100644
--- a/src/IO/ReadSettings.h
+++ b/src/IO/ReadSettings.h
@@ -99,7 +99,7 @@ struct ReadSettings
     bool read_from_filesystem_cache_if_exists_otherwise_bypass_cache = false;
     bool enable_filesystem_cache_log = false;
     /// Don't populate cache when the read is not part of query execution (e.g. background thread).
-    bool avoid_readthrough_cache_outside_query_context = true;
+    bool force_read_through_cache_merges = false;
     size_t filesystem_cache_segments_batch_size = 20;
 
     size_t filesystem_cache_max_download_size = (128UL * 1024 * 1024 * 1024);
diff --git a/src/Interpreters/Cache/FileSegment.cpp b/src/Interpreters/Cache/FileSegment.cpp
index 8bd89465917..7c0505889da 100644
--- a/src/Interpreters/Cache/FileSegment.cpp
+++ b/src/Interpreters/Cache/FileSegment.cpp
@@ -10,6 +10,7 @@
 #include <Common/logger_useful.h>
 #include <Common/scope_guard_safe.h>
 #include <Common/ElapsedTimeProfileEventIncrement.h>
+#include <Common/setThreadName.h>
 
 #include <magic_enum.hpp>
 
@@ -194,7 +195,7 @@ bool FileSegment::isDownloaded() const
 String FileSegment::getCallerId()
 {
     if (!CurrentThread::isInitialized() || CurrentThread::getQueryId().empty())
-        return "None:" + toString(getThreadId());
+        return fmt::format("None:{}:{}", getThreadName(), toString(getThreadId()));
 
     return std::string(CurrentThread::getQueryId()) + ":" + toString(getThreadId());
 }
diff --git a/src/Interpreters/Context.cpp b/src/Interpreters/Context.cpp
index 55a4df10206..36b362e36bb 100644
--- a/src/Interpreters/Context.cpp
+++ b/src/Interpreters/Context.cpp
@@ -5079,6 +5079,7 @@ ReadSettings Context::getReadSettings() const
     res.read_from_filesystem_cache_if_exists_otherwise_bypass_cache = settings.read_from_filesystem_cache_if_exists_otherwise_bypass_cache;
     res.enable_filesystem_cache_log = settings.enable_filesystem_cache_log;
     res.filesystem_cache_segments_batch_size = settings.filesystem_cache_segments_batch_size;
+    res.force_read_through_cache_merges = settings.force_read_through_cache_for_merges;
 
     res.filesystem_cache_max_download_size = settings.filesystem_cache_max_download_size;
     res.skip_download_if_exceeds_query_cache = settings.skip_download_if_exceeds_query_cache;
diff --git a/src/Storages/MergeTree/MergeTreeSequentialSource.cpp b/src/Storages/MergeTree/MergeTreeSequentialSource.cpp
index d0fbc316024..e375e8b0a9f 100644
--- a/src/Storages/MergeTree/MergeTreeSequentialSource.cpp
+++ b/src/Storages/MergeTree/MergeTreeSequentialSource.cpp
@@ -151,7 +151,7 @@ MergeTreeSequentialSource::MergeTreeSequentialSource(
 
     const auto & context = storage.getContext();
     ReadSettings read_settings = context->getReadSettings();
-    read_settings.read_from_filesystem_cache_if_exists_otherwise_bypass_cache = true;
+    read_settings.read_from_filesystem_cache_if_exists_otherwise_bypass_cache = !read_settings.force_read_through_cache_merges;
     /// It does not make sense to use pthread_threadpool for background merges/mutations
     /// And also to preserve backward compatibility
     read_settings.local_fs_method = LocalFSReadMethod::pread;
diff --git a/tests/integration/test_filesystem_cache/test.py b/tests/integration/test_filesystem_cache/test.py
index eb5f896f7a9..c1ba6702dcf 100644
--- a/tests/integration/test_filesystem_cache/test.py
+++ b/tests/integration/test_filesystem_cache/test.py
@@ -19,6 +19,9 @@ def cluster():
             main_configs=[
                 "config.d/storage_conf.xml",
             ],
+            user_configs=[
+                "users.d/cache_on_write_operations.xml",
+            ],
             stay_alive=True,
         )
         cluster.add_instance(
@@ -35,6 +38,17 @@ def cluster():
             ],
             stay_alive=True,
         )
+        cluster.add_instance(
+            "node_force_read_through_cache_on_merge",
+            main_configs=[
+                "config.d/storage_conf.xml",
+            ],
+            user_configs=[
+                "users.d/force_read_through_cache_on_merge.xml",
+                "users.d/cache_on_write_operations.xml",
+            ],
+            stay_alive=True,
+        )
 
         logging.info("Starting cluster...")
         cluster.start()
@@ -323,3 +337,68 @@ def test_custom_cached_disk(cluster):
             "SELECT cache_path FROM system.disks WHERE name = 'custom_cached4'"
         ).strip()
     )
+
+
+def test_force_filesystem_cache_on_merges(cluster):
+    def test(node, forced_read_through_cache_on_merge):
+        node.query(
+            """
+            DROP TABLE IF EXISTS test SYNC;
+
+            CREATE TABLE test (key UInt32, value String)
+            Engine=MergeTree()
+            ORDER BY value
+            SETTINGS disk = disk(
+                type = cache,
+                path = 'force_cache_on_merges',
+                disk = 'hdd_blob',
+                max_file_segment_size = '1Ki',
+                cache_on_write_operations = 1,
+                boundary_alignment = '1Ki',
+                max_size = '10Gi',
+                max_elements = 10000000,
+                load_metadata_threads = 30);
+
+            SYSTEM DROP FILESYSTEM CACHE;
+            INSERT INTO test SELECT * FROM generateRandom('a Int32, b String') LIMIT 1000000;
+            INSERT INTO test SELECT * FROM generateRandom('a Int32, b String') LIMIT 1000000;
+            """
+        )
+        assert int(node.query("SELECT count() FROM system.filesystem_cache")) > 0
+        assert int(node.query("SELECT max(size) FROM system.filesystem_cache")) == 1024
+
+        write_count = int(
+            node.query(
+                "SELECT value FROM system.events WHERE name = 'CachedWriteBufferCacheWriteBytes'"
+            )
+        )
+        assert write_count > 100000
+        assert "" == node.query(
+            "SELECT value FROM system.events WHERE name = 'CachedReadBufferCacheWriteBytes'"
+        )
+
+        node.query("SYSTEM DROP FILESYSTEM CACHE")
+        node.query("OPTIMIZE TABLE test FINAL")
+
+        new_write_count = int(
+            node.query(
+                "SELECT value FROM system.events WHERE name = 'CachedWriteBufferCacheWriteBytes'"
+            )
+        )
+        assert new_write_count >= write_count
+
+        if forced_read_through_cache_on_merge:
+            assert 100000 < int(
+                node.query(
+                    "SELECT value FROM system.events WHERE name = 'CachedReadBufferCacheWriteBytes'"
+                )
+            )
+        else:
+            assert "" == node.query(
+                "SELECT value FROM system.events WHERE name = 'CachedReadBufferCacheWriteBytes'"
+            )
+
+    node = cluster.instances["node_force_read_through_cache_on_merge"]
+    test(node, True)
+    node = cluster.instances["node"]
+    test(node, False)
diff --git a/tests/integration/test_filesystem_cache/users.d/cache_on_write_operations.xml b/tests/integration/test_filesystem_cache/users.d/cache_on_write_operations.xml
new file mode 100644
index 00000000000..5de169edc1e
--- /dev/null
+++ b/tests/integration/test_filesystem_cache/users.d/cache_on_write_operations.xml
@@ -0,0 +1,7 @@
+<clickhouse>
+    <profiles>
+        <default>
+            <enable_filesystem_cache_on_write_operations>1</enable_filesystem_cache_on_write_operations>
+        </default>
+    </profiles>
+</clickhouse>
diff --git a/tests/integration/test_filesystem_cache/users.d/force_read_through_cache_on_merge.xml b/tests/integration/test_filesystem_cache/users.d/force_read_through_cache_on_merge.xml
new file mode 100644
index 00000000000..4d26a1a8bc7
--- /dev/null
+++ b/tests/integration/test_filesystem_cache/users.d/force_read_through_cache_on_merge.xml
@@ -0,0 +1,7 @@
+<clickhouse>
+    <profiles>
+        <default>
+            <force_read_through_cache_for_merges>1</force_read_through_cache_for_merges>
+        </default>
+    </profiles>
+</clickhouse>
diff --git a/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.reference b/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.reference
index 157837983f7..c03b928684b 100644
--- a/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.reference
+++ b/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.reference
@@ -1,62 +1,232 @@
 Using storage policy: s3_cache
+DROP TABLE IF EXISTS test_02241
+CREATE TABLE test_02241 (key UInt32, value String) Engine=MergeTree() ORDER BY key SETTINGS storage_policy='s3_cache', min_bytes_for_wide_part = 10485760, compress_marks=false, compress_primary_key=false, ratio_of_defaults_for_sparse_serialization = 1
+SYSTEM STOP MERGES test_02241
+SYSTEM DROP FILESYSTEM CACHE
+SELECT file_segment_range_begin, file_segment_range_end, size, state
+    FROM
+    (
+        SELECT file_segment_range_begin, file_segment_range_end, size, state, local_path
+        FROM
+        (
+            SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path
+            FROM system.remote_data_paths
+        ) AS data_paths
+        INNER JOIN
+            system.filesystem_cache AS caches
+        ON data_paths.cache_path = caches.cache_path
+    )
+    WHERE endsWith(local_path, 'data.bin')
+    FORMAT Vertical
+SELECT count() FROM (SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path FROM system.remote_data_paths ) AS data_paths INNER JOIN system.filesystem_cache AS caches ON data_paths.cache_path = caches.cache_path
 0
+SELECT count(), sum(size) FROM system.filesystem_cache
 0	0
+INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(100)
+SELECT file_segment_range_begin, file_segment_range_end, size, state
+    FROM
+    (
+        SELECT file_segment_range_begin, file_segment_range_end, size, state, local_path
+        FROM
+        (
+            SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path
+            FROM system.remote_data_paths
+        ) AS data_paths
+        INNER JOIN
+            system.filesystem_cache AS caches
+        ON data_paths.cache_path = caches.cache_path
+    )
+    WHERE endsWith(local_path, 'data.bin')
+    FORMAT Vertical
 Row 1:
 ──────
 file_segment_range_begin: 0
 file_segment_range_end:   745
 size:                     746
 state:                    DOWNLOADED
+SELECT count() FROM (SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path FROM system.remote_data_paths ) AS data_paths INNER JOIN system.filesystem_cache AS caches ON data_paths.cache_path = caches.cache_path
 8
+SELECT count(), sum(size) FROM system.filesystem_cache
 8	1100
+SELECT count() FROM system.filesystem_cache WHERE cache_hits > 0
 0
+SELECT * FROM test_02241 FORMAT Null
+SELECT count() FROM system.filesystem_cache WHERE cache_hits > 0
 2
+SELECT * FROM test_02241 FORMAT Null
+SELECT count() FROM system.filesystem_cache WHERE cache_hits > 0
 2
+SELECT count(), sum(size) size FROM system.filesystem_cache
 8	1100
+SYSTEM DROP FILESYSTEM CACHE
+INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(100, 200)
+SELECT file_segment_range_begin, file_segment_range_end, size, state
+    FROM
+    (
+        SELECT file_segment_range_begin, file_segment_range_end, size, state, local_path
+        FROM
+        (
+            SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path
+            FROM system.remote_data_paths
+        ) AS data_paths
+        INNER JOIN
+            system.filesystem_cache AS caches
+        ON data_paths.cache_path = caches.cache_path
+    )
+    WHERE endsWith(local_path, 'data.bin')
+    FORMAT Vertical;
 Row 1:
 ──────
 file_segment_range_begin: 0
 file_segment_range_end:   1659
 size:                     1660
 state:                    DOWNLOADED
+SELECT count() FROM (SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path FROM system.remote_data_paths ) AS data_paths INNER JOIN system.filesystem_cache AS caches ON data_paths.cache_path = caches.cache_path
 8
+SELECT count(), sum(size) FROM system.filesystem_cache
 8	2014
+SELECT count(), sum(size) FROM system.filesystem_cache
 8	2014
+INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(100) SETTINGS enable_filesystem_cache_on_write_operations=0
+SELECT count(), sum(size) FROM system.filesystem_cache
 8	2014
+INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(100)
+INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(300, 10000)
+SELECT count(), sum(size) FROM system.filesystem_cache
 24	84045
+SYSTEM START MERGES test_02241
+SELECT value FROM system.events WHERE name = 'CachedWriteBufferCacheWriteBytes'
+85146
+SELECT value FROM system.events WHERE name = 'CachedReadBufferCacheWriteBytes'
+OPTIMIZE TABLE test_02241 FINAL
+SELECT value FROM system.events WHERE name = 'CachedWriteBufferCacheWriteBytes'
+251542
+SELECT value FROM system.events WHERE name = 'CachedReadBufferCacheWriteBytes'
+SELECT count(), sum(size) FROM system.filesystem_cache
 32	167243
+ALTER TABLE test_02241 UPDATE value = 'kek' WHERE key = 100
+SELECT count(), sum(size) FROM system.filesystem_cache
 41	250541
+INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(5000000)
+SYSTEM FLUSH LOGS
 INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(5000000)	0
+SELECT count() FROM test_02241
 5010500
+SELECT count() FROM test_02241 WHERE value LIKE '%010%'
 18816
 Using storage policy: local_cache
+DROP TABLE IF EXISTS test_02241
+CREATE TABLE test_02241 (key UInt32, value String) Engine=MergeTree() ORDER BY key SETTINGS storage_policy='local_cache', min_bytes_for_wide_part = 10485760, compress_marks=false, compress_primary_key=false, ratio_of_defaults_for_sparse_serialization = 1
+SYSTEM STOP MERGES test_02241
+SYSTEM DROP FILESYSTEM CACHE
+SELECT file_segment_range_begin, file_segment_range_end, size, state
+    FROM
+    (
+        SELECT file_segment_range_begin, file_segment_range_end, size, state, local_path
+        FROM
+        (
+            SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path
+            FROM system.remote_data_paths
+        ) AS data_paths
+        INNER JOIN
+            system.filesystem_cache AS caches
+        ON data_paths.cache_path = caches.cache_path
+    )
+    WHERE endsWith(local_path, 'data.bin')
+    FORMAT Vertical
+SELECT count() FROM (SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path FROM system.remote_data_paths ) AS data_paths INNER JOIN system.filesystem_cache AS caches ON data_paths.cache_path = caches.cache_path
 0
+SELECT count(), sum(size) FROM system.filesystem_cache
 0	0
+INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(100)
+SELECT file_segment_range_begin, file_segment_range_end, size, state
+    FROM
+    (
+        SELECT file_segment_range_begin, file_segment_range_end, size, state, local_path
+        FROM
+        (
+            SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path
+            FROM system.remote_data_paths
+        ) AS data_paths
+        INNER JOIN
+            system.filesystem_cache AS caches
+        ON data_paths.cache_path = caches.cache_path
+    )
+    WHERE endsWith(local_path, 'data.bin')
+    FORMAT Vertical
 Row 1:
 ──────
 file_segment_range_begin: 0
 file_segment_range_end:   745
 size:                     746
 state:                    DOWNLOADED
+SELECT count() FROM (SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path FROM system.remote_data_paths ) AS data_paths INNER JOIN system.filesystem_cache AS caches ON data_paths.cache_path = caches.cache_path
 8
+SELECT count(), sum(size) FROM system.filesystem_cache
 8	1100
+SELECT count() FROM system.filesystem_cache WHERE cache_hits > 0
 0
+SELECT * FROM test_02241 FORMAT Null
+SELECT count() FROM system.filesystem_cache WHERE cache_hits > 0
 2
+SELECT * FROM test_02241 FORMAT Null
+SELECT count() FROM system.filesystem_cache WHERE cache_hits > 0
 2
+SELECT count(), sum(size) size FROM system.filesystem_cache
 8	1100
+SYSTEM DROP FILESYSTEM CACHE
+INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(100, 200)
+SELECT file_segment_range_begin, file_segment_range_end, size, state
+    FROM
+    (
+        SELECT file_segment_range_begin, file_segment_range_end, size, state, local_path
+        FROM
+        (
+            SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path
+            FROM system.remote_data_paths
+        ) AS data_paths
+        INNER JOIN
+            system.filesystem_cache AS caches
+        ON data_paths.cache_path = caches.cache_path
+    )
+    WHERE endsWith(local_path, 'data.bin')
+    FORMAT Vertical;
 Row 1:
 ──────
 file_segment_range_begin: 0
 file_segment_range_end:   1659
 size:                     1660
 state:                    DOWNLOADED
+SELECT count() FROM (SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path FROM system.remote_data_paths ) AS data_paths INNER JOIN system.filesystem_cache AS caches ON data_paths.cache_path = caches.cache_path
 8
+SELECT count(), sum(size) FROM system.filesystem_cache
 8	2014
+SELECT count(), sum(size) FROM system.filesystem_cache
 8	2014
+INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(100) SETTINGS enable_filesystem_cache_on_write_operations=0
+SELECT count(), sum(size) FROM system.filesystem_cache
 8	2014
+INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(100)
+INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(300, 10000)
+SELECT count(), sum(size) FROM system.filesystem_cache
 24	84045
+SYSTEM START MERGES test_02241
+SELECT value FROM system.events WHERE name = 'CachedWriteBufferCacheWriteBytes'
+81715476
+SELECT value FROM system.events WHERE name = 'CachedReadBufferCacheWriteBytes'
+OPTIMIZE TABLE test_02241 FINAL
+SELECT value FROM system.events WHERE name = 'CachedWriteBufferCacheWriteBytes'
+81881872
+SELECT value FROM system.events WHERE name = 'CachedReadBufferCacheWriteBytes'
+SELECT count(), sum(size) FROM system.filesystem_cache
 32	167243
+ALTER TABLE test_02241 UPDATE value = 'kek' WHERE key = 100
+SELECT count(), sum(size) FROM system.filesystem_cache
 41	250541
+INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(5000000)
+SYSTEM FLUSH LOGS
 INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(5000000)	0
+SELECT count() FROM test_02241
 5010500
+SELECT count() FROM test_02241 WHERE value LIKE '%010%'
 18816
diff --git a/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.sh b/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.sh
index 96f61cf61e8..2b237492e98 100755
--- a/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.sh
+++ b/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.sh
@@ -10,13 +10,13 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
 for STORAGE_POLICY in 's3_cache' 'local_cache'; do
     echo "Using storage policy: $STORAGE_POLICY"
 
-    $CLICKHOUSE_CLIENT --query "DROP TABLE IF EXISTS test_02241"
-    $CLICKHOUSE_CLIENT --query "CREATE TABLE test_02241 (key UInt32, value String) Engine=MergeTree() ORDER BY key SETTINGS storage_policy='$STORAGE_POLICY', min_bytes_for_wide_part = 10485760, compress_marks=false, compress_primary_key=false, ratio_of_defaults_for_sparse_serialization = 1"
-    $CLICKHOUSE_CLIENT --query "SYSTEM STOP MERGES test_02241"
+    $CLICKHOUSE_CLIENT --echo --query "DROP TABLE IF EXISTS test_02241"
+    $CLICKHOUSE_CLIENT --echo --query "CREATE TABLE test_02241 (key UInt32, value String) Engine=MergeTree() ORDER BY key SETTINGS storage_policy='$STORAGE_POLICY', min_bytes_for_wide_part = 10485760, compress_marks=false, compress_primary_key=false, ratio_of_defaults_for_sparse_serialization = 1"
+    $CLICKHOUSE_CLIENT --echo --query "SYSTEM STOP MERGES test_02241"
 
-    $CLICKHOUSE_CLIENT --query "SYSTEM DROP FILESYSTEM CACHE"
+    $CLICKHOUSE_CLIENT --echo --query "SYSTEM DROP FILESYSTEM CACHE"
 
-    $CLICKHOUSE_CLIENT -n --query "SELECT file_segment_range_begin, file_segment_range_end, size, state
+    $CLICKHOUSE_CLIENT --echo -n --query "SELECT file_segment_range_begin, file_segment_range_end, size, state
     FROM
     (
         SELECT file_segment_range_begin, file_segment_range_end, size, state, local_path
@@ -32,12 +32,12 @@ for STORAGE_POLICY in 's3_cache' 'local_cache'; do
     WHERE endsWith(local_path, 'data.bin')
     FORMAT Vertical"
 
-    $CLICKHOUSE_CLIENT --query "SELECT count() FROM (SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path FROM system.remote_data_paths ) AS data_paths INNER JOIN system.filesystem_cache AS caches ON data_paths.cache_path = caches.cache_path"
-    $CLICKHOUSE_CLIENT --query "SELECT count(), sum(size) FROM system.filesystem_cache"
+    $CLICKHOUSE_CLIENT --echo --query "SELECT count() FROM (SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path FROM system.remote_data_paths ) AS data_paths INNER JOIN system.filesystem_cache AS caches ON data_paths.cache_path = caches.cache_path"
+    $CLICKHOUSE_CLIENT --echo --query "SELECT count(), sum(size) FROM system.filesystem_cache"
 
-    $CLICKHOUSE_CLIENT --enable_filesystem_cache_on_write_operations=1 --query "INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(100)"
+    $CLICKHOUSE_CLIENT --echo --enable_filesystem_cache_on_write_operations=1 --query "INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(100)"
 
-    $CLICKHOUSE_CLIENT -n --query "SELECT file_segment_range_begin, file_segment_range_end, size, state
+    $CLICKHOUSE_CLIENT --echo -n --query "SELECT file_segment_range_begin, file_segment_range_end, size, state
     FROM
     (
         SELECT file_segment_range_begin, file_segment_range_end, size, state, local_path
@@ -53,24 +53,24 @@ for STORAGE_POLICY in 's3_cache' 'local_cache'; do
     WHERE endsWith(local_path, 'data.bin')
     FORMAT Vertical"
 
-    $CLICKHOUSE_CLIENT --query "SELECT count() FROM (SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path FROM system.remote_data_paths ) AS data_paths INNER JOIN system.filesystem_cache AS caches ON data_paths.cache_path = caches.cache_path"
-    $CLICKHOUSE_CLIENT --query "SELECT count(), sum(size) FROM system.filesystem_cache"
+    $CLICKHOUSE_CLIENT --echo --query "SELECT count() FROM (SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path FROM system.remote_data_paths ) AS data_paths INNER JOIN system.filesystem_cache AS caches ON data_paths.cache_path = caches.cache_path"
+    $CLICKHOUSE_CLIENT --echo --query "SELECT count(), sum(size) FROM system.filesystem_cache"
 
-    $CLICKHOUSE_CLIENT --query "SELECT count() FROM system.filesystem_cache WHERE cache_hits > 0"
+    $CLICKHOUSE_CLIENT --echo --query "SELECT count() FROM system.filesystem_cache WHERE cache_hits > 0"
 
-    $CLICKHOUSE_CLIENT --query "SELECT * FROM test_02241 FORMAT Null"
-    $CLICKHOUSE_CLIENT --query "SELECT count() FROM system.filesystem_cache WHERE cache_hits > 0"
+    $CLICKHOUSE_CLIENT --echo --query "SELECT * FROM test_02241 FORMAT Null"
+    $CLICKHOUSE_CLIENT --echo --query "SELECT count() FROM system.filesystem_cache WHERE cache_hits > 0"
 
-    $CLICKHOUSE_CLIENT --query "SELECT * FROM test_02241 FORMAT Null"
-    $CLICKHOUSE_CLIENT --query "SELECT count() FROM system.filesystem_cache WHERE cache_hits > 0"
+    $CLICKHOUSE_CLIENT --echo --query "SELECT * FROM test_02241 FORMAT Null"
+    $CLICKHOUSE_CLIENT --echo --query "SELECT count() FROM system.filesystem_cache WHERE cache_hits > 0"
 
-    $CLICKHOUSE_CLIENT --query "SELECT count(), sum(size) size FROM system.filesystem_cache"
+    $CLICKHOUSE_CLIENT --echo --query "SELECT count(), sum(size) size FROM system.filesystem_cache"
 
-    $CLICKHOUSE_CLIENT --query "SYSTEM DROP FILESYSTEM CACHE"
+    $CLICKHOUSE_CLIENT --echo --query "SYSTEM DROP FILESYSTEM CACHE"
 
-    $CLICKHOUSE_CLIENT --enable_filesystem_cache_on_write_operations=1 --query "INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(100, 200)"
+    $CLICKHOUSE_CLIENT --echo --enable_filesystem_cache_on_write_operations=1 --query "INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(100, 200)"
 
-    $CLICKHOUSE_CLIENT -n --query "SELECT file_segment_range_begin, file_segment_range_end, size, state
+    $CLICKHOUSE_CLIENT --echo -n --query "SELECT file_segment_range_begin, file_segment_range_end, size, state
     FROM
     (
         SELECT file_segment_range_begin, file_segment_range_end, size, state, local_path
@@ -86,27 +86,34 @@ for STORAGE_POLICY in 's3_cache' 'local_cache'; do
     WHERE endsWith(local_path, 'data.bin')
     FORMAT Vertical;"
 
-    $CLICKHOUSE_CLIENT --query "SELECT count() FROM (SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path FROM system.remote_data_paths ) AS data_paths INNER JOIN system.filesystem_cache AS caches ON data_paths.cache_path = caches.cache_path"
-    $CLICKHOUSE_CLIENT --query "SELECT count(), sum(size) FROM system.filesystem_cache"
+    $CLICKHOUSE_CLIENT --echo --query "SELECT count() FROM (SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path FROM system.remote_data_paths ) AS data_paths INNER JOIN system.filesystem_cache AS caches ON data_paths.cache_path = caches.cache_path"
+    $CLICKHOUSE_CLIENT --echo --query "SELECT count(), sum(size) FROM system.filesystem_cache"
 
-    $CLICKHOUSE_CLIENT --query "SELECT count(), sum(size) FROM system.filesystem_cache"
-    $CLICKHOUSE_CLIENT --enable_filesystem_cache_on_write_operations=1 --query "INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(100) SETTINGS enable_filesystem_cache_on_write_operations=0"
-    $CLICKHOUSE_CLIENT --query "SELECT count(), sum(size) FROM system.filesystem_cache"
+    $CLICKHOUSE_CLIENT --echo --query "SELECT count(), sum(size) FROM system.filesystem_cache"
+    $CLICKHOUSE_CLIENT --echo --enable_filesystem_cache_on_write_operations=1 --query "INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(100) SETTINGS enable_filesystem_cache_on_write_operations=0"
+    $CLICKHOUSE_CLIENT --echo --query "SELECT count(), sum(size) FROM system.filesystem_cache"
 
-    $CLICKHOUSE_CLIENT --enable_filesystem_cache_on_write_operations=1 --query "INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(100)"
-    $CLICKHOUSE_CLIENT --enable_filesystem_cache_on_write_operations=1 --query "INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(300, 10000)"
-    $CLICKHOUSE_CLIENT --query "SELECT count(), sum(size) FROM system.filesystem_cache"
+    $CLICKHOUSE_CLIENT --echo --enable_filesystem_cache_on_write_operations=1 --query "INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(100)"
+    $CLICKHOUSE_CLIENT --echo --enable_filesystem_cache_on_write_operations=1 --query "INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(300, 10000)"
+    $CLICKHOUSE_CLIENT --echo --query "SELECT count(), sum(size) FROM system.filesystem_cache"
 
-    $CLICKHOUSE_CLIENT --query "SYSTEM START MERGES test_02241"
+    $CLICKHOUSE_CLIENT --echo --query "SYSTEM START MERGES test_02241"
 
-    $CLICKHOUSE_CLIENT --enable_filesystem_cache_on_write_operations=1 --query "OPTIMIZE TABLE test_02241 FINAL"
-    $CLICKHOUSE_CLIENT --query "SELECT count(), sum(size) FROM system.filesystem_cache"
+    $CLICKHOUSE_CLIENT --echo --query "SELECT value FROM system.events WHERE name = 'CachedWriteBufferCacheWriteBytes'"
+    $CLICKHOUSE_CLIENT --echo --query "SELECT value FROM system.events WHERE name = 'CachedReadBufferCacheWriteBytes'"
 
-    $CLICKHOUSE_CLIENT --enable_filesystem_cache_on_write_operations=1 --mutations_sync=2 --query "ALTER TABLE test_02241 UPDATE value = 'kek' WHERE key = 100"
-    $CLICKHOUSE_CLIENT --query "SELECT count(), sum(size) FROM system.filesystem_cache"
-    $CLICKHOUSE_CLIENT --enable_filesystem_cache_on_write_operations=1 --query "INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(5000000)"
+    $CLICKHOUSE_CLIENT --echo --enable_filesystem_cache_on_write_operations=1 --query "OPTIMIZE TABLE test_02241 FINAL"
 
-    $CLICKHOUSE_CLIENT --query "SYSTEM FLUSH LOGS"
+    $CLICKHOUSE_CLIENT --echo --query "SELECT value FROM system.events WHERE name = 'CachedWriteBufferCacheWriteBytes'"
+    $CLICKHOUSE_CLIENT --echo --query "SELECT value FROM system.events WHERE name = 'CachedReadBufferCacheWriteBytes'"
+
+    $CLICKHOUSE_CLIENT --echo --query "SELECT count(), sum(size) FROM system.filesystem_cache"
+
+    $CLICKHOUSE_CLIENT --echo --enable_filesystem_cache_on_write_operations=1 --mutations_sync=2 --query "ALTER TABLE test_02241 UPDATE value = 'kek' WHERE key = 100"
+    $CLICKHOUSE_CLIENT --echo --query "SELECT count(), sum(size) FROM system.filesystem_cache"
+    $CLICKHOUSE_CLIENT --echo --enable_filesystem_cache_on_write_operations=1 --query "INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(5000000)"
+
+    $CLICKHOUSE_CLIENT --echo --query "SYSTEM FLUSH LOGS"
 
     $CLICKHOUSE_CLIENT -n --query "SELECT
         query, ProfileEvents['RemoteFSReadBytes'] > 0 as remote_fs_read
@@ -121,6 +128,6 @@ for STORAGE_POLICY in 's3_cache' 'local_cache'; do
         DESC
     LIMIT 1"
 
-    $CLICKHOUSE_CLIENT --query "SELECT count() FROM test_02241"
-    $CLICKHOUSE_CLIENT --query "SELECT count() FROM test_02241 WHERE value LIKE '%010%'"
+    $CLICKHOUSE_CLIENT --echo --query "SELECT count() FROM test_02241"
+    $CLICKHOUSE_CLIENT --echo --query "SELECT count() FROM test_02241 WHERE value LIKE '%010%'"
 done

From a80747b2385647678771281d815867ef87b580f6 Mon Sep 17 00:00:00 2001
From: Michael Kolupaev <michael.kolupaev@clickhouse.com>
Date: Thu, 22 Feb 2024 15:57:10 +0000
Subject: [PATCH 015/443] Undo something to the test

---
 .../0_stateless/02998_primary_key_skip_columns.reference        | 2 +-
 tests/queries/0_stateless/02998_primary_key_skip_columns.sql    | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/queries/0_stateless/02998_primary_key_skip_columns.reference b/tests/queries/0_stateless/02998_primary_key_skip_columns.reference
index ec44acbd16b..08ccdb83b11 100644
--- a/tests/queries/0_stateless/02998_primary_key_skip_columns.reference
+++ b/tests/queries/0_stateless/02998_primary_key_skip_columns.reference
@@ -15,4 +15,4 @@ Key size: 	2400000
 120
 2334
 19
-Key size ok: 	1	1
+Key size: 	800000
diff --git a/tests/queries/0_stateless/02998_primary_key_skip_columns.sql b/tests/queries/0_stateless/02998_primary_key_skip_columns.sql
index 27672d7854e..b2dadcc5e7c 100644
--- a/tests/queries/0_stateless/02998_primary_key_skip_columns.sql
+++ b/tests/queries/0_stateless/02998_primary_key_skip_columns.sql
@@ -28,6 +28,6 @@ SELECT count() FROM test WHERE b > 7898976344263989848 AND b < 80403209398191531
 SELECT count() FROM test WHERE a > 1849813033528774208 AND a < 4594276315503201760 AND c > 13239894303140990071 AND c < 16179795840886947236;
 SELECT count() FROM test WHERE a > 1849813033528774208 AND a < 4594276315503201760 AND b > 7898976344263989848 AND b < 8040320939819153137 AND c > 13239894303140990071 AND c < 16179795840886947236;
 
-SELECT 'Key size ok: ', (sum(primary_key_bytes_in_memory) as s) >= 800000, s < 1200000 FROM system.parts WHERE database = currentDatabase() AND table = 'test';
+SELECT 'Key size: ', round(sum(primary_key_bytes_in_memory), -5) FROM system.parts WHERE database = currentDatabase() AND table = 'test';
 
 DROP TABLE test;

From a34f42ca22c8a4820e4cbcf67cdd48a3589e3879 Mon Sep 17 00:00:00 2001
From: Azat Khuzhin <a3at.mail@gmail.com>
Date: Sat, 13 Jan 2024 18:48:47 +0300
Subject: [PATCH 016/443] Remove lock from the ReadProgressCallback

It looks redundant (added in 5ef51ed), though it has "fix tests" in the
log message, but CI reports is not available for the commits from that
PR [1], so let's try.

  [1]: https://github.com/ClickHouse/ClickHouse/pull/37543

Also this can be a big problem, since the code under that lock
(throttling or quotas with previous implementation that uses
boost::atomic_shared_ptr) may sleep.

Some numbers:

    run                     | time
    ------------------------|------
    max_threads=100 before  | 23.1
    max_threads=100 after   | 15.1
    max_threads=4500 before | 4.5
    max_threads=4500 after  | 2.3

Query:

    select sum(number) from numbers_mt(2000000) settings max_threads=X, max_block_size = 1

Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>
---
 src/QueryPipeline/ReadProgressCallback.cpp   | 2 --
 src/QueryPipeline/ReadProgressCallback.h     | 1 -
 tests/performance/small_block_contention.xml | 3 +++
 3 files changed, 3 insertions(+), 3 deletions(-)
 create mode 100644 tests/performance/small_block_contention.xml

diff --git a/src/QueryPipeline/ReadProgressCallback.cpp b/src/QueryPipeline/ReadProgressCallback.cpp
index 59843d8791d..e90fc24d882 100644
--- a/src/QueryPipeline/ReadProgressCallback.cpp
+++ b/src/QueryPipeline/ReadProgressCallback.cpp
@@ -126,8 +126,6 @@ bool ReadProgressCallback::onProgress(uint64_t read_rows, uint64_t read_bytes, c
 
         CurrentThread::updatePerformanceCountersIfNeeded();
 
-        std::lock_guard lock(limits_and_quotas_mutex);
-
         /// TODO: Should be done in PipelineExecutor.
         for (const auto & limits : storage_limits)
             limits.local_limits.speed_limits.throttle(progress.read_rows, progress.read_bytes, total_rows, total_stopwatch.elapsedMicroseconds(), limits.local_limits.timeout_overflow_mode);
diff --git a/src/QueryPipeline/ReadProgressCallback.h b/src/QueryPipeline/ReadProgressCallback.h
index 5dbf3344bdf..7dfed9df5da 100644
--- a/src/QueryPipeline/ReadProgressCallback.h
+++ b/src/QueryPipeline/ReadProgressCallback.h
@@ -41,7 +41,6 @@ private:
     /// The total number of bytes to read. For progress bar.
     std::atomic_size_t total_bytes = 0;
 
-    std::mutex limits_and_quotas_mutex;
     Stopwatch total_stopwatch{CLOCK_MONOTONIC_COARSE};  /// Including waiting time
 
     bool update_profile_events = true;
diff --git a/tests/performance/small_block_contention.xml b/tests/performance/small_block_contention.xml
new file mode 100644
index 00000000000..ce1995a0a29
--- /dev/null
+++ b/tests/performance/small_block_contention.xml
@@ -0,0 +1,3 @@
+<test>
+    <query>select sum(number) from numbers_mt(200000) settings max_threads=100, max_block_size = 1 format Null</query>
+</test>

From 9cb1ade3e2967507885f5b0e3deefab2ad40082c Mon Sep 17 00:00:00 2001
From: serxa <sergei@clickhouse.com>
Date: Thu, 22 Feb 2024 17:07:47 +0000
Subject: [PATCH 017/443] fix db iterator wait

---
 src/Common/AsyncLoader.cpp                |  2 +-
 src/Databases/DatabaseOrdinary.cpp        | 20 ++++++++++++++++----
 src/Databases/IDatabase.h                 | 17 +----------------
 src/Interpreters/InterpreterDropQuery.cpp |  2 +-
 4 files changed, 19 insertions(+), 22 deletions(-)

diff --git a/src/Common/AsyncLoader.cpp b/src/Common/AsyncLoader.cpp
index 140194e10b4..4a39454ccbb 100644
--- a/src/Common/AsyncLoader.cpp
+++ b/src/Common/AsyncLoader.cpp
@@ -39,7 +39,7 @@ void logAboutProgress(LoggerPtr log, size_t processed, size_t total, AtomicStopw
 {
     if (total && (processed % PRINT_MESSAGE_EACH_N_OBJECTS == 0 || watch.compareAndRestart(PRINT_MESSAGE_EACH_N_SECONDS)))
     {
-        LOG_INFO(log, "Processed: {}%", static_cast<Int64>(processed * 1000.0 / total) * 0.1);
+        LOG_INFO(log, "Processed: {:.1f}%", static_cast<double>(processed) * 100.0 / total);
         watch.restart();
     }
 }
diff --git a/src/Databases/DatabaseOrdinary.cpp b/src/Databases/DatabaseOrdinary.cpp
index 644bed23350..40e0fb0a0ed 100644
--- a/src/Databases/DatabaseOrdinary.cpp
+++ b/src/Databases/DatabaseOrdinary.cpp
@@ -289,10 +289,22 @@ void DatabaseOrdinary::stopLoading()
 
 DatabaseTablesIteratorPtr DatabaseOrdinary::getTablesIterator(ContextPtr local_context, const DatabaseOnDisk::FilterByNameFunction & filter_by_table_name) const
 {
-    auto result = DatabaseWithOwnTablesBase::getTablesIterator(local_context, filter_by_table_name);
-    std::scoped_lock lock(mutex);
-    typeid_cast<DatabaseTablesSnapshotIterator &>(*result).setLoadTasks(startup_table);
-    return result;
+    // Wait for every table (matching the filter) to be loaded and started up before we make the snapshot.
+    // It is important, because otherwise table might be:
+    //  - not attached and thus will be missed in the snapshot;
+    //  - not started, which is not good for DDL operations.
+    LoadTaskPtrs tasks_to_wait;
+    {
+        std::lock_guard lock(mutex);
+        if (!filter_by_table_name)
+            tasks_to_wait.reserve(startup_table.size());
+        for (const auto & [table_name, task] : startup_table)
+            if (!filter_by_table_name || filter_by_table_name(table_name))
+                tasks_to_wait.emplace_back(task);
+    }
+    waitLoad(currentPoolOr(TablesLoaderForegroundPoolId), tasks_to_wait);
+
+    return DatabaseWithOwnTablesBase::getTablesIterator(local_context, filter_by_table_name);
 }
 
 void DatabaseOrdinary::alterTable(ContextPtr local_context, const StorageID & table_id, const StorageInMemoryMetadata & metadata)
diff --git a/src/Databases/IDatabase.h b/src/Databases/IDatabase.h
index ec380fa759d..75662bfebe3 100644
--- a/src/Databases/IDatabase.h
+++ b/src/Databases/IDatabase.h
@@ -77,17 +77,12 @@ private:
     Tables tables;
     Tables::iterator it;
 
-    // Tasks to wait before returning a table
-    using Tasks = std::unordered_map<String, LoadTaskPtr>;
-    Tasks tasks;
-
 protected:
     DatabaseTablesSnapshotIterator(DatabaseTablesSnapshotIterator && other) noexcept
     : IDatabaseTablesIterator(std::move(other.database_name))
     {
         size_t idx = std::distance(other.tables.begin(), other.it);
         std::swap(tables, other.tables);
-        std::swap(tasks, other.tasks);
         other.it = other.tables.end();
         it = tables.begin();
         std::advance(it, idx);
@@ -110,17 +105,7 @@ public:
 
     const String & name() const override { return it->first; }
 
-    const StoragePtr & table() const override
-    {
-        if (auto task = tasks.find(it->first); task != tasks.end())
-            waitLoad(currentPoolOr(TablesLoaderForegroundPoolId), task->second);
-        return it->second;
-    }
-
-    void setLoadTasks(const Tasks & tasks_)
-    {
-        tasks = tasks_;
-    }
+    const StoragePtr & table() const override { return it->second; }
 };
 
 using DatabaseTablesIteratorPtr = std::unique_ptr<IDatabaseTablesIterator>;
diff --git a/src/Interpreters/InterpreterDropQuery.cpp b/src/Interpreters/InterpreterDropQuery.cpp
index 711100b5de1..72aa4cc63e3 100644
--- a/src/Interpreters/InterpreterDropQuery.cpp
+++ b/src/Interpreters/InterpreterDropQuery.cpp
@@ -417,7 +417,7 @@ BlockIO InterpreterDropQuery::executeToDatabaseImpl(const ASTDropQuery & query,
                     uuids_to_wait.push_back(table_to_wait);
                 }
             }
-           // only if operation is DETACH
+            // only if operation is DETACH
             if ((!drop || !truncate) && query.sync)
             {
                 /// Avoid "some tables are still in use" when sync mode is enabled

From 18741f122eabaeb7903f355958af1e1a88818e83 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Mon, 26 Feb 2024 12:42:13 +0800
Subject: [PATCH 018/443] Move a setting to server setting

---
 src/Core/ServerSettings.h                                  | 2 ++
 src/Core/Settings.h                                        | 1 -
 src/Interpreters/Context.cpp                               | 2 +-
 .../config.d/force_read_through_cache_for_merges.xml       | 3 +++
 tests/integration/test_filesystem_cache/test.py            | 2 +-
 .../users.d/force_read_through_cache_on_merge.xml          | 7 -------
 6 files changed, 7 insertions(+), 10 deletions(-)
 create mode 100644 tests/integration/test_filesystem_cache/config.d/force_read_through_cache_for_merges.xml
 delete mode 100644 tests/integration/test_filesystem_cache/users.d/force_read_through_cache_on_merge.xml

diff --git a/src/Core/ServerSettings.h b/src/Core/ServerSettings.h
index de2a4e9b755..0283b98638f 100644
--- a/src/Core/ServerSettings.h
+++ b/src/Core/ServerSettings.h
@@ -103,6 +103,8 @@ namespace DB
     M(Bool, async_load_databases, false, "Enable asynchronous loading of databases and tables to speedup server startup. Queries to not yet loaded entity will be blocked until load is finished.", 0) \
     M(Bool, display_secrets_in_show_and_select, false, "Allow showing secrets in SHOW and SELECT queries via a format setting and a grant", 0) \
     \
+    M(Bool, force_read_through_cache_for_merges, false, "Force read-through filesystem cache for merges", 0) \
+    \
     M(Seconds, keep_alive_timeout, DEFAULT_HTTP_KEEP_ALIVE_TIMEOUT, "The number of seconds that ClickHouse waits for incoming requests before closing the connection.", 0) \
     M(Seconds, replicated_fetches_http_connection_timeout, 0, "HTTP connection timeout for part fetch requests. Inherited from default profile `http_connection_timeout` if not set explicitly.", 0) \
     M(Seconds, replicated_fetches_http_send_timeout, 0, "HTTP send timeout for part fetch requests. Inherited from default profile `http_send_timeout` if not set explicitly.", 0) \
diff --git a/src/Core/Settings.h b/src/Core/Settings.h
index db060bf712d..433195af9c3 100644
--- a/src/Core/Settings.h
+++ b/src/Core/Settings.h
@@ -769,7 +769,6 @@ class IColumn;
     M(Bool, enable_filesystem_cache_on_write_operations, false, "Write into cache on write operations. To actually work this setting requires be added to disk config too", 0) \
     M(Bool, enable_filesystem_cache_log, false, "Allows to record the filesystem caching log for each query", 0) \
     M(Bool, read_from_filesystem_cache_if_exists_otherwise_bypass_cache, false, "Allow to use the filesystem cache in passive mode - benefit from the existing cache entries, but don't put more entries into the cache. If you set this setting for heavy ad-hoc queries and leave it disabled for short real-time queries, this will allows to avoid cache threshing by too heavy queries and to improve the overall system efficiency.", 0) \
-    M(Bool, force_read_through_cache_for_merges, false, "Force read-through cache for merges", 0) \
     M(Bool, skip_download_if_exceeds_query_cache, true, "Skip download from remote filesystem if exceeds query cache size", 0) \
     M(UInt64, filesystem_cache_max_download_size, (128UL * 1024 * 1024 * 1024), "Max remote filesystem cache size that can be downloaded by a single query", 0) \
     M(Bool, throw_on_error_from_cache_on_write_operations, false, "Ignore error from cache when caching on write operations (INSERT, merges)", 0) \
diff --git a/src/Interpreters/Context.cpp b/src/Interpreters/Context.cpp
index 36b362e36bb..a974eaca067 100644
--- a/src/Interpreters/Context.cpp
+++ b/src/Interpreters/Context.cpp
@@ -5079,7 +5079,7 @@ ReadSettings Context::getReadSettings() const
     res.read_from_filesystem_cache_if_exists_otherwise_bypass_cache = settings.read_from_filesystem_cache_if_exists_otherwise_bypass_cache;
     res.enable_filesystem_cache_log = settings.enable_filesystem_cache_log;
     res.filesystem_cache_segments_batch_size = settings.filesystem_cache_segments_batch_size;
-    res.force_read_through_cache_merges = settings.force_read_through_cache_for_merges;
+    res.force_read_through_cache_merges = getServerSettings().force_read_through_cache_for_merges;
 
     res.filesystem_cache_max_download_size = settings.filesystem_cache_max_download_size;
     res.skip_download_if_exceeds_query_cache = settings.skip_download_if_exceeds_query_cache;
diff --git a/tests/integration/test_filesystem_cache/config.d/force_read_through_cache_for_merges.xml b/tests/integration/test_filesystem_cache/config.d/force_read_through_cache_for_merges.xml
new file mode 100644
index 00000000000..bb2a6e850a4
--- /dev/null
+++ b/tests/integration/test_filesystem_cache/config.d/force_read_through_cache_for_merges.xml
@@ -0,0 +1,3 @@
+<clickhouse>
+    <force_read_through_cache_for_merges>1</force_read_through_cache_for_merges>
+</clickhouse>
diff --git a/tests/integration/test_filesystem_cache/test.py b/tests/integration/test_filesystem_cache/test.py
index c1ba6702dcf..f32fa4e9823 100644
--- a/tests/integration/test_filesystem_cache/test.py
+++ b/tests/integration/test_filesystem_cache/test.py
@@ -42,9 +42,9 @@ def cluster():
             "node_force_read_through_cache_on_merge",
             main_configs=[
                 "config.d/storage_conf.xml",
+                "config.d/force_read_through_cache_for_merges.xml",
             ],
             user_configs=[
-                "users.d/force_read_through_cache_on_merge.xml",
                 "users.d/cache_on_write_operations.xml",
             ],
             stay_alive=True,
diff --git a/tests/integration/test_filesystem_cache/users.d/force_read_through_cache_on_merge.xml b/tests/integration/test_filesystem_cache/users.d/force_read_through_cache_on_merge.xml
deleted file mode 100644
index 4d26a1a8bc7..00000000000
--- a/tests/integration/test_filesystem_cache/users.d/force_read_through_cache_on_merge.xml
+++ /dev/null
@@ -1,7 +0,0 @@
-<clickhouse>
-    <profiles>
-        <default>
-            <force_read_through_cache_for_merges>1</force_read_through_cache_for_merges>
-        </default>
-    </profiles>
-</clickhouse>

From bf5affbe640976d2b73e12f5213a13baacf40619 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Mon, 26 Feb 2024 16:37:09 +0800
Subject: [PATCH 019/443] Fix test

---
 .../02241_filesystem_cache_on_write_operations.sh           | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.sh b/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.sh
index 2b237492e98..ee1d942a421 100755
--- a/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.sh
+++ b/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.sh
@@ -99,14 +99,8 @@ for STORAGE_POLICY in 's3_cache' 'local_cache'; do
 
     $CLICKHOUSE_CLIENT --echo --query "SYSTEM START MERGES test_02241"
 
-    $CLICKHOUSE_CLIENT --echo --query "SELECT value FROM system.events WHERE name = 'CachedWriteBufferCacheWriteBytes'"
-    $CLICKHOUSE_CLIENT --echo --query "SELECT value FROM system.events WHERE name = 'CachedReadBufferCacheWriteBytes'"
-
     $CLICKHOUSE_CLIENT --echo --enable_filesystem_cache_on_write_operations=1 --query "OPTIMIZE TABLE test_02241 FINAL"
 
-    $CLICKHOUSE_CLIENT --echo --query "SELECT value FROM system.events WHERE name = 'CachedWriteBufferCacheWriteBytes'"
-    $CLICKHOUSE_CLIENT --echo --query "SELECT value FROM system.events WHERE name = 'CachedReadBufferCacheWriteBytes'"
-
     $CLICKHOUSE_CLIENT --echo --query "SELECT count(), sum(size) FROM system.filesystem_cache"
 
     $CLICKHOUSE_CLIENT --echo --enable_filesystem_cache_on_write_operations=1 --mutations_sync=2 --query "ALTER TABLE test_02241 UPDATE value = 'kek' WHERE key = 100"

From 277e8d965555b4fcd09a755282666bcae36adae6 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Mon, 26 Feb 2024 14:03:53 +0800
Subject: [PATCH 020/443] Fix usage plain metadata type with new configuration
 option

---
 src/Disks/DiskType.cpp                        | 48 +++++++++++++++++++
 src/Disks/DiskType.h                          | 34 +------------
 src/Disks/ObjectStorages/IObjectStorage.h     |  1 +
 .../ObjectStorages/MetadataStorageFactory.cpp | 36 +++++++++++---
 .../ObjectStorages/MetadataStorageFactory.h   |  7 +++
 .../ObjectStorages/ObjectStorageFactory.cpp   | 43 +++++++++++++----
 src/Disks/ObjectStorages/PlainObjectStorage.h | 29 +++++++++++
 .../RegisterDiskObjectStorage.cpp             | 24 ++--------
 src/Disks/ObjectStorages/S3/S3ObjectStorage.h | 21 --------
 .../configs/disk_s3.xml                       |  7 +++
 .../test_attach_backup_from_s3_plain/test.py  | 25 ++++++----
 11 files changed, 178 insertions(+), 97 deletions(-)
 create mode 100644 src/Disks/ObjectStorages/PlainObjectStorage.h

diff --git a/src/Disks/DiskType.cpp b/src/Disks/DiskType.cpp
index 218b6ee7f26..1778ae8025b 100644
--- a/src/Disks/DiskType.cpp
+++ b/src/Disks/DiskType.cpp
@@ -1,7 +1,27 @@
 #include "DiskType.h"
+#include <Poco/String.h>
+#include <Common/Exception.h>
 
 namespace DB
 {
+namespace ErrorCodes
+{
+    extern const int UNKNOWN_ELEMENT_IN_CONFIG;
+}
+
+MetadataStorageType metadataTypeFromString(const String & type)
+{
+    auto check_type = Poco::toLower(type);
+    if (check_type == "local")
+        return MetadataStorageType::Local;
+    if (check_type == "plain")
+        return MetadataStorageType::Plain;
+    if (check_type == "web")
+        return MetadataStorageType::StaticWeb;
+
+    throw Exception(ErrorCodes::UNKNOWN_ELEMENT_IN_CONFIG,
+                    "MetadataStorageFactory: unknown metadata storage type: {}", type);
+}
 
 bool DataSourceDescription::operator==(const DataSourceDescription & other) const
 {
@@ -14,4 +34,32 @@ bool DataSourceDescription::sameKind(const DataSourceDescription & other) const
         == std::tie(other.type, other.object_storage_type, other.description);
 }
 
+std::string DataSourceDescription::toString() const
+{
+    switch (type)
+    {
+        case DataSourceType::Local:
+            return "local";
+        case DataSourceType::RAM:
+            return "memory";
+        case DataSourceType::ObjectStorage:
+        {
+            switch (object_storage_type)
+            {
+                case ObjectStorageType::S3:
+                    return "s3";
+                case ObjectStorageType::HDFS:
+                    return "hdfs";
+                case ObjectStorageType::Azure:
+                    return "azure_blob_storage";
+                case ObjectStorageType::Local:
+                    return "local_blob_storage";
+                case ObjectStorageType::Web:
+                    return "web";
+                case ObjectStorageType::None:
+                    return "none";
+            }
+        }
+    }
+}
 }
diff --git a/src/Disks/DiskType.h b/src/Disks/DiskType.h
index 15940ea9155..36fe4d83004 100644
--- a/src/Disks/DiskType.h
+++ b/src/Disks/DiskType.h
@@ -17,7 +17,6 @@ enum class ObjectStorageType
 {
     None,
     S3,
-    S3_Plain,
     Azure,
     HDFS,
     Web,
@@ -30,9 +29,9 @@ enum class MetadataStorageType
     Local,
     Plain,
     StaticWeb,
-    Memory,
 };
 
+MetadataStorageType metadataTypeFromString(const String & type);
 String toString(DataSourceType data_source_type);
 
 struct DataSourceDescription
@@ -49,36 +48,7 @@ struct DataSourceDescription
     bool operator==(const DataSourceDescription & other) const;
     bool sameKind(const DataSourceDescription & other) const;
 
-    std::string toString() const
-    {
-        switch (type)
-        {
-            case DataSourceType::Local:
-                return "local";
-            case DataSourceType::RAM:
-                return "memory";
-            case DataSourceType::ObjectStorage:
-            {
-                switch (object_storage_type)
-                {
-                    case ObjectStorageType::S3:
-                        return "s3";
-                    case ObjectStorageType::S3_Plain:
-                        return "s3_plain";
-                    case ObjectStorageType::HDFS:
-                        return "hdfs";
-                    case ObjectStorageType::Azure:
-                        return "azure_blob_storage";
-                    case ObjectStorageType::Local:
-                        return "local_blob_storage";
-                    case ObjectStorageType::Web:
-                        return "web";
-                    case ObjectStorageType::None:
-                        return "none";
-                }
-            }
-        }
-    }
+    std::string toString() const;
 };
 
 }
diff --git a/src/Disks/ObjectStorages/IObjectStorage.h b/src/Disks/ObjectStorages/IObjectStorage.h
index 56c269a3fc5..fde97d82ad1 100644
--- a/src/Disks/ObjectStorages/IObjectStorage.h
+++ b/src/Disks/ObjectStorages/IObjectStorage.h
@@ -218,6 +218,7 @@ public:
 
     virtual bool isReadOnly() const { return false; }
     virtual bool isWriteOnce() const { return false; }
+    virtual bool isPlain() const { return false; }
 
     virtual bool supportParallelWrite() const { return false; }
 
diff --git a/src/Disks/ObjectStorages/MetadataStorageFactory.cpp b/src/Disks/ObjectStorages/MetadataStorageFactory.cpp
index 52a0b9ec268..adc1f84372c 100644
--- a/src/Disks/ObjectStorages/MetadataStorageFactory.cpp
+++ b/src/Disks/ObjectStorages/MetadataStorageFactory.cpp
@@ -32,6 +32,35 @@ void MetadataStorageFactory::registerMetadataStorageType(const std::string & met
     }
 }
 
+std::string MetadataStorageFactory::getCompatibilityMetadataTypeHint(const ObjectStorageType & type)
+{
+    switch (type)
+    {
+        case ObjectStorageType::S3:
+        case ObjectStorageType::HDFS:
+        case ObjectStorageType::Local:
+        case ObjectStorageType::Azure:
+            return "local";
+        case ObjectStorageType::Web:
+            return "web";
+        default:
+            return "";
+    }
+}
+
+std::string MetadataStorageFactory::getMetadataType(
+    const Poco::Util::AbstractConfiguration & config,
+    const std::string & config_prefix,
+    const std::string & compatibility_type_hint)
+{
+    if (compatibility_type_hint.empty() && !config.has(config_prefix + ".metadata_type"))
+    {
+        throw Exception(ErrorCodes::NO_ELEMENTS_IN_CONFIG, "Expected `metadata_type` in config");
+    }
+
+    return config.getString(config_prefix + ".metadata_type", compatibility_type_hint);
+}
+
 MetadataStoragePtr MetadataStorageFactory::create(
     const std::string & name,
     const Poco::Util::AbstractConfiguration & config,
@@ -39,12 +68,7 @@ MetadataStoragePtr MetadataStorageFactory::create(
     ObjectStoragePtr object_storage,
     const std::string & compatibility_type_hint) const
 {
-    if (compatibility_type_hint.empty() && !config.has(config_prefix + ".metadata_type"))
-    {
-        throw Exception(ErrorCodes::NO_ELEMENTS_IN_CONFIG, "Expected `metadata_type` in config");
-    }
-
-    const auto type = config.getString(config_prefix + ".metadata_type", compatibility_type_hint);
+    const auto type = getMetadataType(config, config_prefix, compatibility_type_hint);
     const auto it = registry.find(type);
 
     if (it == registry.end())
diff --git a/src/Disks/ObjectStorages/MetadataStorageFactory.h b/src/Disks/ObjectStorages/MetadataStorageFactory.h
index 5f61125c599..467cd3cef98 100644
--- a/src/Disks/ObjectStorages/MetadataStorageFactory.h
+++ b/src/Disks/ObjectStorages/MetadataStorageFactory.h
@@ -25,6 +25,13 @@ public:
         ObjectStoragePtr object_storage,
         const std::string & compatibility_type_hint) const;
 
+    static std::string getMetadataType(
+        const Poco::Util::AbstractConfiguration & config,
+        const std::string & config_prefix,
+        const std::string & compatibility_type_hint = "");
+
+    static std::string getCompatibilityMetadataTypeHint(const ObjectStorageType & type);
+
 private:
     using Registry = std::unordered_map<String, Creator>;
     Registry registry;
diff --git a/src/Disks/ObjectStorages/ObjectStorageFactory.cpp b/src/Disks/ObjectStorages/ObjectStorageFactory.cpp
index b3626135177..6f6ff199902 100644
--- a/src/Disks/ObjectStorages/ObjectStorageFactory.cpp
+++ b/src/Disks/ObjectStorages/ObjectStorageFactory.cpp
@@ -16,8 +16,10 @@
 #ifndef CLICKHOUSE_KEEPER_STANDALONE_BUILD
 #include <Disks/ObjectStorages/Web/WebObjectStorage.h>
 #include <Disks/ObjectStorages/Local/LocalObjectStorage.h>
+#include <Disks/ObjectStorages/PlainObjectStorage.h>
 #include <Disks/loadLocalDiskConfig.h>
 #endif
+#include <Disks/ObjectStorages/MetadataStorageFactory.h>
 #include <Interpreters/Context.h>
 #include <Common/Macros.h>
 
@@ -32,6 +34,28 @@ namespace ErrorCodes
     extern const int LOGICAL_ERROR;
 }
 
+namespace
+{
+    template <typename BaseObjectStorage, class ...Args>
+    ObjectStoragePtr createObjectStorage(
+        const Poco::Util::AbstractConfiguration & config,
+        const std::string & config_prefix,
+        Args && ...args)
+    {
+        auto compatibility_hint = MetadataStorageFactory::getCompatibilityMetadataTypeHint(ObjectStorageType::S3);
+        auto metadata_type = MetadataStorageFactory::getMetadataType(config, config_prefix, compatibility_hint);
+
+        if (metadataTypeFromString(metadata_type) == MetadataStorageType::Plain)
+        {
+            return std::make_shared<PlainObjectStorage<BaseObjectStorage>>(std::forward<Args>(args)...);
+        }
+        else
+        {
+            return std::make_shared<BaseObjectStorage>(std::forward<Args>(args)...);
+        }
+    }
+}
+
 ObjectStorageFactory & ObjectStorageFactory::instance()
 {
     static ObjectStorageFactory factory;
@@ -129,12 +153,12 @@ void registerS3ObjectStorage(ObjectStorageFactory & factory)
         auto client = getClient(config, config_prefix, context, *settings);
         auto key_generator = getKeyGenerator(disk_type, uri, config, config_prefix);
 
-        auto object_storage = std::make_shared<S3ObjectStorage>(
-            std::move(client), std::move(settings), uri, s3_capabilities, key_generator, name);
+        auto object_storage = createObjectStorage<S3ObjectStorage>(
+            config, config_prefix, std::move(client), std::move(settings), uri, s3_capabilities, key_generator, name);
 
         /// NOTE: should we still perform this check for clickhouse-disks?
         if (!skip_access_check)
-            checkS3Capabilities(*object_storage, s3_capabilities, name, uri.key);
+            checkS3Capabilities(*dynamic_cast<S3ObjectStorage *>(object_storage.get()), s3_capabilities, name, uri.key);
 
         return object_storage;
     });
@@ -165,12 +189,12 @@ void registerS3PlainObjectStorage(ObjectStorageFactory & factory)
         auto client = getClient(config, config_prefix, context, *settings);
         auto key_generator = getKeyGenerator(disk_type, uri, config, config_prefix);
 
-        auto object_storage = std::make_shared<S3PlainObjectStorage>(
+        auto object_storage = std::make_shared<PlainObjectStorage<S3ObjectStorage>>(
             std::move(client), std::move(settings), uri, s3_capabilities, key_generator, name);
 
         /// NOTE: should we still perform this check for clickhouse-disks?
         if (!skip_access_check)
-            checkS3Capabilities(*object_storage, s3_capabilities, name, uri.key);
+            checkS3Capabilities(*dynamic_cast<S3ObjectStorage *>(object_storage.get()), s3_capabilities, name, uri.key);
 
         return object_storage;
     });
@@ -198,7 +222,7 @@ void registerHDFSObjectStorage(ObjectStorageFactory & factory)
             context->getSettingsRef().hdfs_replication
         );
 
-        return std::make_unique<HDFSObjectStorage>(uri, std::move(settings), config);
+        return createObjectStorage<HDFSObjectStorage>(config, config_prefix, uri, std::move(settings), config);
     });
 }
 #endif
@@ -214,7 +238,8 @@ void registerAzureObjectStorage(ObjectStorageFactory & factory)
         bool /* skip_access_check */) -> ObjectStoragePtr
     {
         String container_name = config.getString(config_prefix + ".container_name", "default-container");
-        return std::make_unique<AzureObjectStorage>(
+        return createObjectStorage<AzureObjectStorage>(
+            config, config_prefix,
             name,
             getAzureBlobContainerClient(config, config_prefix),
             getAzureBlobStorageSettings(config, config_prefix, context),
@@ -248,7 +273,7 @@ void registerWebObjectStorage(ObjectStorageFactory & factory)
                 ErrorCodes::BAD_ARGUMENTS, "Bad URI: `{}`. Error: {}", uri, e.what());
         }
 
-        return std::make_shared<WebObjectStorage>(uri, context);
+        return createObjectStorage<WebObjectStorage>(config, config_prefix, uri, context);
     });
 }
 
@@ -266,7 +291,7 @@ void registerLocalObjectStorage(ObjectStorageFactory & factory)
         loadDiskLocalConfig(name, config, config_prefix, context, object_key_prefix, keep_free_space_bytes);
         /// keys are mapped to the fs, object_key_prefix is a directory also
         fs::create_directories(object_key_prefix);
-        return std::make_shared<LocalObjectStorage>(object_key_prefix);
+        return createObjectStorage<LocalObjectStorage>(config, config_prefix, object_key_prefix);
     });
 }
 #endif
diff --git a/src/Disks/ObjectStorages/PlainObjectStorage.h b/src/Disks/ObjectStorages/PlainObjectStorage.h
new file mode 100644
index 00000000000..3a81b85c44b
--- /dev/null
+++ b/src/Disks/ObjectStorages/PlainObjectStorage.h
@@ -0,0 +1,29 @@
+#pragma once
+#include <Disks/ObjectStorages/IObjectStorage.h>
+
+namespace DB
+{
+
+/// Do not encode keys, store as-is, and do not require separate disk for metadata.
+/// But because of this does not support renames/hardlinks/attrs/...
+///
+/// NOTE: This disk has excessive API calls.
+template <typename BaseObjectStorage>
+class PlainObjectStorage : public BaseObjectStorage
+{
+public:
+    template <class ...Args>
+    explicit PlainObjectStorage(Args && ...args)
+        : BaseObjectStorage(std::forward<Args>(args)...) {}
+
+    std::string getName() const override { return "" + BaseObjectStorage::getName(); }
+
+    /// Notes:
+    /// - supports BACKUP to this disk
+    /// - does not support INSERT into MergeTree table on this disk
+    bool isWriteOnce() const override { return true; }
+
+    bool isPlain() const override { return true; }
+};
+
+}
diff --git a/src/Disks/ObjectStorages/RegisterDiskObjectStorage.cpp b/src/Disks/ObjectStorages/RegisterDiskObjectStorage.cpp
index 383a0b079b5..669a0102951 100644
--- a/src/Disks/ObjectStorages/RegisterDiskObjectStorage.cpp
+++ b/src/Disks/ObjectStorages/RegisterDiskObjectStorage.cpp
@@ -10,25 +10,6 @@ namespace DB
 void registerObjectStorages();
 void registerMetadataStorages();
 
-static std::string getCompatibilityMetadataTypeHint(const ObjectStorageType & type)
-{
-    switch (type)
-    {
-        case ObjectStorageType::S3:
-        case ObjectStorageType::HDFS:
-        case ObjectStorageType::Local:
-        case ObjectStorageType::Azure:
-            return "local";
-        case ObjectStorageType::S3_Plain:
-            return "plain";
-        case ObjectStorageType::Web:
-            return "web";
-        case ObjectStorageType::None:
-            return "";
-    }
-    UNREACHABLE();
-}
-
 void registerDiskObjectStorage(DiskFactory & factory, bool global_skip_access_check)
 {
     registerObjectStorages();
@@ -47,7 +28,10 @@ void registerDiskObjectStorage(DiskFactory & factory, bool global_skip_access_ch
         std::string compatibility_metadata_type_hint;
         if (!config.has(config_prefix + ".metadata_type"))
         {
-            compatibility_metadata_type_hint = getCompatibilityMetadataTypeHint(object_storage->getType());
+            if (object_storage->isPlain())
+                compatibility_metadata_type_hint = "plain";
+            else
+                compatibility_metadata_type_hint = MetadataStorageFactory::getCompatibilityMetadataTypeHint(object_storage->getType());
         }
 
         auto metadata_storage = MetadataStorageFactory::instance().create(
diff --git a/src/Disks/ObjectStorages/S3/S3ObjectStorage.h b/src/Disks/ObjectStorages/S3/S3ObjectStorage.h
index ab0fa5bed68..4ece98c5ec4 100644
--- a/src/Disks/ObjectStorages/S3/S3ObjectStorage.h
+++ b/src/Disks/ObjectStorages/S3/S3ObjectStorage.h
@@ -182,27 +182,6 @@ private:
     LoggerPtr log;
 };
 
-/// Do not encode keys, store as-is, and do not require separate disk for metadata.
-/// But because of this does not support renames/hardlinks/attrs/...
-///
-/// NOTE: This disk has excessive API calls.
-class S3PlainObjectStorage : public S3ObjectStorage
-{
-public:
-    std::string getName() const override { return "S3PlainObjectStorage"; }
-
-    template <class ...Args>
-    explicit S3PlainObjectStorage(Args && ...args)
-        : S3ObjectStorage("S3PlainObjectStorage", std::forward<Args>(args)...) {}
-
-    ObjectStorageType getType() const override { return ObjectStorageType::S3_Plain; }
-
-    /// Notes:
-    /// - supports BACKUP to this disk
-    /// - does not support INSERT into MergeTree table on this disk
-    bool isWriteOnce() const override { return true; }
-};
-
 }
 
 #endif
diff --git a/tests/integration/test_attach_backup_from_s3_plain/configs/disk_s3.xml b/tests/integration/test_attach_backup_from_s3_plain/configs/disk_s3.xml
index 779e4b6ae21..3166eea7ccb 100644
--- a/tests/integration/test_attach_backup_from_s3_plain/configs/disk_s3.xml
+++ b/tests/integration/test_attach_backup_from_s3_plain/configs/disk_s3.xml
@@ -8,9 +8,16 @@
                 <access_key_id>minio</access_key_id>
                 <secret_access_key>minio123</secret_access_key>
             </backup_disk_s3_plain>
+            <backup_disk_local_plain>
+                <type>object_storage</type>
+                <object_storage_type>local</object_storage_type>
+                <metadata_type>plain</metadata_type>
+                <path>local_plain/</path>
+            </backup_disk_local_plain>
         </disks>
     </storage_configuration>
     <backups>
         <allowed_disk>backup_disk_s3_plain</allowed_disk>
+        <allowed_disk>backup_disk_local_plain</allowed_disk>
     </backups>
 </clickhouse>
diff --git a/tests/integration/test_attach_backup_from_s3_plain/test.py b/tests/integration/test_attach_backup_from_s3_plain/test.py
index e575c487b7a..4a8da1e6d66 100644
--- a/tests/integration/test_attach_backup_from_s3_plain/test.py
+++ b/tests/integration/test_attach_backup_from_s3_plain/test.py
@@ -20,17 +20,27 @@ def start_cluster():
     finally:
         cluster.shutdown()
 
+s3_disk_def = """disk(type=s3_plain,
+    endpoint='http://minio1:9001/root/data/disks/disk_s3_plain/{backup_name}/',
+    access_key_id='minio',
+    secret_access_key='minio123');"""
+
+local_disk_def = "disk(type=object_storage, object_storage_type = 'local', metadata_type = 'plain'"
 
 @pytest.mark.parametrize(
-    "table_name,backup_name,storage_policy,min_bytes_for_wide_part",
+    "table_name,backup_name,storage_policy,disk_def,min_bytes_for_wide_part",
     [
         pytest.param(
-            "compact", "backup_compact", "s3_backup_compact", int(1e9), id="compact"
+            "compact", "backup_compact_s3", "backup_disk_s3_plain", s3_disk_def, int(1e9), id="compact"
         ),
-        pytest.param("wide", "backup_wide", "s3_backup_wide", int(0), id="wide"),
+        pytest.param("wide", "backup_wide_s3", "backup_disk_s3_plain", s3_disk_def, int(0), id="wide"),
+        pytest.param(
+            "compact", "backup_compact_local", "backup_disk_local_plain", local_disk_def, int(1e9), id="compact"
+        ),
+        pytest.param("wide", "backup_wide_local", "backup_disk_local_plain", local_disk_def, int(0), id="wide"),
     ],
 )
-def test_attach_part(table_name, backup_name, storage_policy, min_bytes_for_wide_part):
+def test_attach_part(table_name, backup_name, storage_policy, disk_def, min_bytes_for_wide_part):
     node.query(
         f"""
     -- Catch any errors (NOTE: warnings are ok)
@@ -45,7 +55,7 @@ def test_attach_part(table_name, backup_name, storage_policy, min_bytes_for_wide
     settings min_bytes_for_wide_part={min_bytes_for_wide_part}
     as select number%5 part, number key from numbers(100);
 
-    backup table ordinary_db.{table_name} TO Disk('backup_disk_s3_plain', '{backup_name}') settings deduplicate_files=0;
+    backup table ordinary_db.{table_name} TO Disk('{storage_policy}', '{backup_name}') settings deduplicate_files=0;
 
     drop table ordinary_db.{table_name};
     attach table ordinary_db.{table_name} (part UInt8, key UInt64)
@@ -53,10 +63,7 @@ def test_attach_part(table_name, backup_name, storage_policy, min_bytes_for_wide
     order by key partition by part
     settings
         max_suspicious_broken_parts=0,
-        disk=disk(type=s3_plain,
-            endpoint='http://minio1:9001/root/data/disks/disk_s3_plain/{backup_name}/',
-            access_key_id='minio',
-            secret_access_key='minio123');
+        disk={disk_def}
     """
     )
 

From 69b5bd02a915ae044b4116de759d11ae80525dc5 Mon Sep 17 00:00:00 2001
From: robot-clickhouse <robot-clickhouse@users.noreply.github.com>
Date: Mon, 26 Feb 2024 09:37:17 +0000
Subject: [PATCH 021/443] Automatic style fix

---
 .../test_attach_backup_from_s3_plain/test.py  | 42 ++++++++++++++++---
 1 file changed, 36 insertions(+), 6 deletions(-)

diff --git a/tests/integration/test_attach_backup_from_s3_plain/test.py b/tests/integration/test_attach_backup_from_s3_plain/test.py
index 4a8da1e6d66..900366b2c9c 100644
--- a/tests/integration/test_attach_backup_from_s3_plain/test.py
+++ b/tests/integration/test_attach_backup_from_s3_plain/test.py
@@ -20,27 +20,57 @@ def start_cluster():
     finally:
         cluster.shutdown()
 
+
 s3_disk_def = """disk(type=s3_plain,
     endpoint='http://minio1:9001/root/data/disks/disk_s3_plain/{backup_name}/',
     access_key_id='minio',
     secret_access_key='minio123');"""
 
-local_disk_def = "disk(type=object_storage, object_storage_type = 'local', metadata_type = 'plain'"
+local_disk_def = (
+    "disk(type=object_storage, object_storage_type = 'local', metadata_type = 'plain'"
+)
+
 
 @pytest.mark.parametrize(
     "table_name,backup_name,storage_policy,disk_def,min_bytes_for_wide_part",
     [
         pytest.param(
-            "compact", "backup_compact_s3", "backup_disk_s3_plain", s3_disk_def, int(1e9), id="compact"
+            "compact",
+            "backup_compact_s3",
+            "backup_disk_s3_plain",
+            s3_disk_def,
+            int(1e9),
+            id="compact",
         ),
-        pytest.param("wide", "backup_wide_s3", "backup_disk_s3_plain", s3_disk_def, int(0), id="wide"),
         pytest.param(
-            "compact", "backup_compact_local", "backup_disk_local_plain", local_disk_def, int(1e9), id="compact"
+            "wide",
+            "backup_wide_s3",
+            "backup_disk_s3_plain",
+            s3_disk_def,
+            int(0),
+            id="wide",
+        ),
+        pytest.param(
+            "compact",
+            "backup_compact_local",
+            "backup_disk_local_plain",
+            local_disk_def,
+            int(1e9),
+            id="compact",
+        ),
+        pytest.param(
+            "wide",
+            "backup_wide_local",
+            "backup_disk_local_plain",
+            local_disk_def,
+            int(0),
+            id="wide",
         ),
-        pytest.param("wide", "backup_wide_local", "backup_disk_local_plain", local_disk_def, int(0), id="wide"),
     ],
 )
-def test_attach_part(table_name, backup_name, storage_policy, disk_def, min_bytes_for_wide_part):
+def test_attach_part(
+    table_name, backup_name, storage_policy, disk_def, min_bytes_for_wide_part
+):
     node.query(
         f"""
     -- Catch any errors (NOTE: warnings are ok)

From ac4af6a4ad3b67860eae79b2ed3320fc5981a954 Mon Sep 17 00:00:00 2001
From: avogar <avogar@clickhouse.com>
Date: Mon, 26 Feb 2024 19:58:49 +0000
Subject: [PATCH 022/443] Don't allow to set max_parallel_replicas to 0 as it
 doesn't make sense

---
 src/Client/ConnectionPoolWithFailover.cpp                | 9 +++++++++
 src/Client/HedgedConnectionsFactory.cpp                  | 3 +++
 src/Client/HedgedConnectionsFactory.h                    | 2 +-
 src/Interpreters/InterpreterSelectQuery.cpp              | 4 ++--
 src/Planner/PlannerJoinTree.cpp                          | 4 ++--
 .../03001_max_parallel_replicas_zero_value.reference     | 0
 .../03001_max_parallel_replicas_zero_value.sql           | 5 +++++
 7 files changed, 22 insertions(+), 5 deletions(-)
 create mode 100644 tests/queries/0_stateless/03001_max_parallel_replicas_zero_value.reference
 create mode 100644 tests/queries/0_stateless/03001_max_parallel_replicas_zero_value.sql

diff --git a/src/Client/ConnectionPoolWithFailover.cpp b/src/Client/ConnectionPoolWithFailover.cpp
index 492fd4ae9e2..46b9741c812 100644
--- a/src/Client/ConnectionPoolWithFailover.cpp
+++ b/src/Client/ConnectionPoolWithFailover.cpp
@@ -191,11 +191,20 @@ std::vector<ConnectionPoolWithFailover::TryResult> ConnectionPoolWithFailover::g
         max_entries = nested_pools.size();
     }
     else if (pool_mode == PoolMode::GET_ONE)
+    {
         max_entries = 1;
+    }
     else if (pool_mode == PoolMode::GET_MANY)
+    {
+        if (settings.max_parallel_replicas == 0)
+            throw Exception(ErrorCodes::BAD_ARGUMENTS, "The value of the setting max_parallel_replicas must be greater than 0");
+
         max_entries = settings.max_parallel_replicas;
+    }
     else
+    {
         throw DB::Exception(DB::ErrorCodes::LOGICAL_ERROR, "Unknown pool allocation mode");
+    }
 
     if (!priority_func)
         priority_func = makeGetPriorityFunc(settings);
diff --git a/src/Client/HedgedConnectionsFactory.cpp b/src/Client/HedgedConnectionsFactory.cpp
index f5b074a0257..a4e5dbf04ac 100644
--- a/src/Client/HedgedConnectionsFactory.cpp
+++ b/src/Client/HedgedConnectionsFactory.cpp
@@ -82,6 +82,9 @@ std::vector<Connection *> HedgedConnectionsFactory::getManyConnections(PoolMode
         }
         case PoolMode::GET_MANY:
         {
+            if (max_parallel_replicas == 0)
+                throw Exception(ErrorCodes::BAD_ARGUMENTS, "The value of the setting max_parallel_replicas must be greater than 0");
+
             max_entries = max_parallel_replicas;
             break;
         }
diff --git a/src/Client/HedgedConnectionsFactory.h b/src/Client/HedgedConnectionsFactory.h
index ce7b553acdd..dd600d58e1e 100644
--- a/src/Client/HedgedConnectionsFactory.h
+++ b/src/Client/HedgedConnectionsFactory.h
@@ -158,7 +158,7 @@ private:
     /// checking the number of requested replicas that are still in process).
     size_t requested_connections_count = 0;
 
-    const size_t max_parallel_replicas = 0;
+    const size_t max_parallel_replicas = 1;
     const bool skip_unavailable_shards = 0;
 };
 
diff --git a/src/Interpreters/InterpreterSelectQuery.cpp b/src/Interpreters/InterpreterSelectQuery.cpp
index d34294b4c4b..fe5e5dc69d1 100644
--- a/src/Interpreters/InterpreterSelectQuery.cpp
+++ b/src/Interpreters/InterpreterSelectQuery.cpp
@@ -871,7 +871,7 @@ bool InterpreterSelectQuery::adjustParallelReplicasAfterAnalysis()
     {
         /// The query could use trivial count if it didn't use parallel replicas, so let's disable it and reanalyze
         context->setSetting("allow_experimental_parallel_reading_from_replicas", Field(0));
-        context->setSetting("max_parallel_replicas", UInt64{0});
+        context->setSetting("max_parallel_replicas", UInt64{1});
         LOG_DEBUG(log, "Disabling parallel replicas to be able to use a trivial count optimization");
         return true;
     }
@@ -909,7 +909,7 @@ bool InterpreterSelectQuery::adjustParallelReplicasAfterAnalysis()
     if (number_of_replicas_to_use <= 1)
     {
         context->setSetting("allow_experimental_parallel_reading_from_replicas", Field(0));
-        context->setSetting("max_parallel_replicas", UInt64{0});
+        context->setSetting("max_parallel_replicas", UInt64{1});
         LOG_DEBUG(log, "Disabling parallel replicas because there aren't enough rows to read");
         return true;
     }
diff --git a/src/Planner/PlannerJoinTree.cpp b/src/Planner/PlannerJoinTree.cpp
index e6a459d0e8a..2b1cd7fb353 100644
--- a/src/Planner/PlannerJoinTree.cpp
+++ b/src/Planner/PlannerJoinTree.cpp
@@ -295,7 +295,7 @@ bool applyTrivialCountIfPossible(
 
         /// The query could use trivial count if it didn't use parallel replicas, so let's disable it
         query_context->setSetting("allow_experimental_parallel_reading_from_replicas", Field(0));
-        query_context->setSetting("max_parallel_replicas", UInt64{0});
+        query_context->setSetting("max_parallel_replicas", UInt64{1});
         LOG_TRACE(getLogger("Planner"), "Disabling parallel replicas to be able to use a trivial count optimization");
 
     }
@@ -756,7 +756,7 @@ JoinTreeQueryPlan buildQueryPlanForTableExpression(QueryTreeNodePtr table_expres
                     {
                         planner_context->getMutableQueryContext()->setSetting(
                             "allow_experimental_parallel_reading_from_replicas", Field(0));
-                        planner_context->getMutableQueryContext()->setSetting("max_parallel_replicas", UInt64{0});
+                        planner_context->getMutableQueryContext()->setSetting("max_parallel_replicas", UInt64{1});
                         LOG_DEBUG(getLogger("Planner"), "Disabling parallel replicas because there aren't enough rows to read");
                     }
                     else if (number_of_replicas_to_use < settings.max_parallel_replicas)
diff --git a/tests/queries/0_stateless/03001_max_parallel_replicas_zero_value.reference b/tests/queries/0_stateless/03001_max_parallel_replicas_zero_value.reference
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/queries/0_stateless/03001_max_parallel_replicas_zero_value.sql b/tests/queries/0_stateless/03001_max_parallel_replicas_zero_value.sql
new file mode 100644
index 00000000000..611aa4777ba
--- /dev/null
+++ b/tests/queries/0_stateless/03001_max_parallel_replicas_zero_value.sql
@@ -0,0 +1,5 @@
+drop table if exists test_d;
+create table test_d engine=Distributed(test_cluster_two_shard_three_replicas_localhost, system, numbers);
+select * from test_d limit 10 settings max_parallel_replicas = 0, prefer_localhost_replica = 0; --{serverError BAD_ARGUMENTS}
+drop table test_d;
+

From 8aa9f36484bbe814a1e3edccc608e71b73915857 Mon Sep 17 00:00:00 2001
From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com>
Date: Mon, 26 Feb 2024 22:05:54 +0100
Subject: [PATCH 023/443] Fix style

---
 src/Client/ConnectionPoolWithFailover.cpp | 1 +
 1 file changed, 1 insertion(+)

diff --git a/src/Client/ConnectionPoolWithFailover.cpp b/src/Client/ConnectionPoolWithFailover.cpp
index 46b9741c812..ad8ed0067d8 100644
--- a/src/Client/ConnectionPoolWithFailover.cpp
+++ b/src/Client/ConnectionPoolWithFailover.cpp
@@ -21,6 +21,7 @@ namespace ErrorCodes
 {
     extern const int LOGICAL_ERROR;
     extern const int ALL_CONNECTION_TRIES_FAILED;
+    extern const int BAD_ARGUMENTS;
 }
 
 

From f264f0a0360baf1413ec38d3f3f30c70595064f4 Mon Sep 17 00:00:00 2001
From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com>
Date: Mon, 26 Feb 2024 22:06:10 +0100
Subject: [PATCH 024/443] Fix style

---
 src/Client/HedgedConnectionsFactory.cpp | 1 +
 1 file changed, 1 insertion(+)

diff --git a/src/Client/HedgedConnectionsFactory.cpp b/src/Client/HedgedConnectionsFactory.cpp
index a4e5dbf04ac..16a03a696bd 100644
--- a/src/Client/HedgedConnectionsFactory.cpp
+++ b/src/Client/HedgedConnectionsFactory.cpp
@@ -19,6 +19,7 @@ namespace ErrorCodes
     extern const int ALL_CONNECTION_TRIES_FAILED;
     extern const int ALL_REPLICAS_ARE_STALE;
     extern const int LOGICAL_ERROR;
+    extern const int BAD_ARGUMENTS;
 }
 
 HedgedConnectionsFactory::HedgedConnectionsFactory(

From f53f43b78d3cf2da6219ea4bdea7018d9811ae54 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Tue, 27 Feb 2024 17:33:47 +0800
Subject: [PATCH 025/443] Fixes for LocalObjectStorage and plain metadata

---
 .../Local/LocalObjectStorage.cpp              | 37 +++++++++++++++++--
 .../ObjectStorages/Local/LocalObjectStorage.h |  4 ++
 .../MetadataStorageFromPlainObjectStorage.cpp |  5 +--
 .../ObjectStorages/ObjectStorageFactory.cpp   | 31 ++++++++++------
 src/Disks/ObjectStorages/PlainObjectStorage.h |  6 +++
 src/Disks/ObjectStorages/S3/DiskS3Utils.cpp   |  6 ---
 src/Disks/ObjectStorages/S3/DiskS3Utils.h     |  1 -
 .../ObjectStorages/S3/S3ObjectStorage.cpp     |  2 +
 .../configs/disk_s3.xml                       |  4 +-
 .../test_attach_backup_from_s3_plain/test.py  |  7 ++--
 10 files changed, 71 insertions(+), 32 deletions(-)

diff --git a/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp b/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp
index 02700b358e0..51c260cc270 100644
--- a/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp
+++ b/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp
@@ -31,6 +31,8 @@ LocalObjectStorage::LocalObjectStorage(String key_prefix_)
         description = *block_device_id;
     else
         description = "/";
+
+    fs::create_directories(getCommonKeyPrefix());
 }
 
 bool LocalObjectStorage::exists(const StoredObject & object) const
@@ -53,6 +55,7 @@ std::unique_ptr<ReadBufferFromFileBase> LocalObjectStorage::readObjects( /// NOL
         return createReadBufferFromFileBase(file_path, modified_settings, read_hint, file_size);
     };
 
+    LOG_TEST(log, "Read object: {}", objects[0].remote_path);
     switch (read_settings.remote_fs_method)
     {
         case RemoteFSReadMethod::read:
@@ -111,8 +114,8 @@ std::unique_ptr<ReadBufferFromFileBase> LocalObjectStorage::readObject( /// NOLI
     if (!file_size)
         file_size = tryGetSizeFromFilePath(path);
 
-    LOG_TEST(log, "Read object: {}", path);
-    return createReadBufferFromFileBase(path, patchSettings(read_settings), read_hint, file_size);
+    LOG_TEST(log, "Read object: {}", object.remote_path);
+    return createReadBufferFromFileBase(object.remote_path, patchSettings(read_settings), read_hint, file_size);
 }
 
 std::unique_ptr<WriteBufferFromFileBase> LocalObjectStorage::writeObject( /// NOLINT
@@ -126,6 +129,7 @@ std::unique_ptr<WriteBufferFromFileBase> LocalObjectStorage::writeObject( /// NO
         throw Exception(ErrorCodes::BAD_ARGUMENTS, "LocalObjectStorage doesn't support append to files");
 
     LOG_TEST(log, "Write object: {}", object.remote_path);
+    fs::create_directories(fs::path(object.remote_path).parent_path());
     return std::make_unique<WriteBufferFromFile>(object.remote_path, buf_size);
 }
 
@@ -157,9 +161,34 @@ void LocalObjectStorage::removeObjectsIfExist(const StoredObjects & objects)
         removeObjectIfExists(object);
 }
 
-ObjectMetadata LocalObjectStorage::getObjectMetadata(const std::string & /* path */) const
+ObjectMetadata LocalObjectStorage::getObjectMetadata(const std::string & path) const
 {
-    throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Metadata is not supported for LocalObjectStorage");
+    ObjectMetadata object_metadata;
+    LOG_TEST(log, "Getting metadata for path: {}", path);
+    object_metadata.size_bytes = fs::file_size(path);
+    object_metadata.last_modified = Poco::Timestamp::fromEpochTime(
+        std::chrono::duration_cast<std::chrono::seconds>(fs::last_write_time(path).time_since_epoch()).count());
+    return object_metadata;
+}
+
+void LocalObjectStorage::listObjects(const std::string & path, RelativePathsWithMetadata & children, int /* max_keys */) const
+{
+    for (const auto & entry : fs::directory_iterator(path))
+    {
+        if (entry.is_directory())
+        {
+            listObjects(entry.path(), children, 0);
+            continue;
+        }
+
+        auto metadata = getObjectMetadata(entry.path());
+        children.emplace_back(entry.path(), std::move(metadata));
+    }
+}
+
+bool LocalObjectStorage::existsOrHasAnyChild(const std::string & path) const
+{
+    return exists(StoredObject(path));
 }
 
 void LocalObjectStorage::copyObject( // NOLINT
diff --git a/src/Disks/ObjectStorages/Local/LocalObjectStorage.h b/src/Disks/ObjectStorages/Local/LocalObjectStorage.h
index ed5f8c1f537..22429a99c76 100644
--- a/src/Disks/ObjectStorages/Local/LocalObjectStorage.h
+++ b/src/Disks/ObjectStorages/Local/LocalObjectStorage.h
@@ -58,6 +58,10 @@ public:
 
     ObjectMetadata getObjectMetadata(const std::string & path) const override;
 
+    void listObjects(const std::string & path, RelativePathsWithMetadata & children, int max_keys) const override;
+
+    bool existsOrHasAnyChild(const std::string & path) const override;
+
     void copyObject( /// NOLINT
         const StoredObject & object_from,
         const StoredObject & object_to,
diff --git a/src/Disks/ObjectStorages/MetadataStorageFromPlainObjectStorage.cpp b/src/Disks/ObjectStorages/MetadataStorageFromPlainObjectStorage.cpp
index b03809f5b39..4b8fc74e956 100644
--- a/src/Disks/ObjectStorages/MetadataStorageFromPlainObjectStorage.cpp
+++ b/src/Disks/ObjectStorages/MetadataStorageFromPlainObjectStorage.cpp
@@ -48,10 +48,7 @@ bool MetadataStorageFromPlainObjectStorage::isDirectory(const std::string & path
     std::string directory = object_key.serialize();
     if (!directory.ends_with('/'))
         directory += '/';
-
-    RelativePathsWithMetadata files;
-    object_storage->listObjects(directory, files, 1);
-    return !files.empty();
+    return object_storage->existsOrHasAnyChild(directory);
 }
 
 uint64_t MetadataStorageFromPlainObjectStorage::getFileSize(const String & path) const
diff --git a/src/Disks/ObjectStorages/ObjectStorageFactory.cpp b/src/Disks/ObjectStorages/ObjectStorageFactory.cpp
index 6f6ff199902..f64c42c1403 100644
--- a/src/Disks/ObjectStorages/ObjectStorageFactory.cpp
+++ b/src/Disks/ObjectStorages/ObjectStorageFactory.cpp
@@ -36,16 +36,24 @@ namespace ErrorCodes
 
 namespace
 {
+    bool isPlainStorage(
+        ObjectStorageType type,
+        const Poco::Util::AbstractConfiguration & config,
+        const std::string & config_prefix)
+    {
+        auto compatibility_hint = MetadataStorageFactory::getCompatibilityMetadataTypeHint(type);
+        auto metadata_type = MetadataStorageFactory::getMetadataType(config, config_prefix, compatibility_hint);
+        return metadataTypeFromString(metadata_type) == MetadataStorageType::Plain;
+    }
+
     template <typename BaseObjectStorage, class ...Args>
     ObjectStoragePtr createObjectStorage(
+        ObjectStorageType type,
         const Poco::Util::AbstractConfiguration & config,
         const std::string & config_prefix,
         Args && ...args)
     {
-        auto compatibility_hint = MetadataStorageFactory::getCompatibilityMetadataTypeHint(ObjectStorageType::S3);
-        auto metadata_type = MetadataStorageFactory::getMetadataType(config, config_prefix, compatibility_hint);
-
-        if (metadataTypeFromString(metadata_type) == MetadataStorageType::Plain)
+        if (isPlainStorage(type, config, config_prefix))
         {
             return std::make_shared<PlainObjectStorage<BaseObjectStorage>>(std::forward<Args>(args)...);
         }
@@ -151,10 +159,10 @@ void registerS3ObjectStorage(ObjectStorageFactory & factory)
         auto s3_capabilities = getCapabilitiesFromConfig(config, config_prefix);
         auto settings = getSettings(config, config_prefix, context);
         auto client = getClient(config, config_prefix, context, *settings);
-        auto key_generator = getKeyGenerator(disk_type, uri, config, config_prefix);
+        auto key_generator = getKeyGenerator(uri, config, config_prefix);
 
         auto object_storage = createObjectStorage<S3ObjectStorage>(
-            config, config_prefix, std::move(client), std::move(settings), uri, s3_capabilities, key_generator, name);
+            ObjectStorageType::S3, config, config_prefix, std::move(client), std::move(settings), uri, s3_capabilities, key_generator, name);
 
         /// NOTE: should we still perform this check for clickhouse-disks?
         if (!skip_access_check)
@@ -187,7 +195,7 @@ void registerS3PlainObjectStorage(ObjectStorageFactory & factory)
         auto s3_capabilities = getCapabilitiesFromConfig(config, config_prefix);
         auto settings = getSettings(config, config_prefix, context);
         auto client = getClient(config, config_prefix, context, *settings);
-        auto key_generator = getKeyGenerator(disk_type, uri, config, config_prefix);
+        auto key_generator = getKeyGenerator(uri, config, config_prefix);
 
         auto object_storage = std::make_shared<PlainObjectStorage<S3ObjectStorage>>(
             std::move(client), std::move(settings), uri, s3_capabilities, key_generator, name);
@@ -222,7 +230,7 @@ void registerHDFSObjectStorage(ObjectStorageFactory & factory)
             context->getSettingsRef().hdfs_replication
         );
 
-        return createObjectStorage<HDFSObjectStorage>(config, config_prefix, uri, std::move(settings), config);
+        return createObjectStorage<HDFSObjectStorage>(ObjectStorageType::HDFS, config, config_prefix, uri, std::move(settings), config);
     });
 }
 #endif
@@ -239,8 +247,7 @@ void registerAzureObjectStorage(ObjectStorageFactory & factory)
     {
         String container_name = config.getString(config_prefix + ".container_name", "default-container");
         return createObjectStorage<AzureObjectStorage>(
-            config, config_prefix,
-            name,
+            ObjectStorageType::Azure, config, config_prefix, name,
             getAzureBlobContainerClient(config, config_prefix),
             getAzureBlobStorageSettings(config, config_prefix, context),
             container_name);
@@ -273,7 +280,7 @@ void registerWebObjectStorage(ObjectStorageFactory & factory)
                 ErrorCodes::BAD_ARGUMENTS, "Bad URI: `{}`. Error: {}", uri, e.what());
         }
 
-        return createObjectStorage<WebObjectStorage>(config, config_prefix, uri, context);
+        return createObjectStorage<WebObjectStorage>(ObjectStorageType::Web, config, config_prefix, uri, context);
     });
 }
 
@@ -291,7 +298,7 @@ void registerLocalObjectStorage(ObjectStorageFactory & factory)
         loadDiskLocalConfig(name, config, config_prefix, context, object_key_prefix, keep_free_space_bytes);
         /// keys are mapped to the fs, object_key_prefix is a directory also
         fs::create_directories(object_key_prefix);
-        return createObjectStorage<LocalObjectStorage>(config, config_prefix, object_key_prefix);
+        return createObjectStorage<LocalObjectStorage>(ObjectStorageType::Local, config, config_prefix, object_key_prefix);
     });
 }
 #endif
diff --git a/src/Disks/ObjectStorages/PlainObjectStorage.h b/src/Disks/ObjectStorages/PlainObjectStorage.h
index 3a81b85c44b..e0907d0b4d8 100644
--- a/src/Disks/ObjectStorages/PlainObjectStorage.h
+++ b/src/Disks/ObjectStorages/PlainObjectStorage.h
@@ -1,5 +1,6 @@
 #pragma once
 #include <Disks/ObjectStorages/IObjectStorage.h>
+#include <Common/ObjectStorageKeyGenerator.h>
 
 namespace DB
 {
@@ -24,6 +25,11 @@ public:
     bool isWriteOnce() const override { return true; }
 
     bool isPlain() const override { return true; }
+
+    ObjectStorageKey generateObjectKeyForPath(const std::string & path) const override
+    {
+        return ObjectStorageKey::createAsRelative(BaseObjectStorage::getCommonKeyPrefix(), path);
+    }
 };
 
 }
diff --git a/src/Disks/ObjectStorages/S3/DiskS3Utils.cpp b/src/Disks/ObjectStorages/S3/DiskS3Utils.cpp
index bb7b53b2d22..4b889f89f90 100644
--- a/src/Disks/ObjectStorages/S3/DiskS3Utils.cpp
+++ b/src/Disks/ObjectStorages/S3/DiskS3Utils.cpp
@@ -15,16 +15,10 @@ namespace ErrorCodes
 }
 
 ObjectStorageKeysGeneratorPtr getKeyGenerator(
-    String type,
     const S3::URI & uri,
     const Poco::Util::AbstractConfiguration & config,
     const String & config_prefix)
 {
-    if (type == "s3_plain")
-        return createObjectStorageKeysGeneratorAsIsWithPrefix(uri.key);
-
-    chassert(type == "s3");
-
     bool storage_metadata_write_full_object_key = DiskObjectStorageMetadata::getWriteFullObjectKeySetting();
     bool send_metadata = config.getBool(config_prefix + ".send_metadata", false);
 
diff --git a/src/Disks/ObjectStorages/S3/DiskS3Utils.h b/src/Disks/ObjectStorages/S3/DiskS3Utils.h
index 29e39d4bc1b..8524a9ccac3 100644
--- a/src/Disks/ObjectStorages/S3/DiskS3Utils.h
+++ b/src/Disks/ObjectStorages/S3/DiskS3Utils.h
@@ -12,7 +12,6 @@ namespace DB
 namespace S3 { struct URI; }
 
 ObjectStorageKeysGeneratorPtr getKeyGenerator(
-    String type,
     const S3::URI & uri,
     const Poco::Util::AbstractConfiguration & config,
     const String & config_prefix);
diff --git a/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp b/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp
index 5771eb1ebe0..b2a9ab8fdc3 100644
--- a/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp
+++ b/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp
@@ -561,6 +561,8 @@ std::unique_ptr<IObjectStorage> S3ObjectStorage::cloneObjectStorage(
 
 ObjectStorageKey S3ObjectStorage::generateObjectKeyForPath(const std::string & path) const
 {
+    if (!key_generator)
+        throw Exception(ErrorCodes::LOGICAL_ERROR, "Key generator is not set");
     return key_generator->generate(path);
 }
 
diff --git a/tests/integration/test_attach_backup_from_s3_plain/configs/disk_s3.xml b/tests/integration/test_attach_backup_from_s3_plain/configs/disk_s3.xml
index 3166eea7ccb..2edabc76c8b 100644
--- a/tests/integration/test_attach_backup_from_s3_plain/configs/disk_s3.xml
+++ b/tests/integration/test_attach_backup_from_s3_plain/configs/disk_s3.xml
@@ -10,9 +10,9 @@
             </backup_disk_s3_plain>
             <backup_disk_local_plain>
                 <type>object_storage</type>
-                <object_storage_type>local</object_storage_type>
+                <object_storage_type>local_blob_storage</object_storage_type>
                 <metadata_type>plain</metadata_type>
-                <path>local_plain/</path>
+                <path>/local_plain/</path>
             </backup_disk_local_plain>
         </disks>
     </storage_configuration>
diff --git a/tests/integration/test_attach_backup_from_s3_plain/test.py b/tests/integration/test_attach_backup_from_s3_plain/test.py
index 4a8da1e6d66..983275cc24f 100644
--- a/tests/integration/test_attach_backup_from_s3_plain/test.py
+++ b/tests/integration/test_attach_backup_from_s3_plain/test.py
@@ -21,11 +21,11 @@ def start_cluster():
         cluster.shutdown()
 
 s3_disk_def = """disk(type=s3_plain,
-    endpoint='http://minio1:9001/root/data/disks/disk_s3_plain/{backup_name}/',
+    endpoint='http://minio1:9001/root/data/disks/disk_s3_plain/{}/',
     access_key_id='minio',
     secret_access_key='minio123');"""
 
-local_disk_def = "disk(type=object_storage, object_storage_type = 'local', metadata_type = 'plain'"
+local_disk_def = "disk(type=object_storage, object_storage_type = 'local_blob_storage', metadata_type = 'plain', path = '/local_plain/{}/')"
 
 @pytest.mark.parametrize(
     "table_name,backup_name,storage_policy,disk_def,min_bytes_for_wide_part",
@@ -41,6 +41,7 @@ local_disk_def = "disk(type=object_storage, object_storage_type = 'local', metad
     ],
 )
 def test_attach_part(table_name, backup_name, storage_policy, disk_def, min_bytes_for_wide_part):
+    disk_definition = disk_def.format(backup_name)
     node.query(
         f"""
     -- Catch any errors (NOTE: warnings are ok)
@@ -63,7 +64,7 @@ def test_attach_part(table_name, backup_name, storage_policy, disk_def, min_byte
     order by key partition by part
     settings
         max_suspicious_broken_parts=0,
-        disk={disk_def}
+        disk={disk_definition}
     """
     )
 

From fb38bd139c433ead685028f232e8c4fad5e566d2 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Tue, 27 Feb 2024 17:38:02 +0800
Subject: [PATCH 026/443] Remove debug logging

---
 src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp b/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp
index 51c260cc270..4ec998a2bb0 100644
--- a/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp
+++ b/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp
@@ -55,7 +55,6 @@ std::unique_ptr<ReadBufferFromFileBase> LocalObjectStorage::readObjects( /// NOL
         return createReadBufferFromFileBase(file_path, modified_settings, read_hint, file_size);
     };
 
-    LOG_TEST(log, "Read object: {}", objects[0].remote_path);
     switch (read_settings.remote_fs_method)
     {
         case RemoteFSReadMethod::read:
@@ -109,10 +108,8 @@ std::unique_ptr<ReadBufferFromFileBase> LocalObjectStorage::readObject( /// NOLI
     std::optional<size_t> read_hint,
     std::optional<size_t> file_size) const
 {
-    const auto & path = object.remote_path;
-
     if (!file_size)
-        file_size = tryGetSizeFromFilePath(path);
+        file_size = tryGetSizeFromFilePath(object.remote_path);
 
     LOG_TEST(log, "Read object: {}", object.remote_path);
     return createReadBufferFromFileBase(object.remote_path, patchSettings(read_settings), read_hint, file_size);

From 978fe9fa1a069a231bb52c66b3898c6ce112a215 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Tue, 27 Feb 2024 17:43:34 +0800
Subject: [PATCH 027/443] Add comments

---
 src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp b/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp
index 4ec998a2bb0..7f34ca48f7f 100644
--- a/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp
+++ b/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp
@@ -126,7 +126,11 @@ std::unique_ptr<WriteBufferFromFileBase> LocalObjectStorage::writeObject( /// NO
         throw Exception(ErrorCodes::BAD_ARGUMENTS, "LocalObjectStorage doesn't support append to files");
 
     LOG_TEST(log, "Write object: {}", object.remote_path);
+
+    /// Unlike real blob storage, in local fs we cannot create a file with non-existing prefix.
+    /// So let's create it.
     fs::create_directories(fs::path(object.remote_path).parent_path());
+
     return std::make_unique<WriteBufferFromFile>(object.remote_path, buf_size);
 }
 
@@ -185,6 +189,8 @@ void LocalObjectStorage::listObjects(const std::string & path, RelativePathsWith
 
 bool LocalObjectStorage::existsOrHasAnyChild(const std::string & path) const
 {
+    /// Unlike real object storage, existance of a prefix path can be checked by
+    /// just checking existence of this prefix directly, so simple exists is enough here.
     return exists(StoredObject(path));
 }
 

From 33788250b1f74384661cd241e2badef82c8fdbf6 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Tue, 27 Feb 2024 18:07:19 +0800
Subject: [PATCH 028/443] Update test.py

---
 tests/integration/test_attach_backup_from_s3_plain/test.py | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/tests/integration/test_attach_backup_from_s3_plain/test.py b/tests/integration/test_attach_backup_from_s3_plain/test.py
index 3a0fa70a715..c2f8936b82c 100644
--- a/tests/integration/test_attach_backup_from_s3_plain/test.py
+++ b/tests/integration/test_attach_backup_from_s3_plain/test.py
@@ -26,9 +26,8 @@ s3_disk_def = """disk(type=s3_plain,
     access_key_id='minio',
     secret_access_key='minio123');"""
 
-local_disk_def = (
-    "disk(type=object_storage, object_storage_type = 'local', metadata_type = 'plain', path = '/local_plain/{}/'"
-)
+local_disk_def = "disk(type=object_storage, object_storage_type = 'local_blob_storage', metadata_type = 'plain', path = '/local_plain/{}/');"
+
 
 @pytest.mark.parametrize(
     "table_name,backup_name,storage_policy,disk_def,min_bytes_for_wide_part",
@@ -67,7 +66,6 @@ local_disk_def = (
         ),
     ],
 )
-
 def test_attach_part(
     table_name, backup_name, storage_policy, disk_def, min_bytes_for_wide_part
 ):

From 58a53b42acb3b25a41e8529186db9df0d4387f77 Mon Sep 17 00:00:00 2001
From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com>
Date: Tue, 27 Feb 2024 14:31:35 +0100
Subject: [PATCH 029/443] Set max_entries to min(max_parallel_replicas, all
 available reolicas)

---
 src/Client/HedgedConnectionsFactory.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Client/HedgedConnectionsFactory.cpp b/src/Client/HedgedConnectionsFactory.cpp
index 16a03a696bd..703cc1f8821 100644
--- a/src/Client/HedgedConnectionsFactory.cpp
+++ b/src/Client/HedgedConnectionsFactory.cpp
@@ -86,7 +86,7 @@ std::vector<Connection *> HedgedConnectionsFactory::getManyConnections(PoolMode
             if (max_parallel_replicas == 0)
                 throw Exception(ErrorCodes::BAD_ARGUMENTS, "The value of the setting max_parallel_replicas must be greater than 0");
 
-            max_entries = max_parallel_replicas;
+            max_entries = std::min(max_parallel_replicas, shuffled_pools.size());
             break;
         }
     }

From 98b27fd45fbe1109442c2313181ca4e8435e2024 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Tue, 27 Feb 2024 23:00:27 +0800
Subject: [PATCH 030/443] Fix style check

---
 src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp | 2 +-
 src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp       | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp b/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp
index 7f34ca48f7f..eba57969580 100644
--- a/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp
+++ b/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp
@@ -189,7 +189,7 @@ void LocalObjectStorage::listObjects(const std::string & path, RelativePathsWith
 
 bool LocalObjectStorage::existsOrHasAnyChild(const std::string & path) const
 {
-    /// Unlike real object storage, existance of a prefix path can be checked by
+    /// Unlike real object storage, existence of a prefix path can be checked by
     /// just checking existence of this prefix directly, so simple exists is enough here.
     return exists(StoredObject(path));
 }
diff --git a/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp b/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp
index b2a9ab8fdc3..eec3a5914fc 100644
--- a/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp
+++ b/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp
@@ -48,6 +48,7 @@ namespace ErrorCodes
 {
     extern const int S3_ERROR;
     extern const int BAD_ARGUMENTS;
+    extern const int LOGICAL_ERROR;
 }
 
 namespace

From 416638461fe832673252445d8fabb3fe554eed49 Mon Sep 17 00:00:00 2001
From: avogar <avogar@clickhouse.com>
Date: Tue, 27 Feb 2024 15:02:13 +0000
Subject: [PATCH 031/443] Try to fix logical error 'Cannot capture column
 because it has incompatible type' in mapContainsKeyLike

---
 src/Functions/array/FunctionArrayMapped.h                     | 4 ++--
 .../03002_map_array_functions_with_low_cardinality.reference  | 1 +
 .../03002_map_array_functions_with_low_cardinality.sql        | 2 ++
 3 files changed, 5 insertions(+), 2 deletions(-)
 create mode 100644 tests/queries/0_stateless/03002_map_array_functions_with_low_cardinality.reference
 create mode 100644 tests/queries/0_stateless/03002_map_array_functions_with_low_cardinality.sql

diff --git a/src/Functions/array/FunctionArrayMapped.h b/src/Functions/array/FunctionArrayMapped.h
index 49ed9d495e2..136d3481771 100644
--- a/src/Functions/array/FunctionArrayMapped.h
+++ b/src/Functions/array/FunctionArrayMapped.h
@@ -355,7 +355,7 @@ public:
                     {
                         arrays.emplace_back(
                             column_tuple->getColumnPtr(j),
-                            recursiveRemoveLowCardinality(type_tuple.getElement(j)),
+                            type_tuple.getElement(j),
                             array_with_type_and_name.name + "." + tuple_names[j]);
                     }
                 }
@@ -363,7 +363,7 @@ public:
                 {
                     arrays.emplace_back(
                         column_array->getDataPtr(),
-                        recursiveRemoveLowCardinality(array_type->getNestedType()),
+                        array_type->getNestedType(),
                         array_with_type_and_name.name);
                 }
 
diff --git a/tests/queries/0_stateless/03002_map_array_functions_with_low_cardinality.reference b/tests/queries/0_stateless/03002_map_array_functions_with_low_cardinality.reference
new file mode 100644
index 00000000000..d00491fd7e5
--- /dev/null
+++ b/tests/queries/0_stateless/03002_map_array_functions_with_low_cardinality.reference
@@ -0,0 +1 @@
+1
diff --git a/tests/queries/0_stateless/03002_map_array_functions_with_low_cardinality.sql b/tests/queries/0_stateless/03002_map_array_functions_with_low_cardinality.sql
new file mode 100644
index 00000000000..8240a8f93f5
--- /dev/null
+++ b/tests/queries/0_stateless/03002_map_array_functions_with_low_cardinality.sql
@@ -0,0 +1,2 @@
+SELECT mapContainsKeyLike(map('aa', toLowCardinality(1), 'bb', toLowCardinality(2)), toLowCardinality('a%'));
+

From 5771e739f0e65baae69f1e7abd42495d5fbc5488 Mon Sep 17 00:00:00 2001
From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com>
Date: Tue, 27 Feb 2024 23:11:29 +0800
Subject: [PATCH 032/443] Update ReadSettings.h

---
 src/IO/ReadSettings.h | 1 -
 1 file changed, 1 deletion(-)

diff --git a/src/IO/ReadSettings.h b/src/IO/ReadSettings.h
index 2c79735317d..846fcd668f0 100644
--- a/src/IO/ReadSettings.h
+++ b/src/IO/ReadSettings.h
@@ -98,7 +98,6 @@ struct ReadSettings
     bool enable_filesystem_cache = true;
     bool read_from_filesystem_cache_if_exists_otherwise_bypass_cache = false;
     bool enable_filesystem_cache_log = false;
-    /// Don't populate cache when the read is not part of query execution (e.g. background thread).
     bool force_read_through_cache_merges = false;
     size_t filesystem_cache_segments_batch_size = 20;
 

From 1eba06dc113881b2845d36a7d3a4703ad64659d7 Mon Sep 17 00:00:00 2001
From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com>
Date: Tue, 27 Feb 2024 23:12:41 +0800
Subject: [PATCH 033/443] Update
 02241_filesystem_cache_on_write_operations.reference

---
 .../02241_filesystem_cache_on_write_operations.reference    | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.reference b/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.reference
index c03b928684b..53566a18edc 100644
--- a/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.reference
+++ b/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.reference
@@ -95,13 +95,7 @@ INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(300, 10000)
 SELECT count(), sum(size) FROM system.filesystem_cache
 24	84045
 SYSTEM START MERGES test_02241
-SELECT value FROM system.events WHERE name = 'CachedWriteBufferCacheWriteBytes'
-85146
-SELECT value FROM system.events WHERE name = 'CachedReadBufferCacheWriteBytes'
 OPTIMIZE TABLE test_02241 FINAL
-SELECT value FROM system.events WHERE name = 'CachedWriteBufferCacheWriteBytes'
-251542
-SELECT value FROM system.events WHERE name = 'CachedReadBufferCacheWriteBytes'
 SELECT count(), sum(size) FROM system.filesystem_cache
 32	167243
 ALTER TABLE test_02241 UPDATE value = 'kek' WHERE key = 100

From ffd69e0e127f64cf90a41d7b710c375ced13f092 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Tue, 27 Feb 2024 23:22:04 +0800
Subject: [PATCH 034/443] Move setting to merge-tree level

---
 src/Core/ServerSettings.h                                     | 3 ---
 src/Interpreters/Context.cpp                                  | 1 -
 src/Storages/MergeTree/MergeTreeSequentialSource.cpp          | 3 ++-
 src/Storages/MergeTree/MergeTreeSettings.h                    | 1 +
 .../config.d/force_read_through_cache_for_merges.xml          | 4 +++-
 5 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/src/Core/ServerSettings.h b/src/Core/ServerSettings.h
index 0283b98638f..0063b3a2bd6 100644
--- a/src/Core/ServerSettings.h
+++ b/src/Core/ServerSettings.h
@@ -102,9 +102,6 @@ namespace DB
     M(UInt64, tables_loader_background_pool_size, 0, "The maximum number of threads that will be used for background async loading of tables. Zero means use all CPUs.", 0) \
     M(Bool, async_load_databases, false, "Enable asynchronous loading of databases and tables to speedup server startup. Queries to not yet loaded entity will be blocked until load is finished.", 0) \
     M(Bool, display_secrets_in_show_and_select, false, "Allow showing secrets in SHOW and SELECT queries via a format setting and a grant", 0) \
-    \
-    M(Bool, force_read_through_cache_for_merges, false, "Force read-through filesystem cache for merges", 0) \
-    \
     M(Seconds, keep_alive_timeout, DEFAULT_HTTP_KEEP_ALIVE_TIMEOUT, "The number of seconds that ClickHouse waits for incoming requests before closing the connection.", 0) \
     M(Seconds, replicated_fetches_http_connection_timeout, 0, "HTTP connection timeout for part fetch requests. Inherited from default profile `http_connection_timeout` if not set explicitly.", 0) \
     M(Seconds, replicated_fetches_http_send_timeout, 0, "HTTP send timeout for part fetch requests. Inherited from default profile `http_send_timeout` if not set explicitly.", 0) \
diff --git a/src/Interpreters/Context.cpp b/src/Interpreters/Context.cpp
index a974eaca067..55a4df10206 100644
--- a/src/Interpreters/Context.cpp
+++ b/src/Interpreters/Context.cpp
@@ -5079,7 +5079,6 @@ ReadSettings Context::getReadSettings() const
     res.read_from_filesystem_cache_if_exists_otherwise_bypass_cache = settings.read_from_filesystem_cache_if_exists_otherwise_bypass_cache;
     res.enable_filesystem_cache_log = settings.enable_filesystem_cache_log;
     res.filesystem_cache_segments_batch_size = settings.filesystem_cache_segments_batch_size;
-    res.force_read_through_cache_merges = getServerSettings().force_read_through_cache_for_merges;
 
     res.filesystem_cache_max_download_size = settings.filesystem_cache_max_download_size;
     res.skip_download_if_exceeds_query_cache = settings.skip_download_if_exceeds_query_cache;
diff --git a/src/Storages/MergeTree/MergeTreeSequentialSource.cpp b/src/Storages/MergeTree/MergeTreeSequentialSource.cpp
index e375e8b0a9f..6b0c5ccb59a 100644
--- a/src/Storages/MergeTree/MergeTreeSequentialSource.cpp
+++ b/src/Storages/MergeTree/MergeTreeSequentialSource.cpp
@@ -151,7 +151,8 @@ MergeTreeSequentialSource::MergeTreeSequentialSource(
 
     const auto & context = storage.getContext();
     ReadSettings read_settings = context->getReadSettings();
-    read_settings.read_from_filesystem_cache_if_exists_otherwise_bypass_cache = !read_settings.force_read_through_cache_merges;
+    read_settings.read_from_filesystem_cache_if_exists_otherwise_bypass_cache = !storage.getSettings()->force_read_through_cache_for_merges;
+
     /// It does not make sense to use pthread_threadpool for background merges/mutations
     /// And also to preserve backward compatibility
     read_settings.local_fs_method = LocalFSReadMethod::pread;
diff --git a/src/Storages/MergeTree/MergeTreeSettings.h b/src/Storages/MergeTree/MergeTreeSettings.h
index b64632b6139..9cb74e76dd5 100644
--- a/src/Storages/MergeTree/MergeTreeSettings.h
+++ b/src/Storages/MergeTree/MergeTreeSettings.h
@@ -191,6 +191,7 @@ struct Settings;
     M(String, remote_fs_zero_copy_zookeeper_path, "/clickhouse/zero_copy", "ZooKeeper path for zero-copy table-independent info.", 0) \
     M(Bool, remote_fs_zero_copy_path_compatible_mode, false, "Run zero-copy in compatible mode during conversion process.", 0) \
     M(Bool, cache_populated_by_fetch, false, "Only available in ClickHouse Cloud", 0) \
+    M(Bool, force_read_through_cache_for_merges, false, "Force read-through filesystem cache for merges", 0) \
     M(Bool, allow_experimental_block_number_column, false, "Enable persisting column _block_number for each row.", 0) \
     M(Bool, allow_experimental_replacing_merge_with_cleanup, false, "Allow experimental CLEANUP merges for ReplacingMergeTree with is_deleted column.", 0) \
     \
diff --git a/tests/integration/test_filesystem_cache/config.d/force_read_through_cache_for_merges.xml b/tests/integration/test_filesystem_cache/config.d/force_read_through_cache_for_merges.xml
index bb2a6e850a4..23d3fdea800 100644
--- a/tests/integration/test_filesystem_cache/config.d/force_read_through_cache_for_merges.xml
+++ b/tests/integration/test_filesystem_cache/config.d/force_read_through_cache_for_merges.xml
@@ -1,3 +1,5 @@
 <clickhouse>
-    <force_read_through_cache_for_merges>1</force_read_through_cache_for_merges>
+    <merge_tree>
+        <force_read_through_cache_for_merges>1</force_read_through_cache_for_merges>
+    </merge_tree>
 </clickhouse>

From cb8390e9c8672bcdead0108be75021d6c6f21331 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Wed, 28 Feb 2024 13:32:43 +0800
Subject: [PATCH 035/443] Fix build

---
 src/Disks/ObjectStorages/ObjectStorageFactory.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Disks/ObjectStorages/ObjectStorageFactory.cpp b/src/Disks/ObjectStorages/ObjectStorageFactory.cpp
index f64c42c1403..d0c2c9ac4f4 100644
--- a/src/Disks/ObjectStorages/ObjectStorageFactory.cpp
+++ b/src/Disks/ObjectStorages/ObjectStorageFactory.cpp
@@ -16,10 +16,10 @@
 #ifndef CLICKHOUSE_KEEPER_STANDALONE_BUILD
 #include <Disks/ObjectStorages/Web/WebObjectStorage.h>
 #include <Disks/ObjectStorages/Local/LocalObjectStorage.h>
-#include <Disks/ObjectStorages/PlainObjectStorage.h>
 #include <Disks/loadLocalDiskConfig.h>
 #endif
 #include <Disks/ObjectStorages/MetadataStorageFactory.h>
+#include <Disks/ObjectStorages/PlainObjectStorage.h>
 #include <Interpreters/Context.h>
 #include <Common/Macros.h>
 

From d2ea882bd8105f5d2e173a6670bf23b2917b3190 Mon Sep 17 00:00:00 2001
From: avogar <avogar@clickhouse.com>
Date: Wed, 28 Feb 2024 21:26:19 +0000
Subject: [PATCH 036/443] Fix deadlock in parallel parsing when lots of rows
 are skipped due to errors

---
 .../Formats/Impl/ParallelParsingInputFormat.cpp      |  4 +++-
 .../03001_parallel_parsing_deadlock.reference        |  0
 .../0_stateless/03001_parallel_parsing_deadlock.sh   | 12 ++++++++++++
 3 files changed, 15 insertions(+), 1 deletion(-)
 create mode 100644 tests/queries/0_stateless/03001_parallel_parsing_deadlock.reference
 create mode 100755 tests/queries/0_stateless/03001_parallel_parsing_deadlock.sh

diff --git a/src/Processors/Formats/Impl/ParallelParsingInputFormat.cpp b/src/Processors/Formats/Impl/ParallelParsingInputFormat.cpp
index 8b6969bbfcc..447adb1ed48 100644
--- a/src/Processors/Formats/Impl/ParallelParsingInputFormat.cpp
+++ b/src/Processors/Formats/Impl/ParallelParsingInputFormat.cpp
@@ -224,7 +224,9 @@ Chunk ParallelParsingInputFormat::read()
             /// skipped all rows. For example, it can happen while using settings
             /// input_format_allow_errors_num/input_format_allow_errors_ratio
             /// and this segment contained only rows with errors.
-            /// Process the next unit.
+            /// Return this empty unit back to segmentator and process the next unit.
+            unit->status = READY_TO_INSERT;
+            segmentator_condvar.notify_all();
             ++reader_ticket_number;
             unit = &processing_units[reader_ticket_number % processing_units.size()];
         }
diff --git a/tests/queries/0_stateless/03001_parallel_parsing_deadlock.reference b/tests/queries/0_stateless/03001_parallel_parsing_deadlock.reference
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/queries/0_stateless/03001_parallel_parsing_deadlock.sh b/tests/queries/0_stateless/03001_parallel_parsing_deadlock.sh
new file mode 100755
index 00000000000..1bf21dfc53b
--- /dev/null
+++ b/tests/queries/0_stateless/03001_parallel_parsing_deadlock.sh
@@ -0,0 +1,12 @@
+#!/usr/bin/env bash
+# Tags: no-fasttest, no-cpu-aarch64
+
+CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
+# shellcheck source=../shell_config.sh
+. "$CURDIR"/../shell_config.sh
+
+DATA_FILE=$CLICKHOUSE_TEST_UNIQUE_NAME.csv
+$CLICKHOUSE_LOCAL -q "select number > 1000000 ? 'error' : toString(number) from numbers(2000000) format CSV" > $DATA_FILE
+$CLICKHOUSE_LOCAL -q "select * from file($DATA_FILE, CSV, 'x UInt64') format Null settings input_format_allow_errors_ratio=1"
+rm $DATA_FILE
+

From 6fbd298b3d7cc06b1f11727263a25bc613f7c295 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Thu, 29 Feb 2024 05:03:09 +0300
Subject: [PATCH 037/443] Revert "Revert "Use `MergeTree` as a default table
 engine""

---
 src/Core/Settings.h                                      | 2 +-
 src/Core/SettingsChangesHistory.h                        | 1 +
 tests/queries/0_stateless/02184_default_table_engine.sql | 2 ++
 3 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/src/Core/Settings.h b/src/Core/Settings.h
index ae6ea165cc9..5f52396d3bb 100644
--- a/src/Core/Settings.h
+++ b/src/Core/Settings.h
@@ -566,7 +566,7 @@ class IColumn;
     M(UInt64, min_free_disk_space_for_temporary_data, 0, "The minimum disk space to keep while writing temporary data used in external sorting and aggregation.", 0) \
     \
     M(DefaultTableEngine, default_temporary_table_engine, DefaultTableEngine::Memory, "Default table engine used when ENGINE is not set in CREATE TEMPORARY statement.",0) \
-    M(DefaultTableEngine, default_table_engine, DefaultTableEngine::None, "Default table engine used when ENGINE is not set in CREATE statement.",0) \
+    M(DefaultTableEngine, default_table_engine, DefaultTableEngine::MergeTree, "Default table engine used when ENGINE is not set in CREATE statement.",0) \
     M(Bool, show_table_uuid_in_table_create_query_if_not_nil, false, "For tables in databases with Engine=Atomic show UUID of the table in its CREATE query.", 0) \
     M(Bool, database_atomic_wait_for_drop_and_detach_synchronously, false, "When executing DROP or DETACH TABLE in Atomic database, wait for table data to be finally dropped or detached.", 0) \
     M(Bool, enable_scalar_subquery_optimization, true, "If it is set to true, prevent scalar subqueries from (de)serializing large scalar values and possibly avoid running the same subquery more than once.", 0) \
diff --git a/src/Core/SettingsChangesHistory.h b/src/Core/SettingsChangesHistory.h
index e8d013d13ec..661e7cb80da 100644
--- a/src/Core/SettingsChangesHistory.h
+++ b/src/Core/SettingsChangesHistory.h
@@ -133,6 +133,7 @@ static std::map<ClickHouseVersion, SettingsChangesHistory::SettingsChanges> sett
               {"function_visible_width_behavior", 0, 1, "We changed the default behavior of `visibleWidth` to be more precise"},
               {"max_estimated_execution_time", 0, 0, "Separate max_execution_time and max_estimated_execution_time"},
               {"iceberg_engine_ignore_schema_evolution", false, false, "Allow to ignore schema evolution in Iceberg table engine"},
+              {"default_table_engine", "None", "MergeTree", "Set default table engine to MergeTree for better usability"},
               {"optimize_injective_functions_in_group_by", false, true, "Replace injective functions by it's arguments in GROUP BY section in analyzer"},
               {"update_insert_deduplication_token_in_dependent_materialized_views", false, false, "Allow to update insert deduplication token with table identifier during insert in dependent materialized views"},
               {"azure_max_unexpected_write_error_retries", 4, 4, "The maximum number of retries in case of unexpected errors during Azure blob storage write"},
diff --git a/tests/queries/0_stateless/02184_default_table_engine.sql b/tests/queries/0_stateless/02184_default_table_engine.sql
index a984ec1b6c9..aff30eeea98 100644
--- a/tests/queries/0_stateless/02184_default_table_engine.sql
+++ b/tests/queries/0_stateless/02184_default_table_engine.sql
@@ -1,3 +1,5 @@
+SET default_table_engine = 'None';
+
 CREATE TABLE table_02184 (x UInt8); --{serverError 119}
 SET default_table_engine = 'Log';
 CREATE TABLE table_02184 (x UInt8);

From 0d4648b535a61561d122c87cf181434215753b35 Mon Sep 17 00:00:00 2001
From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com>
Date: Thu, 29 Feb 2024 10:30:17 +0800
Subject: [PATCH 038/443] Fix clang-tidy

---
 src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp b/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp
index eba57969580..c0b45e1d46a 100644
--- a/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp
+++ b/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp
@@ -32,7 +32,7 @@ LocalObjectStorage::LocalObjectStorage(String key_prefix_)
     else
         description = "/";
 
-    fs::create_directories(getCommonKeyPrefix());
+    fs::create_directories(key_prefix);
 }
 
 bool LocalObjectStorage::exists(const StoredObject & object) const

From 3188c1ebdac52efbdadb8f64a13b0c4b6f4e1acc Mon Sep 17 00:00:00 2001
From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com>
Date: Wed, 28 Feb 2024 13:51:48 +0800
Subject: [PATCH 039/443] Update test.py

---
 tests/integration/test_filesystem_cache/test.py | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/tests/integration/test_filesystem_cache/test.py b/tests/integration/test_filesystem_cache/test.py
index f32fa4e9823..0cb1866f8e4 100644
--- a/tests/integration/test_filesystem_cache/test.py
+++ b/tests/integration/test_filesystem_cache/test.py
@@ -94,12 +94,21 @@ def test_parallel_cache_loading_on_startup(cluster, node_name):
     cache_state = node.query(
         "SELECT key, file_segment_range_begin, size FROM system.filesystem_cache WHERE size > 0 ORDER BY key, file_segment_range_begin, size"
     )
+    keys = (
+        node.query(
+            "SELECT distinct(key) FROM system.filesystem_cache WHERE size > 0 ORDER BY key, file_segment_range_begin, size"
+        )
+        .strip()
+        .splitlines()
+    )
 
     node.restart_clickhouse()
 
-    assert cache_count == int(node.query("SELECT count() FROM system.filesystem_cache"))
+    # < because of additional files loaded into cache on server startup.
+    assert cache_count <= int(node.query("SELECT count() FROM system.filesystem_cache"))
+    keys_set = ",".join(["'" + x + "'" for x in keys])
     assert cache_state == node.query(
-        "SELECT key, file_segment_range_begin, size FROM system.filesystem_cache ORDER BY key, file_segment_range_begin, size"
+        f"SELECT key, file_segment_range_begin, size FROM system.filesystem_cache WHERE key in ({keys_set}) ORDER BY key, file_segment_range_begin, size"
     )
 
     assert node.contains_in_log("Loading filesystem cache with 30 threads")

From f8561b2265b924c64c60bdbc5305785c0f0b6f2e Mon Sep 17 00:00:00 2001
From: Sergei Trifonov <svtrifonov@gmail.com>
Date: Thu, 29 Feb 2024 13:53:27 +0100
Subject: [PATCH 040/443] Revert "Revert "Support resource request canceling""

---
 docs/en/operations/system-tables/scheduler.md |  4 +
 src/Common/Scheduler/ISchedulerNode.h         |  2 +
 src/Common/Scheduler/ISchedulerQueue.h        |  6 ++
 src/Common/Scheduler/Nodes/FairPolicy.h       | 99 ++++++++++---------
 src/Common/Scheduler/Nodes/FifoQueue.h        | 31 ++++--
 src/Common/Scheduler/Nodes/PriorityPolicy.h   | 38 ++++---
 .../tests/gtest_dynamic_resource_manager.cpp  |  1 -
 .../Nodes/tests/gtest_resource_scheduler.cpp  | 63 ++++++++++++
 src/Common/Scheduler/ResourceGuard.h          |  9 +-
 src/Common/Scheduler/ResourceRequest.cpp      | 13 +++
 src/Common/Scheduler/ResourceRequest.h        | 30 +++---
 src/Common/Scheduler/SchedulerRoot.h          | 32 +++---
 .../System/StorageSystemScheduler.cpp         |  4 +
 13 files changed, 224 insertions(+), 108 deletions(-)
 create mode 100644 src/Common/Scheduler/ResourceRequest.cpp

diff --git a/docs/en/operations/system-tables/scheduler.md b/docs/en/operations/system-tables/scheduler.md
index 953db4c28f2..c4de7f76fdc 100644
--- a/docs/en/operations/system-tables/scheduler.md
+++ b/docs/en/operations/system-tables/scheduler.md
@@ -26,7 +26,9 @@ priority:          0
 is_active:         0
 active_children:   0
 dequeued_requests: 67
+canceled_requests: 0
 dequeued_cost:     4692272
+canceled_cost:     0
 busy_periods:      63
 vruntime:          938454.1999999989
 system_vruntime:   ᴺᵁᴸᴸ
@@ -54,7 +56,9 @@ Columns:
 - `is_active` (`UInt8`) - Whether this node is currently active - has resource requests to be dequeued and constraints satisfied.
 - `active_children` (`UInt64`) - The number of children in active state.
 - `dequeued_requests` (`UInt64`) - The total number of resource requests dequeued from this node.
+- `canceled_requests` (`UInt64`) - The total number of resource requests canceled from this node.
 - `dequeued_cost` (`UInt64`) - The sum of costs (e.g. size in bytes) of all requests dequeued from this node.
+- `canceled_cost` (`UInt64`) - The sum of costs (e.g. size in bytes) of all requests canceled from this node.
 - `busy_periods` (`UInt64`) - The total number of deactivations of this node.
 - `vruntime` (`Nullable(Float64)`) - For children of `fair` nodes only. Virtual runtime of a node used by SFQ algorithm to select the next child to process in a max-min fair manner.
 - `system_vruntime` (`Nullable(Float64)`) - For `fair` nodes only. Virtual runtime showing `vruntime` of the last processed resource request. Used during child activation as the new value of `vruntime`.
diff --git a/src/Common/Scheduler/ISchedulerNode.h b/src/Common/Scheduler/ISchedulerNode.h
index 804026d7bf4..20c1f4332da 100644
--- a/src/Common/Scheduler/ISchedulerNode.h
+++ b/src/Common/Scheduler/ISchedulerNode.h
@@ -387,7 +387,9 @@ public:
 
     /// Introspection
     std::atomic<UInt64> dequeued_requests{0};
+    std::atomic<UInt64> canceled_requests{0};
     std::atomic<ResourceCost> dequeued_cost{0};
+    std::atomic<ResourceCost> canceled_cost{0};
     std::atomic<UInt64> busy_periods{0};
 };
 
diff --git a/src/Common/Scheduler/ISchedulerQueue.h b/src/Common/Scheduler/ISchedulerQueue.h
index cbe63bd304a..532f4bf6c63 100644
--- a/src/Common/Scheduler/ISchedulerQueue.h
+++ b/src/Common/Scheduler/ISchedulerQueue.h
@@ -50,6 +50,12 @@ public:
     /// Should be called outside of scheduling subsystem, implementation must be thread-safe.
     virtual void enqueueRequest(ResourceRequest * request) = 0;
 
+    /// Cancel previously enqueued request.
+    /// Returns `false` and does nothing given unknown or already executed request.
+    /// Returns `true` if requests has been found and canceled.
+    /// Should be called outside of scheduling subsystem, implementation must be thread-safe.
+    virtual bool cancelRequest(ResourceRequest * request) = 0;
+
     /// For introspection
     ResourceCost getBudget() const
     {
diff --git a/src/Common/Scheduler/Nodes/FairPolicy.h b/src/Common/Scheduler/Nodes/FairPolicy.h
index c0e187e6fa9..ce2bf729a04 100644
--- a/src/Common/Scheduler/Nodes/FairPolicy.h
+++ b/src/Common/Scheduler/Nodes/FairPolicy.h
@@ -134,56 +134,65 @@ public:
 
     std::pair<ResourceRequest *, bool> dequeueRequest() override
     {
-        if (heap_size == 0)
-            return {nullptr, false};
-
-        // Recursively pull request from child
-        auto [request, child_active] = items.front().child->dequeueRequest();
-        assert(request != nullptr);
-        std::pop_heap(items.begin(), items.begin() + heap_size);
-        Item & current = items[heap_size - 1];
-
-        // SFQ fairness invariant: system vruntime equals last served request start-time
-        assert(current.vruntime >= system_vruntime);
-        system_vruntime = current.vruntime;
-
-        // By definition vruntime is amount of consumed resource (cost) divided by weight
-        current.vruntime += double(request->cost) / current.child->info.weight;
-        max_vruntime = std::max(max_vruntime, current.vruntime);
-
-        if (child_active) // Put active child back in heap after vruntime update
+        // Cycle is required to do deactivations in the case of canceled requests, when dequeueRequest returns `nullptr`
+        while (true)
         {
-            std::push_heap(items.begin(), items.begin() + heap_size);
-        }
-        else // Deactivate child if it is empty, but remember it's vruntime for latter activations
-        {
-            heap_size--;
+            if (heap_size == 0)
+                return {nullptr, false};
 
-            // Store index of this inactive child in `parent.idx`
-            // This enables O(1) search of inactive children instead of O(n)
-            current.child->info.parent.idx = heap_size;
-        }
+            // Recursively pull request from child
+            auto [request, child_active] = items.front().child->dequeueRequest();
+            std::pop_heap(items.begin(), items.begin() + heap_size);
+            Item & current = items[heap_size - 1];
 
-        // Reset any difference between children on busy period end
-        if (heap_size == 0)
-        {
-            // Reset vtime to zero to avoid floating-point error accumulation,
-            // but do not reset too often, because it's O(N)
-            UInt64 ns = clock_gettime_ns();
-            if (last_reset_ns + 1000000000 < ns)
+            if (request)
             {
-                last_reset_ns = ns;
-                for (Item & item : items)
-                    item.vruntime = 0;
-                max_vruntime = 0;
-            }
-            system_vruntime = max_vruntime;
-            busy_periods++;
-        }
+                // SFQ fairness invariant: system vruntime equals last served request start-time
+                assert(current.vruntime >= system_vruntime);
+                system_vruntime = current.vruntime;
 
-        dequeued_requests++;
-        dequeued_cost += request->cost;
-        return {request, heap_size > 0};
+                // By definition vruntime is amount of consumed resource (cost) divided by weight
+                current.vruntime += double(request->cost) / current.child->info.weight;
+                max_vruntime = std::max(max_vruntime, current.vruntime);
+            }
+
+            if (child_active) // Put active child back in heap after vruntime update
+            {
+                std::push_heap(items.begin(), items.begin() + heap_size);
+            }
+            else // Deactivate child if it is empty, but remember it's vruntime for latter activations
+            {
+                heap_size--;
+
+                // Store index of this inactive child in `parent.idx`
+                // This enables O(1) search of inactive children instead of O(n)
+                current.child->info.parent.idx = heap_size;
+            }
+
+            // Reset any difference between children on busy period end
+            if (heap_size == 0)
+            {
+                // Reset vtime to zero to avoid floating-point error accumulation,
+                // but do not reset too often, because it's O(N)
+                UInt64 ns = clock_gettime_ns();
+                if (last_reset_ns + 1000000000 < ns)
+                {
+                    last_reset_ns = ns;
+                    for (Item & item : items)
+                        item.vruntime = 0;
+                    max_vruntime = 0;
+                }
+                system_vruntime = max_vruntime;
+                busy_periods++;
+            }
+
+            if (request)
+            {
+                dequeued_requests++;
+                dequeued_cost += request->cost;
+                return {request, heap_size > 0};
+            }
+        }
     }
 
     bool isActive() override
diff --git a/src/Common/Scheduler/Nodes/FifoQueue.h b/src/Common/Scheduler/Nodes/FifoQueue.h
index 38ae902bc2f..45ed32343ff 100644
--- a/src/Common/Scheduler/Nodes/FifoQueue.h
+++ b/src/Common/Scheduler/Nodes/FifoQueue.h
@@ -39,8 +39,7 @@ public:
 
     void enqueueRequest(ResourceRequest * request) override
     {
-        std::unique_lock lock(mutex);
-        request->enqueue_ns = clock_gettime_ns();
+        std::lock_guard lock(mutex);
         queue_cost += request->cost;
         bool was_empty = requests.empty();
         requests.push_back(request);
@@ -50,7 +49,7 @@ public:
 
     std::pair<ResourceRequest *, bool> dequeueRequest() override
     {
-        std::unique_lock lock(mutex);
+        std::lock_guard lock(mutex);
         if (requests.empty())
             return {nullptr, false};
         ResourceRequest * result = requests.front();
@@ -63,9 +62,29 @@ public:
         return {result, !requests.empty()};
     }
 
+    bool cancelRequest(ResourceRequest * request) override
+    {
+        std::lock_guard lock(mutex);
+        // TODO(serxa): reimplement queue as intrusive list of ResourceRequest to make this O(1) instead of O(N)
+        for (auto i = requests.begin(), e = requests.end(); i != e; ++i)
+        {
+            if (*i == request)
+            {
+                requests.erase(i);
+                if (requests.empty())
+                    busy_periods++;
+                queue_cost -= request->cost;
+                canceled_requests++;
+                canceled_cost += request->cost;
+                return true;
+            }
+        }
+        return false;
+    }
+
     bool isActive() override
     {
-        std::unique_lock lock(mutex);
+        std::lock_guard lock(mutex);
         return !requests.empty();
     }
 
@@ -98,14 +117,14 @@ public:
 
     std::pair<UInt64, Int64> getQueueLengthAndCost()
     {
-        std::unique_lock lock(mutex);
+        std::lock_guard lock(mutex);
         return {requests.size(), queue_cost};
     }
 
 private:
     std::mutex mutex;
     Int64 queue_cost = 0;
-    std::deque<ResourceRequest *> requests;
+    std::deque<ResourceRequest *> requests; // TODO(serxa): reimplement it using intrusive list to avoid allocations/deallocations and O(N) during cancel
 };
 
 }
diff --git a/src/Common/Scheduler/Nodes/PriorityPolicy.h b/src/Common/Scheduler/Nodes/PriorityPolicy.h
index 6d6b15bd063..9b4cfc37f8c 100644
--- a/src/Common/Scheduler/Nodes/PriorityPolicy.h
+++ b/src/Common/Scheduler/Nodes/PriorityPolicy.h
@@ -102,25 +102,31 @@ public:
 
     std::pair<ResourceRequest *, bool> dequeueRequest() override
     {
-        if (items.empty())
-            return {nullptr, false};
-
-        // Recursively pull request from child
-        auto [request, child_active] = items.front().child->dequeueRequest();
-        assert(request != nullptr);
-
-        // Deactivate child if it is empty
-        if (!child_active)
+        // Cycle is required to do deactivations in the case of canceled requests, when dequeueRequest returns `nullptr`
+        while (true)
         {
-            std::pop_heap(items.begin(), items.end());
-            items.pop_back();
             if (items.empty())
-                busy_periods++;
-        }
+                return {nullptr, false};
 
-        dequeued_requests++;
-        dequeued_cost += request->cost;
-        return {request, !items.empty()};
+            // Recursively pull request from child
+            auto [request, child_active] = items.front().child->dequeueRequest();
+
+            // Deactivate child if it is empty
+            if (!child_active)
+            {
+                std::pop_heap(items.begin(), items.end());
+                items.pop_back();
+                if (items.empty())
+                    busy_periods++;
+            }
+
+            if (request)
+            {
+                dequeued_requests++;
+                dequeued_cost += request->cost;
+                return {request, !items.empty()};
+            }
+        }
     }
 
     bool isActive() override
diff --git a/src/Common/Scheduler/Nodes/tests/gtest_dynamic_resource_manager.cpp b/src/Common/Scheduler/Nodes/tests/gtest_dynamic_resource_manager.cpp
index 961a3b6f713..cdf09776077 100644
--- a/src/Common/Scheduler/Nodes/tests/gtest_dynamic_resource_manager.cpp
+++ b/src/Common/Scheduler/Nodes/tests/gtest_dynamic_resource_manager.cpp
@@ -38,7 +38,6 @@ TEST(SchedulerDynamicResourceManager, Smoke)
     {
         ResourceGuard gA(cA->get("res1"), ResourceGuard::PostponeLocking);
         gA.lock();
-        gA.setFailure();
         gA.unlock();
 
         ResourceGuard gB(cB->get("res1"));
diff --git a/src/Common/Scheduler/Nodes/tests/gtest_resource_scheduler.cpp b/src/Common/Scheduler/Nodes/tests/gtest_resource_scheduler.cpp
index 9fefbc02cbd..e76639a4b01 100644
--- a/src/Common/Scheduler/Nodes/tests/gtest_resource_scheduler.cpp
+++ b/src/Common/Scheduler/Nodes/tests/gtest_resource_scheduler.cpp
@@ -4,6 +4,7 @@
 
 #include <Common/Scheduler/Nodes/tests/ResourceTest.h>
 
+#include <barrier>
 #include <future>
 
 using namespace DB;
@@ -73,6 +74,22 @@ struct ResourceHolder
     }
 };
 
+struct MyRequest : public ResourceRequest
+{
+    std::function<void()> on_execute;
+
+    explicit MyRequest(ResourceCost cost_, std::function<void()> on_execute_)
+        : ResourceRequest(cost_)
+        , on_execute(on_execute_)
+    {}
+
+    void execute() override
+    {
+        if (on_execute)
+            on_execute();
+    }
+};
+
 TEST(SchedulerRoot, Smoke)
 {
     ResourceTest t;
@@ -111,3 +128,49 @@ TEST(SchedulerRoot, Smoke)
         EXPECT_TRUE(fc2->requests.contains(&rg.request));
     }
 }
+
+TEST(SchedulerRoot, Cancel)
+{
+    ResourceTest t;
+
+    ResourceHolder r1(t);
+    auto * fc1 = r1.add<ConstraintTest>("/", "<max_requests>1</max_requests>");
+    r1.add<PriorityPolicy>("/prio");
+    auto a = r1.addQueue("/prio/A", "<priority>1</priority>");
+    auto b = r1.addQueue("/prio/B", "<priority>2</priority>");
+    r1.registerResource();
+
+    std::barrier sync(2);
+    std::thread consumer1([&]
+    {
+        std::barrier destruct_sync(2);
+        MyRequest request(1,[&]
+        {
+            sync.arrive_and_wait(); // (A)
+            EXPECT_TRUE(fc1->requests.contains(&request));
+            sync.arrive_and_wait(); // (B)
+            request.finish();
+            destruct_sync.arrive_and_wait(); // (C)
+        });
+        a.queue->enqueueRequest(&request);
+        destruct_sync.arrive_and_wait(); // (C)
+    });
+
+    std::thread consumer2([&]
+    {
+        MyRequest request(1,[&]
+        {
+            FAIL() << "This request must be canceled, but instead executes";
+        });
+        sync.arrive_and_wait(); // (A) wait for request of consumer1 to be inside execute, so that constraint is in violated state and our request will not be executed immediately
+        b.queue->enqueueRequest(&request);
+        bool canceled = b.queue->cancelRequest(&request);
+        EXPECT_TRUE(canceled);
+        sync.arrive_and_wait(); // (B) release request of consumer1 to be finished
+    });
+
+    consumer1.join();
+    consumer2.join();
+
+    EXPECT_TRUE(fc1->requests.empty());
+}
diff --git a/src/Common/Scheduler/ResourceGuard.h b/src/Common/Scheduler/ResourceGuard.h
index dca4041b176..50f665a384b 100644
--- a/src/Common/Scheduler/ResourceGuard.h
+++ b/src/Common/Scheduler/ResourceGuard.h
@@ -71,8 +71,7 @@ public:
             // lock(mutex) is not required because `Dequeued` request cannot be used by the scheduler thread
             chassert(state == Dequeued);
             state = Finished;
-            if (constraint)
-                constraint->finishRequest(this);
+            ResourceRequest::finish();
         }
 
         static Request & local()
@@ -126,12 +125,6 @@ public:
         }
     }
 
-    /// Mark request as unsuccessful; by default request is considered to be successful
-    void setFailure()
-    {
-        request.successful = false;
-    }
-
     ResourceLink link;
     Request & request;
 };
diff --git a/src/Common/Scheduler/ResourceRequest.cpp b/src/Common/Scheduler/ResourceRequest.cpp
new file mode 100644
index 00000000000..26e8084cdfa
--- /dev/null
+++ b/src/Common/Scheduler/ResourceRequest.cpp
@@ -0,0 +1,13 @@
+#include <Common/Scheduler/ResourceRequest.h>
+#include <Common/Scheduler/ISchedulerConstraint.h>
+
+namespace DB
+{
+
+void ResourceRequest::finish()
+{
+    if (constraint)
+        constraint->finishRequest(this);
+}
+
+}
diff --git a/src/Common/Scheduler/ResourceRequest.h b/src/Common/Scheduler/ResourceRequest.h
index 3d2230746f9..f3153ad382c 100644
--- a/src/Common/Scheduler/ResourceRequest.h
+++ b/src/Common/Scheduler/ResourceRequest.h
@@ -14,9 +14,6 @@ class ISchedulerConstraint;
 using ResourceCost = Int64;
 constexpr ResourceCost ResourceCostMax = std::numeric_limits<int>::max();
 
-/// Timestamps (nanoseconds since epoch)
-using ResourceNs = UInt64;
-
 /*
  * Request for a resource consumption. The main moving part of the scheduling subsystem.
  * Resource requests processing workflow:
@@ -31,7 +28,7 @@ using ResourceNs = UInt64;
  *  3) Scheduler calls ISchedulerNode::dequeueRequest() that returns the request.
  *  4) Callback ResourceRequest::execute() is called to provide access to the resource.
  *  5) The resource consumption is happening outside of the scheduling subsystem.
- *  6) request->constraint->finishRequest() is called when consumption is finished.
+ *  6) ResourceRequest::finish() is called when consumption is finished.
  *
  * Steps (5) and (6) can be omitted if constraint is not used by the resource.
  *
@@ -39,7 +36,10 @@ using ResourceNs = UInt64;
  * Request ownership is done outside of the scheduling subsystem.
  * After (6) request can be destructed safely.
  *
- * Request cancelling is not supported yet.
+ * Request can also be canceled before (3) using ISchedulerQueue::cancelRequest().
+ * Returning false means it is too late for request to be canceled. It should be processed in a regular way.
+ * Returning true means successful cancel and therefore steps (4) and (5) are not going to happen
+ * and step (6) MUST be omitted.
  */
 class ResourceRequest
 {
@@ -48,32 +48,20 @@ public:
     /// NOTE: If cost is not known in advance, ResourceBudget should be used (note that every ISchedulerQueue has it)
     ResourceCost cost;
 
-    /// Request outcome
-    /// Should be filled during resource consumption
-    bool successful;
-
     /// Scheduler node to be notified on consumption finish
     /// Auto-filled during request enqueue/dequeue
     ISchedulerConstraint * constraint;
 
-    /// Timestamps for introspection
-    ResourceNs enqueue_ns;
-    ResourceNs execute_ns;
-    ResourceNs finish_ns;
-
     explicit ResourceRequest(ResourceCost cost_ = 1)
     {
         reset(cost_);
     }
 
+    /// ResourceRequest object may be reused again after reset()
     void reset(ResourceCost cost_)
     {
         cost = cost_;
-        successful = true;
         constraint = nullptr;
-        enqueue_ns = 0;
-        execute_ns = 0;
-        finish_ns = 0;
     }
 
     virtual ~ResourceRequest() = default;
@@ -83,6 +71,12 @@ public:
     /// just triggering start of a consumption, not doing the consumption itself
     /// (e.g. setting an std::promise or creating a job in a thread pool)
     virtual void execute() = 0;
+
+    /// Stop resource consumption and notify resource scheduler.
+    /// Should be called when resource consumption is finished by consumer.
+    /// ResourceRequest should not be destructed or reset before calling to `finish()`.
+    /// WARNING: this function MUST not be called if request was canceled.
+    void finish();
 };
 
 }
diff --git a/src/Common/Scheduler/SchedulerRoot.h b/src/Common/Scheduler/SchedulerRoot.h
index 3a23a8df834..ab3f702a422 100644
--- a/src/Common/Scheduler/SchedulerRoot.h
+++ b/src/Common/Scheduler/SchedulerRoot.h
@@ -145,22 +145,27 @@ public:
 
     std::pair<ResourceRequest *, bool> dequeueRequest() override
     {
-        if (current == nullptr) // No active resources
-            return {nullptr, false};
+        while (true)
+        {
+            if (current == nullptr) // No active resources
+                return {nullptr, false};
 
-        // Dequeue request from current resource
-        auto [request, resource_active] = current->root->dequeueRequest();
-        assert(request != nullptr);
+            // Dequeue request from current resource
+            auto [request, resource_active] = current->root->dequeueRequest();
 
-        // Deactivate resource if required
-        if (!resource_active)
-            deactivate(current);
-        else
-            current = current->next; // Just move round-robin pointer
+            // Deactivate resource if required
+            if (!resource_active)
+                deactivate(current);
+            else
+                current = current->next; // Just move round-robin pointer
 
-        dequeued_requests++;
-        dequeued_cost += request->cost;
-        return {request, current != nullptr};
+            if (request == nullptr) // Possible in case of request cancel, just retry
+                continue;
+
+            dequeued_requests++;
+            dequeued_cost += request->cost;
+            return {request, current != nullptr};
+        }
     }
 
     bool isActive() override
@@ -245,7 +250,6 @@ private:
 
     void execute(ResourceRequest * request)
     {
-        request->execute_ns = clock_gettime_ns();
         request->execute();
     }
 
diff --git a/src/Storages/System/StorageSystemScheduler.cpp b/src/Storages/System/StorageSystemScheduler.cpp
index ba07d44dbf9..633bac5d285 100644
--- a/src/Storages/System/StorageSystemScheduler.cpp
+++ b/src/Storages/System/StorageSystemScheduler.cpp
@@ -30,7 +30,9 @@ ColumnsDescription StorageSystemScheduler::getColumnsDescription()
         {"is_active", std::make_shared<DataTypeUInt8>(), "Whether this node is currently active - has resource requests to be dequeued and constraints satisfied."},
         {"active_children", std::make_shared<DataTypeUInt64>(), "The number of children in active state."},
         {"dequeued_requests", std::make_shared<DataTypeUInt64>(), "The total number of resource requests dequeued from this node."},
+        {"canceled_requests", std::make_shared<DataTypeUInt64>(), "The total number of resource requests canceled from this node."},
         {"dequeued_cost", std::make_shared<DataTypeInt64>(), "The sum of costs (e.g. size in bytes) of all requests dequeued from this node."},
+        {"canceled_cost", std::make_shared<DataTypeInt64>(), "The sum of costs (e.g. size in bytes) of all requests canceled from this node."},
         {"busy_periods", std::make_shared<DataTypeUInt64>(), "The total number of deactivations of this node."},
         {"vruntime", std::make_shared<DataTypeNullable>(std::make_shared<DataTypeFloat64>()),
             "For children of `fair` nodes only. Virtual runtime of a node used by SFQ algorithm to select the next child to process in a max-min fair manner."},
@@ -93,7 +95,9 @@ void StorageSystemScheduler::fillData(MutableColumns & res_columns, ContextPtr c
         res_columns[i++]->insert(node->isActive());
         res_columns[i++]->insert(node->activeChildren());
         res_columns[i++]->insert(node->dequeued_requests.load());
+        res_columns[i++]->insert(node->canceled_requests.load());
         res_columns[i++]->insert(node->dequeued_cost.load());
+        res_columns[i++]->insert(node->canceled_cost.load());
         res_columns[i++]->insert(node->busy_periods.load());
 
         Field vruntime;

From 7632c2c33f357c1c616f734c7bf2502ccbfbd496 Mon Sep 17 00:00:00 2001
From: avogar <avogar@clickhouse.com>
Date: Thu, 29 Feb 2024 15:17:12 +0000
Subject: [PATCH 041/443] Remove non-deterministic functions in virtual columns
 filter

---
 src/Storages/MergeTree/MergeTreeData.cpp      |  2 ++
 src/Storages/VirtualColumnUtils.cpp           | 21 +++++++++++++++++++
 ...with_non_deterministic_functions.reference | 11 ++++++++++
 ...lumns_with_non_deterministic_functions.sql |  6 ++++++
 4 files changed, 40 insertions(+)
 create mode 100644 tests/queries/0_stateless/03002_filter_skip_virtual_columns_with_non_deterministic_functions.reference
 create mode 100644 tests/queries/0_stateless/03002_filter_skip_virtual_columns_with_non_deterministic_functions.sql

diff --git a/src/Storages/MergeTree/MergeTreeData.cpp b/src/Storages/MergeTree/MergeTreeData.cpp
index 8aa188cfe5c..6494ed5d844 100644
--- a/src/Storages/MergeTree/MergeTreeData.cpp
+++ b/src/Storages/MergeTree/MergeTreeData.cpp
@@ -1082,6 +1082,8 @@ std::optional<UInt64> MergeTreeData::totalRowsByPartitionPredicateImpl(
     Block virtual_columns_block = getBlockWithVirtualPartColumns(parts, true /* one_part */);
 
     auto filter_dag = VirtualColumnUtils::splitFilterDagForAllowedInputs(filter_actions_dag->getOutputs().at(0), nullptr);
+    if (!filter_dag)
+        return {};
 
     // Generate valid expressions for filtering
     bool valid = true;
diff --git a/src/Storages/VirtualColumnUtils.cpp b/src/Storages/VirtualColumnUtils.cpp
index 33ff6e7104f..3e0ef1d7990 100644
--- a/src/Storages/VirtualColumnUtils.cpp
+++ b/src/Storages/VirtualColumnUtils.cpp
@@ -467,6 +467,23 @@ static bool canEvaluateSubtree(const ActionsDAG::Node * node, const Block & allo
     return true;
 }
 
+static bool isDeterministic(const ActionsDAG::Node * node)
+{
+    if (node->type != ActionsDAG::ActionType::FUNCTION)
+        return true;
+
+    if (!node->function_base->isDeterministic())
+        return false;
+
+    for (const auto * child : node->children)
+    {
+        if (!isDeterministic(child))
+            return false;
+    }
+
+    return true;
+}
+
 static const ActionsDAG::Node * splitFilterNodeForAllowedInputs(
     const ActionsDAG::Node * node,
     const Block * allowed_inputs,
@@ -542,6 +559,10 @@ static const ActionsDAG::Node * splitFilterNodeForAllowedInputs(
                 }
             }
         }
+        else if (!isDeterministic(node))
+        {
+            return nullptr;
+        }
     }
 
     if (allowed_inputs && !canEvaluateSubtree(node, *allowed_inputs))
diff --git a/tests/queries/0_stateless/03002_filter_skip_virtual_columns_with_non_deterministic_functions.reference b/tests/queries/0_stateless/03002_filter_skip_virtual_columns_with_non_deterministic_functions.reference
new file mode 100644
index 00000000000..4c9646d6ffa
--- /dev/null
+++ b/tests/queries/0_stateless/03002_filter_skip_virtual_columns_with_non_deterministic_functions.reference
@@ -0,0 +1,11 @@
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+1
diff --git a/tests/queries/0_stateless/03002_filter_skip_virtual_columns_with_non_deterministic_functions.sql b/tests/queries/0_stateless/03002_filter_skip_virtual_columns_with_non_deterministic_functions.sql
new file mode 100644
index 00000000000..9f8bc6bd3d7
--- /dev/null
+++ b/tests/queries/0_stateless/03002_filter_skip_virtual_columns_with_non_deterministic_functions.sql
@@ -0,0 +1,6 @@
+create table test (number UInt64) engine=MergeTree order by number;
+insert into test select * from numbers(100000000);
+select ignore(number) from test where RAND() > 4292390314 limit 10;
+select count() > 0 from test where RAND() > 4292390314;
+drop table test;
+

From 09a392772d75b38e1b19ad6bd2a863168ea0de5c Mon Sep 17 00:00:00 2001
From: avogar <avogar@clickhouse.com>
Date: Thu, 29 Feb 2024 15:34:45 +0000
Subject: [PATCH 042/443] Use isDeterministicInScopeOfQuery

---
 src/Storages/VirtualColumnUtils.cpp | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/src/Storages/VirtualColumnUtils.cpp b/src/Storages/VirtualColumnUtils.cpp
index 3e0ef1d7990..6d66453442e 100644
--- a/src/Storages/VirtualColumnUtils.cpp
+++ b/src/Storages/VirtualColumnUtils.cpp
@@ -467,17 +467,17 @@ static bool canEvaluateSubtree(const ActionsDAG::Node * node, const Block & allo
     return true;
 }
 
-static bool isDeterministic(const ActionsDAG::Node * node)
+static bool isDeterministicInScopeOfQuery(const ActionsDAG::Node * node)
 {
     if (node->type != ActionsDAG::ActionType::FUNCTION)
         return true;
 
-    if (!node->function_base->isDeterministic())
+    if (!node->function_base->isDeterministicInScopeOfQuery())
         return false;
 
     for (const auto * child : node->children)
     {
-        if (!isDeterministic(child))
+        if (!isDeterministicInScopeOfQuery(child))
             return false;
     }
 
@@ -559,7 +559,7 @@ static const ActionsDAG::Node * splitFilterNodeForAllowedInputs(
                 }
             }
         }
-        else if (!isDeterministic(node))
+        else if (!isDeterministicInScopeOfQuery(node))
         {
             return nullptr;
         }

From 3825cb3ad0d7f2296cf075648d022ef26f1e0cef Mon Sep 17 00:00:00 2001
From: Yakov Olkhovskiy <yakov@clickhouse.com>
Date: Sat, 2 Mar 2024 15:28:45 +0000
Subject: [PATCH 043/443] expand CTE in alter modify query

---
 src/Interpreters/InterpreterAlterQuery.cpp        | 11 +++++++++++
 .../0_stateless/03002_modify_query_cte.reference  |  2 ++
 .../0_stateless/03002_modify_query_cte.sql        | 15 +++++++++++++++
 3 files changed, 28 insertions(+)
 create mode 100644 tests/queries/0_stateless/03002_modify_query_cte.reference
 create mode 100644 tests/queries/0_stateless/03002_modify_query_cte.sql

diff --git a/src/Interpreters/InterpreterAlterQuery.cpp b/src/Interpreters/InterpreterAlterQuery.cpp
index b768593da98..7acaf95becc 100644
--- a/src/Interpreters/InterpreterAlterQuery.cpp
+++ b/src/Interpreters/InterpreterAlterQuery.cpp
@@ -1,3 +1,4 @@
+#include <Interpreters/ApplyWithSubqueryVisitor.h>
 #include <Interpreters/InterpreterAlterQuery.h>
 #include <Interpreters/InterpreterFactory.h>
 
@@ -71,11 +72,15 @@ BlockIO InterpreterAlterQuery::execute()
 
 BlockIO InterpreterAlterQuery::executeToTable(const ASTAlterQuery & alter)
 {
+    ASTSelectWithUnionQuery * modify_query = nullptr;
+
     for (auto & child : alter.command_list->children)
     {
         auto * command_ast = child->as<ASTAlterCommand>();
         if (command_ast->sql_security)
             InterpreterCreateQuery::processSQLSecurityOption(getContext(), command_ast->sql_security->as<ASTSQLSecurity &>());
+        else if (command_ast->type == ASTAlterCommand::MODIFY_QUERY)
+            modify_query = command_ast->select->as<ASTSelectWithUnionQuery>();
     }
 
     BlockIO res;
@@ -123,6 +128,12 @@ BlockIO InterpreterAlterQuery::executeToTable(const ASTAlterQuery & alter)
         throw Exception(ErrorCodes::TABLE_IS_READ_ONLY, "Table is read-only");
     auto table_lock = table->lockForShare(getContext()->getCurrentQueryId(), getContext()->getSettingsRef().lock_acquire_timeout);
 
+    if (modify_query)
+    {
+        // Expand CTE before filling default database
+        ApplyWithSubqueryVisitor().visit(*modify_query);
+    }
+
     /// Add default database to table identifiers that we can encounter in e.g. default expressions, mutation expression, etc.
     AddDefaultDatabaseVisitor visitor(getContext(), table_id.getDatabaseName());
     ASTPtr command_list_ptr = alter.command_list->ptr();
diff --git a/tests/queries/0_stateless/03002_modify_query_cte.reference b/tests/queries/0_stateless/03002_modify_query_cte.reference
new file mode 100644
index 00000000000..a3d66f70f8f
--- /dev/null
+++ b/tests/queries/0_stateless/03002_modify_query_cte.reference
@@ -0,0 +1,2 @@
+CREATE MATERIALIZED VIEW default.mv_03002 TO default.table_03002\n(\n    `ts` DateTime\n)\nAS SELECT ts\nFROM default.table_03002
+CREATE MATERIALIZED VIEW default.mv_03002 TO default.table_03002\n(\n    `ts` DateTime\n)\nAS WITH MY_CTE AS\n    (\n        SELECT ts\n        FROM default.table_03002\n    )\nSELECT *\nFROM\nMY_CTE
diff --git a/tests/queries/0_stateless/03002_modify_query_cte.sql b/tests/queries/0_stateless/03002_modify_query_cte.sql
new file mode 100644
index 00000000000..3a36ce7e7fd
--- /dev/null
+++ b/tests/queries/0_stateless/03002_modify_query_cte.sql
@@ -0,0 +1,15 @@
+
+CREATE TABLE table_03002 (ts DateTime, event_type String) ENGINE = MergeTree ORDER BY (event_type, ts);
+
+CREATE MATERIALIZED VIEW mv_03002 TO table_03002 AS SELECT ts FROM table_03002;
+
+SHOW CREATE TABLE mv_03002;
+
+ALTER TABLE mv_03002 MODIFY QUERY
+WITH MY_CTE AS (SELECT ts FROM table_03002)
+SELECT * FROM MY_CTE;
+
+SHOW CREATE TABLE mv_03002;
+
+DROP TABLE mv_03002;
+DROP TABLE table_03002;

From 17413ded759ebcef809e03a80284f6f805507560 Mon Sep 17 00:00:00 2001
From: Yakov Olkhovskiy <99031427+yakov-olkhovskiy@users.noreply.github.com>
Date: Sat, 2 Mar 2024 11:11:44 -0500
Subject: [PATCH 044/443] Update 03002_modify_query_cte.reference

---
 tests/queries/0_stateless/03002_modify_query_cte.reference | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/queries/0_stateless/03002_modify_query_cte.reference b/tests/queries/0_stateless/03002_modify_query_cte.reference
index a3d66f70f8f..50e4a7c6a07 100644
--- a/tests/queries/0_stateless/03002_modify_query_cte.reference
+++ b/tests/queries/0_stateless/03002_modify_query_cte.reference
@@ -1,2 +1,2 @@
 CREATE MATERIALIZED VIEW default.mv_03002 TO default.table_03002\n(\n    `ts` DateTime\n)\nAS SELECT ts\nFROM default.table_03002
-CREATE MATERIALIZED VIEW default.mv_03002 TO default.table_03002\n(\n    `ts` DateTime\n)\nAS WITH MY_CTE AS\n    (\n        SELECT ts\n        FROM default.table_03002\n    )\nSELECT *\nFROM\nMY_CTE
+CREATE MATERIALIZED VIEW default.mv_03002 TO default.table_03002\n(\n    `ts` DateTime\n)\nAS WITH MY_CTE AS\n    (\n        SELECT ts\n        FROM default.table_03002\n    )\nSELECT *\nFROM MY_CTE

From a6cb302ab54082db5650263d6417052f81f30710 Mon Sep 17 00:00:00 2001
From: serxa <sergei@clickhouse.com>
Date: Sun, 3 Mar 2024 15:48:49 +0000
Subject: [PATCH 045/443] fix 'AddressSanitizer: stack-use-after-return'

---
 src/Common/Scheduler/Nodes/tests/gtest_resource_scheduler.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Common/Scheduler/Nodes/tests/gtest_resource_scheduler.cpp b/src/Common/Scheduler/Nodes/tests/gtest_resource_scheduler.cpp
index e76639a4b01..f8196d15819 100644
--- a/src/Common/Scheduler/Nodes/tests/gtest_resource_scheduler.cpp
+++ b/src/Common/Scheduler/Nodes/tests/gtest_resource_scheduler.cpp
@@ -140,10 +140,10 @@ TEST(SchedulerRoot, Cancel)
     auto b = r1.addQueue("/prio/B", "<priority>2</priority>");
     r1.registerResource();
 
+    std::barrier destruct_sync(2);
     std::barrier sync(2);
     std::thread consumer1([&]
     {
-        std::barrier destruct_sync(2);
         MyRequest request(1,[&]
         {
             sync.arrive_and_wait(); // (A)

From 77fe221665ac8610e5ae42f547771e1877793ad0 Mon Sep 17 00:00:00 2001
From: johnnymatthews <9611008+johnnymatthews@users.noreply.github.com>
Date: Sun, 3 Mar 2024 14:25:25 -0700
Subject: [PATCH 046/443] Adds undocumented rand functions. Prettifies
 markdown.

---
 .../functions/random-functions.md             | 277 +++++++++++++-----
 1 file changed, 206 insertions(+), 71 deletions(-)

diff --git a/docs/en/sql-reference/functions/random-functions.md b/docs/en/sql-reference/functions/random-functions.md
index 6fd31e8d25c..2ce9c75eae4 100644
--- a/docs/en/sql-reference/functions/random-functions.md
+++ b/docs/en/sql-reference/functions/random-functions.md
@@ -11,79 +11,213 @@ elimination](../../sql-reference/functions/index.md#common-subexpression-elimina
 function return different random values.
 
 Related content
+
 - Blog: [Generating random data in ClickHouse](https://clickhouse.com/blog/generating-random-test-distribution-data-for-clickhouse)
 
 :::note
 The random numbers are generated by non-cryptographic algorithms.
 :::
 
-## rand, rand32
+## rand
 
-Returns a random UInt32 number, evenly distributed across the range of all possible UInt32 numbers.
+Returns a random UInt32 number.
+
+### Syntax
+
+```sql
+rand()
+```
+
+### Parameters
+
+None.
+
+### Output
+
+Returns a number of type UInt32.
+
+### Example
+
+```sql
+SELECT rand()
+```
+
+```response
+1569354847
+```
+
+### Implementation details
 
 Uses a linear congruential generator.
 
+## rand32
+
+Returns a random 32-bit unsigned integer (UInt32) number.
+
+### Syntax
+
+```sql
+rand32()
+```
+
+### Parameters
+
+None.
+
+### Output
+
+Returns a number of type UInt32, evenly distributed across the range of all possible UInt32 values.
+
+### Example
+
+```sql
+SELECT rand32();
+```
+
+```response
+2754546224
+```
+
+**Note:** The actual output will be a random number, not the specific number shown in the example.
+
 ## rand64
 
-Returns a random UInt64 number, evenly distributed across the range of all possible UInt64 numbers.
+Returns a random 64-bit unsigned integer (UInt64) number.
 
-Uses a linear congruential generator.
+### Syntax
+
+```sql
+rand64()
+```
+
+### Parameters
+
+None.
+
+### Implementation details
+
+The `rand64` function uses a linear congruential generator, which means that while it appears random, it's not truly random and can be predictable if the initial state is known.
+
+For scenarios where true randomness is crucial, consider using alternative methods like system-level calls or integrating with external libraries.
+
+### Output
+
+Returns a number of type UInt64, evenly distributed across the range of all possible UInt64 values.
+
+### Example
+
+```sql
+SELECT rand64();
+```
+
+```response
+15030268859237645412
+```
+
+**Note:** The actual output will be a random number, not the specific number shown in the example.
 
 ## randCanonical
 
-Returns a random Float64 value, evenly distributed in interval [0, 1).
+Returns a random floating-point number of type Float64, evenly distributed within the closed interval.
+
+### Syntax
+
+```sql
+randCanonical()
+```
+
+### Parameters
+
+None.
+
+### Output
+
+Returns a Float64 value between 0 (inclusive) and 1 (exclusive).
+
+### Example
+
+```sql
+SELECT randCanonical();
+```
+
+```response
+0.3452178901234567
+```
+
+**Note:** The actual output will be a random decimal number between 0 and 1, not the specific number shown in the example.
 
 ## randConstant
 
-Like `rand` but produces a constant column with a random value.
+Generates a single constant column filled with a random value. Unlike `rand`, `randConstant` ensures the same random value appears in every row of the generated column, making it useful for scenarios requiring a consistent random seed across rows in a single query.
 
-**Example**
+### Syntax
 
-``` sql
-SELECT rand(), rand(1), rand(number), randConstant(), randConstant(1), randConstant(number)
-FROM numbers(3)
+```sql
+randConstant([x]);
 ```
 
-Result:
+### Parameters
 
-``` result
-┌─────rand()─┬────rand(1)─┬─rand(number)─┬─randConstant()─┬─randConstant(1)─┬─randConstant(number)─┐
-│ 3047369878 │ 4132449925 │   4044508545 │     2740811946 │      4229401477 │           1924032898 │
-│ 2938880146 │ 1267722397 │   4154983056 │     2740811946 │      4229401477 │           1924032898 │
-│  956619638 │ 4238287282 │   1104342490 │     2740811946 │      4229401477 │           1924032898 │
-└────────────┴────────────┴──────────────┴────────────────┴─────────────────┴──────────────────────┘
+- **[x] (Optional):** An optional expression that influences the generated random value. Even if provided, the resulting value will still be constant within the same query execution. Different queries using the same expression will likely generate different constant values.
+
+### Implementation details
+
+The actual output will be different for each query execution, even with the same optional expression.
+
+The optional parameter may not significantly change the generated value compared to using `randConstant` alone.
+
+### Output
+
+Returns a column of type UInt32 containing the same random value in each row.
+
+### Examples
+
+```sql
+SELECT randConstant() AS random_value;
+```
+
+```response
+| random_value |
+|--------------|
+| 1234567890   |
+```
+
+```sql
+SELECT randConstant(10) AS random_value;
+```
+
+```response
+| random_value |
+|--------------|
+| 9876543210   |
 ```
 
 ## randUniform
 
-Returns a random Float64 drawn uniformly from interval [`min`, `max`) ([continuous uniform distribution](https://en.wikipedia.org/wiki/Continuous_uniform_distribution)).
+Returns a random Float64 drawn uniformly from interval [`min`, `max`].
 
-**Syntax**
+### Syntax
 
-``` sql
+```sql
 randUniform(min, max)
 ```
 
-**Arguments**
+### Parameters
 
 - `min` - `Float64` - left boundary of the range,
 - `max` - `Float64` - right boundary of the range.
 
-**Returned value**
+### Output
 
-- Random number.
+A random number of type [Float64](/docs/en/sql-reference/data-types/float.md).
 
-Type: [Float64](/docs/en/sql-reference/data-types/float.md).
+### Example
 
-**Example**
-
-``` sql
+```sql
 SELECT randUniform(5.5, 10) FROM numbers(5)
 ```
 
-Result:
-
-``` result
+```response
 ┌─randUniform(5.5, 10)─┐
 │    8.094978491443102 │
 │   7.3181248914450885 │
@@ -99,7 +233,7 @@ Returns a random Float64 drawn from a [normal distribution](https://en.wikipedia
 
 **Syntax**
 
-``` sql
+```sql
 randNormal(mean, variance)
 ```
 
@@ -116,13 +250,13 @@ Type: [Float64](/docs/en/sql-reference/data-types/float.md).
 
 **Example**
 
-``` sql
+```sql
 SELECT randNormal(10, 2) FROM numbers(5)
 ```
 
 Result:
 
-``` result
+```result
 ┌──randNormal(10, 2)─┐
 │ 13.389228911709653 │
 │  8.622949707401295 │
@@ -138,7 +272,7 @@ Returns a random Float64 drawn from a [log-normal distribution](https://en.wikip
 
 **Syntax**
 
-``` sql
+```sql
 randLogNormal(mean, variance)
 ```
 
@@ -155,13 +289,13 @@ Type: [Float64](/docs/en/sql-reference/data-types/float.md).
 
 **Example**
 
-``` sql
+```sql
 SELECT randLogNormal(100, 5) FROM numbers(5)
 ```
 
 Result:
 
-``` result
+```result
 ┌─randLogNormal(100, 5)─┐
 │  1.295699673937363e48 │
 │  9.719869109186684e39 │
@@ -177,7 +311,7 @@ Returns a random UInt64 drawn from a [binomial distribution](https://en.wikipedi
 
 **Syntax**
 
-``` sql
+```sql
 randBinomial(experiments, probability)
 ```
 
@@ -194,13 +328,13 @@ Type: [UInt64](/docs/en/sql-reference/data-types/int-uint.md).
 
 **Example**
 
-``` sql
+```sql
 SELECT randBinomial(100, .75) FROM numbers(5)
 ```
 
 Result:
 
-``` result
+```result
 ┌─randBinomial(100, 0.75)─┐
 │                      74 │
 │                      78 │
@@ -216,7 +350,7 @@ Returns a random UInt64 drawn from a [negative binomial distribution](https://en
 
 **Syntax**
 
-``` sql
+```sql
 randNegativeBinomial(experiments, probability)
 ```
 
@@ -233,13 +367,13 @@ Type: [UInt64](/docs/en/sql-reference/data-types/int-uint.md).
 
 **Example**
 
-``` sql
+```sql
 SELECT randNegativeBinomial(100, .75) FROM numbers(5)
 ```
 
 Result:
 
-``` result
+```result
 ┌─randNegativeBinomial(100, 0.75)─┐
 │                              33 │
 │                              32 │
@@ -255,7 +389,7 @@ Returns a random UInt64 drawn from a [Poisson distribution](https://en.wikipedia
 
 **Syntax**
 
-``` sql
+```sql
 randPoisson(n)
 ```
 
@@ -271,13 +405,13 @@ Type: [UInt64](/docs/en/sql-reference/data-types/int-uint.md).
 
 **Example**
 
-``` sql
+```sql
 SELECT randPoisson(10) FROM numbers(5)
 ```
 
 Result:
 
-``` result
+```result
 ┌─randPoisson(10)─┐
 │               8 │
 │               8 │
@@ -293,7 +427,7 @@ Returns a random UInt64 drawn from a [Bernoulli distribution](https://en.wikiped
 
 **Syntax**
 
-``` sql
+```sql
 randBernoulli(probability)
 ```
 
@@ -309,13 +443,13 @@ Type: [UInt64](/docs/en/sql-reference/data-types/int-uint.md).
 
 **Example**
 
-``` sql
+```sql
 SELECT randBernoulli(.75) FROM numbers(5)
 ```
 
 Result:
 
-``` result
+```result
 ┌─randBernoulli(0.75)─┐
 │                   1 │
 │                   1 │
@@ -331,7 +465,7 @@ Returns a random Float64 drawn from a [exponential distribution](https://en.wiki
 
 **Syntax**
 
-``` sql
+```sql
 randExponential(lambda)
 ```
 
@@ -347,13 +481,13 @@ Type: [Float64](/docs/en/sql-reference/data-types/float.md).
 
 **Example**
 
-``` sql
+```sql
 SELECT randExponential(1/10) FROM numbers(5)
 ```
 
 Result:
 
-``` result
+```result
 ┌─randExponential(divide(1, 10))─┐
 │              44.71628934340778 │
 │              4.211013337903262 │
@@ -369,7 +503,7 @@ Returns a random Float64 drawn from a [Chi-square distribution](https://en.wikip
 
 **Syntax**
 
-``` sql
+```sql
 randChiSquared(degree_of_freedom)
 ```
 
@@ -385,13 +519,13 @@ Type: [Float64](/docs/en/sql-reference/data-types/float.md).
 
 **Example**
 
-``` sql
+```sql
 SELECT randChiSquared(10) FROM numbers(5)
 ```
 
 Result:
 
-``` result
+```result
 ┌─randChiSquared(10)─┐
 │ 10.015463656521543 │
 │  9.621799919882768 │
@@ -407,7 +541,7 @@ Returns a random Float64 drawn from a [Student's t-distribution](https://en.wiki
 
 **Syntax**
 
-``` sql
+```sql
 randStudentT(degree_of_freedom)
 ```
 
@@ -423,13 +557,13 @@ Type: [Float64](/docs/en/sql-reference/data-types/float.md).
 
 **Example**
 
-``` sql
+```sql
 SELECT randStudentT(10) FROM numbers(5)
 ```
 
 Result:
 
-``` result
+```result
 ┌─────randStudentT(10)─┐
 │   1.2217309938538725 │
 │   1.7941971681200541 │
@@ -445,7 +579,7 @@ Returns a random Float64 drawn from a [F-distribution](https://en.wikipedia.org/
 
 **Syntax**
 
-``` sql
+```sql
 randFisherF(d1, d2)
 ```
 
@@ -462,13 +596,13 @@ Type: [Float64](/docs/en/sql-reference/data-types/float.md).
 
 **Example**
 
-``` sql
+```sql
 SELECT randFisherF(10, 3) FROM numbers(5)
 ```
 
 Result:
 
-``` result
+```result
 ┌──randFisherF(10, 3)─┐
 │   7.286287504216609 │
 │ 0.26590779413050386 │
@@ -484,7 +618,7 @@ Generates a string of the specified length filled with random bytes (including z
 
 **Syntax**
 
-``` sql
+```sql
 randomString(length)
 ```
 
@@ -502,13 +636,13 @@ Type: [String](../../sql-reference/data-types/string.md).
 
 Query:
 
-``` sql
+```sql
 SELECT randomString(30) AS str, length(str) AS len FROM numbers(2) FORMAT Vertical;
 ```
 
 Result:
 
-``` text
+```text
 Row 1:
 ──────
 str: 3 G  :   pT ?w тi  k aV f6
@@ -526,7 +660,7 @@ Generates a binary string of the specified length filled with random bytes (incl
 
 **Syntax**
 
-``` sql
+```sql
 randomFixedString(length);
 ```
 
@@ -563,7 +697,7 @@ If you pass `length < 0`, the behavior of the function is undefined.
 
 **Syntax**
 
-``` sql
+```sql
 randomPrintableASCII(length)
 ```
 
@@ -579,11 +713,11 @@ Type: [String](../../sql-reference/data-types/string.md)
 
 **Example**
 
-``` sql
+```sql
 SELECT number, randomPrintableASCII(30) as str, length(str) FROM system.numbers LIMIT 3
 ```
 
-``` text
+```text
 ┌─number─┬─str────────────────────────────┬─length(randomPrintableASCII(30))─┐
 │      0 │ SuiCOSTvC0csfABSw=UcSzp2.`rv8x │                               30 │
 │      1 │ 1Ag NlJ &RCN:*>HVPG;PE-nO"SUFD │                               30 │
@@ -597,7 +731,7 @@ Generates a random string of a specified length. Result string contains valid UT
 
 **Syntax**
 
-``` sql
+```sql
 randomStringUTF8(length);
 ```
 
@@ -635,11 +769,12 @@ Flips the bits of String or FixedString `s`, each with probability `prob`.
 
 **Syntax**
 
-``` sql
+```sql
 fuzzBits(s, prob)
 ```
 
 **Arguments**
+
 - `s` - `String` or `FixedString`,
 - `prob` - constant `Float32/64` between 0.0 and 1.0.
 
@@ -649,14 +784,14 @@ Fuzzed string with same type as `s`.
 
 **Example**
 
-``` sql
+```sql
 SELECT fuzzBits(materialize('abacaba'), 0.1)
 FROM numbers(3)
 ```
 
 Result:
 
-``` result
+```result
 ┌─fuzzBits(materialize('abacaba'), 0.1)─┐
 │ abaaaja                               │
 │ a*cjab+                               │

From e98c30c161303d91c483e7928326f0d8efc1f9df Mon Sep 17 00:00:00 2001
From: johnnymatthews <9611008+johnnymatthews@users.noreply.github.com>
Date: Sun, 3 Mar 2024 14:38:59 -0700
Subject: [PATCH 047/443] Reorganizes rand docs page.

---
 .../functions/random-functions.md             | 30 +++++++++----------
 1 file changed, 14 insertions(+), 16 deletions(-)

diff --git a/docs/en/sql-reference/functions/random-functions.md b/docs/en/sql-reference/functions/random-functions.md
index 2ce9c75eae4..b745d2833d3 100644
--- a/docs/en/sql-reference/functions/random-functions.md
+++ b/docs/en/sql-reference/functions/random-functions.md
@@ -36,19 +36,21 @@ None.
 
 Returns a number of type UInt32.
 
+### Implementation details
+
+Uses a linear congruential generator.
+
 ### Example
 
 ```sql
-SELECT rand()
+SELECT rand();
 ```
 
 ```response
 1569354847
 ```
 
-### Implementation details
-
-Uses a linear congruential generator.
+**Note:** The actual output will be a random number, not the specific number shown in the example.
 
 ## rand32
 
@@ -94,16 +96,14 @@ rand64()
 
 None.
 
-### Implementation details
-
-The `rand64` function uses a linear congruential generator, which means that while it appears random, it's not truly random and can be predictable if the initial state is known.
-
-For scenarios where true randomness is crucial, consider using alternative methods like system-level calls or integrating with external libraries.
-
 ### Output
 
 Returns a number of type UInt64, evenly distributed across the range of all possible UInt64 values.
 
+### Implementation details
+
+The `rand64` function uses a linear congruential generator, which means that while it appears random, it's not truly random and can be predictable if the initial state is known. For scenarios where true randomness is crucial, consider using alternative methods like system-level calls or integrating with external libraries.
+
 ### Example
 
 ```sql
@@ -160,16 +160,14 @@ randConstant([x]);
 
 - **[x] (Optional):** An optional expression that influences the generated random value. Even if provided, the resulting value will still be constant within the same query execution. Different queries using the same expression will likely generate different constant values.
 
-### Implementation details
-
-The actual output will be different for each query execution, even with the same optional expression.
-
-The optional parameter may not significantly change the generated value compared to using `randConstant` alone.
-
 ### Output
 
 Returns a column of type UInt32 containing the same random value in each row.
 
+### Implementation details
+
+The actual output will be different for each query execution, even with the same optional expression. The optional parameter may not significantly change the generated value compared to using `randConstant` alone.
+
 ### Examples
 
 ```sql

From 671b0f678afcdcb354a85aa141920bff09e2bcb2 Mon Sep 17 00:00:00 2001
From: M1eyu2018 <857037797@qq.com>
Date: Mon, 4 Mar 2024 10:12:27 +0800
Subject: [PATCH 048/443] Add positional read in libhdfs3

Signed-off-by: M1eyu2018 <857037797@qq.com>
---
 contrib/libhdfs3 | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/contrib/libhdfs3 b/contrib/libhdfs3
index b9598e60167..0d04201c453 160000
--- a/contrib/libhdfs3
+++ b/contrib/libhdfs3
@@ -1 +1 @@
-Subproject commit b9598e6016720a7c088bfe85ce1fa0410f9d2103
+Subproject commit 0d04201c45359f0d0701fb1e8297d25eff7cfecf

From c435d5894f48d37478454b1934d000fb967e2973 Mon Sep 17 00:00:00 2001
From: taiyang-li <654010905@qq.com>
Date: Mon, 4 Mar 2024 14:23:59 +0800
Subject: [PATCH 049/443] remove wrong assertion n quantileGK

---
 .../AggregateFunctionGroupArray.cpp                 | 13 ++++++++-----
 .../AggregateFunctionQuantileGK.cpp                 | 12 ++++--------
 2 files changed, 12 insertions(+), 13 deletions(-)

diff --git a/src/AggregateFunctions/AggregateFunctionGroupArray.cpp b/src/AggregateFunctions/AggregateFunctionGroupArray.cpp
index d72ddb42d9e..6af8b1018dd 100644
--- a/src/AggregateFunctions/AggregateFunctionGroupArray.cpp
+++ b/src/AggregateFunctions/AggregateFunctionGroupArray.cpp
@@ -182,11 +182,14 @@ public:
 
         if constexpr (Trait::sampler == Sampler::NONE)
         {
-            if (limit_num_elems && cur_elems.value.size() >= max_elems)
+            if constexpr (limit_num_elems)
             {
-                if constexpr (Trait::last)
-                    cur_elems.value[(cur_elems.total_values - 1) % max_elems] = row_value;
-                return;
+                if (cur_elems.value.size() >= max_elems)
+                {
+                    if constexpr (Trait::last)
+                        cur_elems.value[(cur_elems.total_values - 1) % max_elems] = row_value;
+                    return;
+                }
             }
 
             cur_elems.value.push_back(row_value, arena);
@@ -236,7 +239,7 @@ public:
 
     void mergeNoSampler(Data & cur_elems, const Data & rhs_elems, Arena * arena) const
     {
-        if (!limit_num_elems)
+        if constexpr (!limit_num_elems)
         {
             if (rhs_elems.value.size())
                 cur_elems.value.insertByOffsets(rhs_elems.value, 0, rhs_elems.value.size(), arena);
diff --git a/src/AggregateFunctions/AggregateFunctionQuantileGK.cpp b/src/AggregateFunctions/AggregateFunctionQuantileGK.cpp
index 2e8ccb2e5e4..26737e43eef 100644
--- a/src/AggregateFunctions/AggregateFunctionQuantileGK.cpp
+++ b/src/AggregateFunctions/AggregateFunctionQuantileGK.cpp
@@ -144,7 +144,7 @@ public:
             count = other.count;
             compressed = other.compressed;
 
-            sampled.resize(other.sampled.size());
+            sampled.resize_exact(other.sampled.size());
             memcpy(sampled.data(), other.sampled.data(), sizeof(Stats) * other.sampled.size());
             return;
         }
@@ -180,7 +180,7 @@ public:
             compress();
 
             backup_sampled.clear();
-            backup_sampled.reserve(sampled.size() + other.sampled.size());
+            backup_sampled.reserve_exact(sampled.size() + other.sampled.size());
             double merged_relative_error = std::max(relative_error, other.relative_error);
             size_t merged_count = count + other.count;
             Int64 additional_self_delta = static_cast<Int64>(std::floor(2 * other.relative_error * other.count));
@@ -268,11 +268,7 @@ public:
 
         size_t sampled_len = 0;
         readBinaryLittleEndian(sampled_len, buf);
-        if (sampled_len > compress_threshold)
-            throw Exception(
-                ErrorCodes::INCORRECT_DATA, "The number of elements {} for quantileGK exceeds {}", sampled_len, compress_threshold);
-
-        sampled.resize(sampled_len);
+        sampled.resize_exact(sampled_len);
 
         for (size_t i = 0; i < sampled_len; ++i)
         {
@@ -317,7 +313,7 @@ private:
             ::sort(head_sampled.begin(), head_sampled.end());
 
         backup_sampled.clear();
-        backup_sampled.reserve(sampled.size() + head_sampled.size());
+        backup_sampled.reserve_exact(sampled.size() + head_sampled.size());
 
         size_t sample_idx = 0;
         size_t ops_idx = 0;

From 6fbfd42a0522fe4161d367e3d923f2480c1df21a Mon Sep 17 00:00:00 2001
From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com>
Date: Mon, 4 Mar 2024 16:13:44 +0800
Subject: [PATCH 050/443] Update
 02241_filesystem_cache_on_write_operations.reference

---
 .../02241_filesystem_cache_on_write_operations.reference    | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.reference b/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.reference
index 53566a18edc..186dcc1eeb2 100644
--- a/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.reference
+++ b/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.reference
@@ -205,13 +205,7 @@ INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(300, 10000)
 SELECT count(), sum(size) FROM system.filesystem_cache
 24	84045
 SYSTEM START MERGES test_02241
-SELECT value FROM system.events WHERE name = 'CachedWriteBufferCacheWriteBytes'
-81715476
-SELECT value FROM system.events WHERE name = 'CachedReadBufferCacheWriteBytes'
 OPTIMIZE TABLE test_02241 FINAL
-SELECT value FROM system.events WHERE name = 'CachedWriteBufferCacheWriteBytes'
-81881872
-SELECT value FROM system.events WHERE name = 'CachedReadBufferCacheWriteBytes'
 SELECT count(), sum(size) FROM system.filesystem_cache
 32	167243
 ALTER TABLE test_02241 UPDATE value = 'kek' WHERE key = 100

From a7db6688edb50f894457c414b207c25548bb18d3 Mon Sep 17 00:00:00 2001
From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com>
Date: Mon, 4 Mar 2024 18:24:24 +0800
Subject: [PATCH 051/443] Update ObjectStorageFactory.cpp

---
 src/Disks/ObjectStorages/ObjectStorageFactory.cpp | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/src/Disks/ObjectStorages/ObjectStorageFactory.cpp b/src/Disks/ObjectStorages/ObjectStorageFactory.cpp
index 47c02f87b23..a0578ac4454 100644
--- a/src/Disks/ObjectStorages/ObjectStorageFactory.cpp
+++ b/src/Disks/ObjectStorages/ObjectStorageFactory.cpp
@@ -246,12 +246,11 @@ void registerAzureObjectStorage(ObjectStorageFactory & factory)
         bool /* skip_access_check */) -> ObjectStoragePtr
     {
         AzureBlobStorageEndpoint endpoint = processAzureBlobStorageEndpoint(config, config_prefix);
-        return std::make_unique<AzureObjectStorage>(
+        return createObjectStorage<AzureObjectStorage>(
             ObjectStorageType::Azure, config, config_prefix, name,
             getAzureBlobContainerClient(config, config_prefix),
             getAzureBlobStorageSettings(config, config_prefix, context),
             endpoint.prefix.empty() ? endpoint.container_name : endpoint.container_name + "/" + endpoint.prefix);
-
     });
 }
 #endif

From b0050566e22d10ca621a33c1b4fedb987ad2620c Mon Sep 17 00:00:00 2001
From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com>
Date: Tue, 5 Mar 2024 12:14:56 +0800
Subject: [PATCH 052/443] Fix style check

---
 src/Disks/IO/ReadBufferFromRemoteFSGather.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp b/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp
index fe8d63b053d..298000ac015 100644
--- a/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp
+++ b/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp
@@ -21,7 +21,7 @@ namespace
     {
         return settings.remote_fs_cache && settings.enable_filesystem_cache;
     }
-  
+
     bool withPageCache(const ReadSettings & settings, bool with_file_cache)
     {
         return settings.page_cache && !with_file_cache && settings.use_page_cache_for_disks_without_file_cache;

From 2ee846b393d79f3f0d9710ddf910552ba1e040cd Mon Sep 17 00:00:00 2001
From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com>
Date: Tue, 5 Mar 2024 14:07:56 +0800
Subject: [PATCH 053/443] Fix build

---
 src/Disks/IO/ReadBufferFromRemoteFSGather.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp b/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp
index 298000ac015..f72e6634465 100644
--- a/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp
+++ b/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp
@@ -17,7 +17,7 @@ using namespace DB;
 
 namespace
 {
-    bool withCache(const ReadSettings & settings)
+    bool withFileCache(const ReadSettings & settings)
     {
         return settings.remote_fs_cache && settings.enable_filesystem_cache;
     }

From 47ad21dd257ff1a5751d191dfd311a7950a93111 Mon Sep 17 00:00:00 2001
From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com>
Date: Tue, 5 Mar 2024 12:17:04 +0100
Subject: [PATCH 054/443] Remove extra empty line

---
 .../03002_map_array_functions_with_low_cardinality.sql           | 1 -
 1 file changed, 1 deletion(-)

diff --git a/tests/queries/0_stateless/03002_map_array_functions_with_low_cardinality.sql b/tests/queries/0_stateless/03002_map_array_functions_with_low_cardinality.sql
index 8240a8f93f5..8820a433da8 100644
--- a/tests/queries/0_stateless/03002_map_array_functions_with_low_cardinality.sql
+++ b/tests/queries/0_stateless/03002_map_array_functions_with_low_cardinality.sql
@@ -1,2 +1 @@
 SELECT mapContainsKeyLike(map('aa', toLowCardinality(1), 'bb', toLowCardinality(2)), toLowCardinality('a%'));
-

From bfb703b579fa192dc58e51ea842067e7e379e949 Mon Sep 17 00:00:00 2001
From: Blargian <shaunstruwig.ss@gmail.com>
Date: Tue, 5 Mar 2024 15:38:42 +0100
Subject: [PATCH 055/443] Add mortonEncode and mortonDecode to documentation

---
 .../functions/encoding-functions.md           | 203 ++++++++++++++++++
 1 file changed, 203 insertions(+)

diff --git a/docs/en/sql-reference/functions/encoding-functions.md b/docs/en/sql-reference/functions/encoding-functions.md
index 618dd3f4b4f..7fd77ce3a6a 100644
--- a/docs/en/sql-reference/functions/encoding-functions.md
+++ b/docs/en/sql-reference/functions/encoding-functions.md
@@ -433,3 +433,206 @@ Result:
 │ [0,1,2,3,4,5,6,7] │
 └───────────────────┘
 ```
+
+## mortonEncode
+
+Calculates the Morton encoding (ZCurve) for a list of unsigned integers.
+
+The function has two modes of operation:
+- Simple
+- Expanded
+
+### Simple mode
+
+Accepts up to 8 unsigned integers as arguments and produces a UInt64 code.
+
+**Syntax**
+
+```sql
+mortonEncode(args)
+```
+
+**Parameters**
+
+- `args`: up to 8 [unsigned integers](../../sql-reference/data-types/int-uint.md) or columns of the aforementioned type.
+
+**Returned value**
+
+- A UInt64 code
+
+Type: [UInt64](../../sql-reference/data-types/int-uint.md)
+
+**Example**
+
+Query:
+
+```sql
+SELECT mortonEncode(1, 2, 3);
+```
+
+```response
+53
+```
+
+### Expanded mode
+
+Accepts a range mask ([tuple](../../sql-reference/data-types/tuple.md)) as a first argument and up to 8 [unsigned integers](../../sql-reference/data-types/int-uint.md) as other arguments.
+
+Each number in the mask configures the amount of range expansion:
+1 - no expansion
+2 - 2x expansion
+3 - 3x expansion
+...
+Up to 8x expansion.
+
+**Syntax**
+
+```sql
+mortonEncode(range_mask, args)
+```
+
+**Parameters**
+- `range_mask`: 1-8.
+- `args`: up to 8 [unsigned integers](../../sql-reference/data-types/int-uint.md) or columns of the aforementioned type.
+
+Note: when using columns for `args` the provided `range_mask` tuple should still be a constant. 
+
+**Returned value**
+
+- A UInt64 code
+
+Type: [UInt64](../../sql-reference/data-types/int-uint.md)
+
+
+**Example**
+
+Range expansion can be beneficial when you need a similar distribution for arguments with wildly different ranges (or cardinality)
+For example: 'IP Address' (0...FFFFFFFF) and 'Country code' (0...FF).
+
+Query:
+
+```sql
+SELECT mortonEncode((1,2), 1024, 16);
+```
+
+```response
+1572864
+```
+
+Note: tuple size must be equal to the number of the other arguments.
+
+**Example**
+
+Morton encoding for one argument is always the argument itself:
+
+Query:
+
+```sql
+SELECT mortonEncode(1);
+```
+
+```response
+1
+```
+
+**Example**
+
+It is also possible to expand one argument too:
+
+Query:
+
+```sql
+SELECT mortonEncode(tuple(2), 128);
+```
+
+```response
+32768
+```
+
+**implementation details**
+
+Please note that you can fit only so much bits of information into Morton code as [UInt64](../../sql-reference/data-types/int-uint.md) has. Two arguments will have a range of maximum 2^32 (64/2) each, three arguments a range of max 2^21 (64/3) each and so on. All overflow will be clamped to zero.
+
+## mortonDecode
+
+Decodes a Morton encoding (ZCurve) into the corresponding unsigned integer tuple.
+
+As with the `mortonEncode` function, this function has two modes of operation:
+- Simple
+- Expanded
+
+### Simple mode
+
+Accepts a resulting tuple size as the first argument and the code as the second argument.
+
+**Syntax**
+
+```sql
+mortonDecode(tuple_size, code)
+```
+
+**Parameters**
+- `tuple_size`: integer value no more than 8.
+- `code`: [UInt64](../../sql-reference/data-types/int-uint.md) code.
+
+**Returned value**
+
+- [tuple](../../sql-reference/data-types/tuple.md) of the specified size.
+
+Type: [UInt64](../../sql-reference/data-types/int-uint.md)
+
+**Example**
+
+Query:
+
+```sql
+SELECT mortonDecode(3, 53);
+```
+
+```response
+["1","2","3"]
+```
+
+### Expanded mode
+
+Accepts a range mask (tuple) as a first argument and the code as the second argument.
+Each number in the mask configures the amount of range shrink
+1 - no shrink
+2 - 2x shrink
+3 - 3x shrink
+...
+Up to 8x shrink.
+
+Range expansion can be beneficial when you need a similar distribution for arguments with wildly different ranges (or cardinality)
+For example: 'IP Address' (0...FFFFFFFF) and 'Country code' (0...FF).
+As with the encode function, this is limited to 8 numbers at most.
+
+**Example**
+
+Query:
+
+```sql
+SELECT mortonDecode(1, 1);
+```
+
+```response
+["1"]
+```
+
+**Example**
+
+It is also possible to shrink one argument:
+
+Query:
+
+```sql
+SELECT mortonDecode(tuple(2), 32768);
+```
+
+```response
+["128"]
+```
+
+
+
+

From 7930a26df136c8a2e10f839ea4738a338dbb6c9e Mon Sep 17 00:00:00 2001
From: Blargian <shaunstruwig.ss@gmail.com>
Date: Tue, 5 Mar 2024 15:47:45 +0100
Subject: [PATCH 056/443] Fix formatting of compression/expansion levels

---
 .../functions/encoding-functions.md           | 24 +++++++++----------
 1 file changed, 12 insertions(+), 12 deletions(-)

diff --git a/docs/en/sql-reference/functions/encoding-functions.md b/docs/en/sql-reference/functions/encoding-functions.md
index 7fd77ce3a6a..0cb459b8e07 100644
--- a/docs/en/sql-reference/functions/encoding-functions.md
+++ b/docs/en/sql-reference/functions/encoding-functions.md
@@ -478,12 +478,12 @@ SELECT mortonEncode(1, 2, 3);
 
 Accepts a range mask ([tuple](../../sql-reference/data-types/tuple.md)) as a first argument and up to 8 [unsigned integers](../../sql-reference/data-types/int-uint.md) as other arguments.
 
-Each number in the mask configures the amount of range expansion:
-1 - no expansion
-2 - 2x expansion
-3 - 3x expansion
-...
-Up to 8x expansion.
+Each number in the mask configures the amount of range expansion:<br/>
+1 - no expansion<br/>
+2 - 2x expansion<br/>
+3 - 3x expansion<br/>
+...<br/>
+Up to 8x expansion.<br/>
 
 **Syntax**
 
@@ -596,12 +596,12 @@ SELECT mortonDecode(3, 53);
 ### Expanded mode
 
 Accepts a range mask (tuple) as a first argument and the code as the second argument.
-Each number in the mask configures the amount of range shrink
-1 - no shrink
-2 - 2x shrink
-3 - 3x shrink
-...
-Up to 8x shrink.
+Each number in the mask configures the amount of range shrink:<br/>
+1 - no shrink<br/>
+2 - 2x shrink<br/> 
+3 - 3x shrink<br/>
+...<br/>
+Up to 8x shrink.<br/>
 
 Range expansion can be beneficial when you need a similar distribution for arguments with wildly different ranges (or cardinality)
 For example: 'IP Address' (0...FFFFFFFF) and 'Country code' (0...FF).

From 45509607ad139c099c5a4d5fea07ac34149dcf2d Mon Sep 17 00:00:00 2001
From: Blargian <shaunstruwig.ss@gmail.com>
Date: Tue, 5 Mar 2024 15:50:30 +0100
Subject: [PATCH 057/443] Fix spelling mistake

---
 docs/en/sql-reference/functions/encoding-functions.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/en/sql-reference/functions/encoding-functions.md b/docs/en/sql-reference/functions/encoding-functions.md
index 0cb459b8e07..28431c84add 100644
--- a/docs/en/sql-reference/functions/encoding-functions.md
+++ b/docs/en/sql-reference/functions/encoding-functions.md
@@ -551,7 +551,7 @@ SELECT mortonEncode(tuple(2), 128);
 
 **implementation details**
 
-Please note that you can fit only so much bits of information into Morton code as [UInt64](../../sql-reference/data-types/int-uint.md) has. Two arguments will have a range of maximum 2^32 (64/2) each, three arguments a range of max 2^21 (64/3) each and so on. All overflow will be clamped to zero.
+Please note that you can fit only so many bits of information into Morton code as [UInt64](../../sql-reference/data-types/int-uint.md) has. Two arguments will have a range of maximum 2^32 (64/2) each, three arguments a range of max 2^21 (64/3) each and so on. All overflow will be clamped to zero.
 
 ## mortonDecode
 

From aa6b70e5f2187be71b6bce835ecff0aa0c0bfca7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <benjamin.antal@clickhouse.com>
Date: Tue, 5 Mar 2024 16:55:08 +0000
Subject: [PATCH 058/443] Add documentation to `simpleJSON` functions

---
 .../sql-reference/functions/json-functions.md | 392 +++++++++++++++---
 1 file changed, 342 insertions(+), 50 deletions(-)

diff --git a/docs/en/sql-reference/functions/json-functions.md b/docs/en/sql-reference/functions/json-functions.md
index 2c837ff4a42..246cb8972fb 100644
--- a/docs/en/sql-reference/functions/json-functions.md
+++ b/docs/en/sql-reference/functions/json-functions.md
@@ -5,80 +5,372 @@ sidebar_label: JSON
 ---
 
 There are two sets of functions to parse JSON.
-   - `visitParam*` (`simpleJSON*`) is made to parse a special very limited subset of a JSON, but these functions are extremely fast.
+   - `simpleJSON*` (`visitParam*`) is made to parse a special very limited subset of a JSON, but these functions are extremely fast.
    - `JSONExtract*` is made to parse normal JSON.
 
-# visitParam functions
+# simpleJSON/visitParam functions
 
 ClickHouse has special functions for working with simplified JSON. All these JSON functions are based on strong assumptions about what the JSON can be, but they try to do as little as possible to get the job done.
 
 The following assumptions are made:
 
 1.  The field name (function argument) must be a constant.
-2.  The field name is somehow canonically encoded in JSON. For example: `visitParamHas('{"abc":"def"}', 'abc') = 1`, but `visitParamHas('{"\\u0061\\u0062\\u0063":"def"}', 'abc') = 0`
+2.  The field name is somehow canonically encoded in JSON. For example: `simpleJSONHas('{"abc":"def"}', 'abc') = 1`, but `simpleJSONHas('{"\\u0061\\u0062\\u0063":"def"}', 'abc') = 0`
 3.  Fields are searched for on any nesting level, indiscriminately. If there are multiple matching fields, the first occurrence is used.
 4.  The JSON does not have space characters outside of string literals.
 
-## visitParamHas(params, name)
+## simpleJSONHas
 
-Checks whether there is a field with the `name` name.
+Checks whether there is a field named `field_name`.  The result is `UInt8`.
 
-Alias: `simpleJSONHas`.
+**Syntax**
 
-## visitParamExtractUInt(params, name)
-
-Parses UInt64 from the value of the field named `name`. If this is a string field, it tries to parse a number from the beginning of the string. If the field does not exist, or it exists but does not contain a number, it returns 0.
-
-Alias: `simpleJSONExtractUInt`.
-
-## visitParamExtractInt(params, name)
-
-The same as for Int64.
-
-Alias: `simpleJSONExtractInt`.
-
-## visitParamExtractFloat(params, name)
-
-The same as for Float64.
-
-Alias: `simpleJSONExtractFloat`.
-
-## visitParamExtractBool(params, name)
-
-Parses a true/false value. The result is UInt8.
-
-Alias: `simpleJSONExtractBool`.
-
-## visitParamExtractRaw(params, name)
-
-Returns the value of a field, including separators.
-
-Alias: `simpleJSONExtractRaw`.
-
-Examples:
-
-``` sql
-visitParamExtractRaw('{"abc":"\\n\\u0000"}', 'abc') = '"\\n\\u0000"';
-visitParamExtractRaw('{"abc":{"def":[1,2,3]}}', 'abc') = '{"def":[1,2,3]}';
+```sql
+simpleJSONHas(json, field_name)
 ```
 
-## visitParamExtractString(params, name)
+**Parameters**
 
-Parses the string in double quotes. The value is unescaped. If unescaping failed, it returns an empty string.
+- `json`: The JSON in which the field is searched for. [String](../../sql-reference/data-types/string.md#string)
+- `field_name`: The name of the field to search for. [String literal](../syntax#string)
 
-Alias: `simpleJSONExtractString`.
+**Returned value**
 
-Examples:
+It returns `1` if the field exists, `0` otherwise.
 
-``` sql
-visitParamExtractString('{"abc":"\\n\\u0000"}', 'abc') = '\n\0';
-visitParamExtractString('{"abc":"\\u263a"}', 'abc') = '☺';
-visitParamExtractString('{"abc":"\\u263"}', 'abc') = '';
-visitParamExtractString('{"abc":"hello}', 'abc') = '';
+**Example**
+
+Query:
+
+```sql
+CREATE TABLE jsons
+(
+    `json` String
+)
+ENGINE = Memory;
+
+INSERT INTO jsons VALUES ('{"foo":"true","qux":1}');
+
+SELECT simpleJSONHas(json, 'foo') FROM jsons;
+SELECT simpleJSONHas(json, 'bar') FROM jsons;
 ```
 
+```response
+1
+0
+```
+## simpleJSONExtractUInt
+
+Parses `UInt64` from the value of the field named `field_name`. If this is a string field, it tries to parse a number from the beginning of the string. If the field does not exist, or it exists but does not contain a number, it returns `0`.
+
+**Syntax**
+
+```sql
+simpleJSONExtractUInt(json, field_name)
+```
+
+**Parameters**
+
+- `json`: The JSON in which the field is searched for. [String](../../sql-reference/data-types/string.md#string)
+- `field_name`: The name of the field to search for. [String literal](../syntax#string)
+
+**Returned value**
+
+It returns the number parsed from the field if the field exists and contains a number, `0` otherwise.
+
+**Example**
+
+Query:
+
+```sql
+CREATE TABLE jsons
+(
+    `json` String
+)
+ENGINE = Memory;
+
+INSERT INTO jsons VALUES ('{"foo":"4e3"}');
+INSERT INTO jsons VALUES ('{"foo":3.4}');
+INSERT INTO jsons VALUES ('{"foo":5}');
+INSERT INTO jsons VALUES ('{"foo":"not1number"}');
+INSERT INTO jsons VALUES ('{"baz":2}');
+
+SELECT simpleJSONExtractUInt(json, 'foo') FROM jsons ORDER BY json;
+```
+
+```response
+0
+4
+0
+3
+5
+```
+
+## simpleJSONExtractInt
+
+Parses `Int64` from the value of the field named `field_name`. If this is a string field, it tries to parse a number from the beginning of the string. If the field does not exist, or it exists but does not contain a number, it returns `0`.
+
+**Syntax**
+
+```sql
+simpleJSONExtractInt(json, field_name)
+```
+
+**Parameters**
+
+- `json`: The JSON in which the field is searched for. [String](../../sql-reference/data-types/string.md#string)
+- `field_name`: The name of the field to search for. [String literal](../syntax#string)
+
+**Returned value**
+
+It returns the number parsed from the field if the field exists and contains a number, `0` otherwise.
+
+**Example**
+
+Query:
+
+```sql
+CREATE TABLE jsons
+(
+    `json` String
+)
+ENGINE = Memory;
+
+INSERT INTO jsons VALUES ('{"foo":"-4e3"}');
+INSERT INTO jsons VALUES ('{"foo":-3.4}');
+INSERT INTO jsons VALUES ('{"foo":5}');
+INSERT INTO jsons VALUES ('{"foo":"not1number"}');
+INSERT INTO jsons VALUES ('{"baz":2}');
+
+SELECT simpleJSONExtractInt(json, 'foo') FROM jsons ORDER BY json;
+```
+
+```response
+0
+-4
+0
+-3
+5
+```
+
+## simpleJSONExtractFloat
+
+Parses `Float64` from the value of the field named `field_name`. If this is a string field, it tries to parse a number from the beginning of the string. If the field does not exist, or it exists but does not contain a number, it returns `0`.
+
+**Syntax**
+
+```sql
+simpleJSONExtractFloat(json, field_name)
+```
+
+**Parameters**
+
+- `json`: The JSON in which the field is searched for. [String](../../sql-reference/data-types/string.md#string)
+- `field_name`: The name of the field to search for. [String literal](../syntax#string)
+
+**Returned value**
+
+It returns the number parsed from the field if the field exists and contains a number, `0` otherwise.
+
+**Example**
+
+Query:
+
+```sql
+CREATE TABLE jsons
+(
+    `json` String
+)
+ENGINE = Memory;
+
+INSERT INTO jsons VALUES ('{"foo":"-4e3"}');
+INSERT INTO jsons VALUES ('{"foo":-3.4}');
+INSERT INTO jsons VALUES ('{"foo":5}');
+INSERT INTO jsons VALUES ('{"foo":"not1number"}');
+INSERT INTO jsons VALUES ('{"baz":2}');
+
+SELECT simpleJSONExtractFloat(json, 'foo') FROM jsons ORDER BY json;
+```
+
+```response
+0
+-4000
+0
+-3.4
+5
+```
+
+## simpleJSONExtractBool
+
+Parses a true/false value from the value of the field named `field_name`. The result is `UInt8`.
+
+**Syntax**
+
+```sql
+simpleJSONExtractBool(json, field_name)
+```
+
+**Parameters**
+
+- `json`: The JSON in which the field is searched for. [String](../../sql-reference/data-types/string.md#string)
+- `field_name`: The name of the field to search for. [String literal](../syntax#string)
+
+**Returned value**
+
+It returns `1` if the value of the field is `true`, `0` otherwise. This means this function will return `0` including (and not only) in the following cases:
+ - If the field doesn't exists.
+ - If the field contains `true` as a string, e.g.: `{"field":"true"}`.
+ - If the field contains `1` as a numerical value.
+
+**Example**
+
+Query:
+
+```sql
+CREATE TABLE jsons
+(
+    `json` String
+)
+ENGINE = Memory;
+
+INSERT INTO jsons VALUES ('{"foo":false,"bar":true}');
+INSERT INTO jsons VALUES ('{"foo":"true","qux":1}');
+
+SELECT simpleJSONExtractBool(json, 'bar') FROM jsons ORDER BY json;
+SELECT simpleJSONExtractBool(json, 'foo') FROM jsons ORDER BY json;
+```
+
+```response
+0
+1
+0
+0
+```
+
+## simpleJSONExtractRaw
+
+Returns the value of the field named `field_name` as a `String`, including separators.
+
+**Syntax**
+
+```sql
+simpleJSONExtractRaw(json, field_name)
+```
+
+**Parameters**
+
+- `json`: The JSON in which the field is searched for. [String](../../sql-reference/data-types/string.md#string)
+- `field_name`: The name of the field to search for. [String literal](../syntax#string)
+
+**Returned value**
+
+It returns the value of the field as a [`String`](../../sql-reference/data-types/string.md#string), including separators if the field exists, or an emtpy `String` otherwise.
+
+**Example**
+
+Query:
+
+```sql
+CREATE TABLE jsons
+(
+    `json` String
+)
+ENGINE = Memory;
+
+INSERT INTO jsons VALUES ('{"foo":"-4e3"}');
+INSERT INTO jsons VALUES ('{"foo":-3.4}');
+INSERT INTO jsons VALUES ('{"foo":5}');
+INSERT INTO jsons VALUES ('{"foo":{"def":[1,2,3]}}');
+INSERT INTO jsons VALUES ('{"baz":2}');
+
+SELECT simpleJSONExtractRaw(json, 'foo') FROM jsons ORDER BY json;
+```
+
+```response
+
+"-4e3"
+-3.4
+5
+{"def":[1,2,3]}
+```
+
+## simpleJSONExtractString
+
+Parses `String` in double quotes from the value of the field named `field_name`.
+
+**Syntax**
+
+```sql
+simpleJSONExtractString(json, field_name)
+```
+
+**Parameters**
+
+- `json`: The JSON in which the field is searched for. [String](../../sql-reference/data-types/string.md#string)
+- `field_name`: The name of the field to search for. [String literal](../syntax#string)
+
+**Returned value**
+
+It returns the value of a field as a [`String`](../../sql-reference/data-types/string.md#string), including separators. The value is unescaped. It returns an empty `String`: if the field doesn't contain a double quoted string, if unescaping fails or if the field doesn't exist.
+
+**Implementation details**
+
 There is currently no support for code points in the format `\uXXXX\uYYYY` that are not from the basic multilingual plane (they are converted to CESU-8 instead of UTF-8).
 
+**Example**
+
+Query:
+
+```sql
+CREATE TABLE jsons
+(
+    `json` String
+)
+ENGINE = Memory;
+
+INSERT INTO jsons VALUES ('{"foo":"\\n\\u0000"}');
+INSERT INTO jsons VALUES ('{"foo":"\\u263"}');
+INSERT INTO jsons VALUES ('{"foo":"\\u263a"}');
+INSERT INTO jsons VALUES ('{"foo":"hello}');
+
+SELECT simpleJSONExtractString(json, 'foo') FROM jsons ORDER BY json;
+```
+
+```response
+\n\0
+
+☺
+
+```
+
+## visitParamHas
+
+This function is [an alias of `simpleJSONHas`](./json-functions#simplejsonhas).
+
+## visitParamExtractUInt
+
+This function is [an alias of `simpleJSONExtractUInt`](./json-functions#simplejsonextractuint).
+
+## visitParamExtractInt
+
+This function is [an alias of `simpleJSONExtractInt`](./json-functions#simplejsonextractint).
+
+## visitParamExtractFloat
+
+This function is [an alias of `simpleJSONExtractFloat`](./json-functions#simplejsonextractfloat).
+
+## visitParamExtractBool
+
+This function is [an alias of `simpleJSONExtractBool`](./json-functions#simplejsonextractbool).
+
+## visitParamExtractRaw
+
+This function is [an alias of `simpleJSONExtractRaw`](./json-functions#simplejsonextractraw).
+
+## visitParamExtractString
+
+This function is [an alias of `simpleJSONExtractString`](./json-functions#simplejsonextractstring).
+
 # JSONExtract functions
 
 The following functions are based on [simdjson](https://github.com/lemire/simdjson) designed for more complex JSON parsing requirements.

From 981c507d8007a4f7761a83a2ecfa0956a364317d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <benjamin.antal@clickhouse.com>
Date: Tue, 5 Mar 2024 17:01:54 +0000
Subject: [PATCH 059/443] Add example to `sin`.

---
 docs/en/sql-reference/functions/math-functions.md | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/docs/en/sql-reference/functions/math-functions.md b/docs/en/sql-reference/functions/math-functions.md
index b27668caf0c..fc659891b5c 100644
--- a/docs/en/sql-reference/functions/math-functions.md
+++ b/docs/en/sql-reference/functions/math-functions.md
@@ -299,6 +299,18 @@ sin(x)
 
 Type: [Float*](../../sql-reference/data-types/float.md).
 
+**Example**
+
+Query:
+
+```sql
+SELECT sin(1.23);
+```
+
+```response
+0.9424888019316975
+```
+
 ## cos
 
 Returns the cosine of the argument.

From 57670a69be7aee37141aad13c3c9509ea2a40162 Mon Sep 17 00:00:00 2001
From: Shaun Struwig <41984034+Blargian@users.noreply.github.com>
Date: Wed, 6 Mar 2024 04:15:57 +0100
Subject: [PATCH 060/443] Add mortonEncode, mortonDecode and related to
 spelling exceptions

---
 utils/check-style/aspell-ignore/en/aspell-dict.txt | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/utils/check-style/aspell-ignore/en/aspell-dict.txt b/utils/check-style/aspell-ignore/en/aspell-dict.txt
index f61448b2f35..6257b2fcd95 100644
--- a/utils/check-style/aspell-ignore/en/aspell-dict.txt
+++ b/utils/check-style/aspell-ignore/en/aspell-dict.txt
@@ -260,6 +260,7 @@ ExactEdgeLengthRads
 ExecutablePool
 ExtType
 ExternalDistributed
+FFFFFFFF
 FFFD
 FIPS
 FOSDEM
@@ -546,6 +547,8 @@ MinIO
 MinMax
 MindsDB
 Mongodb
+mortonDecode
+mortonEncode
 MsgPack
 MultiPolygon
 Multiline
@@ -2741,6 +2744,7 @@ xz
 yaml
 yandex
 youtube
+ZCurve
 zLib
 zLinux
 zabbix

From 68a3ca37c40db7f9b928d9f20bde6912ba6bd7da Mon Sep 17 00:00:00 2001
From: Shaun Struwig <41984034+Blargian@users.noreply.github.com>
Date: Wed, 6 Mar 2024 06:42:01 +0100
Subject: [PATCH 061/443] Add examples using columns

---
 .../functions/encoding-functions.md           | 85 +++++++++++++++++++
 1 file changed, 85 insertions(+)

diff --git a/docs/en/sql-reference/functions/encoding-functions.md b/docs/en/sql-reference/functions/encoding-functions.md
index 28431c84add..c81b3e35317 100644
--- a/docs/en/sql-reference/functions/encoding-functions.md
+++ b/docs/en/sql-reference/functions/encoding-functions.md
@@ -515,6 +515,8 @@ Query:
 SELECT mortonEncode((1,2), 1024, 16);
 ```
 
+Result:
+
 ```response
 1572864
 ```
@@ -531,6 +533,8 @@ Query:
 SELECT mortonEncode(1);
 ```
 
+Result:
+
 ```response
 1
 ```
@@ -545,10 +549,49 @@ Query:
 SELECT mortonEncode(tuple(2), 128);
 ```
 
+Result:
+
 ```response
 32768
 ```
 
+**Example**
+
+You can also use column names in the function.
+
+Query:
+
+First create the table and insert some data.
+
+```sql
+create table morton_numbers(
+    n1 UInt32,
+    n2 UInt32,
+    n3 UInt16,
+    n4 UInt16,
+    n5 UInt8,
+    n6 UInt8,
+    n7 UInt8,
+    n8 UInt8
+)
+Engine=MergeTree()
+ORDER BY n1 SETTINGS index_granularity = 8192, index_granularity_bytes = '10Mi';
+insert into morton_numbers (*) values(1,2,3,4,5,6,7,8);
+```
+Use column names instead of constants as function arguments to `mortonEncode`
+
+Query:
+
+```sql
+SELECT mortonEncode(n1, n2, n3, n4, n5, n6, n7, n8) FROM morton_numbers;
+```
+
+Result:
+
+```response
+2155374165
+```
+
 **implementation details**
 
 Please note that you can fit only so many bits of information into Morton code as [UInt64](../../sql-reference/data-types/int-uint.md) has. Two arguments will have a range of maximum 2^32 (64/2) each, three arguments a range of max 2^21 (64/3) each and so on. All overflow will be clamped to zero.
@@ -589,6 +632,8 @@ Query:
 SELECT mortonDecode(3, 53);
 ```
 
+Result:
+
 ```response
 ["1","2","3"]
 ```
@@ -615,6 +660,8 @@ Query:
 SELECT mortonDecode(1, 1);
 ```
 
+Result:
+
 ```response
 ["1"]
 ```
@@ -629,10 +676,48 @@ Query:
 SELECT mortonDecode(tuple(2), 32768);
 ```
 
+Result:
+
 ```response
 ["128"]
 ```
 
+**Example**
+
+You can also use column names in the function.
+
+First create the table and insert some data.
+
+Query:
+```sql
+create table morton_numbers(
+    n1 UInt32,
+    n2 UInt32,
+    n3 UInt16,
+    n4 UInt16,
+    n5 UInt8,
+    n6 UInt8,
+    n7 UInt8,
+    n8 UInt8
+)
+Engine=MergeTree()
+ORDER BY n1 SETTINGS index_granularity = 8192, index_granularity_bytes = '10Mi';
+insert into morton_numbers (*) values(1,2,3,4,5,6,7,8);
+```
+Use column names instead of constants as function arguments to `mortonDecode`
+
+Query:
+
+```sql
+select untuple(mortonDecode(8, mortonEncode(n1, n2, n3, n4, n5, n6, n7, n8))) from morton_numbers;
+```
+
+Result:
+
+```response
+1	2	3	4	5	6	7	8
+```
+
 
 
 

From 6d4514c045cc565919f9c8384710eee89354f0f3 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Wed, 6 Mar 2024 16:55:48 +0800
Subject: [PATCH 062/443] Fix test

---
 src/Storages/System/StorageSystemDisks.cpp       | 10 +++++++++-
 tests/integration/test_backup_restore_s3/test.py | 12 ++++++------
 2 files changed, 15 insertions(+), 7 deletions(-)

diff --git a/src/Storages/System/StorageSystemDisks.cpp b/src/Storages/System/StorageSystemDisks.cpp
index 30d64156b22..0f8a6640f2c 100644
--- a/src/Storages/System/StorageSystemDisks.cpp
+++ b/src/Storages/System/StorageSystemDisks.cpp
@@ -25,6 +25,8 @@ StorageSystemDisks::StorageSystemDisks(const StorageID & table_id_)
         {"unreserved_space", std::make_shared<DataTypeUInt64>()},
         {"keep_free_space", std::make_shared<DataTypeUInt64>()},
         {"type", std::make_shared<DataTypeString>()},
+        {"object_storage_type", std::make_shared<DataTypeString>()},
+        {"metadata_type", std::make_shared<DataTypeString>()},
         {"is_encrypted", std::make_shared<DataTypeUInt8>()},
         {"is_read_only", std::make_shared<DataTypeUInt8>()},
         {"is_write_once", std::make_shared<DataTypeUInt8>()},
@@ -53,6 +55,8 @@ Pipe StorageSystemDisks::read(
     MutableColumnPtr col_unreserved = ColumnUInt64::create();
     MutableColumnPtr col_keep = ColumnUInt64::create();
     MutableColumnPtr col_type = ColumnString::create();
+    MutableColumnPtr col_object_storage_type = ColumnString::create();
+    MutableColumnPtr col_metadata_type = ColumnString::create();
     MutableColumnPtr col_is_encrypted = ColumnUInt8::create();
     MutableColumnPtr col_is_read_only = ColumnUInt8::create();
     MutableColumnPtr col_is_write_once = ColumnUInt8::create();
@@ -69,7 +73,9 @@ Pipe StorageSystemDisks::read(
         col_unreserved->insert(disk_ptr->getUnreservedSpace().value_or(std::numeric_limits<UInt64>::max()));
         col_keep->insert(disk_ptr->getKeepingFreeSpace());
         auto data_source_description = disk_ptr->getDataSourceDescription();
-        col_type->insert(data_source_description.toString());
+        col_type->insert(data_source_description.type);
+        col_object_storage_type->insert(data_source_description.object_storage_type);
+        col_metadata_type->insert(data_source_description.metadata_type);
         col_is_encrypted->insert(data_source_description.is_encrypted);
         col_is_read_only->insert(disk_ptr->isReadOnly());
         col_is_write_once->insert(disk_ptr->isWriteOnce());
@@ -91,6 +97,8 @@ Pipe StorageSystemDisks::read(
     res_columns.emplace_back(std::move(col_unreserved));
     res_columns.emplace_back(std::move(col_keep));
     res_columns.emplace_back(std::move(col_type));
+    res_columns.emplace_back(std::move(col_object_storage_type));
+    res_columns.emplace_back(std::move(col_metadata_type));
     res_columns.emplace_back(std::move(col_is_encrypted));
     res_columns.emplace_back(std::move(col_is_read_only));
     res_columns.emplace_back(std::move(col_is_write_once));
diff --git a/tests/integration/test_backup_restore_s3/test.py b/tests/integration/test_backup_restore_s3/test.py
index 4d3ee8200a3..95e264107e4 100644
--- a/tests/integration/test_backup_restore_s3/test.py
+++ b/tests/integration/test_backup_restore_s3/test.py
@@ -124,15 +124,15 @@ def check_backup_and_restore(
 def check_system_tables(backup_query_id=None):
     disks = [
         tuple(disk.split("\t"))
-        for disk in node.query("SELECT name, type FROM system.disks").split("\n")
+        for disk in node.query("SELECT name, type, object_storage_type, metadata_type FROM system.disks").split("\n")
         if disk
     ]
     expected_disks = (
-        ("default", "local"),
-        ("disk_s3", "s3"),
-        ("disk_s3_cache", "s3"),
-        ("disk_s3_other_bucket", "s3"),
-        ("disk_s3_plain", "s3_plain"),
+        ("default", "local", "", ""),
+        ("disk_s3", "object_storage", "s3", "local"),
+        ("disk_s3_cache", "object_storage", "s3", "local"),
+        ("disk_s3_other_bucket", "object_storage", "s3", "local"),
+        ("disk_s3_plain", "object_storage", "s3", "plain"),
     )
     assert len(expected_disks) == len(disks)
     for expected_disk in expected_disks:

From be98c95f586762cdf20a6375917e30f296175593 Mon Sep 17 00:00:00 2001
From: robot-clickhouse <robot-clickhouse@users.noreply.github.com>
Date: Wed, 6 Mar 2024 09:12:26 +0000
Subject: [PATCH 063/443] Automatic style fix

---
 tests/integration/test_backup_restore_s3/test.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/tests/integration/test_backup_restore_s3/test.py b/tests/integration/test_backup_restore_s3/test.py
index 95e264107e4..452a9143067 100644
--- a/tests/integration/test_backup_restore_s3/test.py
+++ b/tests/integration/test_backup_restore_s3/test.py
@@ -124,7 +124,9 @@ def check_backup_and_restore(
 def check_system_tables(backup_query_id=None):
     disks = [
         tuple(disk.split("\t"))
-        for disk in node.query("SELECT name, type, object_storage_type, metadata_type FROM system.disks").split("\n")
+        for disk in node.query(
+            "SELECT name, type, object_storage_type, metadata_type FROM system.disks"
+        ).split("\n")
         if disk
     ]
     expected_disks = (

From 56fb61e1866e81e9a00b9b98299ddc56a54f5394 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <benjamin.antal@clickhouse.com>
Date: Wed, 6 Mar 2024 10:53:39 +0000
Subject: [PATCH 064/443] Do not duplicate the first category in case of
 multiple categories in `FunctionDocumentation`

---
 src/Common/FunctionDocumentation.cpp | 1 +
 1 file changed, 1 insertion(+)

diff --git a/src/Common/FunctionDocumentation.cpp b/src/Common/FunctionDocumentation.cpp
index 2aad23b90b7..0dc5b48f9d1 100644
--- a/src/Common/FunctionDocumentation.cpp
+++ b/src/Common/FunctionDocumentation.cpp
@@ -36,6 +36,7 @@ std::string FunctionDocumentation::categoriesAsString() const
 
     auto it = categories.begin();
     std::string res = *it;
+    ++it;
     for (; it != categories.end(); ++it)
         res += ", " + *it;
     return res;

From 6f726865baf3fea606e7ff46e5d8cd98bda94f5c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <benjamin.antal@clickhouse.com>
Date: Wed, 6 Mar 2024 11:10:02 +0000
Subject: [PATCH 065/443] Add inline docs to functions

---
 src/Functions/sin.cpp                         | 10 +++++-
 src/Functions/visitParamExtractBool.cpp       | 30 +++++++++++++++++-
 src/Functions/visitParamExtractFloat.cpp      | 31 ++++++++++++++++++-
 src/Functions/visitParamExtractInt.cpp        | 31 ++++++++++++++++++-
 src/Functions/visitParamExtractRaw.cpp        | 30 +++++++++++++++++-
 src/Functions/visitParamExtractString.cpp     | 30 +++++++++++++++++-
 src/Functions/visitParamExtractUInt.cpp       | 31 ++++++++++++++++++-
 src/Functions/visitParamHas.cpp               | 23 +++++++++++++-
 ...new_functions_must_be_documented.reference |  8 -----
 9 files changed, 208 insertions(+), 16 deletions(-)

diff --git a/src/Functions/sin.cpp b/src/Functions/sin.cpp
index dc75f4800c0..914f431adb4 100644
--- a/src/Functions/sin.cpp
+++ b/src/Functions/sin.cpp
@@ -13,7 +13,15 @@ using FunctionSin = FunctionMathUnary<UnaryFunctionVectorized<SinName, sin>>;
 
 REGISTER_FUNCTION(Sin)
 {
-    factory.registerFunction<FunctionSin>({}, FunctionFactory::CaseInsensitive);
+    factory.registerFunction<FunctionSin>(
+        FunctionDocumentation{
+            .description = "Returns the sine of the argument.",
+            .syntax = "sin(x)",
+            .arguments = {{"x", "The number whose sine will be returned. (U)Int*, Float* or Decimal*."}},
+            .returned_value = "The sine of x.",
+            .examples = {{.name = "simple", .query = "SELECT sin(1.23)", .result = "0.9424888019316975"}},
+            .categories{"Mathematical", "Trigonometric"}},
+        FunctionFactory::CaseInsensitive);
 }
 
 }
diff --git a/src/Functions/visitParamExtractBool.cpp b/src/Functions/visitParamExtractBool.cpp
index 31763fe54ce..2c413ec13bb 100644
--- a/src/Functions/visitParamExtractBool.cpp
+++ b/src/Functions/visitParamExtractBool.cpp
@@ -21,7 +21,35 @@ using FunctionSimpleJSONExtractBool = FunctionsStringSearch<ExtractParamImpl<Nam
 
 REGISTER_FUNCTION(VisitParamExtractBool)
 {
-    factory.registerFunction<FunctionSimpleJSONExtractBool>();
+    factory.registerFunction<FunctionSimpleJSONExtractBool>(FunctionDocumentation{
+        .description = "Parses a true/false value from the value of the field named field_name. The result is UInt8.",
+        .syntax = "simpleJSONExtractBool(json, field_name)",
+        .arguments
+        = {{"json", "The JSON in which the field is searched for. String."},
+           {"field_name", "The name of the field to search for. String literal."}},
+        .returned_value
+        = R"(It returns 1 if the value of the field is true, 0 otherwise. This means this function will return 0 including (and not only) in the following cases:
+ - If the field doesn't exists.
+ - If the field contains true as a string, e.g.: {"field":"true"}.
+ - If the field contains 1 as a numerical value.)",
+        .examples
+        = {{.name = "simple",
+            .query = R"(CREATE TABLE jsons
+(
+    json String
+)
+ENGINE = Memory;
+
+INSERT INTO jsons VALUES ('{"foo":false,"bar":true}');
+INSERT INTO jsons VALUES ('{"foo":"true","qux":1}');
+
+SELECT simpleJSONExtractBool(json, 'bar') FROM jsons ORDER BY json;
+SELECT simpleJSONExtractBool(json, 'foo') FROM jsons ORDER BY json;)",
+            .result = R"(0
+1
+0
+0)"}},
+        .categories{"JSON"}});
     factory.registerAlias("visitParamExtractBool", "simpleJSONExtractBool");
 }
 
diff --git a/src/Functions/visitParamExtractFloat.cpp b/src/Functions/visitParamExtractFloat.cpp
index 6f6d5274050..fc839142cc7 100644
--- a/src/Functions/visitParamExtractFloat.cpp
+++ b/src/Functions/visitParamExtractFloat.cpp
@@ -11,7 +11,36 @@ using FunctionSimpleJSONExtractFloat = FunctionsStringSearch<ExtractParamImpl<Na
 
 REGISTER_FUNCTION(VisitParamExtractFloat)
 {
-    factory.registerFunction<FunctionSimpleJSONExtractFloat>();
+    factory.registerFunction<FunctionSimpleJSONExtractFloat>(FunctionDocumentation{
+        .description
+        = "Parses Float64 from the value of the field named field_name. If this is a string field, it tries to parse a number from the "
+          "beginning of the string. If the field does not exist, or it exists but does not contain a number, it returns 0.",
+        .syntax = "simpleJSONExtractFloat(json, field_name)",
+        .arguments
+        = {{"json", "The JSON in which the field is searched for. String."},
+           {"field_name", "The name of the field to search for. String literal."}},
+        .returned_value = "It returns the number parsed from the field if the field exists and contains a number, 0 otherwise.",
+        .examples
+        = {{.name = "simple",
+            .query = R"(CREATE TABLE jsons
+(
+    json String
+)
+ENGINE = Memory;
+
+INSERT INTO jsons VALUES ('{"foo":"-4e3"}');
+INSERT INTO jsons VALUES ('{"foo":-3.4}');
+INSERT INTO jsons VALUES ('{"foo":5}');
+INSERT INTO jsons VALUES ('{"foo":"not1number"}');
+INSERT INTO jsons VALUES ('{"baz":2}');
+
+SELECT simpleJSONExtractFloat(json, 'foo') FROM jsons ORDER BY json;)",
+            .result = R"(0
+-4000
+0
+-3.4
+5)"}},
+        .categories{"JSON"}});
     factory.registerAlias("visitParamExtractFloat", "simpleJSONExtractFloat");
 }
 
diff --git a/src/Functions/visitParamExtractInt.cpp b/src/Functions/visitParamExtractInt.cpp
index e020c43e8b4..4588fc55c52 100644
--- a/src/Functions/visitParamExtractInt.cpp
+++ b/src/Functions/visitParamExtractInt.cpp
@@ -11,7 +11,36 @@ using FunctionSimpleJSONExtractInt = FunctionsStringSearch<ExtractParamImpl<Name
 
 REGISTER_FUNCTION(VisitParamExtractInt)
 {
-    factory.registerFunction<FunctionSimpleJSONExtractInt>();
+    factory.registerFunction<FunctionSimpleJSONExtractInt>(FunctionDocumentation{
+        .description
+        = "Parses Int64 from the value of the field named field_name. If this is a string field, it tries to parse a number from the "
+          "beginning of the string. If the field does not exist, or it exists but does not contain a number, it returns 0.",
+        .syntax = "simpleJSONExtractInt(json, field_name)",
+        .arguments
+        = {{"json", "The JSON in which the field is searched for. String."},
+           {"field_name", "The name of the field to search for. String literal."}},
+        .returned_value = "It returns the number parsed from the field if the field exists and contains a number, 0 otherwise.",
+        .examples
+        = {{.name = "simple",
+            .query = R"(CREATE TABLE jsons
+(
+    json String
+)
+ENGINE = Memory;
+
+INSERT INTO jsons VALUES ('{"foo":"-4e3"}');
+INSERT INTO jsons VALUES ('{"foo":-3.4}');
+INSERT INTO jsons VALUES ('{"foo":5}');
+INSERT INTO jsons VALUES ('{"foo":"not1number"}');
+INSERT INTO jsons VALUES ('{"baz":2}');
+
+SELECT simpleJSONExtractInt(json, 'foo') FROM jsons ORDER BY json;)",
+            .result = R"(0
+-4
+0
+-3
+5)"}},
+        .categories{"JSON"}});
     factory.registerAlias("visitParamExtractInt", "simpleJSONExtractInt");
 }
 
diff --git a/src/Functions/visitParamExtractRaw.cpp b/src/Functions/visitParamExtractRaw.cpp
index 74a83170545..296429423fe 100644
--- a/src/Functions/visitParamExtractRaw.cpp
+++ b/src/Functions/visitParamExtractRaw.cpp
@@ -61,7 +61,35 @@ using FunctionSimpleJSONExtractRaw = FunctionsStringSearchToString<ExtractParamT
 
 REGISTER_FUNCTION(VisitParamExtractRaw)
 {
-    factory.registerFunction<FunctionSimpleJSONExtractRaw>();
+    factory.registerFunction<FunctionSimpleJSONExtractRaw>(FunctionDocumentation{
+        .description = "Returns the value of the field named field_name as a String, including separators.",
+        .syntax = "simpleJSONExtractRaw(json, field_name)",
+        .arguments
+        = {{"json", "The JSON in which the field is searched for. String."},
+           {"field_name", "The name of the field to search for. String literal."}},
+        .returned_value
+        = "It returns the value of the field as a String including separators if the field exists, or an emtpy String otherwise.",
+        .examples
+        = {{.name = "simple",
+            .query = R"(CREATE TABLE jsons
+(
+    json String
+)
+ENGINE = Memory;
+
+INSERT INTO jsons VALUES ('{"foo":"-4e3"}');
+INSERT INTO jsons VALUES ('{"foo":-3.4}');
+INSERT INTO jsons VALUES ('{"foo":5}');
+INSERT INTO jsons VALUES ('{"foo":{"def":[1,2,3]}}');
+INSERT INTO jsons VALUES ('{"baz":2}');
+
+SELECT simpleJSONExtractRaw(json, 'foo') FROM jsons ORDER BY json;)",
+            .result = R"(
+"-4e3"
+-3.4
+5
+{"def":[1,2,3]})"}},
+        .categories{"JSON"}});
     factory.registerAlias("visitParamExtractRaw", "simpleJSONExtractRaw");
 }
 
diff --git a/src/Functions/visitParamExtractString.cpp b/src/Functions/visitParamExtractString.cpp
index 50d5f345189..8dae10638f8 100644
--- a/src/Functions/visitParamExtractString.cpp
+++ b/src/Functions/visitParamExtractString.cpp
@@ -22,7 +22,35 @@ using FunctionSimpleJSONExtractString = FunctionsStringSearchToString<ExtractPar
 
 REGISTER_FUNCTION(VisitParamExtractString)
 {
-    factory.registerFunction<FunctionSimpleJSONExtractString>();
+    factory.registerFunction<FunctionSimpleJSONExtractString>(FunctionDocumentation{
+        .description = R"(Parses String in double quotes from the value of the field named field_name.
+
+        There is currently no support for code points in the format \uXXXX\uYYYY that are not from the basic multilingual plane (they are converted to CESU-8 instead of UTF-8).)",
+        .syntax = "simpleJSONExtractString(json, field_name)",
+        .arguments
+        = {{"json", "The JSON in which the field is searched for. String."},
+           {"field_name", "The name of the field to search for. String literal."}},
+        .returned_value = "It returns the value of a field as a String, including separators. The value is unescaped. It returns an empty "
+                          "String: if the field doesn't contain a double quoted string, if unescaping fails or if the field doesn't exist.",
+        .examples
+        = {{.name = "simple",
+            .query = R"(CREATE TABLE jsons
+(
+    json String
+)
+ENGINE = Memory;
+
+INSERT INTO jsons VALUES ('{"foo":"\\n\\u0000"}');
+INSERT INTO jsons VALUES ('{"foo":"\\u263"}');
+INSERT INTO jsons VALUES ('{"foo":"\\u263a"}');
+INSERT INTO jsons VALUES ('{"foo":"hello}');
+
+SELECT simpleJSONExtractString(json, 'foo') FROM jsons ORDER BY json;)",
+            .result = R"(\n\0
+
+☺
+)"}},
+        .categories{"JSON"}});
     factory.registerAlias("visitParamExtractString", "simpleJSONExtractString");
 }
 
diff --git a/src/Functions/visitParamExtractUInt.cpp b/src/Functions/visitParamExtractUInt.cpp
index fb58e417f34..777df9fdd24 100644
--- a/src/Functions/visitParamExtractUInt.cpp
+++ b/src/Functions/visitParamExtractUInt.cpp
@@ -12,7 +12,36 @@ using FunctionSimpleJSONExtractUInt = FunctionsStringSearch<ExtractParamImpl<Nam
 
 REGISTER_FUNCTION(VisitParamExtractUInt)
 {
-    factory.registerFunction<FunctionSimpleJSONExtractUInt>();
+    factory.registerFunction<FunctionSimpleJSONExtractUInt>(FunctionDocumentation{
+        .description
+        = "Parses UInt64 from the value of the field named field_name. If this is a string field, it tries to parse a number from the "
+          "beginning of the string. If the field does not exist, or it exists but does not contain a number, it returns 0.",
+        .syntax = "simpleJSONExtractUInt(json, field_name)",
+        .arguments
+        = {{"json", "The JSON in which the field is searched for. String."},
+           {"field_name", "The name of the field to search for. String literal."}},
+        .returned_value = "It returns the number parsed from the field if the field exists and contains a number, 0 otherwise.",
+        .examples
+        = {{.name = "simple",
+            .query = R"(CREATE TABLE jsons
+(
+    json String
+)
+ENGINE = Memory;
+
+INSERT INTO jsons VALUES ('{"foo":"4e3"}');
+INSERT INTO jsons VALUES ('{"foo":3.4}');
+INSERT INTO jsons VALUES ('{"foo":5}');
+INSERT INTO jsons VALUES ('{"foo":"not1number"}');
+INSERT INTO jsons VALUES ('{"baz":2}');
+
+SELECT simpleJSONExtractUInt(json, 'foo') FROM jsons ORDER BY json;)",
+            .result = R"(0
+4
+0
+3
+5)"}},
+        .categories{"JSON"}});
     factory.registerAlias("visitParamExtractUInt", "simpleJSONExtractUInt");
 }
 
diff --git a/src/Functions/visitParamHas.cpp b/src/Functions/visitParamHas.cpp
index 1ed1f1d16e7..09fec782980 100644
--- a/src/Functions/visitParamHas.cpp
+++ b/src/Functions/visitParamHas.cpp
@@ -21,7 +21,28 @@ using FunctionSimpleJSONHas = FunctionsStringSearch<ExtractParamImpl<NameSimpleJ
 
 REGISTER_FUNCTION(VisitParamHas)
 {
-    factory.registerFunction<FunctionSimpleJSONHas>();
+    factory.registerFunction<FunctionSimpleJSONHas>(FunctionDocumentation{
+        .description = "Checks whether there is a field named field_name.  The result is UInt8.",
+        .syntax = "simpleJSONHas(json, field_name)",
+        .arguments
+        = {{"json", "The JSON in which the field is searched for. String."},
+           {"field_name", "The name of the field to search for. String literal."}},
+        .returned_value = "It returns 1 if the field exists, 0 otherwise.",
+        .examples
+        = {{.name = "simple",
+            .query = R"(CREATE TABLE jsons
+(
+    json String
+)
+ENGINE = Memory;
+
+INSERT INTO jsons VALUES ('{"foo":"true","qux":1}');
+
+SELECT simpleJSONHas(json, 'foo') FROM jsons;
+SELECT simpleJSONHas(json, 'bar') FROM jsons;)",
+            .result = R"(1
+0)"}},
+        .categories{"JSON"}});
     factory.registerAlias("visitParamHas", "simpleJSONHas");
 }
 
diff --git a/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference b/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference
index 379eea4dbbb..0a11e8b5034 100644
--- a/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference
+++ b/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference
@@ -643,14 +643,6 @@ shardNum
 showCertificate
 sigmoid
 sign
-simpleJSONExtractBool
-simpleJSONExtractFloat
-simpleJSONExtractInt
-simpleJSONExtractRaw
-simpleJSONExtractString
-simpleJSONExtractUInt
-simpleJSONHas
-sin
 sinh
 sipHash128
 sipHash128Keyed

From 5b94f9b4115e3b7e03118b4a4f4999139e58511e Mon Sep 17 00:00:00 2001
From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com>
Date: Wed, 6 Mar 2024 15:31:19 +0100
Subject: [PATCH 066/443] Check children first

---
 src/Storages/VirtualColumnUtils.cpp | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/src/Storages/VirtualColumnUtils.cpp b/src/Storages/VirtualColumnUtils.cpp
index 6d66453442e..e8441b96782 100644
--- a/src/Storages/VirtualColumnUtils.cpp
+++ b/src/Storages/VirtualColumnUtils.cpp
@@ -469,18 +469,18 @@ static bool canEvaluateSubtree(const ActionsDAG::Node * node, const Block & allo
 
 static bool isDeterministicInScopeOfQuery(const ActionsDAG::Node * node)
 {
-    if (node->type != ActionsDAG::ActionType::FUNCTION)
-        return true;
-
-    if (!node->function_base->isDeterministicInScopeOfQuery())
-        return false;
-
     for (const auto * child : node->children)
     {
         if (!isDeterministicInScopeOfQuery(child))
             return false;
     }
 
+    if (node->type != ActionsDAG::ActionType::FUNCTION)
+        return true;
+
+    if (!node->function_base->isDeterministicInScopeOfQuery())
+        return false;
+
     return true;
 }
 

From 526f162082dfbb4ad2fb5d3d807dfd2ad9b54bdd Mon Sep 17 00:00:00 2001
From: avogar <avogar@clickhouse.com>
Date: Thu, 29 Feb 2024 18:20:47 +0000
Subject: [PATCH 067/443] Fix logical error on bad compatibility setting value
 type

---
 src/Core/Settings.cpp                                         | 4 ++++
 .../03003_compatibility_setting_bad_value.reference           | 0
 .../0_stateless/03003_compatibility_setting_bad_value.sql     | 2 ++
 3 files changed, 6 insertions(+)
 create mode 100644 tests/queries/0_stateless/03003_compatibility_setting_bad_value.reference
 create mode 100644 tests/queries/0_stateless/03003_compatibility_setting_bad_value.sql

diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp
index a38197b9eeb..fb456b46d89 100644
--- a/src/Core/Settings.cpp
+++ b/src/Core/Settings.cpp
@@ -114,7 +114,11 @@ std::vector<String> Settings::getAllRegisteredNames() const
 void Settings::set(std::string_view name, const Field & value)
 {
     if (name == "compatibility")
+    {
+        if (value.getType() != Field::Types::Which::String)
+            throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unexpected type of value for setting 'compatibility'. Expected String, got {}", value.getTypeName());
         applyCompatibilitySetting(value.get<String>());
+    }
     /// If we change setting that was changed by compatibility setting before
     /// we should remove it from settings_changed_by_compatibility_setting,
     /// otherwise the next time we will change compatibility setting
diff --git a/tests/queries/0_stateless/03003_compatibility_setting_bad_value.reference b/tests/queries/0_stateless/03003_compatibility_setting_bad_value.reference
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/queries/0_stateless/03003_compatibility_setting_bad_value.sql b/tests/queries/0_stateless/03003_compatibility_setting_bad_value.sql
new file mode 100644
index 00000000000..9a6f4e7944a
--- /dev/null
+++ b/tests/queries/0_stateless/03003_compatibility_setting_bad_value.sql
@@ -0,0 +1,2 @@
+select 42 settings compatibility=NULL;  -- {clientError BAD_GET}
+

From bdb76d9dd4b42ab4f40db0d371165665171afb4d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <benjamin.antal@clickhouse.com>
Date: Wed, 6 Mar 2024 16:30:22 +0000
Subject: [PATCH 068/443] Fix aspell errors

---
 docs/en/sql-reference/functions/json-functions.md  |  2 +-
 utils/check-style/aspell-ignore/en/aspell-dict.txt | 12 +++++++++---
 2 files changed, 10 insertions(+), 4 deletions(-)

diff --git a/docs/en/sql-reference/functions/json-functions.md b/docs/en/sql-reference/functions/json-functions.md
index 246cb8972fb..e920ab82988 100644
--- a/docs/en/sql-reference/functions/json-functions.md
+++ b/docs/en/sql-reference/functions/json-functions.md
@@ -264,7 +264,7 @@ simpleJSONExtractRaw(json, field_name)
 
 **Returned value**
 
-It returns the value of the field as a [`String`](../../sql-reference/data-types/string.md#string), including separators if the field exists, or an emtpy `String` otherwise.
+It returns the value of the field as a [`String`](../../sql-reference/data-types/string.md#string), including separators if the field exists, or an empty `String` otherwise.
 
 **Example**
 
diff --git a/utils/check-style/aspell-ignore/en/aspell-dict.txt b/utils/check-style/aspell-ignore/en/aspell-dict.txt
index 3614bcb7452..917b2cdcc71 100644
--- a/utils/check-style/aspell-ignore/en/aspell-dict.txt
+++ b/utils/check-style/aspell-ignore/en/aspell-dict.txt
@@ -1,4 +1,4 @@
-personal_ws-1.1 en 2724
+personal_ws-1.1 en 2758 
 AArch
 ACLs
 ALTERs
@@ -843,7 +843,6 @@ SendScalars
 ShareAlike
 SharedMergeTree
 Shortkeys
-Shortkeys
 SimHash
 Simhash
 SimpleAggregateFunction
@@ -1703,7 +1702,6 @@ hyperscan
 hypot
 hyvor
 iTerm
-iTerm
 icosahedron
 icudata
 idempotency
@@ -2327,6 +2325,14 @@ shortcircuit
 shortkeys
 shoutout
 simdjson
+simpleJSON
+simpleJSONExtractBool
+simpleJSONExtractFloat
+simpleJSONExtractInt
+simpleJSONExtractRaw
+simpleJSONExtractString
+simpleJSONExtractUInt
+simpleJSONHas
 simpleLinearRegression
 simpleaggregatefunction
 simplelinearregression

From 77a980373a1dab7c49e5713ba7050d218c1250c7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <benjamin.antal@clickhouse.com>
Date: Wed, 6 Mar 2024 16:31:27 +0000
Subject: [PATCH 069/443] Fix typo in inline doc

---
 src/Functions/visitParamExtractRaw.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Functions/visitParamExtractRaw.cpp b/src/Functions/visitParamExtractRaw.cpp
index 296429423fe..3cdc5001e13 100644
--- a/src/Functions/visitParamExtractRaw.cpp
+++ b/src/Functions/visitParamExtractRaw.cpp
@@ -68,7 +68,7 @@ REGISTER_FUNCTION(VisitParamExtractRaw)
         = {{"json", "The JSON in which the field is searched for. String."},
            {"field_name", "The name of the field to search for. String literal."}},
         .returned_value
-        = "It returns the value of the field as a String including separators if the field exists, or an emtpy String otherwise.",
+        = "It returns the value of the field as a String including separators if the field exists, or an empty String otherwise.",
         .examples
         = {{.name = "simple",
             .query = R"(CREATE TABLE jsons

From 06a7665d0e780b52b4c1e2c1115ed41050d2d03a Mon Sep 17 00:00:00 2001
From: Shaun Struwig <41984034+Blargian@users.noreply.github.com>
Date: Wed, 6 Mar 2024 18:19:52 +0100
Subject: [PATCH 070/443] Update encoding-functions.md

Add missing "Result:" after query
---
 docs/en/sql-reference/functions/encoding-functions.md | 1 +
 1 file changed, 1 insertion(+)

diff --git a/docs/en/sql-reference/functions/encoding-functions.md b/docs/en/sql-reference/functions/encoding-functions.md
index c81b3e35317..4f6da764b3c 100644
--- a/docs/en/sql-reference/functions/encoding-functions.md
+++ b/docs/en/sql-reference/functions/encoding-functions.md
@@ -469,6 +469,7 @@ Query:
 ```sql
 SELECT mortonEncode(1, 2, 3);
 ```
+Result:
 
 ```response
 53

From f77b5963748c321975d8bd131e794dcc57002fc8 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Thu, 7 Mar 2024 16:17:27 +0800
Subject: [PATCH 071/443] Fix test

---
 .../integration/test_filesystem_cache/test.py | 47 ++++++++++++-------
 1 file changed, 29 insertions(+), 18 deletions(-)

diff --git a/tests/integration/test_filesystem_cache/test.py b/tests/integration/test_filesystem_cache/test.py
index 0cb1866f8e4..63316aba57e 100644
--- a/tests/integration/test_filesystem_cache/test.py
+++ b/tests/integration/test_filesystem_cache/test.py
@@ -350,6 +350,20 @@ def test_custom_cached_disk(cluster):
 
 def test_force_filesystem_cache_on_merges(cluster):
     def test(node, forced_read_through_cache_on_merge):
+        def to_int(value):
+            if value == "":
+                return 0
+            else:
+                return int(value)
+
+        r_cache_count = to_int(node.query(
+            "SELECT value FROM system.events WHERE name = 'CachedReadBufferCacheWriteBytes'"
+        ))
+
+        w_cache_count = to_int(node.query(
+            "SELECT value FROM system.events WHERE name = 'CachedWriteBufferCacheWriteBytes'"
+        ))
+
         node.query(
             """
             DROP TABLE IF EXISTS test SYNC;
@@ -376,36 +390,33 @@ def test_force_filesystem_cache_on_merges(cluster):
         assert int(node.query("SELECT count() FROM system.filesystem_cache")) > 0
         assert int(node.query("SELECT max(size) FROM system.filesystem_cache")) == 1024
 
-        write_count = int(
+        w_cache_count_2 = int(
             node.query(
                 "SELECT value FROM system.events WHERE name = 'CachedWriteBufferCacheWriteBytes'"
             )
         )
-        assert write_count > 100000
-        assert "" == node.query(
-            "SELECT value FROM system.events WHERE name = 'CachedReadBufferCacheWriteBytes'"
+        assert w_cache_count_2 > w_cache_count
+
+        r_cache_count_2 = to_int(
+            node.query(
+                "SELECT value FROM system.events WHERE name = 'CachedReadBufferCacheWriteBytes'"
+            )
         )
+        assert r_cache_count_2 == r_cache_count
 
         node.query("SYSTEM DROP FILESYSTEM CACHE")
         node.query("OPTIMIZE TABLE test FINAL")
 
-        new_write_count = int(
+        r_cache_count_3 = to_int(
             node.query(
-                "SELECT value FROM system.events WHERE name = 'CachedWriteBufferCacheWriteBytes'"
-            )
-        )
-        assert new_write_count >= write_count
-
-        if forced_read_through_cache_on_merge:
-            assert 100000 < int(
-                node.query(
-                    "SELECT value FROM system.events WHERE name = 'CachedReadBufferCacheWriteBytes'"
-                )
-            )
-        else:
-            assert "" == node.query(
                 "SELECT value FROM system.events WHERE name = 'CachedReadBufferCacheWriteBytes'"
             )
+        )
+
+        if forced_read_through_cache_on_merge:
+            assert r_cache_count_3 > r_cache_count
+        else:
+            assert r_cache_count_3 == r_cache_count
 
     node = cluster.instances["node_force_read_through_cache_on_merge"]
     test(node, True)

From 50b84954e4810c94c1397504a64ca96e1a0fed55 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Thu, 7 Mar 2024 16:29:38 +0800
Subject: [PATCH 072/443] Update .reference

---
 .../0_stateless/02117_show_create_table_system.reference        | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tests/queries/0_stateless/02117_show_create_table_system.reference b/tests/queries/0_stateless/02117_show_create_table_system.reference
index 7382b24afbc..5081527ceef 100644
--- a/tests/queries/0_stateless/02117_show_create_table_system.reference
+++ b/tests/queries/0_stateless/02117_show_create_table_system.reference
@@ -195,6 +195,8 @@ CREATE TABLE system.disks
     `unreserved_space` UInt64,
     `keep_free_space` UInt64,
     `type` String,
+    `object_storage_type` String,
+    `metadata_type` String,
     `is_encrypted` UInt8,
     `is_read_only` UInt8,
     `is_write_once` UInt8,

From c7f5b1631c359c61b6e4c74727092df73e956922 Mon Sep 17 00:00:00 2001
From: robot-clickhouse <robot-clickhouse@users.noreply.github.com>
Date: Thu, 7 Mar 2024 08:30:34 +0000
Subject: [PATCH 073/443] Automatic style fix

---
 tests/integration/test_filesystem_cache/test.py | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/tests/integration/test_filesystem_cache/test.py b/tests/integration/test_filesystem_cache/test.py
index 63316aba57e..c44d817c57c 100644
--- a/tests/integration/test_filesystem_cache/test.py
+++ b/tests/integration/test_filesystem_cache/test.py
@@ -356,13 +356,17 @@ def test_force_filesystem_cache_on_merges(cluster):
             else:
                 return int(value)
 
-        r_cache_count = to_int(node.query(
-            "SELECT value FROM system.events WHERE name = 'CachedReadBufferCacheWriteBytes'"
-        ))
+        r_cache_count = to_int(
+            node.query(
+                "SELECT value FROM system.events WHERE name = 'CachedReadBufferCacheWriteBytes'"
+            )
+        )
 
-        w_cache_count = to_int(node.query(
-            "SELECT value FROM system.events WHERE name = 'CachedWriteBufferCacheWriteBytes'"
-        ))
+        w_cache_count = to_int(
+            node.query(
+                "SELECT value FROM system.events WHERE name = 'CachedWriteBufferCacheWriteBytes'"
+            )
+        )
 
         node.query(
             """

From 31ed1966e3c5388e601edd6e97c0497153bb7196 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Thu, 7 Mar 2024 16:44:10 +0800
Subject: [PATCH 074/443] Fix build

---
 src/Disks/ObjectStorages/ObjectStorageFactory.cpp | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/src/Disks/ObjectStorages/ObjectStorageFactory.cpp b/src/Disks/ObjectStorages/ObjectStorageFactory.cpp
index 9d7e714445a..46136ad7b12 100644
--- a/src/Disks/ObjectStorages/ObjectStorageFactory.cpp
+++ b/src/Disks/ObjectStorages/ObjectStorageFactory.cpp
@@ -166,7 +166,7 @@ void registerS3ObjectStorage(ObjectStorageFactory & factory)
 
         /// NOTE: should we still perform this check for clickhouse-disks?
         if (!skip_access_check)
-            checkS3Capabilities(*object_storage, s3_capabilities, name);
+            checkS3Capabilities(*dynamic_cast<S3ObjectStorage *>(object_storage.get()), s3_capabilities, name);
 
         return object_storage;
     });
@@ -202,7 +202,7 @@ void registerS3PlainObjectStorage(ObjectStorageFactory & factory)
 
         /// NOTE: should we still perform this check for clickhouse-disks?
         if (!skip_access_check)
-            checkS3Capabilities(*object_storage, s3_capabilities, name);
+            checkS3Capabilities(*dynamic_cast<S3ObjectStorage *>(object_storage.get()), s3_capabilities, name);
 
         return object_storage;
     });

From 6c69e7d4dcfdfa21cfcaa103fc1cc7c53dfe0291 Mon Sep 17 00:00:00 2001
From: HowePa <2873679104@qq.com>
Date: Thu, 7 Mar 2024 20:29:04 +0800
Subject: [PATCH 075/443] detect output format by file extension in
 clickhouse-local

---
 programs/local/LocalServer.cpp | 17 ++++++++++++++++-
 1 file changed, 16 insertions(+), 1 deletion(-)

diff --git a/programs/local/LocalServer.cpp b/programs/local/LocalServer.cpp
index 68f0e52ce08..20974dd9751 100644
--- a/programs/local/LocalServer.cpp
+++ b/programs/local/LocalServer.cpp
@@ -327,6 +327,14 @@ static bool checkIfStdinIsRegularFile()
     return fstat(STDIN_FILENO, &file_stat) == 0 && S_ISREG(file_stat.st_mode);
 }
 
+
+static bool checkIfStdoutIsRegularFile()
+{
+    struct stat file_stat;
+    return fstat(STDOUT_FILENO, &file_stat) == 0 && S_ISREG(file_stat.st_mode);
+}
+
+
 std::string LocalServer::getInitialCreateTableQuery()
 {
     if (!config().has("table-structure") && !config().has("table-file") && !config().has("table-data-format") && (!checkIfStdinIsRegularFile() || queries.empty()))
@@ -638,7 +646,14 @@ void LocalServer::processConfig()
     if (config().has("macros"))
         global_context->setMacros(std::make_unique<Macros>(config(), "macros", log));
 
-    format = config().getString("output-format", config().getString("format", is_interactive ? "PrettyCompact" : "TSV"));
+    if (!config().has("output-format") && !config().has("format") && checkIfStdoutIsRegularFile())
+    {
+        std::optional<String> format_from_file_name;
+        format_from_file_name = FormatFactory::instance().tryGetFormatFromFileDescriptor(STDOUT_FILENO);
+        format = format_from_file_name ? *format_from_file_name : "TSV";
+    }
+    else
+        format = config().getString("output-format", config().getString("format", is_interactive ? "PrettyCompact" : "TSV"));
     insert_format = "Values";
 
     /// Setting value from cmd arg overrides one from config

From 6d5fd2857ed50047d8acf48766165aa815ca30b9 Mon Sep 17 00:00:00 2001
From: HowePa <2873679104@qq.com>
Date: Thu, 7 Mar 2024 20:29:42 +0800
Subject: [PATCH 076/443] detect output format by file extension in
 clickhouse-client

---
 programs/client/Client.cpp | 17 ++++++++++++++++-
 1 file changed, 16 insertions(+), 1 deletion(-)

diff --git a/programs/client/Client.cpp b/programs/client/Client.cpp
index a2bd6b6016a..fac34003553 100644
--- a/programs/client/Client.cpp
+++ b/programs/client/Client.cpp
@@ -50,6 +50,7 @@
 #include <Functions/registerFunctions.h>
 #include <AggregateFunctions/registerAggregateFunctions.h>
 #include <Formats/registerFormats.h>
+#include <Formats/FormatFactory.h>
 
 namespace fs = std::filesystem;
 using namespace std::literals;
@@ -1137,6 +1138,13 @@ void Client::processOptions(const OptionsDescription & options_description,
 }
 
 
+static bool checkIfStdoutIsRegularFile()
+{
+    struct stat file_stat;
+    return fstat(STDOUT_FILENO, &file_stat) == 0 && S_ISREG(file_stat.st_mode);
+}
+
+
 void Client::processConfig()
 {
     if (!queries.empty() && config().has("queries-file"))
@@ -1173,7 +1181,14 @@ void Client::processConfig()
     pager = config().getString("pager", "");
 
     is_default_format = !config().has("vertical") && !config().has("format");
-    if (config().has("vertical"))
+    if (is_default_format && checkIfStdoutIsRegularFile())
+    {
+        is_default_format = false;
+        std::optional<String> format_from_file_name;
+        format_from_file_name = FormatFactory::instance().tryGetFormatFromFileDescriptor(STDOUT_FILENO);
+        format = format_from_file_name ? *format_from_file_name : "TabSeparated";
+    }
+    else if (config().has("vertical"))
         format = config().getString("format", "Vertical");
     else
         format = config().getString("format", is_interactive ? "PrettyCompact" : "TabSeparated");

From 112c1efb7da2619cb67a48ff7fbe65ecea8e44a9 Mon Sep 17 00:00:00 2001
From: HowePa <2873679104@qq.com>
Date: Thu, 7 Mar 2024 20:30:24 +0800
Subject: [PATCH 077/443] test detect output format by file extension

---
 ..._output_format_by_file_extension.reference | 20 +++++++++++++++++++
 ..._detect_output_format_by_file_extension.sh | 13 ++++++++++++
 2 files changed, 33 insertions(+)
 create mode 100644 tests/queries/0_stateless/02181_detect_output_format_by_file_extension.reference
 create mode 100755 tests/queries/0_stateless/02181_detect_output_format_by_file_extension.sh

diff --git a/tests/queries/0_stateless/02181_detect_output_format_by_file_extension.reference b/tests/queries/0_stateless/02181_detect_output_format_by_file_extension.reference
new file mode 100644
index 00000000000..7b36cc96f5e
--- /dev/null
+++ b/tests/queries/0_stateless/02181_detect_output_format_by_file_extension.reference
@@ -0,0 +1,20 @@
+0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+0
+1
+2
+3
+4
+5
+6
+7
+8
+9
diff --git a/tests/queries/0_stateless/02181_detect_output_format_by_file_extension.sh b/tests/queries/0_stateless/02181_detect_output_format_by_file_extension.sh
new file mode 100755
index 00000000000..ec1edd710a1
--- /dev/null
+++ b/tests/queries/0_stateless/02181_detect_output_format_by_file_extension.sh
@@ -0,0 +1,13 @@
+#!/usr/bin/env bash
+# Tags: no-parallel, no-fasttest
+
+CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
+# shellcheck source=../shell_config.sh
+. "$CURDIR"/../shell_config.sh
+
+
+$CLICKHOUSE_LOCAL -q "select * from numbers(10)" > $CLICKHOUSE_TMP/data.parquet
+$CLICKHOUSE_LOCAL -q "select * from table" < $CLICKHOUSE_TMP/data.parquet 
+
+$CLICKHOUSE_CLIENT -q "select * from numbers(10)" > $CLICKHOUSE_TMP/data.parquet
+$CLICKHOUSE_LOCAL -q "select * from table" < $CLICKHOUSE_TMP/data.parquet 

From 7fd13df8a5055892d2f8cdc83dcb900c19c87a95 Mon Sep 17 00:00:00 2001
From: Han Fei <hanfei19910905@gmail.com>
Date: Thu, 7 Mar 2024 17:09:55 +0100
Subject: [PATCH 078/443] check memory limit periodically

---
 programs/keeper/Keeper.cpp                | 10 +++++++++
 programs/server/Server.cpp                |  1 +
 src/Common/CgroupsMemoryUsageObserver.cpp | 26 +++++++++++++++++------
 src/Common/CgroupsMemoryUsageObserver.h   |  7 ++++--
 4 files changed, 35 insertions(+), 9 deletions(-)

diff --git a/programs/keeper/Keeper.cpp b/programs/keeper/Keeper.cpp
index 8972c82eab8..76dd8cb15a5 100644
--- a/programs/keeper/Keeper.cpp
+++ b/programs/keeper/Keeper.cpp
@@ -10,6 +10,7 @@
 #include <IO/UseSSL.h>
 #include <Core/ServerUUID.h>
 #include <Common/logger_useful.h>
+#include <Common/CgroupsMemoryUsageObserver.h>
 #include <Common/ErrorHandlers.h>
 #include <Common/assertProcessUserMatchesDataOwner.h>
 #include <Common/makeSocketAddress.h>
@@ -623,6 +624,15 @@ try
     buildLoggers(config(), logger());
     main_config_reloader->start();
 
+    std::optional<CgroupsMemoryUsageObserver> observer;
+    auto cgroups_memory_observer_wait_time = config().getUInt64("keeper_server.cgroups_memory_observer_wait_time", 1);
+    if (cgroups_memory_observer_wait_time > 0)
+    {
+        observer.emplace(std::chrono::seconds(cgroups_memory_observer_wait_time));
+        observer->startThread();
+    }
+
+
     LOG_INFO(log, "Ready for connections.");
 
     waitForTerminationRequest();
diff --git a/programs/server/Server.cpp b/programs/server/Server.cpp
index c45291ba52c..6b282893dee 100644
--- a/programs/server/Server.cpp
+++ b/programs/server/Server.cpp
@@ -1362,6 +1362,7 @@ try
                 cgroups_memory_usage_observer->setLimits(
                     static_cast<uint64_t>(max_server_memory_usage * hard_limit_ratio),
                     static_cast<uint64_t>(max_server_memory_usage * soft_limit_ratio));
+                cgroups_memory_usage_observer->startThread();
             }
 
             size_t merges_mutations_memory_usage_soft_limit = new_server_settings.merges_mutations_memory_usage_soft_limit;
diff --git a/src/Common/CgroupsMemoryUsageObserver.cpp b/src/Common/CgroupsMemoryUsageObserver.cpp
index 9bed6b191e4..5f24c2553b5 100644
--- a/src/Common/CgroupsMemoryUsageObserver.cpp
+++ b/src/Common/CgroupsMemoryUsageObserver.cpp
@@ -8,7 +8,9 @@
 #include <IO/ReadBufferFromFile.h>
 #include <IO/ReadBufferFromFileDescriptor.h>
 #include <IO/ReadHelpers.h>
+#include <Interpreters/Context.h>
 #include <base/cgroupsv2.h>
+#include <base/getMemoryAmount.h>
 #include <base/sleep.h>
 
 #include <filesystem>
@@ -48,11 +50,10 @@ CgroupsMemoryUsageObserver::~CgroupsMemoryUsageObserver()
 
 void CgroupsMemoryUsageObserver::setLimits(uint64_t hard_limit_, uint64_t soft_limit_)
 {
+    std::lock_guard<std::mutex> lock(set_limit_mutex);
     if (hard_limit_ == hard_limit && soft_limit_ == soft_limit)
         return;
 
-    stopThread();
-
     hard_limit = hard_limit_;
     soft_limit = soft_limit_;
 
@@ -94,8 +95,6 @@ void CgroupsMemoryUsageObserver::setLimits(uint64_t hard_limit_, uint64_t soft_l
         }
     };
 
-    startThread();
-
     LOG_INFO(log, "Set new limits, soft limit: {}, hard limit: {}", ReadableSize(soft_limit_), ReadableSize(hard_limit_));
 }
 
@@ -277,7 +276,7 @@ void CgroupsMemoryUsageObserver::stopThread()
 void CgroupsMemoryUsageObserver::runThread()
 {
     setThreadName("CgrpMemUsgObsr");
-
+    last_memory_amount = getMemoryAmount();
     std::unique_lock lock(thread_mutex);
     while (true)
     {
@@ -286,8 +285,21 @@ void CgroupsMemoryUsageObserver::runThread()
 
         try
         {
-            uint64_t memory_usage = file.readMemoryUsage();
-            processMemoryUsage(memory_usage);
+            uint64_t memory_limit = getMemoryAmount();
+            if (memory_limit != last_memory_amount)
+            {
+                last_memory_amount = memory_limit;
+                /// if we find memory amount changes, we just reload config.
+                /// Reloading config will check the memory amount again and calculate soft/hard limit again.
+                auto global_context = getContext()->getGlobalContext();
+                global_context->reloadConfig();
+            }
+            std::lock_guard<std::mutex> set_limit_lock(set_limit_mutex);
+            if (soft_limit > 0 && hard_limit > 0)
+            {
+                uint64_t memory_usage = file.readMemoryUsage();
+                processMemoryUsage(memory_usage);
+            }
         }
         catch (...)
         {
diff --git a/src/Common/CgroupsMemoryUsageObserver.h b/src/Common/CgroupsMemoryUsageObserver.h
index 28bf08c82b5..6edf2e2049d 100644
--- a/src/Common/CgroupsMemoryUsageObserver.h
+++ b/src/Common/CgroupsMemoryUsageObserver.h
@@ -1,6 +1,7 @@
 #pragma once
 
 #include <Common/ThreadPool.h>
+#include <Interpreters/Context_fwd.h>
 
 #include <atomic>
 #include <chrono>
@@ -14,7 +15,7 @@ namespace DB
 /// - When the soft memory limit is hit, drop jemalloc cache.
 /// - When the hard memory limit is hit, update MemoryTracking metric to throw memory exceptions faster.
 #if defined(OS_LINUX)
-class CgroupsMemoryUsageObserver
+class CgroupsMemoryUsageObserver : public WithContext
 {
 public:
     enum class CgroupsVersion
@@ -27,6 +28,7 @@ public:
     ~CgroupsMemoryUsageObserver();
 
     void setLimits(uint64_t hard_limit_, uint64_t soft_limit_);
+    void startThread();
 
     size_t getHardLimit() const { return hard_limit; }
     size_t getSoftLimit() const { return soft_limit; }
@@ -64,16 +66,17 @@ private:
 
     File file;
 
-    void startThread();
     void stopThread();
 
     void runThread();
     void processMemoryUsage(uint64_t usage);
 
     std::mutex thread_mutex;
+    std::mutex set_limit_mutex;
     std::condition_variable cond;
     ThreadFromGlobalPool thread;
     bool quit = false;
+    uint64_t last_memory_amount;
 };
 
 #else

From bd194aab41401492c5d628269df53e68243a1211 Mon Sep 17 00:00:00 2001
From: johnnymatthews <9611008+johnnymatthews@users.noreply.github.com>
Date: Thu, 7 Mar 2024 12:55:21 -0400
Subject: [PATCH 079/443] Adds makeDateTime64 function.

---
 .../functions/date-time-functions.md          | 137 ++++++++++++++----
 .../functions/other-functions.md              |  65 ++++++++-
 2 files changed, 170 insertions(+), 32 deletions(-)

diff --git a/docs/en/sql-reference/functions/date-time-functions.md b/docs/en/sql-reference/functions/date-time-functions.md
index 41503abfa2f..12f0c996ce7 100644
--- a/docs/en/sql-reference/functions/date-time-functions.md
+++ b/docs/en/sql-reference/functions/date-time-functions.md
@@ -26,66 +26,115 @@ SELECT
 
 ## makeDate
 
-Creates a [Date](../../sql-reference/data-types/date.md)
-- from a year, month and day argument, or
-- from a year and day of year argument.
+Creates a [Date](../../sql-reference/data-types/date.md) from either one of the following sets of arguments:
 
-**Syntax**
+- a year, month, and day.
+- a year and day of year.
 
-``` sql
-makeDate(year, month, day);
-makeDate(year, day_of_year);
+### Syntax
+
+Using a year, month, and day:
+
+```sql
+makeDate(year, month, day)
 ```
 
-Alias:
-- `MAKEDATE(year, month, day);`
-- `MAKEDATE(year, day_of_year);`
+Using a year and day of year:
 
-**Arguments**
+```sql
+makeDate(year, day_of_year)
+```
+
+### Arguments
 
 - `year` — Year. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md).
 - `month` — Month. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md).
 - `day` — Day. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md).
 - `day_of_year` — Day of the year. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md).
 
-**Returned value**
+### Returned values
 
-- A date created from the arguments.
+A date created from the arguments.
 
 Type: [Date](../../sql-reference/data-types/date.md).
 
-**Example**
+### Examples
 
 Create a Date from a year, month and day:
 
-``` sql
+```sql
 SELECT makeDate(2023, 2, 28) AS Date;
 ```
 
-Result:
-
-``` text
-┌───────date─┐
-│ 2023-02-28 │
-└────────────┘
+```response
+2023-02-28
 ```
 
-Create a Date from a year and day of year argument:
+Create a Date from a year and day of year:
 
 ``` sql
 SELECT makeDate(2023, 42) AS Date;
 ```
 
-Result:
-
-``` text
-┌───────date─┐
-│ 2023-02-11 │
-└────────────┘
+```response
+2023-02-11
 ```
+
 ## makeDate32
 
-Like [makeDate](#makeDate) but produces a [Date32](../../sql-reference/data-types/date32.md).
+Creates a date of type [Date32](../../sql-reference/data-types/date32.md) from either one of the following sets of arguments:
+
+- a year, month, and day.
+- a year and day of year.
+
+### Syntax
+
+Using a year, month, and day:
+
+```sql
+makeDate32(year, month, day)
+```
+
+Using a year and day of year:
+
+```sql
+makeDate32(year, day_of_year)
+```
+
+### Arguments
+
+- `year` — Year. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md).
+- `month` — Month. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md).
+- `day` — Day. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md).
+- `day_of_year` — Day of the year. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md).
+
+### Returned values
+
+A date created from the arguments.
+
+Type: [Date32](../../sql-reference/data-types/date32.md).
+
+### Examples
+
+Create a date from a year, month, and day:
+
+```sql
+SELECT makeDate32(2024, 1, 1);
+```
+
+```response
+2024-01-01
+```
+
+Create a Date from a year and day of year:
+
+``` sql
+SELECT makeDate32(2024, 100);
+```
+
+```response
+2024-04-09
+```
 
 ## makeDateTime
 
@@ -129,12 +178,38 @@ Result:
 
 ## makeDateTime64
 
-Like [makeDateTime](#makedatetime) but produces a [DateTime64](../../sql-reference/data-types/datetime64.md).
+Create a [DateTime64](../../sql-reference/data-types/datetime64.md) data type value from its components (year, month, day, hour, minute, second, and optionally, subsecond precision).
+
+The DateTime64 data type stores both the date and time components in a single 64-bit integer value. The precision of the time component is configurable, allowing you to store time values with subsecond precision up to nanoseconds.
 
 **Syntax**
 
+```sql
+makeDateTime64(year, month, day, hour, minute, second[, fraction[, precision]])
+```
+
+**Arguments**
+
+- `year` — [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md). The year component (0-9999).
+- `month` — Month. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md). The month component (1-12).
+- `day` — Day. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md). The day component (1-31).
+- `hour` — Hour. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md). The hour component (0-23).
+- `minute` — Minute. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md). The minute component (0-59).
+- `second` — Second. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md). The second component (0-59).
+- `subsecond_precision` (optional) [Integer](../../sql-reference/data-types/int-uint.md): The precision of the subsecond component (0-9, where 0 means no subsecond precision, and 9 means nanosecond precision).
+
+**Returned value**
+
+A date and time element of type [DateTime64](../../sql-reference/data-types/datetime64.md) with created from the supplied arguments.
+
+**Example**
+
 ``` sql
-makeDateTime64(year, month, day, hour, minute, second[, fraction[, precision[, timezone]]])
+SELECT makeDateTime64(2023, 5, 15, 10, 30, 45, 779, 5);
+```
+
+```response
+2023-05-15 10:30:45.00779
 ```
 
 ## timestamp
diff --git a/docs/en/sql-reference/functions/other-functions.md b/docs/en/sql-reference/functions/other-functions.md
index 739b688a0d2..10ceedad9aa 100644
--- a/docs/en/sql-reference/functions/other-functions.md
+++ b/docs/en/sql-reference/functions/other-functions.md
@@ -1866,7 +1866,7 @@ As you can see, `runningAccumulate` merges states for each group of rows separat
 
 ## joinGet
 
-The function lets you extract data from the table the same way as from a [dictionary](../../sql-reference/dictionaries/index.md).
+Allows you to extract data from a specific column in a Join table, similar to how you would access a value from a dictionary.
 
 Gets the data from [Join](../../engines/table-engines/special/join.md#creating-a-table) tables using the specified join key.
 
@@ -1927,6 +1927,69 @@ Result:
 └──────────────────────────────────────────────────┘
 ```
 
+## joinGetOrNull
+
+Allows you to extract data from a specific column in a Join table, similar to how you would access a value from a dictionary.
+
+Gets the data from [Join](../../engines/table-engines/special/join.md#creating-a-table) tables using the specified join key.
+
+Only supports tables created with the `ENGINE = Join(ANY, LEFT, <join_keys>)` statement.
+
+### Syntax
+
+```sql
+joinGet(join_storage_table_name, `value_column`, join_keys)
+```
+
+### Parameters
+
+- `join_storage_table_name` — an [identifier](../../sql-reference/syntax.md#syntax-identifiers) indicating where the search is performed. The identifier is searched in the default database (see setting `default_database` in the config file). To override the default database, use `USE db_name` or specify the database and the table through the separator `db_name.db_table` as in the example.
+- `value_column` — name of the column of the table that contains required data.
+- `join_keys` — list of keys.
+
+### Returned value
+
+Returns a list of values corresponded to list of keys.
+
+If certain does not exist in source table then `0` or `null` will be returned based on [join_use_nulls](../../operations/settings/settings.md#join_use_nulls) setting.
+
+More info about `join_use_nulls` in [Join operation](../../engines/table-engines/special/join.md).
+
+**Example**
+
+Input table:
+
+``` sql
+CREATE DATABASE db_test
+CREATE TABLE db_test.id_val(`id` UInt32, `val` UInt32) ENGINE = Join(ANY, LEFT, id) SETTINGS join_use_nulls = 1
+INSERT INTO db_test.id_val VALUES (1,11)(2,12)(4,13)
+```
+
+``` text
+┌─id─┬─val─┐
+│  4 │  13 │
+│  2 │  12 │
+│  1 │  11 │
+└────┴─────┘
+```
+
+Query:
+
+``` sql
+SELECT joinGet(db_test.id_val, 'val', toUInt32(number)) from numbers(4) SETTINGS join_use_nulls = 1
+```
+
+Result:
+
+``` text
+┌─joinGet(db_test.id_val, 'val', toUInt32(number))─┐
+│                                                0 │
+│                                               11 │
+│                                               12 │
+│                                                0 │
+└──────────────────────────────────────────────────┘
+```
+
 ## catboostEvaluate(path_to_model, feature_1, feature_2, …, feature_n)
 
 :::note

From a00a1fd7b4c9a4e83eeb746da781b27c74dcd0b2 Mon Sep 17 00:00:00 2001
From: johnnymatthews <9611008+johnnymatthews@users.noreply.github.com>
Date: Thu, 7 Mar 2024 14:45:15 -0400
Subject: [PATCH 080/443] Adds readWKT docs.

---
 .../en/sql-reference/functions/geo/polygon.md | 56 +++++++++++++++++++
 1 file changed, 56 insertions(+)

diff --git a/docs/en/sql-reference/functions/geo/polygon.md b/docs/en/sql-reference/functions/geo/polygon.md
index 4a8653965c2..35e2280e5cc 100644
--- a/docs/en/sql-reference/functions/geo/polygon.md
+++ b/docs/en/sql-reference/functions/geo/polygon.md
@@ -53,6 +53,62 @@ String starting with `POLYGON`
 
 Polygon
 
+## readWKTPoint
+
+The `readWKTPoint` function in ClickHouse parses a Well-Known Text (WKT) representation of a Point geometry and returns a point in the internal ClickHouse format.
+
+### Syntax
+
+```sql
+readWKTPoint(wkt_string)
+```
+
+### Arguments
+
+- `wkt_string`: The input WKT string representing a Point geometry.
+
+### Returned value
+
+The function returns a ClickHouse internal representation of the Point geometry.
+
+### Example
+
+```sql
+SELECT readWKTPoint('POINT (1.2 3.4)');
+```
+
+```response
+(1.2,3.4)
+```
+
+## readWKTRing
+
+Parses a Well-Known Text (WKT) representation of a Polygon geometry and returns a ring (closed linestring) in the internal ClickHouse format.
+
+### Syntax
+
+```sql
+readWKTRing(wkt_string)
+```
+
+### Arguments
+
+- `wkt_string`: The input WKT string representing a Polygon geometry.
+
+### Returned value
+
+The function returns a ClickHouse internal representation of the ring (closed linestring) geometry.
+
+### Example
+
+```sql
+SELECT readWKTRing('LINESTRING (1 1, 2 2, 3 3, 1 1)');
+```
+
+```response
+[(1,1),(2,2),(3,3),(1,1)]
+```
+
 ## polygonsWithinSpherical
 
 Returns true or false depending on whether or not one polygon lies completely inside another polygon. Reference https://www.boost.org/doc/libs/1_62_0/libs/geometry/doc/html/geometry/reference/algorithms/within/within_2.html

From 2bc4d27ac6f42af79120dac72b25db6e36ee4b42 Mon Sep 17 00:00:00 2001
From: Nikita Mikhaylov <nikitamikhaylov@clickhouse.com>
Date: Thu, 7 Mar 2024 19:24:39 +0000
Subject: [PATCH 081/443] Bye bye

---
 docker/packager/README.md                     |    1 -
 docs/en/operations/backup.md                  |    6 +-
 .../operations/utilities/clickhouse-copier.md |  187 --
 docs/en/operations/utilities/index.md         |    2 -
 .../sql-reference/statements/alter/column.md  |    2 +-
 docs/ru/getting-started/tutorial.md           |    4 -
 docs/ru/operations/backup.md                  |    6 -
 .../operations/utilities/clickhouse-copier.md |  183 --
 docs/ru/operations/utilities/index.md         |    1 -
 .../sql-reference/statements/alter/column.md  |    4 +-
 docs/zh/getting-started/tutorial.md           |    2 -
 docs/zh/operations/backup.md                  |    6 -
 .../operations/utilities/clickhouse-copier.md |  172 --
 docs/zh/operations/utilities/index.md         |    1 -
 docs/zh/sql-reference/statements/alter.md     |    2 +-
 packages/clickhouse-server.yaml               |    2 -
 programs/CMakeLists.txt                       |    2 -
 programs/copier/Aliases.h                     |   15 -
 programs/copier/CMakeLists.txt                |   28 -
 programs/copier/ClusterCopier.cpp             | 2076 -----------------
 programs/copier/ClusterCopier.h               |  240 --
 programs/copier/ClusterCopierApp.cpp          |  252 --
 programs/copier/ClusterCopierApp.h            |   99 -
 programs/copier/ClusterPartition.h            |   22 -
 programs/copier/Internals.cpp                 |  280 ---
 programs/copier/Internals.h                   |  198 --
 programs/copier/ShardPartition.cpp            |   70 -
 programs/copier/ShardPartition.h              |   54 -
 programs/copier/ShardPartitionPiece.cpp       |   64 -
 programs/copier/ShardPartitionPiece.h         |   43 -
 programs/copier/StatusAccumulator.cpp         |   48 -
 programs/copier/StatusAccumulator.h           |   27 -
 programs/copier/TaskCluster.cpp               |   74 -
 programs/copier/TaskCluster.h                 |   51 -
 programs/copier/TaskShard.cpp                 |   37 -
 programs/copier/TaskShard.h                   |   56 -
 programs/copier/TaskTable.cpp                 |  222 --
 programs/copier/TaskTable.h                   |  173 --
 programs/copier/ZooKeeperStaff.h              |  221 --
 programs/copier/clickhouse-copier.cpp         |    1 -
 .../testdata/configs/xml/config.xml           |    2 +-
 .../testdata/configs/yaml/config.yaml         |    2 +-
 .../testdata/configs/yandex_xml/config.xml    |    2 +-
 programs/install/Install.cpp                  |    1 -
 programs/main.cpp                             |    2 -
 programs/server/config.xml                    |    2 +-
 programs/server/config.yaml.example           |    2 +-
 src/Storages/StorageDistributed.h             |    1 -
 tests/integration/README.md                   |    2 +-
 .../test_cluster_copier/__init__.py           |    0
 .../configs/conf.d/clusters.xml               |   73 -
 .../configs/conf.d/clusters_trivial.xml       |   20 -
 .../configs/conf.d/ddl.xml                    |    5 -
 .../configs/conf.d/query_log.xml              |   14 -
 .../configs/config-copier.xml                 |   11 -
 .../test_cluster_copier/configs/users.xml     |   34 -
 .../configs_three_nodes/conf.d/clusters.xml   |   27 -
 .../configs_three_nodes/conf.d/ddl.xml        |    5 -
 .../configs_three_nodes/config-copier.xml     |   27 -
 .../configs_three_nodes/users.xml             |   32 -
 .../configs_two_nodes/conf.d/clusters.xml     |   22 -
 .../configs_two_nodes/conf.d/ddl.xml          |    5 -
 .../conf.d/storage_configuration.xml          |   34 -
 .../configs_two_nodes/config-copier.xml       |   19 -
 .../configs_two_nodes/users.xml               |   32 -
 .../test_cluster_copier/task0_description.xml |   95 -
 .../task_drop_target_partition.xml            |   41 -
 .../task_month_to_week_description.xml        |   99 -
 .../test_cluster_copier/task_no_arg.xml       |   39 -
 .../test_cluster_copier/task_no_index.xml     |  109 -
 .../task_non_partitioned_table.xml            |   39 -
 .../test_cluster_copier/task_self_copy.xml    |   63 -
 .../test_cluster_copier/task_skip_index.xml   |   39 -
 .../test_cluster_copier/task_taxi_data.xml    |   42 -
 .../task_test_block_size.xml                  |  101 -
 .../test_cluster_copier/task_trivial.xml      |   63 -
 .../task_trivial_without_arguments.xml        |   63 -
 .../test_cluster_copier/task_ttl_columns.xml  |   39 -
 .../task_ttl_move_to_volume.xml               |   39 -
 .../task_with_different_schema.xml            |   39 -
 tests/integration/test_cluster_copier/test.py |  653 ------
 .../test_cluster_copier/test_three_nodes.py   |  286 ---
 .../test_cluster_copier/test_trivial.py       |  227 --
 .../test_cluster_copier/test_two_nodes.py     |  597 -----
 .../test_config_xml_full/configs/config.xml   |    2 +-
 85 files changed, 12 insertions(+), 7973 deletions(-)
 delete mode 100644 docs/en/operations/utilities/clickhouse-copier.md
 delete mode 100644 docs/ru/operations/utilities/clickhouse-copier.md
 delete mode 100644 docs/zh/operations/utilities/clickhouse-copier.md
 delete mode 100644 programs/copier/Aliases.h
 delete mode 100644 programs/copier/CMakeLists.txt
 delete mode 100644 programs/copier/ClusterCopier.cpp
 delete mode 100644 programs/copier/ClusterCopier.h
 delete mode 100644 programs/copier/ClusterCopierApp.cpp
 delete mode 100644 programs/copier/ClusterCopierApp.h
 delete mode 100644 programs/copier/ClusterPartition.h
 delete mode 100644 programs/copier/Internals.cpp
 delete mode 100644 programs/copier/Internals.h
 delete mode 100644 programs/copier/ShardPartition.cpp
 delete mode 100644 programs/copier/ShardPartition.h
 delete mode 100644 programs/copier/ShardPartitionPiece.cpp
 delete mode 100644 programs/copier/ShardPartitionPiece.h
 delete mode 100644 programs/copier/StatusAccumulator.cpp
 delete mode 100644 programs/copier/StatusAccumulator.h
 delete mode 100644 programs/copier/TaskCluster.cpp
 delete mode 100644 programs/copier/TaskCluster.h
 delete mode 100644 programs/copier/TaskShard.cpp
 delete mode 100644 programs/copier/TaskShard.h
 delete mode 100644 programs/copier/TaskTable.cpp
 delete mode 100644 programs/copier/TaskTable.h
 delete mode 100644 programs/copier/ZooKeeperStaff.h
 delete mode 100644 programs/copier/clickhouse-copier.cpp
 delete mode 100644 tests/integration/test_cluster_copier/__init__.py
 delete mode 100644 tests/integration/test_cluster_copier/configs/conf.d/clusters.xml
 delete mode 100644 tests/integration/test_cluster_copier/configs/conf.d/clusters_trivial.xml
 delete mode 100644 tests/integration/test_cluster_copier/configs/conf.d/ddl.xml
 delete mode 100644 tests/integration/test_cluster_copier/configs/conf.d/query_log.xml
 delete mode 100644 tests/integration/test_cluster_copier/configs/config-copier.xml
 delete mode 100644 tests/integration/test_cluster_copier/configs/users.xml
 delete mode 100644 tests/integration/test_cluster_copier/configs_three_nodes/conf.d/clusters.xml
 delete mode 100644 tests/integration/test_cluster_copier/configs_three_nodes/conf.d/ddl.xml
 delete mode 100644 tests/integration/test_cluster_copier/configs_three_nodes/config-copier.xml
 delete mode 100644 tests/integration/test_cluster_copier/configs_three_nodes/users.xml
 delete mode 100644 tests/integration/test_cluster_copier/configs_two_nodes/conf.d/clusters.xml
 delete mode 100644 tests/integration/test_cluster_copier/configs_two_nodes/conf.d/ddl.xml
 delete mode 100644 tests/integration/test_cluster_copier/configs_two_nodes/conf.d/storage_configuration.xml
 delete mode 100644 tests/integration/test_cluster_copier/configs_two_nodes/config-copier.xml
 delete mode 100644 tests/integration/test_cluster_copier/configs_two_nodes/users.xml
 delete mode 100644 tests/integration/test_cluster_copier/task0_description.xml
 delete mode 100644 tests/integration/test_cluster_copier/task_drop_target_partition.xml
 delete mode 100644 tests/integration/test_cluster_copier/task_month_to_week_description.xml
 delete mode 100644 tests/integration/test_cluster_copier/task_no_arg.xml
 delete mode 100644 tests/integration/test_cluster_copier/task_no_index.xml
 delete mode 100644 tests/integration/test_cluster_copier/task_non_partitioned_table.xml
 delete mode 100644 tests/integration/test_cluster_copier/task_self_copy.xml
 delete mode 100644 tests/integration/test_cluster_copier/task_skip_index.xml
 delete mode 100644 tests/integration/test_cluster_copier/task_taxi_data.xml
 delete mode 100644 tests/integration/test_cluster_copier/task_test_block_size.xml
 delete mode 100644 tests/integration/test_cluster_copier/task_trivial.xml
 delete mode 100644 tests/integration/test_cluster_copier/task_trivial_without_arguments.xml
 delete mode 100644 tests/integration/test_cluster_copier/task_ttl_columns.xml
 delete mode 100644 tests/integration/test_cluster_copier/task_ttl_move_to_volume.xml
 delete mode 100644 tests/integration/test_cluster_copier/task_with_different_schema.xml
 delete mode 100644 tests/integration/test_cluster_copier/test.py
 delete mode 100644 tests/integration/test_cluster_copier/test_three_nodes.py
 delete mode 100644 tests/integration/test_cluster_copier/test_trivial.py
 delete mode 100644 tests/integration/test_cluster_copier/test_two_nodes.py

diff --git a/docker/packager/README.md b/docker/packager/README.md
index e0b7f38ea58..3604e8585a4 100644
--- a/docker/packager/README.md
+++ b/docker/packager/README.md
@@ -28,7 +28,6 @@ lrwxrwxrwx 1 root root        10  clickhouse-benchmark -> clickhouse
 lrwxrwxrwx 1 root root        10  clickhouse-clang -> clickhouse
 lrwxrwxrwx 1 root root        10  clickhouse-client -> clickhouse
 lrwxrwxrwx 1 root root        10  clickhouse-compressor -> clickhouse
-lrwxrwxrwx 1 root root        10  clickhouse-copier -> clickhouse
 lrwxrwxrwx 1 root root        10  clickhouse-extract-from-config -> clickhouse
 lrwxrwxrwx 1 root root        10  clickhouse-format -> clickhouse
 lrwxrwxrwx 1 root root        10  clickhouse-lld -> clickhouse
diff --git a/docs/en/operations/backup.md b/docs/en/operations/backup.md
index 2d9bf2a2ee8..8639af468c2 100644
--- a/docs/en/operations/backup.md
+++ b/docs/en/operations/backup.md
@@ -170,7 +170,7 @@ RESTORE TABLE test.table PARTITIONS '2', '3'
 
 ### Backups as tar archives
 
-Backups can also be stored as tar archives. The functionality is the same as for zip, except that a password is not supported. 
+Backups can also be stored as tar archives. The functionality is the same as for zip, except that a password is not supported.
 
 Write a backup as a tar:
 ```
@@ -444,10 +444,6 @@ Often data that is ingested into ClickHouse is delivered through some sort of pe
 
 Some local filesystems provide snapshot functionality (for example, [ZFS](https://en.wikipedia.org/wiki/ZFS)), but they might not be the best choice for serving live queries. A possible solution is to create additional replicas with this kind of filesystem and exclude them from the [Distributed](../engines/table-engines/special/distributed.md) tables that are used for `SELECT` queries. Snapshots on such replicas will be out of reach of any queries that modify data. As a bonus, these replicas might have special hardware configurations with more disks attached per server, which would be cost-effective.
 
-### clickhouse-copier {#clickhouse-copier}
-
-[clickhouse-copier](../operations/utilities/clickhouse-copier.md) is a versatile tool that was initially created to re-shard petabyte-sized tables. It can also be used for backup and restore purposes because it reliably copies data between ClickHouse tables and clusters.
-
 For smaller volumes of data, a simple `INSERT INTO ... SELECT ...` to remote tables might work as well.
 
 ### Manipulations with Parts {#manipulations-with-parts}
diff --git a/docs/en/operations/utilities/clickhouse-copier.md b/docs/en/operations/utilities/clickhouse-copier.md
deleted file mode 100644
index 0d329487504..00000000000
--- a/docs/en/operations/utilities/clickhouse-copier.md
+++ /dev/null
@@ -1,187 +0,0 @@
----
-slug: /en/operations/utilities/clickhouse-copier
-sidebar_position: 59
-sidebar_label: clickhouse-copier
----
-
-# clickhouse-copier 
-
-Copies data from the tables in one cluster to tables in another (or the same) cluster.
-
-:::note    
-To get a consistent copy, the data in the source tables and partitions should not change during the entire process.
-:::
-
-You can run multiple `clickhouse-copier` instances on different servers to perform the same job. ClickHouse Keeper, or ZooKeeper, is used for syncing the processes.
-
-After starting, `clickhouse-copier`:
-
-- Connects to ClickHouse Keeper and receives:
-
-    - Copying jobs.
-    - The state of the copying jobs.
-
-- It performs the jobs.
-
-    Each running process chooses the “closest” shard of the source cluster and copies the data into the destination cluster, resharding the data if necessary.
-
-`clickhouse-copier` tracks the changes in ClickHouse Keeper and applies them on the fly.
-
-To reduce network traffic, we recommend running `clickhouse-copier` on the same server where the source data is located.
-
-## Running Clickhouse-copier {#running-clickhouse-copier}
-
-The utility should be run manually:
-
-``` bash
-$ clickhouse-copier --daemon --config keeper.xml --task-path /task/path --base-dir /path/to/dir
-```
-
-Parameters:
-
-- `daemon` — Starts `clickhouse-copier` in daemon mode.
-- `config` — The path to the `keeper.xml` file with the parameters for the connection to ClickHouse Keeper.
-- `task-path` — The path to the ClickHouse Keeper node. This node is used for syncing `clickhouse-copier` processes and storing tasks. Tasks are stored in `$task-path/description`.
-- `task-file` — Optional path to file with task configuration for initial upload to ClickHouse Keeper.
-- `task-upload-force` — Force upload `task-file` even if node already exists. Default is false.
-- `base-dir` — The path to logs and auxiliary files. When it starts, `clickhouse-copier` creates `clickhouse-copier_YYYYMMHHSS_<PID>` subdirectories in `$base-dir`. If this parameter is omitted, the directories are created in the directory where `clickhouse-copier` was launched.
-
-## Format of keeper.xml {#format-of-zookeeper-xml}
-
-``` xml
-<clickhouse>
-    <logger>
-        <level>trace</level>
-        <size>100M</size>
-        <count>3</count>
-    </logger>
-
-    <zookeeper>
-        <node index="1">
-            <host>127.0.0.1</host>
-            <port>2181</port>
-        </node>
-    </zookeeper>
-</clickhouse>
-```
-
-## Configuration of Copying Tasks {#configuration-of-copying-tasks}
-
-``` xml
-<clickhouse>
-    <!-- Configuration of clusters as in an ordinary server config -->
-    <remote_servers>
-        <source_cluster>
-            <!--
-                source cluster & destination clusters accept exactly the same
-                parameters as parameters for the usual Distributed table
-                see https://clickhouse.com/docs/en/engines/table-engines/special/distributed/
-            -->
-            <shard>
-                <internal_replication>false</internal_replication>
-                    <replica>
-                        <host>127.0.0.1</host>
-                        <port>9000</port>
-                        <!--
-                        <user>default</user>
-                        <password>default</password>
-                        <secure>1</secure>
-                        -->
-                    </replica>
-            </shard>
-            ...
-        </source_cluster>
-
-        <destination_cluster>
-        ...
-        </destination_cluster>
-    </remote_servers>
-
-    <!-- How many simultaneously active workers are possible. If you run more workers superfluous workers will sleep. -->
-    <max_workers>2</max_workers>
-
-    <!-- Setting used to fetch (pull) data from source cluster tables -->
-    <settings_pull>
-        <readonly>1</readonly>
-    </settings_pull>
-
-    <!-- Setting used to insert (push) data to destination cluster tables -->
-    <settings_push>
-        <readonly>0</readonly>
-    </settings_push>
-
-    <!-- Common setting for fetch (pull) and insert (push) operations. Also, copier process context uses it.
-         They are overlaid by <settings_pull/> and <settings_push/> respectively. -->
-    <settings>
-        <connect_timeout>3</connect_timeout>
-        <!-- Sync insert is set forcibly, leave it here just in case. -->
-        <distributed_foreground_insert>1</distributed_foreground_insert>
-    </settings>
-
-    <!-- Copying tasks description.
-         You could specify several table task in the same task description (in the same ZooKeeper node), they will be performed
-         sequentially.
-    -->
-    <tables>
-        <!-- A table task, copies one table. -->
-        <table_hits>
-            <!-- Source cluster name (from <remote_servers/> section) and tables in it that should be copied -->
-            <cluster_pull>source_cluster</cluster_pull>
-            <database_pull>test</database_pull>
-            <table_pull>hits</table_pull>
-
-            <!-- Destination cluster name and tables in which the data should be inserted -->
-            <cluster_push>destination_cluster</cluster_push>
-            <database_push>test</database_push>
-            <table_push>hits2</table_push>
-
-            <!-- Engine of destination tables.
-                 If destination tables have not be created, workers create them using columns definition from source tables and engine
-                 definition from here.
-
-                 NOTE: If the first worker starts insert data and detects that destination partition is not empty then the partition will
-                 be dropped and refilled, take it into account if you already have some data in destination tables. You could directly
-                 specify partitions that should be copied in <enabled_partitions/>, they should be in quoted format like partition column of
-                 system.parts table.
-            -->
-            <engine>
-            ENGINE=ReplicatedMergeTree('/clickhouse/tables/{cluster}/{shard}/hits2', '{replica}')
-            PARTITION BY toMonday(date)
-            ORDER BY (CounterID, EventDate)
-            </engine>
-
-            <!-- Sharding key used to insert data to destination cluster -->
-            <sharding_key>jumpConsistentHash(intHash64(UserID), 2)</sharding_key>
-
-            <!-- Optional expression that filter data while pull them from source servers -->
-            <where_condition>CounterID != 0</where_condition>
-
-            <!-- This section specifies partitions that should be copied, other partition will be ignored.
-                 Partition names should have the same format as
-                 partition column of system.parts table (i.e. a quoted text).
-                 Since partition key of source and destination cluster could be different,
-                 these partition names specify destination partitions.
-
-                 NOTE: In spite of this section is optional (if it is not specified, all partitions will be copied),
-                 it is strictly recommended to specify them explicitly.
-                 If you already have some ready partitions on destination cluster they
-                 will be removed at the start of the copying since they will be interpeted
-                 as unfinished data from the previous copying!!!
-            -->
-            <enabled_partitions>
-                <partition>'2018-02-26'</partition>
-                <partition>'2018-03-05'</partition>
-                ...
-            </enabled_partitions>
-        </table_hits>
-
-        <!-- Next table to copy. It is not copied until previous table is copying. -->
-        <table_visits>
-        ...
-        </table_visits>
-        ...
-    </tables>
-</clickhouse>
-```
-
-`clickhouse-copier` tracks the changes in `/task/path/description` and applies them on the fly. For instance, if you change the value of `max_workers`, the number of processes running tasks will also change.
diff --git a/docs/en/operations/utilities/index.md b/docs/en/operations/utilities/index.md
index 8959073d00e..912a5b9ccb1 100644
--- a/docs/en/operations/utilities/index.md
+++ b/docs/en/operations/utilities/index.md
@@ -2,13 +2,11 @@
 slug: /en/operations/utilities/
 sidebar_position: 56
 sidebar_label: List of tools and utilities
-pagination_next: 'en/operations/utilities/clickhouse-copier'
 ---
 
 # List of tools and utilities
 
 - [clickhouse-local](../../operations/utilities/clickhouse-local.md) — Allows running SQL queries on data without starting the ClickHouse server, similar to how `awk` does this.
-- [clickhouse-copier](../../operations/utilities/clickhouse-copier.md) — Copies (and reshards) data from one cluster to another cluster.
 - [clickhouse-benchmark](../../operations/utilities/clickhouse-benchmark.md) — Loads server with the custom queries and settings.
 - [clickhouse-format](../../operations/utilities/clickhouse-format.md) — Enables formatting input queries.
 - [ClickHouse obfuscator](../../operations/utilities/clickhouse-obfuscator.md) — Obfuscates data.
diff --git a/docs/en/sql-reference/statements/alter/column.md b/docs/en/sql-reference/statements/alter/column.md
index 0989c151d18..a23710b12bd 100644
--- a/docs/en/sql-reference/statements/alter/column.md
+++ b/docs/en/sql-reference/statements/alter/column.md
@@ -335,7 +335,7 @@ The `ALTER` query lets you create and delete separate elements (columns) in nest
 
 There is no support for deleting columns in the primary key or the sampling key (columns that are used in the `ENGINE` expression). Changing the type for columns that are included in the primary key is only possible if this change does not cause the data to be modified (for example, you are allowed to add values to an Enum or to change a type from `DateTime` to `UInt32`).
 
-If the `ALTER` query is not sufficient to make the table changes you need, you can create a new table, copy the data to it using the [INSERT SELECT](/docs/en/sql-reference/statements/insert-into.md/#inserting-the-results-of-select) query, then switch the tables using the [RENAME](/docs/en/sql-reference/statements/rename.md/#rename-table) query and delete the old table. You can use the [clickhouse-copier](/docs/en/operations/utilities/clickhouse-copier.md) as an alternative to the `INSERT SELECT` query.
+If the `ALTER` query is not sufficient to make the table changes you need, you can create a new table, copy the data to it using the [INSERT SELECT](/docs/en/sql-reference/statements/insert-into.md/#inserting-the-results-of-select) query, then switch the tables using the [RENAME](/docs/en/sql-reference/statements/rename.md/#rename-table) query and delete the old table.
 
 The `ALTER` query blocks all reads and writes for the table. In other words, if a long `SELECT` is running at the time of the `ALTER` query, the `ALTER` query will wait for it to complete. At the same time, all new queries to the same table will wait while this `ALTER` is running.
 
diff --git a/docs/ru/getting-started/tutorial.md b/docs/ru/getting-started/tutorial.md
index 34064b6cf2f..8c827137e6d 100644
--- a/docs/ru/getting-started/tutorial.md
+++ b/docs/ru/getting-started/tutorial.md
@@ -585,10 +585,6 @@ ENGINE = Distributed(perftest_3shards_1replicas, tutorial, hits_local, rand());
 INSERT INTO tutorial.hits_all SELECT * FROM tutorial.hits_v1;
 ```
 
-:::danger Внимание!
-Этот подход не годится для сегментирования больших таблиц. Есть инструмент [clickhouse-copier](../operations/utilities/clickhouse-copier.md), специально предназначенный для перераспределения любых больших таблиц.
-:::
-
 Как и следовало ожидать, вычислительно сложные запросы работают втрое быстрее, если они выполняются на трёх серверах, а не на одном.
 
 В данном случае мы использовали кластер из трёх сегментов с одной репликой для каждого.
diff --git a/docs/ru/operations/backup.md b/docs/ru/operations/backup.md
index 9ff13bbc8a6..50ee6b45e09 100644
--- a/docs/ru/operations/backup.md
+++ b/docs/ru/operations/backup.md
@@ -24,12 +24,6 @@ sidebar_label: "Резервное копирование данных"
 
 Некоторые локальные файловые системы позволяют делать снимки (например, [ZFS](https://en.wikipedia.org/wiki/ZFS)), но они могут быть не лучшим выбором для обслуживания живых запросов. Возможным решением является создание дополнительных реплик с такой файловой системой и исключение их из [Distributed](../engines/table-engines/special/distributed.md) таблиц, используемых для запросов `SELECT`. Снимки на таких репликах будут недоступны для запросов, изменяющих данные. В качестве бонуса, эти реплики могут иметь особые конфигурации оборудования с большим количеством дисков, подключенных к серверу, что будет экономически эффективным.
 
-## clickhouse-copier {#clickhouse-copier}
-
-[clickhouse-copier](utilities/clickhouse-copier.md) — это универсальный инструмент, который изначально был создан для перешардирования таблиц с петабайтами данных. Его также можно использовать для резервного копирования и восстановления, поскольку он надёжно копирует данные между таблицами и кластерами ClickHouse.
-
-Для небольших объёмов данных можно применять `INSERT INTO ... SELECT ...` в удалённые таблицы.
-
 ## Манипуляции с партициями {#manipuliatsii-s-partitsiiami}
 
 ClickHouse позволяет использовать запрос `ALTER TABLE ... FREEZE PARTITION ...` для создания локальной копии партиций таблицы. Это реализуется с помощью жестких ссылок (hardlinks) на каталог `/var/lib/clickhouse/shadow/`, поэтому такая копия обычно не занимает дополнительное место на диске для старых данных. Созданные копии файлов не обрабатываются сервером ClickHouse, поэтому вы можете просто оставить их там: у вас будет простая резервная копия, которая не требует дополнительной внешней системы, однако при аппаратных проблемах вы можете утратить и актуальные данные и сохраненную копию. По этой причине, лучше удаленно скопировать их в другое место, а затем удалить локальную копию. Распределенные файловые системы и хранилища объектов по-прежнему являются хорошими вариантами для этого, однако можно использовать и обычные присоединенные файловые серверы с достаточно большой ёмкостью (в этом случае передача будет происходить через сетевую файловую систему или, возможно, [rsync](https://en.wikipedia.org/wiki/Rsync)).
diff --git a/docs/ru/operations/utilities/clickhouse-copier.md b/docs/ru/operations/utilities/clickhouse-copier.md
deleted file mode 100644
index da86ef2d35d..00000000000
--- a/docs/ru/operations/utilities/clickhouse-copier.md
+++ /dev/null
@@ -1,183 +0,0 @@
----
-slug: /ru/operations/utilities/clickhouse-copier
-sidebar_position: 59
-sidebar_label: clickhouse-copier
----
-
-# clickhouse-copier {#clickhouse-copier}
-
-Копирует данные из таблиц одного кластера в таблицы другого (или этого же) кластера.
-
-Можно запустить несколько `clickhouse-copier` для разных серверах для выполнения одного и того же задания. Для синхронизации между процессами используется ZooKeeper.
-
-После запуска, `clickhouse-copier`:
-
--   Соединяется с ZooKeeper и получает:
-
-    -   Задания на копирование.
-    -   Состояние заданий на копирование.
-
--   Выполняет задания.
-
-        Каждый запущенный процесс выбирает "ближайший" шард исходного кластера и копирует данные в кластер назначения, при необходимости перешардируя их.
-
-`clickhouse-copier` отслеживает изменения в ZooKeeper и применяет их «на лету».
-
-Для снижения сетевого трафика рекомендуем запускать `clickhouse-copier` на том же сервере, где находятся исходные данные.
-
-## Запуск Clickhouse-copier {#zapusk-clickhouse-copier}
-
-Утилиту следует запускать вручную следующим образом:
-
-``` bash
-$ clickhouse-copier --daemon --config zookeeper.xml --task-path /task/path --base-dir /path/to/dir
-```
-
-Параметры запуска:
-
--   `daemon` - запускает `clickhouse-copier` в режиме демона.
--   `config` - путь к файлу `zookeeper.xml` с параметрами соединения с ZooKeeper.
--   `task-path` - путь к ноде ZooKeeper. Нода используется для синхронизации между процессами `clickhouse-copier` и для хранения заданий. Задания хранятся в `$task-path/description`.
--   `task-file` - необязательный путь к файлу с описанием конфигурация заданий для загрузки в ZooKeeper.
--   `task-upload-force` - Загрузить `task-file` в ZooKeeper даже если уже было загружено.
--   `base-dir` - путь к логам и вспомогательным файлам. При запуске `clickhouse-copier` создает в `$base-dir` подкаталоги `clickhouse-copier_YYYYMMHHSS_<PID>`. Если параметр не указан, то каталоги будут создаваться в каталоге, где `clickhouse-copier` был запущен.
-
-## Формат Zookeeper.xml {#format-zookeeper-xml}
-
-``` xml
-<clickhouse>
-    <logger>
-        <level>trace</level>
-        <size>100M</size>
-        <count>3</count>
-    </logger>
-
-    <zookeeper>
-        <node index="1">
-            <host>127.0.0.1</host>
-            <port>2181</port>
-        </node>
-    </zookeeper>
-</clickhouse>
-```
-
-## Конфигурация заданий на копирование {#konfiguratsiia-zadanii-na-kopirovanie}
-
-``` xml
-<clickhouse>
-    <!-- Configuration of clusters as in an ordinary server config -->
-    <remote_servers>
-        <source_cluster>
-		    <!--
-                source cluster & destination clusters accept exactly the same
-                parameters as parameters for the usual Distributed table
-                see https://clickhouse.com/docs/ru/engines/table-engines/special/distributed/
-            -->
-            <shard>
-                <internal_replication>false</internal_replication>
-                    <replica>
-                        <host>127.0.0.1</host>
-                        <port>9000</port>
-						<!--
-                        <user>default</user>
-                        <password>default</password>
-                        <secure>1</secure>
-                        -->
-                    </replica>
-            </shard>
-            ...
-        </source_cluster>
-
-        <destination_cluster>
-        ...
-        </destination_cluster>
-    </remote_servers>
-
-    <!-- How many simultaneously active workers are possible. If you run more workers superfluous workers will sleep. -->
-    <max_workers>2</max_workers>
-
-    <!-- Setting used to fetch (pull) data from source cluster tables -->
-    <settings_pull>
-        <readonly>1</readonly>
-    </settings_pull>
-
-    <!-- Setting used to insert (push) data to destination cluster tables -->
-    <settings_push>
-        <readonly>0</readonly>
-    </settings_push>
-
-    <!-- Common setting for fetch (pull) and insert (push) operations. Also, copier process context uses it.
-         They are overlaid by <settings_pull/> and <settings_push/> respectively. -->
-    <settings>
-        <connect_timeout>3</connect_timeout>
-        <!-- Sync insert is set forcibly, leave it here just in case. -->
-        <distributed_foreground_insert>1</distributed_foreground_insert>
-    </settings>
-
-    <!-- Copying tasks description.
-         You could specify several table task in the same task description (in the same ZooKeeper node), they will be performed
-         sequentially.
-    -->
-    <tables>
-        <!-- A table task, copies one table. -->
-        <table_hits>
-            <!-- Source cluster name (from <remote_servers/> section) and tables in it that should be copied -->
-            <cluster_pull>source_cluster</cluster_pull>
-            <database_pull>test</database_pull>
-            <table_pull>hits</table_pull>
-
-            <!-- Destination cluster name and tables in which the data should be inserted -->
-            <cluster_push>destination_cluster</cluster_push>
-            <database_push>test</database_push>
-            <table_push>hits2</table_push>
-
-            <!-- Engine of destination tables.
-                 If destination tables have not be created, workers create them using columns definition from source tables and engine
-                 definition from here.
-
-                 NOTE: If the first worker starts insert data and detects that destination partition is not empty then the partition will
-                 be dropped and refilled, take it into account if you already have some data in destination tables. You could directly
-                 specify partitions that should be copied in <enabled_partitions/>, they should be in quoted format like partition column of
-                 system.parts table.
-            -->
-            <engine>
-            ENGINE=ReplicatedMergeTree('/clickhouse/tables/{cluster}/{shard}/hits2', '{replica}')
-            PARTITION BY toMonday(date)
-            ORDER BY (CounterID, EventDate)
-            </engine>
-
-            <!-- Sharding key used to insert data to destination cluster -->
-            <sharding_key>jumpConsistentHash(intHash64(UserID), 2)</sharding_key>
-
-            <!-- Optional expression that filter data while pull them from source servers -->
-            <where_condition>CounterID != 0</where_condition>
-
-            <!-- This section specifies partitions that should be copied, other partition will be ignored.
-                 Partition names should have the same format as
-                 partition column of system.parts table (i.e. a quoted text).
-                 Since partition key of source and destination cluster could be different,
-                 these partition names specify destination partitions.
-
-                 NOTE: In spite of this section is optional (if it is not specified, all partitions will be copied),
-                 it is strictly recommended to specify them explicitly.
-                 If you already have some ready partitions on destination cluster they
-                 will be removed at the start of the copying since they will be interpeted
-                 as unfinished data from the previous copying!!!
-            -->
-            <enabled_partitions>
-                <partition>'2018-02-26'</partition>
-                <partition>'2018-03-05'</partition>
-                ...
-            </enabled_partitions>
-        </table_hits>
-
-        <!-- Next table to copy. It is not copied until previous table is copying. -->
-        <table_visits>
-        ...
-        </table_visits>
-        ...
-    </tables>
-</clickhouse>
-```
-
-`clickhouse-copier` отслеживает изменения `/task/path/description` и применяет их «на лету». Если вы поменяете, например, значение `max_workers`, то количество процессов, выполняющих задания, также изменится.
diff --git a/docs/ru/operations/utilities/index.md b/docs/ru/operations/utilities/index.md
index 9eb90a3037c..e4b01a0276d 100644
--- a/docs/ru/operations/utilities/index.md
+++ b/docs/ru/operations/utilities/index.md
@@ -7,7 +7,6 @@ sidebar_position: 56
 # Утилиты ClickHouse {#utility-clickhouse}
 
 -   [clickhouse-local](clickhouse-local.md) - позволяет выполнять SQL-запросы над данными без остановки сервера ClickHouse, подобно утилите `awk`.
--   [clickhouse-copier](clickhouse-copier.md) - копирует (и перешардирует) данные с одного кластера на другой.
 -   [clickhouse-benchmark](../../operations/utilities/clickhouse-benchmark.md) — устанавливает соединение с сервером ClickHouse и запускает циклическое выполнение указанных запросов.
 -   [clickhouse-format](../../operations/utilities/clickhouse-format.md) — позволяет форматировать входящие запросы.
 -   [ClickHouse obfuscator](../../operations/utilities/clickhouse-obfuscator.md) — обфусцирует данные.
diff --git a/docs/ru/sql-reference/statements/alter/column.md b/docs/ru/sql-reference/statements/alter/column.md
index 385a9835eca..2ea045f4ae3 100644
--- a/docs/ru/sql-reference/statements/alter/column.md
+++ b/docs/ru/sql-reference/statements/alter/column.md
@@ -94,7 +94,7 @@ RENAME COLUMN [IF EXISTS] name to new_name
 
 Переименовывает столбец `name` в `new_name`. Если указано выражение `IF EXISTS`, то запрос не будет возвращать ошибку при условии, что столбец `name` не существует. Поскольку переименование не затрагивает физические данные колонки, запрос выполняется практически мгновенно.
 
-**ЗАМЕЧЕНИЕ**: Столбцы, являющиеся частью основного ключа или ключа сортировки (заданные с помощью `ORDER BY` или `PRIMARY KEY`), не могут быть переименованы. Попытка переименовать эти слобцы приведет к `SQL Error [524]`. 
+**ЗАМЕЧЕНИЕ**: Столбцы, являющиеся частью основного ключа или ключа сортировки (заданные с помощью `ORDER BY` или `PRIMARY KEY`), не могут быть переименованы. Попытка переименовать эти слобцы приведет к `SQL Error [524]`.
 
 Пример:
 
@@ -254,7 +254,7 @@ SELECT groupArray(x), groupArray(s) FROM tmp;
 
 Отсутствует возможность удалять столбцы, входящие в первичный ключ или ключ для сэмплирования (в общем, входящие в выражение `ENGINE`). Изменение типа у столбцов, входящих в первичный ключ возможно только в том случае, если это изменение не приводит к изменению данных (например, разрешено добавление значения в Enum или изменение типа с `DateTime` на `UInt32`).
 
-Если возможностей запроса `ALTER` не хватает для нужного изменения таблицы, вы можете создать новую таблицу, скопировать туда данные с помощью запроса [INSERT SELECT](../insert-into.md#inserting-the-results-of-select), затем поменять таблицы местами с помощью запроса [RENAME](../rename.md#rename-table), и удалить старую таблицу. В качестве альтернативы для запроса `INSERT SELECT`, можно использовать инструмент [clickhouse-copier](../../../sql-reference/statements/alter/index.md).
+Если возможностей запроса `ALTER` не хватает для нужного изменения таблицы, вы можете создать новую таблицу, скопировать туда данные с помощью запроса [INSERT SELECT](../insert-into.md#inserting-the-results-of-select), затем поменять таблицы местами с помощью запроса [RENAME](../rename.md#rename-table), и удалить старую таблицу.
 
 Запрос `ALTER` блокирует все чтения и записи для таблицы. То есть если на момент запроса `ALTER` выполнялся долгий `SELECT`, то запрос `ALTER` сначала дождётся его выполнения. И в это время все новые запросы к той же таблице будут ждать, пока завершится этот `ALTER`.
 
diff --git a/docs/zh/getting-started/tutorial.md b/docs/zh/getting-started/tutorial.md
index 989cf5f57d8..d0c9bda83ef 100644
--- a/docs/zh/getting-started/tutorial.md
+++ b/docs/zh/getting-started/tutorial.md
@@ -582,8 +582,6 @@ ENGINE = Distributed(perftest_3shards_1replicas, tutorial, hits_local, rand());
 INSERT INTO tutorial.hits_all SELECT * FROM tutorial.hits_v1;
 ```
 
-!!! warning "注意:"
-    这种方法不适合大型表的分片。 有一个单独的工具 [clickhouse-copier](../operations/utilities/clickhouse-copier.md) 这可以重新分片任意大表。
 
 正如您所期望的那样，如果计算量大的查询使用3台服务器而不是一个，则运行速度快N倍。
 
diff --git a/docs/zh/operations/backup.md b/docs/zh/operations/backup.md
index 6d491f9c2f7..48e852b4228 100644
--- a/docs/zh/operations/backup.md
+++ b/docs/zh/operations/backup.md
@@ -24,12 +24,6 @@ sidebar_label: "\u6570\u636E\u5907\u4EFD"
 
 某些本地文件系统提供快照功能（例如, [ZFS](https://en.wikipedia.org/wiki/ZFS)），但它们可能不是提供实时查询的最佳选择。 一个可能的解决方案是使用这种文件系统创建额外的副本，并将它们与用于`SELECT` 查询的 [分布式](../engines/table-engines/special/distributed.md) 表分离。 任何修改数据的查询都无法访问此类副本上的快照。 作为回报，这些副本可能具有特殊的硬件配置，每个服务器附加更多的磁盘，这将是经济高效的。
 
-## clickhouse-copier {#clickhouse-copier}
-
-[clickhouse-copier](utilities/clickhouse-copier.md) 是一个多功能工具，最初创建它是为了用于重新切分pb大小的表。 因为它能够在ClickHouse表和集群之间可靠地复制数据，所以它也可用于备份和还原数据。
-
-对于较小的数据量，一个简单的 `INSERT INTO ... SELECT ...` 到远程表也可以工作。
-
 ## part操作 {#manipulations-with-parts}
 
 ClickHouse允许使用 `ALTER TABLE ... FREEZE PARTITION ...` 查询以创建表分区的本地副本。 这是利用硬链接(hardlink)到 `/var/lib/clickhouse/shadow/` 文件夹中实现的，所以它通常不会因为旧数据而占用额外的磁盘空间。 创建的文件副本不由ClickHouse服务器处理，所以你可以把它们留在那里：你将有一个简单的备份，不需要任何额外的外部系统，但它仍然容易出现硬件问题。 出于这个原因，最好将它们远程复制到另一个位置，然后删除本地副本。 分布式文件系统和对象存储仍然是一个不错的选择，但是具有足够大容量的正常附加文件服务器也可以工作（在这种情况下，传输将通过网络文件系统或者也许是 [rsync](https://en.wikipedia.org/wiki/Rsync) 来进行).
diff --git a/docs/zh/operations/utilities/clickhouse-copier.md b/docs/zh/operations/utilities/clickhouse-copier.md
deleted file mode 100644
index b01edd9257c..00000000000
--- a/docs/zh/operations/utilities/clickhouse-copier.md
+++ /dev/null
@@ -1,172 +0,0 @@
----
-slug: /zh/operations/utilities/clickhouse-copier
----
-# clickhouse-copier {#clickhouse-copier}
-
-将数据从一个群集中的表复制到另一个（或相同）群集中的表。
-
-您可以运行多个 `clickhouse-copier` 不同服务器上的实例执行相同的作业。 ZooKeeper用于同步进程。
-
-开始后, `clickhouse-copier`:
-
--   连接到ZooKeeper并且接收:
-
-    -   复制作业。
-    -   复制作业的状态。
-
--   它执行的工作。
-
-        每个正在运行的进程都会选择源集群的“最接近”分片，然后将数据复制到目标集群，并在必要时重新分片数据。
-
-`clickhouse-copier` 跟踪ZooKeeper中的更改，并实时应用它们。
-
-为了减少网络流量，我们建议运行 `clickhouse-copier` 在源数据所在的同一服务器上。
-
-## 运行Clickhouse-copier {#running-clickhouse-copier}
-
-该实用程序应手动运行:
-
-``` bash
-clickhouse-copier --daemon --config zookeeper.xml --task-path /task/path --base-dir /path/to/dir
-```
-
-参数:
-
--   `daemon` — 在守护进程模式下启动`clickhouse-copier`。
--   `config` — `zookeeper.xml`文件的路径，其中包含用于连接ZooKeeper的参数。
--   `task-path` — ZooKeeper节点的路径。 该节点用于同步`clickhouse-copier`进程和存储任务。 任务存储在`$task-path/description`中。
--   `task-file` — 可选的非必须参数, 指定一个包含任务配置的参数文件, 用于初始上传到ZooKeeper。
--   `task-upload-force` — 即使节点已经存在，也强制上载`task-file`。
--   `base-dir` — 日志和辅助文件的路径。 启动时，`clickhouse-copier`在`$base-dir`中创建`clickhouse-copier_YYYYMMHHSS_<PID>`子目录。 如果省略此参数，则会在启动`clickhouse-copier`的目录中创建目录。
-
-
-
-## Zookeeper.xml格式 {#format-of-zookeeper-xml}
-
-``` xml
-<clickhouse>
-    <logger>
-        <level>trace</level>
-        <size>100M</size>
-        <count>3</count>
-    </logger>
-
-    <zookeeper>
-        <node index="1">
-            <host>127.0.0.1</host>
-            <port>2181</port>
-        </node>
-    </zookeeper>
-</clickhouse>
-```
-
-## 复制任务的配置 {#configuration-of-copying-tasks}
-
-``` xml
-<clickhouse>
-    <!-- Configuration of clusters as in an ordinary server config -->
-    <remote_servers>
-        <source_cluster>
-            <shard>
-                <internal_replication>false</internal_replication>
-                    <replica>
-                        <host>127.0.0.1</host>
-                        <port>9000</port>
-                    </replica>
-            </shard>
-            ...
-        </source_cluster>
-
-        <destination_cluster>
-        ...
-        </destination_cluster>
-    </remote_servers>
-
-    <!-- How many simultaneously active workers are possible. If you run more workers superfluous workers will sleep. -->
-    <max_workers>2</max_workers>
-
-    <!-- Setting used to fetch (pull) data from source cluster tables -->
-    <settings_pull>
-        <readonly>1</readonly>
-    </settings_pull>
-
-    <!-- Setting used to insert (push) data to destination cluster tables -->
-    <settings_push>
-        <readonly>0</readonly>
-    </settings_push>
-
-    <!-- Common setting for fetch (pull) and insert (push) operations. Also, copier process context uses it.
-         They are overlaid by <settings_pull/> and <settings_push/> respectively. -->
-    <settings>
-        <connect_timeout>3</connect_timeout>
-        <!-- Sync insert is set forcibly, leave it here just in case. -->
-        <distributed_foreground_insert>1</distributed_foreground_insert>
-    </settings>
-
-    <!-- Copying tasks description.
-         You could specify several table task in the same task description (in the same ZooKeeper node), they will be performed
-         sequentially.
-    -->
-    <tables>
-        <!-- A table task, copies one table. -->
-        <table_hits>
-            <!-- Source cluster name (from <remote_servers/> section) and tables in it that should be copied -->
-            <cluster_pull>source_cluster</cluster_pull>
-            <database_pull>test</database_pull>
-            <table_pull>hits</table_pull>
-
-            <!-- Destination cluster name and tables in which the data should be inserted -->
-            <cluster_push>destination_cluster</cluster_push>
-            <database_push>test</database_push>
-            <table_push>hits2</table_push>
-
-            <!-- Engine of destination tables.
-                 If destination tables have not be created, workers create them using columns definition from source tables and engine
-                 definition from here.
-
-                 NOTE: If the first worker starts insert data and detects that destination partition is not empty then the partition will
-                 be dropped and refilled, take it into account if you already have some data in destination tables. You could directly
-                 specify partitions that should be copied in <enabled_partitions/>, they should be in quoted format like partition column of
-                 system.parts table.
-            -->
-            <engine>
-            ENGINE=ReplicatedMergeTree('/clickhouse/tables/{cluster}/{shard}/hits2', '{replica}')
-            PARTITION BY toMonday(date)
-            ORDER BY (CounterID, EventDate)
-            </engine>
-
-            <!-- Sharding key used to insert data to destination cluster -->
-            <sharding_key>jumpConsistentHash(intHash64(UserID), 2)</sharding_key>
-
-            <!-- Optional expression that filter data while pull them from source servers -->
-            <where_condition>CounterID != 0</where_condition>
-
-            <!-- This section specifies partitions that should be copied, other partition will be ignored.
-                 Partition names should have the same format as
-                 partition column of system.parts table (i.e. a quoted text).
-                 Since partition key of source and destination cluster could be different,
-                 these partition names specify destination partitions.
-
-                 NOTE: In spite of this section is optional (if it is not specified, all partitions will be copied),
-                 it is strictly recommended to specify them explicitly.
-                 If you already have some ready partitions on destination cluster they
-                 will be removed at the start of the copying since they will be interpeted
-                 as unfinished data from the previous copying!!!
-            -->
-            <enabled_partitions>
-                <partition>'2018-02-26'</partition>
-                <partition>'2018-03-05'</partition>
-                ...
-            </enabled_partitions>
-        </table_hits>
-
-        <!-- Next table to copy. It is not copied until previous table is copying. -->
-        <table_visits>
-        ...
-        </table_visits>
-        ...
-    </tables>
-</clickhouse>
-```
-
-`clickhouse-copier` 跟踪更改 `/task/path/description` 并在飞行中应用它们。 例如，如果你改变的值 `max_workers`，运行任务的进程数也会发生变化。
diff --git a/docs/zh/operations/utilities/index.md b/docs/zh/operations/utilities/index.md
index af158baf275..cebe312450c 100644
--- a/docs/zh/operations/utilities/index.md
+++ b/docs/zh/operations/utilities/index.md
@@ -4,5 +4,4 @@ slug: /zh/operations/utilities/
 # 实用工具 {#clickhouse-utility}
 
 -   [本地查询](clickhouse-local.md) —  在不停止ClickHouse服务的情况下，对数据执行查询操作(类似于 `awk` 命令)。
--   [跨集群复制](clickhouse-copier.md) —  在不同集群间复制数据。
 -   [性能测试](clickhouse-benchmark.md) — 连接到Clickhouse服务器，执行性能测试。
diff --git a/docs/zh/sql-reference/statements/alter.md b/docs/zh/sql-reference/statements/alter.md
index 002d5102fa3..48665ae04ab 100644
--- a/docs/zh/sql-reference/statements/alter.md
+++ b/docs/zh/sql-reference/statements/alter.md
@@ -150,7 +150,7 @@ ALTER TABLE visits MODIFY COLUMN browser Array(String)
 
 不支持对primary key或者sampling key中的列（在 `ENGINE` 表达式中用到的列）进行删除操作。改变包含在primary key中的列的类型时，如果操作不会导致数据的变化（例如，往Enum中添加一个值，或者将`DateTime` 类型改成 `UInt32`），那么这种操作是可行的。
 
-如果 `ALTER` 操作不足以完成你想要的表变动操作，你可以创建一张新的表，通过 [INSERT SELECT](../../sql-reference/statements/insert-into.md#inserting-the-results-of-select)将数据拷贝进去，然后通过  [RENAME](../../sql-reference/statements/misc.md#misc_operations-rename)将新的表改成和原有表一样的名称，并删除原有的表。你可以使用 [clickhouse-copier](../../operations/utilities/clickhouse-copier.md) 代替 `INSERT SELECT`。
+如果 `ALTER` 操作不足以完成你想要的表变动操作，你可以创建一张新的表，通过 [INSERT SELECT](../../sql-reference/statements/insert-into.md#inserting-the-results-of-select)将数据拷贝进去，然后通过  [RENAME](../../sql-reference/statements/misc.md#misc_operations-rename)将新的表改成和原有表一样的名称，并删除原有的表。
 
  `ALTER` 操作会阻塞对表的所有读写操作。换句话说，当一个大的 `SELECT` 语句和 `ALTER`同时执行时，`ALTER`会等待，直到 `SELECT` 执行结束。与此同时，当 `ALTER` 运行时，新的 sql 语句将会等待。
 
diff --git a/packages/clickhouse-server.yaml b/packages/clickhouse-server.yaml
index 7894129b8e3..dc183ead102 100644
--- a/packages/clickhouse-server.yaml
+++ b/packages/clickhouse-server.yaml
@@ -50,8 +50,6 @@ contents:
   dst: /etc/init.d/clickhouse-server
 - src: clickhouse-server.service
   dst: /lib/systemd/system/clickhouse-server.service
-- src: root/usr/bin/clickhouse-copier
-  dst: /usr/bin/clickhouse-copier
 - src: root/usr/bin/clickhouse-server
   dst: /usr/bin/clickhouse-server
 # clickhouse-keeper part
diff --git a/programs/CMakeLists.txt b/programs/CMakeLists.txt
index 62bcf068879..d945fdf4a6f 100644
--- a/programs/CMakeLists.txt
+++ b/programs/CMakeLists.txt
@@ -122,7 +122,6 @@ add_subdirectory (local)
 add_subdirectory (benchmark)
 add_subdirectory (extract-from-config)
 add_subdirectory (compressor)
-add_subdirectory (copier)
 add_subdirectory (format)
 add_subdirectory (obfuscator)
 add_subdirectory (install)
@@ -200,7 +199,6 @@ clickhouse_program_install(clickhouse-server server)
 clickhouse_program_install(clickhouse-client client chc)
 clickhouse_program_install(clickhouse-local local chl ch)
 clickhouse_program_install(clickhouse-benchmark benchmark)
-clickhouse_program_install(clickhouse-copier copier)
 clickhouse_program_install(clickhouse-extract-from-config extract-from-config)
 clickhouse_program_install(clickhouse-compressor compressor)
 clickhouse_program_install(clickhouse-format format)
diff --git a/programs/copier/Aliases.h b/programs/copier/Aliases.h
deleted file mode 100644
index 02be3441acd..00000000000
--- a/programs/copier/Aliases.h
+++ /dev/null
@@ -1,15 +0,0 @@
-#pragma once
-
-#include <base/types.h>
-
-#include <Poco/Util/AbstractConfiguration.h>
-
-#include <utility>
-
-namespace DB
-{
-    using ConfigurationPtr = Poco::AutoPtr<Poco::Util::AbstractConfiguration>;
-
-    using DatabaseAndTableName = std::pair<String, String>;
-    using ListOfDatabasesAndTableNames = std::vector<DatabaseAndTableName>;
-}
diff --git a/programs/copier/CMakeLists.txt b/programs/copier/CMakeLists.txt
deleted file mode 100644
index 2c17e70bc5e..00000000000
--- a/programs/copier/CMakeLists.txt
+++ /dev/null
@@ -1,28 +0,0 @@
-set(CLICKHOUSE_COPIER_SOURCES
-        "${CMAKE_CURRENT_SOURCE_DIR}/ClusterCopierApp.cpp"
-        "${CMAKE_CURRENT_SOURCE_DIR}/ClusterCopier.cpp"
-        "${CMAKE_CURRENT_SOURCE_DIR}/Internals.cpp"
-        "${CMAKE_CURRENT_SOURCE_DIR}/ShardPartition.cpp"
-        "${CMAKE_CURRENT_SOURCE_DIR}/ShardPartitionPiece.cpp"
-        "${CMAKE_CURRENT_SOURCE_DIR}/StatusAccumulator.cpp"
-        "${CMAKE_CURRENT_SOURCE_DIR}/TaskCluster.cpp"
-        "${CMAKE_CURRENT_SOURCE_DIR}/TaskShard.cpp"
-        "${CMAKE_CURRENT_SOURCE_DIR}/TaskTable.cpp")
-
-set (CLICKHOUSE_COPIER_LINK
-        PRIVATE
-                clickhouse_common_zookeeper
-                clickhouse_common_config
-                clickhouse_parsers
-                clickhouse_functions
-                clickhouse_table_functions
-                clickhouse_aggregate_functions
-                string_utils
-
-        PUBLIC
-                daemon
-)
-
-set(CLICKHOUSE_COPIER_INCLUDE SYSTEM PRIVATE ${CMAKE_CURRENT_SOURCE_DIR})
-
-clickhouse_program_add(copier)
diff --git a/programs/copier/ClusterCopier.cpp b/programs/copier/ClusterCopier.cpp
deleted file mode 100644
index 59505d08f5c..00000000000
--- a/programs/copier/ClusterCopier.cpp
+++ /dev/null
@@ -1,2076 +0,0 @@
-#include "ClusterCopier.h"
-
-#include "Internals.h"
-#include "StatusAccumulator.h"
-
-#include <Common/ZooKeeper/ZooKeeper.h>
-#include <Common/ZooKeeper/KeeperException.h>
-#include <Common/randomSeed.h>
-#include <Common/setThreadName.h>
-#include <Common/CurrentMetrics.h>
-#include <Interpreters/InterpreterInsertQuery.h>
-#include <Interpreters/InterpreterSelectWithUnionQuery.h>
-#include <Parsers/ASTFunction.h>
-#include <Processors/Transforms/ExpressionTransform.h>
-#include <QueryPipeline/QueryPipelineBuilder.h>
-#include <QueryPipeline/Chain.h>
-#include <Processors/Executors/PullingPipelineExecutor.h>
-#include <Processors/Executors/PushingPipelineExecutor.h>
-#include <Processors/Sources/RemoteSource.h>
-#include <Processors/QueryPlan/QueryPlan.h>
-#include <Processors/QueryPlan/BuildQueryPipelineSettings.h>
-#include <Processors/QueryPlan/Optimizations/QueryPlanOptimizationSettings.h>
-
-namespace CurrentMetrics
-{
-    extern const Metric LocalThread;
-    extern const Metric LocalThreadActive;
-    extern const Metric LocalThreadScheduled;
-}
-
-namespace DB
-{
-
-namespace ErrorCodes
-{
-    extern const int NOT_IMPLEMENTED;
-    extern const int LOGICAL_ERROR;
-    extern const int UNFINISHED;
-    extern const int BAD_ARGUMENTS;
-}
-
-
-void ClusterCopier::init()
-{
-    auto zookeeper = getContext()->getZooKeeper();
-
-    task_description_watch_callback = [this] (const Coordination::WatchResponse & response)
-    {
-        if (response.error != Coordination::Error::ZOK)
-            return;
-        UInt64 version = ++task_description_version;
-        LOG_INFO(log, "Task description should be updated, local version {}", version);
-    };
-
-    task_description_path = task_zookeeper_path + "/description";
-    task_cluster = std::make_unique<TaskCluster>(task_zookeeper_path, working_database_name);
-
-    reloadTaskDescription();
-
-    task_cluster->loadTasks(*task_cluster_current_config);
-    getContext()->setClustersConfig(task_cluster_current_config, false, task_cluster->clusters_prefix);
-
-    /// Set up shards and their priority
-    task_cluster->random_engine.seed(randomSeed());
-    for (auto & task_table : task_cluster->table_tasks)
-    {
-        task_table.cluster_pull = getContext()->getCluster(task_table.cluster_pull_name);
-        task_table.cluster_push = getContext()->getCluster(task_table.cluster_push_name);
-        task_table.initShards(task_cluster->random_engine);
-    }
-
-    LOG_INFO(log, "Will process {} table tasks", task_cluster->table_tasks.size());
-
-    /// Do not initialize tables, will make deferred initialization in process()
-
-    zookeeper->createAncestors(getWorkersPathVersion() + "/");
-    zookeeper->createAncestors(getWorkersPath() + "/");
-    /// Init status node
-    zookeeper->createIfNotExists(task_zookeeper_path + "/status", "{}");
-}
-
-template <typename T>
-decltype(auto) ClusterCopier::retry(T && func, UInt64 max_tries)
-{
-    std::exception_ptr exception;
-
-    if (max_tries == 0)
-        throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot perform zero retries");
-
-    for (UInt64 try_number = 1; try_number <= max_tries; ++try_number)
-    {
-        try
-        {
-            return func();
-        }
-        catch (...)
-        {
-            exception = std::current_exception();
-            if (try_number < max_tries)
-            {
-                tryLogCurrentException(log, "Will retry");
-                std::this_thread::sleep_for(retry_delay_ms);
-            }
-        }
-    }
-
-    std::rethrow_exception(exception);
-}
-
-
-void ClusterCopier::discoverShardPartitions(const ConnectionTimeouts & timeouts, const TaskShardPtr & task_shard)
-{
-    TaskTable & task_table = task_shard->task_table;
-
-    LOG_INFO(log, "Discover partitions of shard {}", task_shard->getDescription());
-
-    auto get_partitions = [&] () { return getShardPartitions(timeouts, *task_shard); };
-    auto existing_partitions_names = retry(get_partitions, 60);
-    Strings filtered_partitions_names;
-    Strings missing_partitions;
-
-    /// Check that user specified correct partition names
-    auto check_partition_format = [] (const DataTypePtr & type, const String & partition_text_quoted)
-    {
-        MutableColumnPtr column_dummy = type->createColumn();
-        ReadBufferFromString rb(partition_text_quoted);
-
-        try
-        {
-            type->getDefaultSerialization()->deserializeTextQuoted(*column_dummy, rb, FormatSettings());
-        }
-        catch (Exception & e)
-        {
-            throw Exception(ErrorCodes::BAD_ARGUMENTS, "Partition {} has incorrect format. {}", partition_text_quoted, e.displayText());
-        }
-    };
-
-    if (task_table.has_enabled_partitions)
-    {
-        /// Process partition in order specified by <enabled_partitions/>
-        for (const String & partition_name : task_table.enabled_partitions)
-        {
-            /// Check that user specified correct partition names
-            check_partition_format(task_shard->partition_key_column.type, partition_name);
-
-            auto it = existing_partitions_names.find(partition_name);
-
-            /// Do not process partition if it is not in enabled_partitions list
-            if (it == existing_partitions_names.end())
-            {
-                missing_partitions.emplace_back(partition_name);
-                continue;
-            }
-
-            filtered_partitions_names.emplace_back(*it);
-        }
-
-        for (const String & partition_name : existing_partitions_names)
-        {
-            if (!task_table.enabled_partitions_set.contains(partition_name))
-            {
-                LOG_INFO(log, "Partition {} will not be processed, since it is not in enabled_partitions of {}", partition_name, task_table.table_id);
-            }
-        }
-    }
-    else
-    {
-        for (const String & partition_name : existing_partitions_names)
-            filtered_partitions_names.emplace_back(partition_name);
-    }
-
-    for (const String & partition_name : filtered_partitions_names)
-    {
-        const size_t number_of_splits = task_table.number_of_splits;
-        task_shard->partition_tasks.emplace(partition_name, ShardPartition(*task_shard, partition_name, number_of_splits));
-        task_shard->checked_partitions.emplace(partition_name, true);
-
-        auto shard_partition_it = task_shard->partition_tasks.find(partition_name);
-        PartitionPieces & shard_partition_pieces = shard_partition_it->second.pieces;
-
-        for (size_t piece_number = 0; piece_number < number_of_splits; ++piece_number)
-        {
-            bool res = checkPresentPartitionPiecesOnCurrentShard(timeouts, *task_shard, partition_name, piece_number);
-            shard_partition_pieces.emplace_back(shard_partition_it->second, piece_number, res);
-        }
-    }
-
-    if (!missing_partitions.empty())
-    {
-        WriteBufferFromOwnString ss;
-        for (const String & missing_partition : missing_partitions)
-            ss << " " << missing_partition;
-
-        LOG_WARNING(log, "There are no {} partitions from enabled_partitions in shard {} :{}", missing_partitions.size(), task_shard->getDescription(), ss.str());
-    }
-
-    LOG_INFO(log, "Will copy {} partitions from shard {}", task_shard->partition_tasks.size(), task_shard->getDescription());
-}
-
-void ClusterCopier::discoverTablePartitions(const ConnectionTimeouts & timeouts, TaskTable & task_table, UInt64 num_threads)
-{
-    /// Fetch partitions list from a shard
-    {
-        ThreadPool thread_pool(CurrentMetrics::LocalThread, CurrentMetrics::LocalThreadActive, CurrentMetrics::LocalThreadScheduled, num_threads ? num_threads : 2 * getNumberOfPhysicalCPUCores());
-
-        for (const TaskShardPtr & task_shard : task_table.all_shards)
-            thread_pool.scheduleOrThrowOnError([this, timeouts, task_shard]()
-            {
-                setThreadName("DiscoverPartns");
-                discoverShardPartitions(timeouts, task_shard);
-            });
-
-        LOG_INFO(log, "Waiting for {} setup jobs", thread_pool.active());
-        thread_pool.wait();
-    }
-}
-
-void ClusterCopier::uploadTaskDescription(const std::string & task_path, const std::string & task_file, const bool force)
-{
-    auto local_task_description_path = task_path + "/description";
-
-    String task_config_str;
-    {
-        ReadBufferFromFile in(task_file);
-        readStringUntilEOF(task_config_str, in);
-    }
-    if (task_config_str.empty())
-        return;
-
-    auto zookeeper = getContext()->getZooKeeper();
-
-    zookeeper->createAncestors(local_task_description_path);
-    auto code = zookeeper->tryCreate(local_task_description_path, task_config_str, zkutil::CreateMode::Persistent);
-    if (code != Coordination::Error::ZOK && force)
-        zookeeper->createOrUpdate(local_task_description_path, task_config_str, zkutil::CreateMode::Persistent);
-
-    LOG_INFO(log, "Task description {} uploaded to {} with result {} ({})",
-        ((code != Coordination::Error::ZOK && !force) ? "not " : ""), local_task_description_path, code, Coordination::errorMessage(code));
-}
-
-void ClusterCopier::reloadTaskDescription()
-{
-    auto zookeeper = getContext()->getZooKeeper();
-    task_description_watch_zookeeper = zookeeper;
-
-    Coordination::Stat stat{};
-
-    /// It will throw exception if such a node doesn't exist.
-    auto task_config_str = zookeeper->get(task_description_path, &stat);
-
-    LOG_INFO(log, "Loading task description");
-    task_cluster_current_config = getConfigurationFromXMLString(task_config_str);
-
-    /// Setup settings
-    task_cluster->reloadSettings(*task_cluster_current_config);
-    getContext()->setSettings(task_cluster->settings_common);
-}
-
-void ClusterCopier::updateConfigIfNeeded()
-{
-    UInt64 version_to_update = task_description_version;
-    bool is_outdated_version = task_description_current_version != version_to_update;
-    bool is_expired_session  = !task_description_watch_zookeeper || task_description_watch_zookeeper->expired();
-
-    if (!is_outdated_version && !is_expired_session)
-        return;
-
-    LOG_INFO(log, "Updating task description");
-    reloadTaskDescription();
-
-    task_description_current_version = version_to_update;
-}
-
-void ClusterCopier::process(const ConnectionTimeouts & timeouts)
-{
-    for (TaskTable & task_table : task_cluster->table_tasks)
-    {
-        LOG_INFO(log, "Process table task {} with {} shards, {} of them are local ones", task_table.table_id, task_table.all_shards.size(), task_table.local_shards.size());
-
-        if (task_table.all_shards.empty())
-            continue;
-
-        /// Discover partitions of each shard and total set of partitions
-        if (!task_table.has_enabled_partitions)
-        {
-            /// If there are no specified enabled_partitions, we must discover them manually
-            discoverTablePartitions(timeouts, task_table);
-
-            /// After partitions of each shard are initialized, initialize cluster partitions
-            for (const TaskShardPtr & task_shard : task_table.all_shards)
-            {
-                for (const auto & partition_elem : task_shard->partition_tasks)
-                {
-                    const String & partition_name = partition_elem.first;
-                    task_table.cluster_partitions.emplace(partition_name, ClusterPartition{});
-                }
-            }
-
-            for (auto & partition_elem : task_table.cluster_partitions)
-            {
-                const String & partition_name = partition_elem.first;
-
-                for (const TaskShardPtr & task_shard : task_table.all_shards)
-                    task_shard->checked_partitions.emplace(partition_name);
-
-                task_table.ordered_partition_names.emplace_back(partition_name);
-            }
-        }
-        else
-        {
-            /// If enabled_partitions are specified, assume that each shard has all partitions
-            /// We will refine partition set of each shard in future
-
-            for (const String & partition_name : task_table.enabled_partitions)
-            {
-                task_table.cluster_partitions.emplace(partition_name, ClusterPartition{});
-                task_table.ordered_partition_names.emplace_back(partition_name);
-            }
-        }
-
-        task_table.watch.restart();
-
-        /// Retry table processing
-        bool table_is_done = false;
-        for (UInt64 num_table_tries = 1; num_table_tries <= max_table_tries; ++num_table_tries)
-        {
-            if (tryProcessTable(timeouts, task_table))
-            {
-                table_is_done = true;
-                break;
-            }
-        }
-
-        if (!table_is_done)
-        {
-            throw Exception(ErrorCodes::UNFINISHED, "Too many tries to process table {}. Abort remaining execution",
-                            task_table.table_id);
-        }
-    }
-}
-
-/// Protected section
-
-
-/*
- * Creates task worker node and checks maximum number of workers not to exceed the limit.
- * To achieve this we have to check version of workers_version_path node and create current_worker_path
- * node atomically.
- * */
-
-zkutil::EphemeralNodeHolder::Ptr ClusterCopier::createTaskWorkerNodeAndWaitIfNeed(
-    const zkutil::ZooKeeperPtr & zookeeper,
-    const String & description,
-    bool unprioritized)
-{
-    std::chrono::milliseconds current_sleep_time = retry_delay_ms;
-    static constexpr std::chrono::milliseconds max_sleep_time(30000); // 30 sec
-
-    if (unprioritized)
-        std::this_thread::sleep_for(current_sleep_time);
-
-    String workers_version_path = getWorkersPathVersion();
-    String workers_path         = getWorkersPath();
-    String current_worker_path  = getCurrentWorkerNodePath();
-
-    UInt64 num_bad_version_errors = 0;
-
-    while (true)
-    {
-        updateConfigIfNeeded();
-
-        Coordination::Stat stat;
-        zookeeper->get(workers_version_path, &stat);
-        auto version = stat.version;
-        zookeeper->get(workers_path, &stat);
-
-        if (static_cast<UInt64>(stat.numChildren) >= task_cluster->max_workers)
-        {
-            LOG_INFO(log, "Too many workers ({}, maximum {}). Postpone processing {}", stat.numChildren, task_cluster->max_workers, description);
-
-            if (unprioritized)
-                current_sleep_time = std::min(max_sleep_time, current_sleep_time + retry_delay_ms);
-
-            std::this_thread::sleep_for(current_sleep_time);
-            num_bad_version_errors = 0;
-        }
-        else
-        {
-            Coordination::Requests ops;
-            ops.emplace_back(zkutil::makeSetRequest(workers_version_path, description, version));
-            ops.emplace_back(zkutil::makeCreateRequest(current_worker_path, description, zkutil::CreateMode::Ephemeral));
-            Coordination::Responses responses;
-            auto code = zookeeper->tryMulti(ops, responses);
-
-            if (code == Coordination::Error::ZOK || code == Coordination::Error::ZNODEEXISTS)
-                return zkutil::EphemeralNodeHolder::existing(current_worker_path, *zookeeper);
-
-            if (code == Coordination::Error::ZBADVERSION)
-            {
-                ++num_bad_version_errors;
-
-                /// Try to make fast retries
-                if (num_bad_version_errors > 3)
-                {
-                    LOG_INFO(log, "A concurrent worker has just been added, will check free worker slots again");
-                    std::chrono::milliseconds random_sleep_time(std::uniform_int_distribution<int>(1, 1000)(task_cluster->random_engine));
-                    std::this_thread::sleep_for(random_sleep_time);
-                    num_bad_version_errors = 0;
-                }
-            }
-            else
-                throw Coordination::Exception(code);
-        }
-    }
-}
-
-
-bool ClusterCopier::checkPartitionPieceIsClean(
-        const zkutil::ZooKeeperPtr & zookeeper,
-        const CleanStateClock & clean_state_clock,
-        const String & task_status_path)
-{
-    LogicalClock task_start_clock;
-
-    Coordination::Stat stat{};
-    if (zookeeper->exists(task_status_path, &stat))
-        task_start_clock = LogicalClock(stat.mzxid);
-
-    return clean_state_clock.is_clean() && (!task_start_clock.hasHappened() || clean_state_clock.discovery_zxid <= task_start_clock);
-}
-
-
-bool ClusterCopier::checkAllPiecesInPartitionAreDone(const TaskTable & task_table, const String & partition_name, const TasksShard & shards_with_partition)
-{
-    bool answer = true;
-    for (size_t piece_number = 0; piece_number < task_table.number_of_splits; ++piece_number)
-    {
-        bool piece_is_done = checkPartitionPieceIsDone(task_table, partition_name, piece_number, shards_with_partition);
-        if (!piece_is_done)
-            LOG_INFO(log, "Partition {} piece {} is not already done.", partition_name, piece_number);
-        answer &= piece_is_done;
-    }
-
-    return answer;
-}
-
-
-/* The same as function above
- * Assume that we don't know on which shards do we have partition certain piece.
- * We'll check them all (I mean shards that contain the whole partition)
- * And shards that don't have certain piece MUST mark that piece is_done true.
- * */
-bool ClusterCopier::checkPartitionPieceIsDone(const TaskTable & task_table, const String & partition_name,
-                               size_t piece_number, const TasksShard & shards_with_partition)
-{
-    LOG_INFO(log, "Check that all shards processed partition {} piece {} successfully", partition_name, piece_number);
-
-    auto zookeeper = getContext()->getZooKeeper();
-
-    /// Collect all shards that contain partition piece number piece_number.
-    Strings piece_status_paths;
-    for (const auto & shard : shards_with_partition)
-    {
-        ShardPartition & task_shard_partition = shard->partition_tasks.find(partition_name)->second;
-        ShardPartitionPiece & shard_partition_piece = task_shard_partition.pieces[piece_number];
-        piece_status_paths.emplace_back(shard_partition_piece.getShardStatusPath());
-    }
-
-    std::vector<int64_t> zxid1, zxid2;
-
-    try
-    {
-        std::vector<zkutil::ZooKeeper::FutureGet> get_futures;
-        for (const String & path : piece_status_paths)
-            get_futures.emplace_back(zookeeper->asyncGet(path));
-
-        // Check that state is Finished and remember zxid
-        for (auto & future : get_futures)
-        {
-            auto res = future.get();
-
-            TaskStateWithOwner status = TaskStateWithOwner::fromString(res.data);
-            if (status.state != TaskState::Finished)
-            {
-                LOG_INFO(log, "The task {} is being rewritten by {}. Partition piece will be rechecked", res.data, status.owner);
-                return false;
-            }
-
-            zxid1.push_back(res.stat.pzxid);
-        }
-
-        const String piece_is_dirty_flag_path = task_table.getCertainPartitionPieceIsDirtyPath(partition_name, piece_number);
-        const String piece_is_dirty_cleaned_path = task_table.getCertainPartitionPieceIsCleanedPath(partition_name, piece_number);
-        const String piece_task_status_path = task_table.getCertainPartitionPieceTaskStatusPath(partition_name, piece_number);
-
-        CleanStateClock clean_state_clock (zookeeper, piece_is_dirty_flag_path, piece_is_dirty_cleaned_path);
-
-        const bool is_clean = checkPartitionPieceIsClean(zookeeper, clean_state_clock, piece_task_status_path);
-
-
-        if (!is_clean)
-        {
-            LOG_INFO(log, "Partition {} become dirty", partition_name);
-            return false;
-        }
-
-        get_futures.clear();
-        for (const String & path : piece_status_paths)
-            get_futures.emplace_back(zookeeper->asyncGet(path));
-
-        // Remember zxid of states again
-        for (auto & future : get_futures)
-        {
-            auto res = future.get();
-            zxid2.push_back(res.stat.pzxid);
-        }
-    }
-    catch (const Coordination::Exception & e)
-    {
-        LOG_INFO(log, "A ZooKeeper error occurred while checking partition {} piece number {}. Will recheck the partition. Error: {}", partition_name, toString(piece_number), e.displayText());
-        return false;
-    }
-
-    // If all task is finished and zxid is not changed then partition could not become dirty again
-    for (UInt64 shard_num = 0; shard_num < piece_status_paths.size(); ++shard_num)
-    {
-        if (zxid1[shard_num] != zxid2[shard_num])
-        {
-            LOG_INFO(log, "The task {} is being modified now. Partition piece will be rechecked", piece_status_paths[shard_num]);
-            return false;
-        }
-    }
-
-    LOG_INFO(log, "Partition {} piece number {} is copied successfully", partition_name, toString(piece_number));
-    return true;
-}
-
-
-TaskStatus ClusterCopier::tryMoveAllPiecesToDestinationTable(const TaskTable & task_table, const String & partition_name)
-{
-    bool inject_fault = false;
-    if (move_fault_probability > 0)
-    {
-        double value = std::uniform_real_distribution<>(0, 1)(task_table.task_cluster.random_engine);
-        inject_fault = value < move_fault_probability;
-    }
-
-    LOG_INFO(log, "Try to move {} to destination table", partition_name);
-
-    auto zookeeper = getContext()->getZooKeeper();
-
-    const auto current_partition_attach_is_active = task_table.getPartitionAttachIsActivePath(partition_name);
-    const auto current_partition_attach_is_done   = task_table.getPartitionAttachIsDonePath(partition_name);
-
-    /// Create ephemeral node to mark that we are active and process the partition
-    zookeeper->createAncestors(current_partition_attach_is_active);
-    zkutil::EphemeralNodeHolderPtr partition_attach_node_holder;
-    try
-    {
-        partition_attach_node_holder = zkutil::EphemeralNodeHolder::create(current_partition_attach_is_active, *zookeeper, host_id);
-    }
-    catch (const Coordination::Exception & e)
-    {
-        if (e.code == Coordination::Error::ZNODEEXISTS)
-        {
-            LOG_INFO(log, "Someone is already moving pieces {}", current_partition_attach_is_active);
-            return TaskStatus::Active;
-        }
-
-        throw;
-    }
-
-
-    /// Exit if task has been already processed;
-    /// create blocking node to signal cleaning up if it is abandoned
-    {
-        String status_data;
-        if (zookeeper->tryGet(current_partition_attach_is_done, status_data))
-        {
-            TaskStateWithOwner status = TaskStateWithOwner::fromString(status_data);
-            if (status.state == TaskState::Finished)
-            {
-                LOG_INFO(log, "All pieces for partition from this task {} has been successfully moved to destination table by {}", current_partition_attach_is_active, status.owner);
-                return TaskStatus::Finished;
-            }
-
-            /// Task is abandoned, because previously we created ephemeral node, possibly in other copier's process.
-            /// Initialize DROP PARTITION
-            LOG_INFO(log, "Moving piece for partition {} has not been successfully finished by {}. Will try to move by myself.", current_partition_attach_is_active, status.owner);
-
-            /// Remove is_done marker.
-            zookeeper->remove(current_partition_attach_is_done);
-        }
-    }
-
-
-    /// Try start processing, create node about it
-    {
-        String start_state = TaskStateWithOwner::getData(TaskState::Started, host_id);
-        zookeeper->create(current_partition_attach_is_done, start_state, zkutil::CreateMode::Persistent);
-    }
-
-
-    /// Try to drop destination partition in original table
-    if (task_table.allow_to_drop_target_partitions)
-    {
-        DatabaseAndTableName original_table = task_table.table_push;
-
-        WriteBufferFromOwnString ss;
-        ss << "ALTER TABLE " << getQuotedTable(original_table) << ((partition_name == "'all'") ? " DROP PARTITION ID " : " DROP PARTITION ") << partition_name;
-
-        UInt64 num_shards_drop_partition = executeQueryOnCluster(task_table.cluster_push, ss.str(), task_cluster->settings_push, ClusterExecutionMode::ON_EACH_SHARD);
-        if (num_shards_drop_partition != task_table.cluster_push->getShardCount())
-            return TaskStatus::Error;
-
-        LOG_INFO(log, "Drop partition {} in original table {} have been executed successfully on {} shards of {}",
-            partition_name, getQuotedTable(original_table), num_shards_drop_partition, task_table.cluster_push->getShardCount());
-    }
-
-    /// Move partition to original destination table.
-    for (size_t current_piece_number = 0; current_piece_number < task_table.number_of_splits; ++current_piece_number)
-    {
-        LOG_INFO(log, "Trying to move partition {} piece {} to original table", partition_name, toString(current_piece_number));
-
-        ASTPtr query_alter_ast;
-        String query_alter_ast_string;
-
-        DatabaseAndTableName original_table = task_table.table_push;
-        DatabaseAndTableName helping_table = DatabaseAndTableName(original_table.first,
-                                                                  original_table.second + "_piece_" +
-                                                                  toString(current_piece_number));
-
-        Settings settings_push = task_cluster->settings_push;
-        ClusterExecutionMode execution_mode = ClusterExecutionMode::ON_EACH_NODE;
-
-        if (settings_push.alter_sync == 1)
-            execution_mode = ClusterExecutionMode::ON_EACH_SHARD;
-
-        query_alter_ast_string += " ALTER TABLE " + getQuotedTable(original_table) +
-                                  ((partition_name == "'all'") ? " ATTACH PARTITION ID " : " ATTACH PARTITION ") + partition_name +
-                                  " FROM " + getQuotedTable(helping_table);
-
-        LOG_INFO(log, "Executing ALTER query: {}", query_alter_ast_string);
-
-        try
-        {
-            /// Try attach partition on each shard
-            UInt64 num_nodes = executeQueryOnCluster(
-                task_table.cluster_push,
-                query_alter_ast_string,
-                task_cluster->settings_push,
-                execution_mode);
-
-            if (settings_push.alter_sync == 1)
-            {
-                LOG_INFO(
-                    log,
-                    "Destination tables {} have been executed alter query successfully on {} shards of {}",
-                    getQuotedTable(task_table.table_push),
-                    num_nodes,
-                    task_table.cluster_push->getShardCount());
-
-                if (num_nodes != task_table.cluster_push->getShardCount())
-                    return TaskStatus::Error;
-            }
-            else
-            {
-                LOG_INFO(log, "Number of nodes that executed ALTER query successfully : {}", toString(num_nodes));
-            }
-        }
-        catch (...)
-        {
-            LOG_INFO(log, "Error while moving partition {} piece {} to original table", partition_name, toString(current_piece_number));
-            LOG_WARNING(log, "In case of non-replicated tables it can cause duplicates.");
-            throw;
-        }
-
-        if (inject_fault)
-            throw Exception(ErrorCodes::UNFINISHED, "Copy fault injection is activated");
-    }
-
-    /// Create node to signal that we finished moving
-    /// Also increment a counter of processed partitions
-    {
-        const auto state_finished = TaskStateWithOwner::getData(TaskState::Finished, host_id);
-        const auto task_status = task_zookeeper_path + "/status";
-
-        /// Try until success
-        while (true)
-        {
-            Coordination::Stat stat;
-            auto status_json = zookeeper->get(task_status, &stat);
-            auto statuses = StatusAccumulator::fromJSON(status_json);
-
-            /// Increment status for table.
-            (*statuses)[task_table.name_in_config].processed_partitions_count += 1;
-            auto statuses_to_commit = StatusAccumulator::serializeToJSON(statuses);
-
-            Coordination::Requests ops;
-            ops.emplace_back(zkutil::makeSetRequest(current_partition_attach_is_done, state_finished, 0));
-            ops.emplace_back(zkutil::makeSetRequest(task_status, statuses_to_commit, stat.version));
-
-            Coordination::Responses responses;
-            Coordination::Error code = zookeeper->tryMulti(ops, responses);
-
-            if (code == Coordination::Error::ZOK)
-                break;
-        }
-    }
-
-    return TaskStatus::Finished;
-}
-
-/// This is needed to create internal Distributed table
-/// Removes column's TTL expression from `CREATE` query
-/// Removes MATEREALIZED or ALIAS columns not to copy additional and useless data over the network.
-/// Removes data skipping indices.
-ASTPtr ClusterCopier::removeAliasMaterializedAndTTLColumnsFromCreateQuery(const ASTPtr & query_ast, bool allow_to_copy_alias_and_materialized_columns)
-{
-    const ASTs & column_asts = query_ast->as<ASTCreateQuery &>().columns_list->columns->children;
-    auto new_columns = std::make_shared<ASTExpressionList>();
-
-    for (const ASTPtr & column_ast : column_asts)
-    {
-        const auto & column = column_ast->as<ASTColumnDeclaration &>();
-
-        /// Skip this columns
-        if (!column.default_specifier.empty() && !allow_to_copy_alias_and_materialized_columns)
-        {
-            ColumnDefaultKind kind = columnDefaultKindFromString(column.default_specifier);
-            if (kind == ColumnDefaultKind::Materialized || kind == ColumnDefaultKind::Alias)
-                continue;
-        }
-
-        /// Remove TTL on columns definition.
-        auto new_column_ast = column_ast->clone();
-        auto & new_column = new_column_ast->as<ASTColumnDeclaration &>();
-        if (new_column.ttl)
-            new_column.ttl.reset();
-
-        new_columns->children.emplace_back(new_column_ast);
-    }
-
-    ASTPtr new_query_ast = query_ast->clone();
-    auto & new_query = new_query_ast->as<ASTCreateQuery &>();
-
-    auto new_columns_list = std::make_shared<ASTColumns>();
-    new_columns_list->set(new_columns_list->columns, new_columns);
-
-    /// Skip indices and projections are not needed, because distributed table doesn't support it.
-
-    new_query.replace(new_query.columns_list, new_columns_list);
-
-    return new_query_ast;
-}
-
-/// Replaces ENGINE and table name in a create query
-std::shared_ptr<ASTCreateQuery> rewriteCreateQueryStorage(const ASTPtr & create_query_ast,
-                                                          const DatabaseAndTableName & new_table,
-                                                          const ASTPtr & new_storage_ast)
-{
-    const auto & create = create_query_ast->as<ASTCreateQuery &>();
-    auto res = std::make_shared<ASTCreateQuery>(create);
-
-    if (create.storage == nullptr || new_storage_ast == nullptr)
-        throw Exception(ErrorCodes::LOGICAL_ERROR, "Storage is not specified");
-
-    res->setDatabase(new_table.first);
-    res->setTable(new_table.second);
-
-    res->children.clear();
-    res->set(res->columns_list, create.columns_list->clone());
-    res->set(res->storage, new_storage_ast->clone());
-    /// Just to make it better and don't store additional flag like `is_table_created` somewhere else
-    res->if_not_exists = true;
-
-    return res;
-}
-
-
-bool ClusterCopier::tryDropPartitionPiece(
-        ShardPartition & task_partition,
-        const size_t current_piece_number,
-        const zkutil::ZooKeeperPtr & zookeeper,
-        const CleanStateClock & clean_state_clock)
-{
-    if (is_safe_mode)
-        throw Exception(ErrorCodes::NOT_IMPLEMENTED, "DROP PARTITION is prohibited in safe mode");
-
-    TaskTable & task_table = task_partition.task_shard.task_table;
-    ShardPartitionPiece & partition_piece = task_partition.pieces[current_piece_number];
-
-    const String current_shards_path                  = partition_piece.getPartitionPieceShardsPath();
-    const String current_partition_active_workers_dir = partition_piece.getPartitionPieceActiveWorkersPath();
-    const String is_dirty_flag_path                   = partition_piece.getPartitionPieceIsDirtyPath();
-    const String dirty_cleaner_path                   = partition_piece.getPartitionPieceCleanerPath();
-    const String is_dirty_cleaned_path                = partition_piece.getPartitionPieceIsCleanedPath();
-
-    zkutil::EphemeralNodeHolder::Ptr cleaner_holder;
-    try
-    {
-        cleaner_holder = zkutil::EphemeralNodeHolder::create(dirty_cleaner_path, *zookeeper, host_id);
-    }
-    catch (const Coordination::Exception & e)
-    {
-        if (e.code == Coordination::Error::ZNODEEXISTS)
-        {
-            LOG_INFO(log, "Partition {} piece {} is cleaning now by somebody, sleep", task_partition.name, toString(current_piece_number));
-            std::this_thread::sleep_for(retry_delay_ms);
-            return false;
-        }
-
-        throw;
-    }
-
-    Coordination::Stat stat{};
-    if (zookeeper->exists(current_partition_active_workers_dir, &stat))
-    {
-        if (stat.numChildren != 0)
-        {
-            LOG_INFO(log, "Partition {} contains {} active workers while trying to drop it. Going to sleep.", task_partition.name, stat.numChildren);
-            std::this_thread::sleep_for(retry_delay_ms);
-            return false;
-        }
-        else
-        {
-            zookeeper->remove(current_partition_active_workers_dir);
-        }
-    }
-
-    {
-        zkutil::EphemeralNodeHolder::Ptr active_workers_lock;
-        try
-        {
-            active_workers_lock = zkutil::EphemeralNodeHolder::create(current_partition_active_workers_dir, *zookeeper, host_id);
-        }
-        catch (const Coordination::Exception & e)
-        {
-            if (e.code == Coordination::Error::ZNODEEXISTS)
-            {
-                LOG_INFO(log, "Partition {} is being filled now by somebody, sleep", task_partition.name);
-                return false;
-            }
-
-            throw;
-        }
-
-        // Lock the dirty flag
-        zookeeper->set(is_dirty_flag_path, host_id, clean_state_clock.discovery_version.value());
-        zookeeper->tryRemove(partition_piece.getPartitionPieceCleanStartPath());
-        CleanStateClock my_clock(zookeeper, is_dirty_flag_path, is_dirty_cleaned_path);
-
-        /// Remove all status nodes
-        {
-            Strings children;
-            if (zookeeper->tryGetChildren(current_shards_path, children) == Coordination::Error::ZOK)
-                for (const auto & child : children)
-                {
-                    zookeeper->removeRecursive(current_shards_path + "/" + child);
-                }
-        }
-
-
-        DatabaseAndTableName original_table = task_table.table_push;
-        DatabaseAndTableName helping_table = DatabaseAndTableName(original_table.first, original_table.second + "_piece_" + toString(current_piece_number));
-
-        String query = "ALTER TABLE " + getQuotedTable(helping_table);
-        query += ((task_partition.name == "'all'") ? " DROP PARTITION ID " : " DROP PARTITION ")  + task_partition.name + "";
-
-        /// TODO: use this statement after servers will be updated up to 1.1.54310
-        // query += " DROP PARTITION ID '" + task_partition.name + "'";
-
-        ClusterPtr & cluster_push = task_table.cluster_push;
-        Settings settings_push = task_cluster->settings_push;
-
-        /// It is important, DROP PARTITION must be done synchronously
-        settings_push.alter_sync = 2;
-
-        LOG_INFO(log, "Execute distributed DROP PARTITION: {}", query);
-        /// We have to drop partition_piece on each replica
-        size_t num_shards = executeQueryOnCluster(
-                cluster_push, query,
-                settings_push,
-                ClusterExecutionMode::ON_EACH_NODE);
-
-        LOG_INFO(log, "DROP PARTITION was successfully executed on {} nodes of a cluster.", num_shards);
-
-        /// Update the locking node
-        if (!my_clock.is_stale())
-        {
-            zookeeper->set(is_dirty_flag_path, host_id, my_clock.discovery_version.value());
-            if (my_clock.clean_state_version)
-                zookeeper->set(is_dirty_cleaned_path, host_id, my_clock.clean_state_version.value());
-            else
-                zookeeper->create(is_dirty_cleaned_path, host_id, zkutil::CreateMode::Persistent);
-        }
-        else
-        {
-            LOG_INFO(log, "Clean state is altered when dropping the partition, cowardly bailing");
-            /// clean state is stale
-            return false;
-        }
-
-        LOG_INFO(log, "Partition {} piece {} was dropped on cluster {}", task_partition.name, toString(current_piece_number), task_table.cluster_push_name);
-        if (zookeeper->tryCreate(current_shards_path, host_id, zkutil::CreateMode::Persistent) == Coordination::Error::ZNODEEXISTS)
-            zookeeper->set(current_shards_path, host_id);
-    }
-
-    LOG_INFO(log, "Partition {} piece {} is safe for work now.", task_partition.name, toString(current_piece_number));
-    return true;
-}
-
-bool ClusterCopier::tryProcessTable(const ConnectionTimeouts & timeouts, TaskTable & task_table)
-{
-    /// Create destination table
-    TaskStatus task_status = TaskStatus::Error;
-
-    task_status = tryCreateDestinationTable(timeouts, task_table);
-    /// Exit if success
-    if (task_status != TaskStatus::Finished)
-    {
-        LOG_WARNING(log, "Create destination table failed ");
-        return false;
-    }
-
-    /// Set all_partitions_count for table in Zookeeper
-    auto zookeeper = getContext()->getZooKeeper();
-    while (true)
-    {
-        Coordination::Stat stat;
-        auto status_json = zookeeper->get(task_zookeeper_path + "/status", &stat);
-        auto statuses = StatusAccumulator::fromJSON(status_json);
-
-        /// Exit if someone already set the initial value for this table.
-        if (statuses->find(task_table.name_in_config) != statuses->end())
-            break;
-        (*statuses)[task_table.name_in_config] = StatusAccumulator::TableStatus
-        {
-            /*all_partitions_count=*/task_table.ordered_partition_names.size(),
-            /*processed_partition_count=*/0
-        };
-
-        auto statuses_to_commit = StatusAccumulator::serializeToJSON(statuses);
-        auto error = zookeeper->trySet(task_zookeeper_path + "/status", statuses_to_commit, stat.version);
-        if (error == Coordination::Error::ZOK)
-            break;
-    }
-
-
-    /// An heuristic: if previous shard is already done, then check next one without sleeps due to max_workers constraint
-    bool previous_shard_is_instantly_finished = false;
-
-    /// Process each partition that is present in cluster
-    for (const String & partition_name : task_table.ordered_partition_names)
-    {
-        if (!task_table.cluster_partitions.contains(partition_name))
-            throw Exception(ErrorCodes::LOGICAL_ERROR, "There are no expected partition {}. It is a bug", partition_name);
-
-        ClusterPartition & cluster_partition = task_table.cluster_partitions[partition_name];
-
-        Stopwatch watch;
-        /// We will check all the shards of the table and check if they contain current partition.
-        TasksShard expected_shards;
-        UInt64 num_failed_shards = 0;
-
-        ++cluster_partition.total_tries;
-
-        LOG_INFO(log, "Processing partition {} for the whole cluster", partition_name);
-
-        /// Process each source shard having current partition and copy current partition
-        /// NOTE: shards are sorted by "distance" to current host
-        bool has_shard_to_process = false;
-        for (const TaskShardPtr & shard : task_table.all_shards)
-        {
-            /// Does shard have a node with current partition?
-            if (!shard->partition_tasks.contains(partition_name))
-            {
-                /// If not, did we check existence of that partition previously?
-                if (!shard->checked_partitions.contains(partition_name))
-                {
-                    auto check_shard_has_partition = [&] () { return checkShardHasPartition(timeouts, *shard, partition_name); };
-                    bool has_partition = retry(check_shard_has_partition);
-
-                    shard->checked_partitions.emplace(partition_name);
-
-                    if (has_partition)
-                    {
-                        const size_t number_of_splits = task_table.number_of_splits;
-                        shard->partition_tasks.emplace(partition_name, ShardPartition(*shard, partition_name, number_of_splits));
-                        LOG_INFO(log, "Discovered partition {} in shard {}", partition_name, shard->getDescription());
-                        /// To save references in the future.
-                        auto shard_partition_it = shard->partition_tasks.find(partition_name);
-                        PartitionPieces & shard_partition_pieces = shard_partition_it->second.pieces;
-
-                        for (size_t piece_number = 0; piece_number < number_of_splits; ++piece_number)
-                        {
-                            auto res = checkPresentPartitionPiecesOnCurrentShard(timeouts, *shard, partition_name, piece_number);
-                            shard_partition_pieces.emplace_back(shard_partition_it->second, piece_number, res);
-                        }
-                    }
-                    else
-                    {
-                        LOG_INFO(log, "Found that shard {} does not contain current partition {}", shard->getDescription(), partition_name);
-                        continue;
-                    }
-                }
-                else
-                {
-                    /// We have already checked that partition, but did not discover it
-                    previous_shard_is_instantly_finished = true;
-                    continue;
-                }
-            }
-
-            auto it_shard_partition = shard->partition_tasks.find(partition_name);
-            /// Previously when we discovered that shard does not contain current partition, we skipped it.
-            /// At this moment partition have to be present.
-            if (it_shard_partition == shard->partition_tasks.end())
-                throw Exception(ErrorCodes::LOGICAL_ERROR, "There are no such partition in a shard. This is a bug.");
-            auto & partition = it_shard_partition->second;
-
-            expected_shards.emplace_back(shard);
-
-            /// Do not sleep if there is a sequence of already processed shards to increase startup
-            bool is_unprioritized_task = !previous_shard_is_instantly_finished && shard->priority.is_remote;
-            task_status = TaskStatus::Error;
-            bool was_error = false;
-            has_shard_to_process = true;
-            for (UInt64 try_num = 1; try_num <= max_shard_partition_tries; ++try_num)
-            {
-                task_status = tryProcessPartitionTask(timeouts, partition, is_unprioritized_task);
-
-                /// Exit if success
-                if (task_status == TaskStatus::Finished)
-                    break;
-
-                was_error = true;
-
-                /// Skip if the task is being processed by someone
-                if (task_status == TaskStatus::Active)
-                    break;
-
-                /// Repeat on errors
-                std::this_thread::sleep_for(retry_delay_ms);
-            }
-
-            if (task_status == TaskStatus::Error)
-                ++num_failed_shards;
-
-            previous_shard_is_instantly_finished = !was_error;
-        }
-
-        cluster_partition.elapsed_time_seconds += watch.elapsedSeconds();
-
-        /// Check that whole cluster partition is done
-        /// Firstly check the number of failed partition tasks, then look into ZooKeeper and ensure that each partition is done
-        bool partition_copying_is_done = num_failed_shards == 0;
-        try
-        {
-            partition_copying_is_done =
-                    !has_shard_to_process
-                    || (partition_copying_is_done && checkAllPiecesInPartitionAreDone(task_table, partition_name, expected_shards));
-        }
-        catch (...)
-        {
-            tryLogCurrentException(log);
-            partition_copying_is_done = false;
-        }
-
-
-        bool partition_moving_is_done = false;
-        /// Try to move only if all pieces were copied.
-        if (partition_copying_is_done)
-        {
-            for (UInt64 try_num = 0; try_num < max_shard_partition_piece_tries_for_alter; ++try_num)
-            {
-                try
-                {
-                    auto res = tryMoveAllPiecesToDestinationTable(task_table, partition_name);
-                    /// Exit and mark current task is done.
-                    if (res == TaskStatus::Finished)
-                    {
-                        partition_moving_is_done = true;
-                        break;
-                    }
-
-                    /// Exit if this task is active.
-                    if (res == TaskStatus::Active)
-                        break;
-
-                    /// Repeat on errors.
-                    std::this_thread::sleep_for(retry_delay_ms);
-                }
-                catch (...)
-                {
-                    tryLogCurrentException(log, "Some error occurred while moving pieces to destination table for partition " + partition_name);
-                }
-            }
-        }
-
-        if (partition_copying_is_done && partition_moving_is_done)
-        {
-            task_table.finished_cluster_partitions.emplace(partition_name);
-
-            task_table.bytes_copied += cluster_partition.bytes_copied;
-            task_table.rows_copied += cluster_partition.rows_copied;
-            double elapsed = cluster_partition.elapsed_time_seconds;
-
-            LOG_INFO(log, "It took {} seconds to copy partition {}: {} uncompressed bytes, {} rows and {} source blocks are copied",
-                elapsed, partition_name,
-                formatReadableSizeWithDecimalSuffix(cluster_partition.bytes_copied),
-                formatReadableQuantity(cluster_partition.rows_copied),
-                cluster_partition.blocks_copied);
-
-            if (cluster_partition.rows_copied)
-            {
-                LOG_INFO(log, "Average partition speed: {} per second.", formatReadableSizeWithDecimalSuffix(cluster_partition.bytes_copied / elapsed));
-            }
-
-            if (task_table.rows_copied)
-            {
-                LOG_INFO(log, "Average table {} speed: {} per second.", task_table.table_id, formatReadableSizeWithDecimalSuffix(task_table.bytes_copied / elapsed));
-            }
-        }
-    }
-
-    UInt64 required_partitions = task_table.cluster_partitions.size();
-    UInt64 finished_partitions = task_table.finished_cluster_partitions.size();
-    bool table_is_done = finished_partitions >= required_partitions;
-
-    if (!table_is_done)
-    {
-        LOG_INFO(log, "Table {} is not processed yet. Copied {} of {}, will retry", task_table.table_id, finished_partitions, required_partitions);
-    }
-    else
-    {
-        /// Delete helping tables in case that whole table is done
-        dropHelpingTables(task_table);
-    }
-
-    return table_is_done;
-}
-
-TaskStatus ClusterCopier::tryCreateDestinationTable(const ConnectionTimeouts & timeouts, TaskTable & task_table)
-{
-    /// Try create original table (if not exists) on each shard
-
-    //TaskTable & task_table = task_shard.task_table;
-    const TaskShardPtr task_shard = task_table.all_shards.at(0);
-    /// We need to update table definitions for each part, it could be changed after ALTER
-    task_shard->current_pull_table_create_query = getCreateTableForPullShard(timeouts, *task_shard);
-    try
-    {
-        auto create_query_push_ast
-            = rewriteCreateQueryStorage(task_shard->current_pull_table_create_query, task_table.table_push, task_table.engine_push_ast);
-        auto & create = create_query_push_ast->as<ASTCreateQuery &>();
-        create.if_not_exists = true;
-        InterpreterCreateQuery::prepareOnClusterQuery(create, getContext(), task_table.cluster_push_name);
-        String query = queryToString(create_query_push_ast);
-
-        LOG_INFO(log, "Create destination tables. Query: {}", query);
-        UInt64 shards = executeQueryOnCluster(task_table.cluster_push, query, task_cluster->settings_push, ClusterExecutionMode::ON_EACH_NODE);
-        LOG_INFO(
-            log,
-            "Destination tables {} have been created on {} shards of {}",
-            getQuotedTable(task_table.table_push),
-            shards,
-            task_table.cluster_push->getShardCount());
-    }
-    catch (...)
-    {
-        tryLogCurrentException(log, "Error while creating original table. Maybe we are not first.");
-    }
-
-    return TaskStatus::Finished;
-}
-
-/// Job for copying partition from particular shard.
-TaskStatus ClusterCopier::tryProcessPartitionTask(const ConnectionTimeouts & timeouts, ShardPartition & task_partition, bool is_unprioritized_task)
-{
-    TaskStatus res;
-
-    try
-    {
-        res = iterateThroughAllPiecesInPartition(timeouts, task_partition, is_unprioritized_task);
-    }
-    catch (...)
-    {
-        tryLogCurrentException(log, "An error occurred while processing partition " + task_partition.name);
-        res = TaskStatus::Error;
-    }
-
-    /// At the end of each task check if the config is updated
-    try
-    {
-        updateConfigIfNeeded();
-    }
-    catch (...)
-    {
-        tryLogCurrentException(log, "An error occurred while updating the config");
-    }
-
-    return res;
-}
-
-TaskStatus ClusterCopier::iterateThroughAllPiecesInPartition(const ConnectionTimeouts & timeouts, ShardPartition & task_partition,
-                                                       bool is_unprioritized_task)
-{
-    const size_t total_number_of_pieces = task_partition.task_shard.task_table.number_of_splits;
-
-    TaskStatus res{TaskStatus::Finished};
-
-    bool was_failed_pieces = false;
-    bool was_active_pieces = false;
-
-    for (size_t piece_number = 0; piece_number < total_number_of_pieces; piece_number++)
-    {
-        for (UInt64 try_num = 0; try_num < max_shard_partition_tries; ++try_num)
-        {
-            LOG_INFO(log, "Attempt number {} to process partition {} piece number {} on shard number {} with index {}.",
-                try_num, task_partition.name, piece_number,
-                task_partition.task_shard.numberInCluster(),
-                task_partition.task_shard.indexInCluster());
-
-            res = processPartitionPieceTaskImpl(timeouts, task_partition, piece_number, is_unprioritized_task);
-
-            /// Exit if success
-            if (res == TaskStatus::Finished)
-                break;
-
-            /// Skip if the task is being processed by someone
-            if (res == TaskStatus::Active)
-                break;
-
-            /// Repeat on errors
-            std::this_thread::sleep_for(retry_delay_ms);
-        }
-
-        was_active_pieces |= (res == TaskStatus::Active);
-        was_failed_pieces |= (res == TaskStatus::Error);
-    }
-
-    if (was_failed_pieces)
-        return TaskStatus::Error;
-
-    if (was_active_pieces)
-        return TaskStatus::Active;
-
-    return TaskStatus::Finished;
-}
-
-
-TaskStatus ClusterCopier::processPartitionPieceTaskImpl(
-        const ConnectionTimeouts & timeouts, ShardPartition & task_partition,
-        const size_t current_piece_number, bool is_unprioritized_task)
-{
-    TaskShard & task_shard = task_partition.task_shard;
-    TaskTable & task_table = task_shard.task_table;
-    ClusterPartition & cluster_partition  = task_table.getClusterPartition(task_partition.name);
-    ShardPartitionPiece & partition_piece = task_partition.pieces[current_piece_number];
-
-    const size_t number_of_splits = task_table.number_of_splits;
-    const String primary_key_comma_separated = task_table.primary_key_comma_separated;
-
-    /// We need to update table definitions for each partition, it could be changed after ALTER
-    createShardInternalTables(timeouts, task_shard, true);
-
-    auto split_table_for_current_piece = task_shard.list_of_split_tables_on_shard[current_piece_number];
-
-    auto zookeeper = getContext()->getZooKeeper();
-
-    const String piece_is_dirty_flag_path          = partition_piece.getPartitionPieceIsDirtyPath();
-    const String piece_is_dirty_cleaned_path       = partition_piece.getPartitionPieceIsCleanedPath();
-    const String current_task_piece_is_active_path = partition_piece.getActiveWorkerPath();
-    const String current_task_piece_status_path    = partition_piece.getShardStatusPath();
-
-    /// Auxiliary functions:
-
-    /// Creates is_dirty node to initialize DROP PARTITION
-    auto create_is_dirty_node = [&] (const CleanStateClock & clock)
-    {
-        if (clock.is_stale())
-            LOG_INFO(log, "Clean state clock is stale while setting dirty flag, cowardly bailing");
-        else if (!clock.is_clean())
-            LOG_INFO(log, "Thank you, Captain Obvious");
-        else if (clock.discovery_version)
-        {
-            LOG_INFO(log, "Updating clean state clock");
-            zookeeper->set(piece_is_dirty_flag_path, host_id, clock.discovery_version.value());
-        }
-        else
-        {
-            LOG_INFO(log, "Creating clean state clock");
-            zookeeper->create(piece_is_dirty_flag_path, host_id, zkutil::CreateMode::Persistent);
-        }
-    };
-
-    /// Returns SELECT query filtering current partition and applying user filter
-    auto get_select_query = [&] (const DatabaseAndTableName & from_table, const String & fields, bool enable_splitting, String limit = "")
-    {
-        String query;
-        query += "WITH " + task_partition.name + " AS partition_key ";
-        query += "SELECT " + fields + " FROM " + getQuotedTable(from_table);
-
-        if (enable_splitting && experimental_use_sample_offset)
-            query += " SAMPLE 1/" + toString(number_of_splits) + " OFFSET " + toString(current_piece_number) + "/" + toString(number_of_splits);
-
-        /// TODO: Bad, it is better to rewrite with ASTLiteral(partition_key_field)
-        query += " WHERE (" + queryToString(task_table.engine_push_partition_key_ast) + " = partition_key)";
-
-        if (enable_splitting && !experimental_use_sample_offset)
-            query += " AND ( cityHash64(" + primary_key_comma_separated + ") %" + toString(number_of_splits) + " = " + toString(current_piece_number) + " )";
-
-        if (!task_table.where_condition_str.empty())
-            query += " AND (" + task_table.where_condition_str + ")";
-
-        if (!limit.empty())
-            query += " LIMIT " + limit;
-
-        query += " FORMAT Native";
-
-        ParserQuery p_query(query.data() + query.size());
-
-        const auto & settings = getContext()->getSettingsRef();
-        return parseQuery(p_query, query, settings.max_query_size, settings.max_parser_depth);
-    };
-
-    /// Load balancing
-    auto worker_node_holder = createTaskWorkerNodeAndWaitIfNeed(zookeeper, current_task_piece_status_path, is_unprioritized_task);
-
-    LOG_INFO(log, "Processing {}", current_task_piece_status_path);
-
-    const String piece_status_path = partition_piece.getPartitionPieceShardsPath();
-
-    CleanStateClock clean_state_clock(zookeeper, piece_is_dirty_flag_path, piece_is_dirty_cleaned_path);
-
-    const bool is_clean = checkPartitionPieceIsClean(zookeeper, clean_state_clock, piece_status_path);
-
-    /// Do not start if partition piece is dirty, try to clean it
-    if (is_clean)
-    {
-        LOG_INFO(log, "Partition {} piece {} appears to be clean", task_partition.name, current_piece_number);
-        zookeeper->createAncestors(current_task_piece_status_path);
-    }
-    else
-    {
-        LOG_INFO(log, "Partition {} piece {} is dirty, try to drop it", task_partition.name, current_piece_number);
-
-        try
-        {
-            tryDropPartitionPiece(task_partition, current_piece_number, zookeeper, clean_state_clock);
-        }
-        catch (...)
-        {
-            tryLogCurrentException(log, "An error occurred when clean partition");
-        }
-
-        return TaskStatus::Error;
-    }
-
-    /// Create ephemeral node to mark that we are active and process the partition
-    zookeeper->createAncestors(current_task_piece_is_active_path);
-    zkutil::EphemeralNodeHolderPtr partition_task_node_holder;
-    try
-    {
-        partition_task_node_holder = zkutil::EphemeralNodeHolder::create(current_task_piece_is_active_path, *zookeeper, host_id);
-    }
-    catch (const Coordination::Exception & e)
-    {
-        if (e.code == Coordination::Error::ZNODEEXISTS)
-        {
-            LOG_INFO(log, "Someone is already processing {}", current_task_piece_is_active_path);
-            return TaskStatus::Active;
-        }
-
-        throw;
-    }
-
-    /// Exit if task has been already processed;
-    /// create blocking node to signal cleaning up if it is abandoned
-    {
-        String status_data;
-        if (zookeeper->tryGet(current_task_piece_status_path, status_data))
-        {
-            TaskStateWithOwner status = TaskStateWithOwner::fromString(status_data);
-            if (status.state == TaskState::Finished)
-            {
-                LOG_INFO(log, "Task {} has been successfully executed by {}", current_task_piece_status_path, status.owner);
-                return TaskStatus::Finished;
-            }
-
-            /// Task is abandoned, because previously we created ephemeral node, possibly in other copier's process.
-            /// Initialize DROP PARTITION
-            LOG_INFO(log, "Task {} has not been successfully finished by {}. Partition will be dropped and refilled.", current_task_piece_status_path, status.owner);
-
-            create_is_dirty_node(clean_state_clock);
-            return TaskStatus::Error;
-        }
-    }
-
-
-    /// Try create table (if not exists) on each shard
-    /// We have to create this table even in case that partition piece is empty
-    /// This is significant, because we will have simpler code
-    {
-        /// 1) Get columns description from any replica of destination cluster
-        /// 2) Change ENGINE, database and table name
-        /// 3) Create helping table on the whole destination cluster
-        auto & settings_push = task_cluster->settings_push;
-
-        auto connection = task_table.cluster_push->getAnyShardInfo().pool->get(timeouts, settings_push, true);
-        String create_query = getRemoteCreateTable(task_shard.task_table.table_push, *connection, settings_push);
-
-        ParserCreateQuery parser_create_query;
-        auto create_query_ast = parseQuery(parser_create_query, create_query, settings_push.max_query_size, settings_push.max_parser_depth);
-        /// Define helping table database and name for current partition piece
-        DatabaseAndTableName database_and_table_for_current_piece
-        {
-            task_table.table_push.first,
-            task_table.table_push.second + "_piece_" + toString(current_piece_number)
-        };
-
-
-        auto new_engine_push_ast = task_table.engine_push_ast;
-        if (task_table.isReplicatedTable())
-            new_engine_push_ast = task_table.rewriteReplicatedCreateQueryToPlain();
-
-        /// Take columns definition from destination table, new database and table name, and new engine (non replicated variant of MergeTree)
-        auto create_query_push_ast = rewriteCreateQueryStorage(create_query_ast, database_and_table_for_current_piece, new_engine_push_ast);
-        String query = queryToString(create_query_push_ast);
-
-        LOG_INFO(log, "Create destination tables. Query: {}", query);
-        UInt64 shards = executeQueryOnCluster(task_table.cluster_push, query, task_cluster->settings_push, ClusterExecutionMode::ON_EACH_NODE);
-        LOG_INFO(
-            log,
-            "Destination tables {} have been created on {} shards of {}",
-            getQuotedTable(task_table.table_push),
-            shards,
-            task_table.cluster_push->getShardCount());
-    }
-
-
-    /// Exit if current piece is absent on this shard. Also mark it as finished, because we will check
-    /// whether each shard have processed each partitition (and its pieces).
-    if (partition_piece.is_absent_piece)
-    {
-        String state_finished = TaskStateWithOwner::getData(TaskState::Finished, host_id);
-        auto res = zookeeper->tryCreate(current_task_piece_status_path, state_finished, zkutil::CreateMode::Persistent);
-        if (res == Coordination::Error::ZNODEEXISTS)
-            LOG_INFO(log, "Partition {} piece {} is absent on current replica of a shard. But other replicas have already marked it as done.", task_partition.name, current_piece_number);
-        if (res == Coordination::Error::ZOK)
-            LOG_INFO(log, "Partition {} piece {} is absent on current replica of a shard. Will mark it as done. Other replicas will do the same.", task_partition.name, current_piece_number);
-        return TaskStatus::Finished;
-    }
-
-    /// Check that destination partition is empty if we are first worker
-    /// NOTE: this check is incorrect if pull and push tables have different partition key!
-    String clean_start_status;
-    if (!zookeeper->tryGet(partition_piece.getPartitionPieceCleanStartPath(), clean_start_status) || clean_start_status != "ok")
-    {
-        zookeeper->createIfNotExists(partition_piece.getPartitionPieceCleanStartPath(), "");
-        auto checker = zkutil::EphemeralNodeHolder::create(partition_piece.getPartitionPieceCleanStartPath() + "/checker",
-                                                           *zookeeper, host_id);
-        // Maybe we are the first worker
-
-        ASTPtr query_select_ast = get_select_query(split_table_for_current_piece, "count()", /* enable_splitting= */ true);
-        UInt64 count;
-        {
-            auto local_context = Context::createCopy(context);
-            // Use pull (i.e. readonly) settings, but fetch data from destination servers
-            local_context->setSettings(task_cluster->settings_pull);
-            local_context->setSetting("skip_unavailable_shards", true);
-
-            InterpreterSelectWithUnionQuery select(query_select_ast, local_context, SelectQueryOptions{});
-            QueryPlan plan;
-            select.buildQueryPlan(plan);
-            auto builder = std::move(*plan.buildQueryPipeline(
-                QueryPlanOptimizationSettings::fromContext(local_context),
-                BuildQueryPipelineSettings::fromContext(local_context)));
-
-            Block block = getBlockWithAllStreamData(std::move(builder));
-            count = (block) ? block.safeGetByPosition(0).column->getUInt(0) : 0;
-        }
-
-        if (count != 0)
-        {
-            LOG_INFO(log, "Partition {} piece {} is not empty. In contains {} rows.", task_partition.name, current_piece_number, count);
-            Coordination::Stat stat_shards{};
-            zookeeper->get(partition_piece.getPartitionPieceShardsPath(), &stat_shards);
-
-            /// NOTE: partition is still fresh if dirt discovery happens before cleaning
-            if (stat_shards.numChildren == 0)
-            {
-                LOG_WARNING(log, "There are no workers for partition {} piece {}, but destination table contains {} rows. Partition will be dropped and refilled.", task_partition.name, toString(current_piece_number), count);
-
-                create_is_dirty_node(clean_state_clock);
-                return TaskStatus::Error;
-            }
-        }
-        zookeeper->set(partition_piece.getPartitionPieceCleanStartPath(), "ok");
-    }
-    /// At this point, we need to sync that the destination table is clean
-    /// before any actual work
-
-    /// Try start processing, create node about it
-    {
-        String start_state = TaskStateWithOwner::getData(TaskState::Started, host_id);
-        CleanStateClock new_clean_state_clock(zookeeper, piece_is_dirty_flag_path, piece_is_dirty_cleaned_path);
-        if (clean_state_clock != new_clean_state_clock)
-        {
-            LOG_INFO(log, "Partition {} piece {} clean state changed, cowardly bailing", task_partition.name, toString(current_piece_number));
-            return TaskStatus::Error;
-        }
-        else if (!new_clean_state_clock.is_clean())
-        {
-            LOG_INFO(log, "Partition {} piece {} is dirty and will be dropped and refilled", task_partition.name, toString(current_piece_number));
-            create_is_dirty_node(new_clean_state_clock);
-            return TaskStatus::Error;
-        }
-        zookeeper->create(current_task_piece_status_path, start_state, zkutil::CreateMode::Persistent);
-    }
-
-
-    /// Do the copying
-    {
-        bool inject_fault = false;
-        if (copy_fault_probability > 0)
-        {
-            double value = std::uniform_real_distribution<>(0, 1)(task_table.task_cluster.random_engine);
-            inject_fault = value < copy_fault_probability;
-        }
-
-        // Select all fields
-        ASTPtr query_select_ast = get_select_query(task_shard.table_read_shard, "*", /* enable_splitting= */ true, inject_fault ? "1" : "");
-
-        LOG_INFO(log, "Executing SELECT query and pull from {}: {}", task_shard.getDescription(), queryToString(query_select_ast));
-
-        ASTPtr query_insert_ast;
-        {
-            String query;
-            query += "INSERT INTO " + getQuotedTable(split_table_for_current_piece) + " FORMAT Native  ";
-
-            ParserQuery p_query(query.data() + query.size());
-            const auto & settings = getContext()->getSettingsRef();
-            query_insert_ast = parseQuery(p_query, query, settings.max_query_size, settings.max_parser_depth);
-
-            LOG_INFO(log, "Executing INSERT query: {}", query);
-        }
-
-        try
-        {
-            auto context_select = Context::createCopy(context);
-            context_select->setSettings(task_cluster->settings_pull);
-
-            auto context_insert = Context::createCopy(context);
-            context_insert->setSettings(task_cluster->settings_push);
-
-            /// Custom INSERT SELECT implementation
-            QueryPipeline input;
-            QueryPipeline output;
-            {
-                BlockIO io_insert = InterpreterFactory::instance().get(query_insert_ast, context_insert)->execute();
-
-                InterpreterSelectWithUnionQuery select(query_select_ast, context_select, SelectQueryOptions{});
-                QueryPlan plan;
-                select.buildQueryPlan(plan);
-                auto builder = std::move(*plan.buildQueryPipeline(
-                    QueryPlanOptimizationSettings::fromContext(context_select),
-                    BuildQueryPipelineSettings::fromContext(context_select)));
-
-                output = std::move(io_insert.pipeline);
-
-                /// Add converting actions to make it possible to copy blocks with slightly different schema
-                const auto & select_block = builder.getHeader();
-                const auto & insert_block = output.getHeader();
-                auto actions_dag = ActionsDAG::makeConvertingActions(
-                        select_block.getColumnsWithTypeAndName(),
-                        insert_block.getColumnsWithTypeAndName(),
-                        ActionsDAG::MatchColumnsMode::Position);
-
-                auto actions = std::make_shared<ExpressionActions>(actions_dag, ExpressionActionsSettings::fromContext(getContext()));
-
-                builder.addSimpleTransform([&](const Block & header)
-                {
-                    return std::make_shared<ExpressionTransform>(header, actions);
-                });
-                input = QueryPipelineBuilder::getPipeline(std::move(builder));
-            }
-
-            /// Fail-fast optimization to abort copying when the current clean state expires
-            std::future<Coordination::ExistsResponse> future_is_dirty_checker;
-
-            Stopwatch watch(CLOCK_MONOTONIC_COARSE);
-            constexpr UInt64 check_period_milliseconds = 500;
-
-            /// Will asynchronously check that ZooKeeper connection and is_dirty flag appearing while copying data
-            auto cancel_check = [&] ()
-            {
-                if (zookeeper->expired())
-                    throw Exception(ErrorCodes::UNFINISHED, "ZooKeeper session is expired, cancel INSERT SELECT");
-
-                if (!future_is_dirty_checker.valid())
-                    future_is_dirty_checker = zookeeper->asyncExists(piece_is_dirty_flag_path);
-
-                /// check_period_milliseconds should less than average insert time of single block
-                /// Otherwise, the insertion will slow a little bit
-                if (watch.elapsedMilliseconds() >= check_period_milliseconds)
-                {
-                    Coordination::ExistsResponse status = future_is_dirty_checker.get();
-
-                    if (status.error != Coordination::Error::ZNONODE)
-                    {
-                        LogicalClock dirt_discovery_epoch (status.stat.mzxid);
-                        if (dirt_discovery_epoch == clean_state_clock.discovery_zxid)
-                            return false;
-                        throw Exception(ErrorCodes::UNFINISHED, "Partition is dirty, cancel INSERT SELECT");
-                    }
-                }
-
-                return false;
-            };
-
-            /// Update statistics
-            /// It is quite rough: bytes_copied don't take into account DROP PARTITION.
-            auto update_stats = [&cluster_partition] (const Block & block)
-            {
-                cluster_partition.bytes_copied += block.bytes();
-                cluster_partition.rows_copied += block.rows();
-                cluster_partition.blocks_copied += 1;
-            };
-
-            /// Main work is here
-            PullingPipelineExecutor pulling_executor(input);
-            PushingPipelineExecutor pushing_executor(output);
-
-            Block data;
-            bool is_cancelled = false;
-            while (pulling_executor.pull(data))
-            {
-                if (cancel_check())
-                {
-                    is_cancelled = true;
-                    pushing_executor.cancel();
-                    pushing_executor.cancel();
-                    break;
-                }
-                pushing_executor.push(data);
-                update_stats(data);
-            }
-
-            if (!is_cancelled)
-                pushing_executor.finish();
-
-            // Just in case
-            if (future_is_dirty_checker.valid())
-                future_is_dirty_checker.get();
-
-            if (inject_fault)
-                throw Exception(ErrorCodes::UNFINISHED, "Copy fault injection is activated");
-        }
-        catch (...)
-        {
-            tryLogCurrentException(log, "An error occurred during copying, partition will be marked as dirty");
-            create_is_dirty_node(clean_state_clock);
-            return TaskStatus::Error;
-        }
-    }
-
-    LOG_INFO(log, "Partition {} piece {} copied. But not moved to original destination table.", task_partition.name, toString(current_piece_number));
-
-    /// Finalize the processing, change state of current partition task (and also check is_dirty flag)
-    {
-        String state_finished = TaskStateWithOwner::getData(TaskState::Finished, host_id);
-        CleanStateClock new_clean_state_clock (zookeeper, piece_is_dirty_flag_path, piece_is_dirty_cleaned_path);
-        if (clean_state_clock != new_clean_state_clock)
-        {
-            LOG_INFO(log, "Partition {} piece {} clean state changed, cowardly bailing", task_partition.name, toString(current_piece_number));
-            return TaskStatus::Error;
-        }
-        else if (!new_clean_state_clock.is_clean())
-        {
-            LOG_INFO(log, "Partition {} piece {} became dirty and will be dropped and refilled", task_partition.name, toString(current_piece_number));
-            create_is_dirty_node(new_clean_state_clock);
-            return TaskStatus::Error;
-        }
-        zookeeper->set(current_task_piece_status_path, state_finished, 0);
-    }
-
-    return TaskStatus::Finished;
-}
-
-void ClusterCopier::dropAndCreateLocalTable(const ASTPtr & create_ast)
-{
-    const auto & create = create_ast->as<ASTCreateQuery &>();
-    dropLocalTableIfExists({create.getDatabase(), create.getTable()});
-
-    auto create_context = Context::createCopy(getContext());
-
-    InterpreterCreateQuery interpreter(create_ast, create_context);
-    interpreter.execute();
-}
-
-void ClusterCopier::dropLocalTableIfExists(const DatabaseAndTableName & table_name) const
-{
-    auto drop_ast = std::make_shared<ASTDropQuery>();
-    drop_ast->if_exists = true;
-    drop_ast->setDatabase(table_name.first);
-    drop_ast->setTable(table_name.second);
-
-    auto drop_context = Context::createCopy(getContext());
-
-    InterpreterDropQuery interpreter(drop_ast, drop_context);
-    interpreter.execute();
-}
-
-void ClusterCopier::dropHelpingTablesByPieceNumber(const TaskTable & task_table, size_t current_piece_number)
-{
-    LOG_INFO(log, "Removing helping tables piece {}", current_piece_number);
-
-    DatabaseAndTableName original_table = task_table.table_push;
-    DatabaseAndTableName helping_table
-        = DatabaseAndTableName(original_table.first, original_table.second + "_piece_" + toString(current_piece_number));
-
-    String query = "DROP TABLE IF EXISTS " + getQuotedTable(helping_table);
-
-    const ClusterPtr & cluster_push = task_table.cluster_push;
-    Settings settings_push = task_cluster->settings_push;
-
-    LOG_INFO(log, "Execute distributed DROP TABLE: {}", query);
-
-    /// We have to drop partition_piece on each replica
-    UInt64 num_nodes = executeQueryOnCluster(cluster_push, query, settings_push, ClusterExecutionMode::ON_EACH_NODE);
-
-    LOG_INFO(log, "DROP TABLE query was successfully executed on {} nodes.", toString(num_nodes));
-}
-
-void ClusterCopier::dropHelpingTables(const TaskTable & task_table)
-{
-    LOG_INFO(log, "Removing helping tables");
-    for (size_t current_piece_number = 0; current_piece_number < task_table.number_of_splits; ++current_piece_number)
-    {
-        dropHelpingTablesByPieceNumber(task_table, current_piece_number);
-    }
-}
-
-void ClusterCopier::dropParticularPartitionPieceFromAllHelpingTables(const TaskTable & task_table, const String & partition_name)
-{
-    LOG_INFO(log, "Try drop partition partition from all helping tables.");
-    for (size_t current_piece_number = 0; current_piece_number < task_table.number_of_splits; ++current_piece_number)
-    {
-        DatabaseAndTableName original_table = task_table.table_push;
-        DatabaseAndTableName helping_table = DatabaseAndTableName(original_table.first, original_table.second + "_piece_" + toString(current_piece_number));
-
-        String query = "ALTER TABLE " + getQuotedTable(helping_table) + ((partition_name == "'all'") ? " DROP PARTITION ID " : " DROP PARTITION ") + partition_name;
-
-        const ClusterPtr & cluster_push = task_table.cluster_push;
-        Settings settings_push = task_cluster->settings_push;
-
-        LOG_INFO(log, "Execute distributed DROP PARTITION: {}", query);
-        /// We have to drop partition_piece on each replica
-        UInt64 num_nodes = executeQueryOnCluster(
-                cluster_push, query,
-                settings_push,
-                ClusterExecutionMode::ON_EACH_NODE);
-
-        LOG_INFO(log, "DROP PARTITION query was successfully executed on {} nodes.", toString(num_nodes));
-    }
-    LOG_INFO(log, "All helping tables dropped partition {}", partition_name);
-}
-
-String ClusterCopier::getRemoteCreateTable(const DatabaseAndTableName & table, Connection & connection, const Settings & settings)
-{
-    auto remote_context = Context::createCopy(context);
-    remote_context->setSettings(settings);
-
-    String query = "SHOW CREATE TABLE " + getQuotedTable(table);
-
-    QueryPipelineBuilder builder;
-    builder.init(Pipe(std::make_shared<RemoteSource>(
-            std::make_shared<RemoteQueryExecutor>(connection, query, InterpreterShowCreateQuery::getSampleBlock(), remote_context), false, false, /* async_query_sending= */ false)));
-    Block block = getBlockWithAllStreamData(std::move(builder));
-    return typeid_cast<const ColumnString &>(*block.safeGetByPosition(0).column).getDataAt(0).toString();
-}
-
-
-ASTPtr ClusterCopier::getCreateTableForPullShard(const ConnectionTimeouts & timeouts, TaskShard & task_shard)
-{
-    /// Fetch and parse (possibly) new definition
-    auto connection_entry = task_shard.info.pool->get(timeouts, task_cluster->settings_pull, true);
-    String create_query_pull_str = getRemoteCreateTable(
-            task_shard.task_table.table_pull,
-            *connection_entry,
-            task_cluster->settings_pull);
-
-    ParserCreateQuery parser_create_query;
-    const auto & settings = getContext()->getSettingsRef();
-    return parseQuery(parser_create_query, create_query_pull_str, settings.max_query_size, settings.max_parser_depth);
-}
-
-
-/// If it is implicitly asked to create split Distributed table for certain piece on current shard, we will do it.
-void ClusterCopier::createShardInternalTables(const ConnectionTimeouts & timeouts,
-        TaskShard & task_shard, bool create_split)
-{
-    TaskTable & task_table = task_shard.task_table;
-
-    /// We need to update table definitions for each part, it could be changed after ALTER
-    task_shard.current_pull_table_create_query = getCreateTableForPullShard(timeouts, task_shard);
-
-    /// Create local Distributed tables:
-    ///  a table fetching data from current shard and a table inserting data to the whole destination cluster
-    String read_shard_prefix = ".read_shard_" + toString(task_shard.indexInCluster()) + ".";
-    String split_shard_prefix = ".split.";
-    task_shard.table_read_shard = DatabaseAndTableName(working_database_name, read_shard_prefix + task_table.table_id);
-    task_shard.main_table_split_shard = DatabaseAndTableName(working_database_name, split_shard_prefix + task_table.table_id);
-
-    for (const auto & piece_number : collections::range(0, task_table.number_of_splits))
-    {
-        task_shard.list_of_split_tables_on_shard[piece_number] =
-                DatabaseAndTableName(working_database_name, split_shard_prefix + task_table.table_id + "_piece_" + toString(piece_number));
-    }
-
-    /// Create special cluster with single shard
-    String shard_read_cluster_name = read_shard_prefix + task_table.cluster_pull_name;
-    ClusterPtr cluster_pull_current_shard = task_table.cluster_pull->getClusterWithSingleShard(task_shard.indexInCluster());
-    getContext()->setCluster(shard_read_cluster_name, cluster_pull_current_shard);
-
-    auto storage_shard_ast = createASTStorageDistributed(shard_read_cluster_name, task_table.table_pull.first, task_table.table_pull.second);
-
-    auto create_query_ast = removeAliasMaterializedAndTTLColumnsFromCreateQuery(
-        task_shard.current_pull_table_create_query,
-        task_table.allow_to_copy_alias_and_materialized_columns);
-
-    auto create_table_pull_ast = rewriteCreateQueryStorage(create_query_ast, task_shard.table_read_shard, storage_shard_ast);
-    dropAndCreateLocalTable(create_table_pull_ast);
-
-    if (create_split)
-    {
-        auto create_table_split_piece_ast = rewriteCreateQueryStorage(
-                create_query_ast,
-                task_shard.main_table_split_shard,
-                task_table.main_engine_split_ast);
-
-        dropAndCreateLocalTable(create_table_split_piece_ast);
-
-        /// Create auxiliary split tables for each piece
-        for (const auto & piece_number : collections::range(0, task_table.number_of_splits))
-        {
-            const auto & storage_piece_split_ast = task_table.auxiliary_engine_split_asts[piece_number];
-
-            create_table_split_piece_ast = rewriteCreateQueryStorage(
-                    create_query_ast,
-                    task_shard.list_of_split_tables_on_shard[piece_number],
-                    storage_piece_split_ast);
-
-            dropAndCreateLocalTable(create_table_split_piece_ast);
-        }
-    }
-
-}
-
-
-std::set<String> ClusterCopier::getShardPartitions(const ConnectionTimeouts & timeouts, TaskShard & task_shard)
-{
-    std::set<String> res;
-
-    createShardInternalTables(timeouts, task_shard, false);
-
-    TaskTable & task_table = task_shard.task_table;
-
-    const String & partition_name = queryToString(task_table.engine_push_partition_key_ast);
-
-    if (partition_name == "'all'")
-    {
-        res.emplace("'all'");
-        return res;
-    }
-
-    String query;
-    {
-        WriteBufferFromOwnString wb;
-        wb << "SELECT " << partition_name << " AS partition FROM "
-           << getQuotedTable(task_shard.table_read_shard) << " GROUP BY partition ORDER BY partition DESC";
-        query = wb.str();
-    }
-
-    ParserQuery parser_query(query.data() + query.size());
-    const auto & settings = getContext()->getSettingsRef();
-    ASTPtr query_ast = parseQuery(parser_query, query, settings.max_query_size, settings.max_parser_depth);
-
-    LOG_INFO(log, "Computing destination partition set, executing query: {}", query);
-
-    auto local_context = Context::createCopy(context);
-    local_context->setSettings(task_cluster->settings_pull);
-    InterpreterSelectWithUnionQuery select(query_ast, local_context, SelectQueryOptions{});
-    QueryPlan plan;
-    select.buildQueryPlan(plan);
-    auto builder = std::move(*plan.buildQueryPipeline(
-        QueryPlanOptimizationSettings::fromContext(local_context),
-        BuildQueryPipelineSettings::fromContext(local_context)));
-
-    Block block = getBlockWithAllStreamData(std::move(builder));
-
-    if (block)
-    {
-        ColumnWithTypeAndName & column = block.getByPosition(0);
-        task_shard.partition_key_column = column;
-
-        for (size_t i = 0; i < column.column->size(); ++i)
-        {
-            WriteBufferFromOwnString wb;
-            column.type->getDefaultSerialization()->serializeTextQuoted(*column.column, i, wb, FormatSettings());
-            res.emplace(wb.str());
-        }
-    }
-
-    LOG_INFO(log, "There are {} destination partitions in shard {}", res.size(), task_shard.getDescription());
-
-    return res;
-}
-
-bool ClusterCopier::checkShardHasPartition(const ConnectionTimeouts & timeouts,
-        TaskShard & task_shard, const String & partition_quoted_name)
-{
-    createShardInternalTables(timeouts, task_shard, false);
-
-    TaskTable & task_table = task_shard.task_table;
-
-    WriteBufferFromOwnString ss;
-    ss << "WITH " + partition_quoted_name + " AS partition_key ";
-    ss << "SELECT 1 FROM " << getQuotedTable(task_shard.table_read_shard);
-    ss << " WHERE (" << queryToString(task_table.engine_push_partition_key_ast) << " = partition_key)";
-    if (!task_table.where_condition_str.empty())
-        ss << " AND (" << task_table.where_condition_str << ")";
-    ss << " LIMIT 1";
-    auto query = ss.str();
-
-    ParserQuery parser_query(query.data() + query.size());
-    const auto & settings = getContext()->getSettingsRef();
-    ASTPtr query_ast = parseQuery(parser_query, query, settings.max_query_size, settings.max_parser_depth);
-
-    LOG_INFO(log, "Checking shard {} for partition {} existence, executing query: {}",
-        task_shard.getDescription(), partition_quoted_name, query_ast->formatForErrorMessage());
-
-    auto local_context = Context::createCopy(context);
-    local_context->setSettings(task_cluster->settings_pull);
-    auto pipeline = InterpreterFactory::instance().get(query_ast, local_context)->execute().pipeline;
-    PullingPipelineExecutor executor(pipeline);
-    Block block;
-    executor.pull(block);
-    return block.rows() != 0;
-}
-
-bool ClusterCopier::checkPresentPartitionPiecesOnCurrentShard(const ConnectionTimeouts & timeouts,
-                           TaskShard & task_shard, const String & partition_quoted_name, size_t current_piece_number)
-{
-    createShardInternalTables(timeouts, task_shard, false);
-
-    TaskTable & task_table = task_shard.task_table;
-    const size_t number_of_splits = task_table.number_of_splits;
-    const String & primary_key_comma_separated = task_table.primary_key_comma_separated;
-
-    UNUSED(primary_key_comma_separated);
-
-    std::string query;
-
-    query += "WITH " + partition_quoted_name + " AS partition_key ";
-    query += "SELECT 1 FROM " + getQuotedTable(task_shard.table_read_shard);
-
-    if (experimental_use_sample_offset)
-        query += " SAMPLE 1/" + toString(number_of_splits) + " OFFSET " + toString(current_piece_number) + "/" + toString(number_of_splits);
-
-    query += " WHERE (" + queryToString(task_table.engine_push_partition_key_ast) + " = partition_key)";
-
-    if (!experimental_use_sample_offset)
-        query += " AND (cityHash64(" + primary_key_comma_separated + ") % "
-                 + std::to_string(number_of_splits) + " = " + std::to_string(current_piece_number) + " )";
-
-    if (!task_table.where_condition_str.empty())
-        query += " AND (" + task_table.where_condition_str + ")";
-
-    query += " LIMIT 1";
-
-    LOG_INFO(log, "Checking shard {} for partition {} piece {} existence, executing query: {}", task_shard.getDescription(), partition_quoted_name, std::to_string(current_piece_number), query);
-
-    ParserQuery parser_query(query.data() + query.size());
-    const auto & settings = getContext()->getSettingsRef();
-    ASTPtr query_ast = parseQuery(parser_query, query, settings.max_query_size, settings.max_parser_depth);
-
-    auto local_context = Context::createCopy(context);
-    local_context->setSettings(task_cluster->settings_pull);
-    auto pipeline = InterpreterFactory::instance().get(query_ast, local_context)->execute().pipeline;
-    PullingPipelineExecutor executor(pipeline);
-    Block result;
-    executor.pull(result);
-    if (result.rows() != 0)
-        LOG_INFO(log, "Partition {} piece number {} is PRESENT on shard {}", partition_quoted_name, std::to_string(current_piece_number), task_shard.getDescription());
-    else
-        LOG_INFO(log, "Partition {} piece number {} is ABSENT on shard {}", partition_quoted_name, std::to_string(current_piece_number), task_shard.getDescription());
-    return result.rows() != 0;
-}
-
-
-/** Executes simple query (without output streams, for example DDL queries) on each shard of the cluster
-  * Returns number of shards for which at least one replica executed query successfully
-  */
-UInt64 ClusterCopier::executeQueryOnCluster(
-        const ClusterPtr & cluster,
-        const String & query,
-        const Settings & current_settings,
-        ClusterExecutionMode execution_mode) const
-{
-    ClusterPtr cluster_for_query = cluster;
-    if (execution_mode == ClusterExecutionMode::ON_EACH_NODE)
-        cluster_for_query = cluster->getClusterWithReplicasAsShards(current_settings);
-
-    std::vector<std::shared_ptr<Connection>> connections;
-    connections.reserve(cluster->getShardCount());
-
-    std::atomic<UInt64> successfully_executed = 0;
-
-    for (const auto & replicas : cluster_for_query->getShardsAddresses())
-    {
-        for (const auto & node : replicas)
-        {
-            try
-            {
-                connections.emplace_back(std::make_shared<Connection>(
-                    node.host_name, node.port, node.default_database,
-                    node.user, node.password, ssh::SSHKey(), node.quota_key, node.cluster, node.cluster_secret,
-                    "ClusterCopier", node.compression, node.secure
-                ));
-
-                /// We execute only Alter, Create and Drop queries.
-                const auto header = Block{};
-
-                /// For unknown reason global context is passed to IStorage::read() method
-                /// So, task_identifier is passed as constructor argument. It is more obvious.
-                auto remote_query_executor = std::make_shared<RemoteQueryExecutor>(
-                        *connections.back(), query, header, getContext(),
-                        /*throttler=*/nullptr, Scalars(), Tables(), QueryProcessingStage::Complete);
-
-                try
-                {
-                    remote_query_executor->sendQuery();
-                }
-                catch (...)
-                {
-                    LOG_WARNING(log, "Node with address {} seems to be unreachable.", node.host_name);
-                    continue;
-                }
-
-                while (true)
-                {
-                    auto block = remote_query_executor->readBlock();
-                    if (!block)
-                        break;
-                }
-
-                remote_query_executor->finish();
-                ++successfully_executed;
-                break;
-            }
-            catch (...)
-            {
-                LOG_WARNING(log, "An error occurred while processing query: {}", query);
-                tryLogCurrentException(log);
-                continue;
-            }
-        }
-    }
-
-    return successfully_executed.load();
-}
-
-}
diff --git a/programs/copier/ClusterCopier.h b/programs/copier/ClusterCopier.h
deleted file mode 100644
index 01f8b30f546..00000000000
--- a/programs/copier/ClusterCopier.h
+++ /dev/null
@@ -1,240 +0,0 @@
-#pragma once
-
-#include "Aliases.h"
-#include "Internals.h"
-#include "TaskCluster.h"
-#include "TaskShard.h"
-#include "TaskTable.h"
-#include "ShardPartition.h"
-#include "ShardPartitionPiece.h"
-#include "ZooKeeperStaff.h"
-
-
-namespace DB
-{
-
-class ClusterCopier : WithMutableContext
-{
-public:
-    ClusterCopier(const String & task_path_,
-                  const String & host_id_,
-                  const String & proxy_database_name_,
-                  ContextMutablePtr context_,
-                  LoggerRawPtr log_)
-            : WithMutableContext(context_),
-            task_zookeeper_path(task_path_),
-            host_id(host_id_),
-            working_database_name(proxy_database_name_),
-            log(log_) {}
-
-    void init();
-
-    template <typename T>
-    decltype(auto) retry(T && func, UInt64 max_tries = 100);
-
-    void discoverShardPartitions(const ConnectionTimeouts & timeouts, const TaskShardPtr & task_shard);
-
-    /// Compute set of partitions, assume set of partitions aren't changed during the processing
-    void discoverTablePartitions(const ConnectionTimeouts & timeouts, TaskTable & task_table, UInt64 num_threads = 0);
-
-    void uploadTaskDescription(const std::string & task_path, const std::string & task_file, bool force);
-
-    void reloadTaskDescription();
-
-    void updateConfigIfNeeded();
-
-    void process(const ConnectionTimeouts & timeouts);
-
-    /// Disables DROP PARTITION commands that used to clear data after errors
-    void setSafeMode(bool is_safe_mode_ = true)
-    {
-        is_safe_mode = is_safe_mode_;
-    }
-
-    void setCopyFaultProbability(double copy_fault_probability_)
-    {
-        copy_fault_probability = copy_fault_probability_;
-    }
-
-    void setMoveFaultProbability(double move_fault_probability_)
-    {
-        move_fault_probability = move_fault_probability_;
-    }
-
-    void setExperimentalUseSampleOffset(bool value)
-    {
-        experimental_use_sample_offset = value;
-    }
-
-    void setMaxTableTries(UInt64 tries)
-    {
-        max_table_tries = tries;
-    }
-    void setMaxShardPartitionTries(UInt64 tries)
-    {
-        max_shard_partition_tries = tries;
-    }
-    void setMaxShardPartitionPieceTriesForAlter(UInt64 tries)
-    {
-        max_shard_partition_piece_tries_for_alter = tries;
-    }
-    void setRetryDelayMs(std::chrono::milliseconds ms)
-    {
-        retry_delay_ms = ms;
-    }
-
-protected:
-
-    String getWorkersPath() const
-    {
-        return task_cluster->task_zookeeper_path + "/task_active_workers";
-    }
-
-    String getWorkersPathVersion() const
-    {
-        return getWorkersPath() + "_version";
-    }
-
-    String getCurrentWorkerNodePath() const
-    {
-        return getWorkersPath() + "/" + host_id;
-    }
-
-    zkutil::EphemeralNodeHolder::Ptr createTaskWorkerNodeAndWaitIfNeed(
-            const zkutil::ZooKeeperPtr & zookeeper,
-            const String & description,
-            bool unprioritized);
-
-    /*
-     * Checks that partition piece or some other entity is clean.
-     * The only requirement is that you have to pass is_dirty_flag_path and is_dirty_cleaned_path to the function.
-     * And is_dirty_flag_path is a parent of is_dirty_cleaned_path.
-     * */
-    static bool checkPartitionPieceIsClean(
-            const zkutil::ZooKeeperPtr & zookeeper,
-            const CleanStateClock & clean_state_clock,
-            const String & task_status_path);
-
-    bool checkAllPiecesInPartitionAreDone(const TaskTable & task_table, const String & partition_name, const TasksShard & shards_with_partition);
-
-    /** Checks that the whole partition of a table was copied. We should do it carefully due to dirty lock.
-     * State of some task could change during the processing.
-     * We have to ensure that all shards have the finished state and there is no dirty flag.
-     * Moreover, we have to check status twice and check zxid, because state can change during the checking.
-     */
-
-    /* The same as function above
-     * Assume that we don't know on which shards do we have partition certain piece.
-     * We'll check them all (I mean shards that contain the whole partition)
-     * And shards that don't have certain piece MUST mark that piece is_done true.
-     * */
-    bool checkPartitionPieceIsDone(const TaskTable & task_table, const String & partition_name,
-                                   size_t piece_number, const TasksShard & shards_with_partition);
-
-
-    /*Alter successful insertion to helping tables it will move all pieces to destination table*/
-    TaskStatus tryMoveAllPiecesToDestinationTable(const TaskTable & task_table, const String & partition_name);
-
-    /// Removes MATERIALIZED and ALIAS columns from create table query
-    static ASTPtr removeAliasMaterializedAndTTLColumnsFromCreateQuery(const ASTPtr & query_ast, bool allow_to_copy_alias_and_materialized_columns);
-
-    bool tryDropPartitionPiece(ShardPartition & task_partition, size_t current_piece_number,
-            const zkutil::ZooKeeperPtr & zookeeper, const CleanStateClock & clean_state_clock);
-
-    bool tryProcessTable(const ConnectionTimeouts & timeouts, TaskTable & task_table);
-
-    TaskStatus tryCreateDestinationTable(const ConnectionTimeouts & timeouts, TaskTable & task_table);
-    /// Job for copying partition from particular shard.
-    TaskStatus tryProcessPartitionTask(const ConnectionTimeouts & timeouts,
-                                       ShardPartition & task_partition,
-                                       bool is_unprioritized_task);
-
-    TaskStatus iterateThroughAllPiecesInPartition(const ConnectionTimeouts & timeouts,
-                                                  ShardPartition & task_partition,
-                                                  bool is_unprioritized_task);
-
-    TaskStatus processPartitionPieceTaskImpl(const ConnectionTimeouts & timeouts,
-                                             ShardPartition & task_partition,
-                                             size_t current_piece_number,
-                                             bool is_unprioritized_task);
-
-    void dropAndCreateLocalTable(const ASTPtr & create_ast);
-
-    void dropLocalTableIfExists(const DatabaseAndTableName & table_name) const;
-
-    void dropHelpingTables(const TaskTable & task_table);
-
-    void dropHelpingTablesByPieceNumber(const TaskTable & task_table, size_t current_piece_number);
-
-    /// Is used for usage less disk space.
-    /// After all pieces were successfully moved to original destination
-    /// table we can get rid of partition pieces (partitions in helping tables).
-    void dropParticularPartitionPieceFromAllHelpingTables(const TaskTable & task_table, const String & partition_name);
-
-    String getRemoteCreateTable(const DatabaseAndTableName & table, Connection & connection, const Settings & settings);
-
-    ASTPtr getCreateTableForPullShard(const ConnectionTimeouts & timeouts, TaskShard & task_shard);
-
-    /// If it is implicitly asked to create split Distributed table for certain piece on current shard, we will do it.
-    void createShardInternalTables(const ConnectionTimeouts & timeouts, TaskShard & task_shard, bool create_split = true);
-
-    std::set<String> getShardPartitions(const ConnectionTimeouts & timeouts, TaskShard & task_shard);
-
-    bool checkShardHasPartition(const ConnectionTimeouts & timeouts, TaskShard & task_shard, const String & partition_quoted_name);
-
-    bool checkPresentPartitionPiecesOnCurrentShard(const ConnectionTimeouts & timeouts,
-             TaskShard & task_shard, const String & partition_quoted_name, size_t current_piece_number);
-
-    /*
-     * This class is used in executeQueryOnCluster function
-     * You can execute query on each shard (no sense it is executed on each replica of a shard or not)
-     * or you can execute query on each replica on each shard.
-     * First mode is useful for INSERTS queries.
-     * */
-    enum ClusterExecutionMode
-    {
-        ON_EACH_SHARD,
-        ON_EACH_NODE
-    };
-
-    /** Executes simple query (without output streams, for example DDL queries) on each shard of the cluster
-      * Returns number of shards for which at least one replica executed query successfully
-      */
-    UInt64 executeQueryOnCluster(
-            const ClusterPtr & cluster,
-            const String & query,
-            const Settings & current_settings,
-            ClusterExecutionMode execution_mode = ClusterExecutionMode::ON_EACH_SHARD) const;
-
-private:
-    String task_zookeeper_path;
-    String task_description_path;
-    String host_id;
-    String working_database_name;
-
-    /// Auto update config stuff
-    UInt64 task_description_current_version = 1;
-    std::atomic<UInt64> task_description_version{1};
-    Coordination::WatchCallback task_description_watch_callback;
-    /// ZooKeeper session used to set the callback
-    zkutil::ZooKeeperPtr task_description_watch_zookeeper;
-
-    ConfigurationPtr task_cluster_initial_config;
-    ConfigurationPtr task_cluster_current_config;
-
-    std::unique_ptr<TaskCluster> task_cluster;
-
-    bool is_safe_mode = false;
-    double copy_fault_probability = 0.0;
-    double move_fault_probability = 0.0;
-
-    bool experimental_use_sample_offset{false};
-
-    LoggerRawPtr log;
-
-    UInt64 max_table_tries = 3;
-    UInt64 max_shard_partition_tries = 3;
-    UInt64 max_shard_partition_piece_tries_for_alter = 10;
-    std::chrono::milliseconds retry_delay_ms{1000};
-};
-}
diff --git a/programs/copier/ClusterCopierApp.cpp b/programs/copier/ClusterCopierApp.cpp
deleted file mode 100644
index fdf07dec61a..00000000000
--- a/programs/copier/ClusterCopierApp.cpp
+++ /dev/null
@@ -1,252 +0,0 @@
-#include "ClusterCopierApp.h"
-#include <Common/ZooKeeper/ZooKeeper.h>
-#include <Common/StatusFile.h>
-#include <Common/TerminalSize.h>
-#include <Databases/registerDatabases.h>
-#include <IO/ConnectionTimeouts.h>
-#include <Interpreters/registerInterpreters.h>
-#include <Formats/registerFormats.h>
-#include <Common/scope_guard_safe.h>
-#include <unistd.h>
-#include <filesystem>
-
-namespace fs = std::filesystem;
-
-namespace DB
-{
-
-/// ClusterCopierApp
-
-void ClusterCopierApp::initialize(Poco::Util::Application & self)
-{
-    is_help = config().has("help");
-    if (is_help)
-        return;
-
-    config_xml_path = config().getString("config-file");
-    task_path = config().getString("task-path");
-    log_level = config().getString("log-level", "info");
-    is_safe_mode = config().has("safe-mode");
-    is_status_mode = config().has("status");
-    if (config().has("copy-fault-probability"))
-        copy_fault_probability = std::max(std::min(config().getDouble("copy-fault-probability"), 1.0), 0.0);
-    if (config().has("move-fault-probability"))
-        move_fault_probability = std::max(std::min(config().getDouble("move-fault-probability"), 1.0), 0.0);
-    base_dir = (config().has("base-dir")) ? config().getString("base-dir") : fs::current_path().string();
-
-    max_table_tries = std::max<size_t>(config().getUInt("max-table-tries", 3), 1);
-    max_shard_partition_tries = std::max<size_t>(config().getUInt("max-shard-partition-tries", 3), 1);
-    max_shard_partition_piece_tries_for_alter = std::max<size_t>(config().getUInt("max-shard-partition-piece-tries-for-alter", 10), 1);
-    retry_delay_ms = std::chrono::milliseconds(std::max<size_t>(config().getUInt("retry-delay-ms", 1000), 100));
-
-    if (config().has("experimental-use-sample-offset"))
-        experimental_use_sample_offset = config().getBool("experimental-use-sample-offset");
-
-    // process_id is '<hostname>#<start_timestamp>_<pid>'
-    time_t timestamp = Poco::Timestamp().epochTime();
-    auto curr_pid = Poco::Process::id();
-
-    process_id = std::to_string(DateLUT::serverTimezoneInstance().toNumYYYYMMDDhhmmss(timestamp)) + "_" + std::to_string(curr_pid);
-    host_id = escapeForFileName(getFQDNOrHostName()) + '#' + process_id;
-    process_path = fs::weakly_canonical(fs::path(base_dir) / ("clickhouse-copier_" + process_id));
-    fs::create_directories(process_path);
-
-    /// Override variables for BaseDaemon
-    if (config().has("log-level"))
-        config().setString("logger.level", config().getString("log-level"));
-
-    if (config().has("base-dir") || !config().has("logger.log"))
-        config().setString("logger.log", fs::path(process_path) / "log.log");
-
-    if (config().has("base-dir") || !config().has("logger.errorlog"))
-        config().setString("logger.errorlog", fs::path(process_path) / "log.err.log");
-
-    Base::initialize(self);
-}
-
-
-void ClusterCopierApp::handleHelp(const std::string &, const std::string &)
-{
-    uint16_t terminal_width = 0;
-    if (isatty(STDIN_FILENO))
-        terminal_width = getTerminalWidth();
-
-    Poco::Util::HelpFormatter help_formatter(options());
-    if (terminal_width)
-        help_formatter.setWidth(terminal_width);
-    help_formatter.setCommand(commandName());
-    help_formatter.setHeader("Copies tables from one cluster to another");
-    help_formatter.setUsage("--config-file <config-file> --task-path <task-path>");
-    help_formatter.format(std::cerr);
-    help_formatter.setFooter("See also: https://clickhouse.com/docs/en/operations/utilities/clickhouse-copier/");
-
-    stopOptionsProcessing();
-}
-
-
-void ClusterCopierApp::defineOptions(Poco::Util::OptionSet & options)
-{
-    Base::defineOptions(options);
-
-    options.addOption(Poco::Util::Option("task-path", "", "path to task in ZooKeeper")
-                          .argument("task-path").binding("task-path"));
-    options.addOption(Poco::Util::Option("task-file", "", "path to task file for uploading in ZooKeeper to task-path")
-                          .argument("task-file").binding("task-file"));
-    options.addOption(Poco::Util::Option("task-upload-force", "", "Force upload task-file even node already exists. Default is false.")
-                          .argument("task-upload-force").binding("task-upload-force"));
-    options.addOption(Poco::Util::Option("safe-mode", "", "disables ALTER DROP PARTITION in case of errors")
-                          .binding("safe-mode"));
-    options.addOption(Poco::Util::Option("copy-fault-probability", "", "the copying fails with specified probability (used to test partition state recovering)")
-                          .argument("copy-fault-probability").binding("copy-fault-probability"));
-    options.addOption(Poco::Util::Option("move-fault-probability", "", "the moving fails with specified probability (used to test partition state recovering)")
-                              .argument("move-fault-probability").binding("move-fault-probability"));
-    options.addOption(Poco::Util::Option("log-level", "", "sets log level")
-                          .argument("log-level").binding("log-level"));
-    options.addOption(Poco::Util::Option("base-dir", "", "base directory for copiers, consecutive copier launches will populate /base-dir/launch_id/* directories")
-                          .argument("base-dir").binding("base-dir"));
-    options.addOption(Poco::Util::Option("experimental-use-sample-offset", "", "Use SAMPLE OFFSET query instead of cityHash64(PRIMARY KEY) % n == k")
-                          .argument("experimental-use-sample-offset").binding("experimental-use-sample-offset"));
-    options.addOption(Poco::Util::Option("status", "", "Get for status for current execution").binding("status"));
-
-    options.addOption(Poco::Util::Option("max-table-tries", "", "Number of tries for the copy table task")
-                          .argument("max-table-tries").binding("max-table-tries"));
-    options.addOption(Poco::Util::Option("max-shard-partition-tries", "", "Number of tries for the copy one partition task")
-                          .argument("max-shard-partition-tries").binding("max-shard-partition-tries"));
-    options.addOption(Poco::Util::Option("max-shard-partition-piece-tries-for-alter", "", "Number of tries for final ALTER ATTACH to destination table")
-                          .argument("max-shard-partition-piece-tries-for-alter").binding("max-shard-partition-piece-tries-for-alter"));
-    options.addOption(Poco::Util::Option("retry-delay-ms", "", "Delay between task retries")
-                          .argument("retry-delay-ms").binding("retry-delay-ms"));
-
-    using Me = std::decay_t<decltype(*this)>;
-    options.addOption(Poco::Util::Option("help", "", "produce this help message").binding("help")
-                          .callback(Poco::Util::OptionCallback<Me>(this, &Me::handleHelp)));
-}
-
-
-void ClusterCopierApp::mainImpl()
-{
-    /// Status command
-    {
-        if (is_status_mode)
-        {
-            SharedContextHolder shared_context = Context::createShared();
-            auto context = Context::createGlobal(shared_context.get());
-            context->makeGlobalContext();
-            SCOPE_EXIT_SAFE(context->shutdown());
-
-            auto zookeeper = context->getZooKeeper();
-            auto status_json = zookeeper->get(task_path + "/status");
-
-            LOG_INFO(&logger(), "{}", status_json);
-            std::cout << status_json << std::endl;
-
-            context->resetZooKeeper();
-            return;
-        }
-    }
-    StatusFile status_file(process_path + "/status", StatusFile::write_full_info);
-    ThreadStatus thread_status;
-
-    auto * log = &logger();
-    LOG_INFO(log, "Starting clickhouse-copier (id {}, host_id {}, path {}, revision {})", process_id, host_id, process_path, ClickHouseRevision::getVersionRevision());
-
-    SharedContextHolder shared_context = Context::createShared();
-    auto context = Context::createGlobal(shared_context.get());
-    context->makeGlobalContext();
-    SCOPE_EXIT_SAFE(context->shutdown());
-
-    context->setConfig(loaded_config.configuration);
-    context->setApplicationType(Context::ApplicationType::LOCAL);
-    context->setPath(process_path + "/");
-
-    registerInterpreters();
-    registerFunctions();
-    registerAggregateFunctions();
-    registerTableFunctions();
-    registerDatabases();
-    registerStorages();
-    registerDictionaries();
-    registerDisks(/* global_skip_access_check= */ true);
-    registerFormats();
-
-    static const std::string default_database = "_local";
-    DatabaseCatalog::instance().attachDatabase(default_database, std::make_shared<DatabaseMemory>(default_database, context));
-    context->setCurrentDatabase(default_database);
-
-    /// Disable queries logging, since:
-    /// - There are bits that is not allowed for global context, like adding factories info (for the query_log)
-    /// - And anyway it is useless for copier.
-    context->setSetting("log_queries", false);
-
-    auto local_context = Context::createCopy(context);
-
-    /// Initialize query scope just in case.
-    CurrentThread::QueryScope query_scope(local_context);
-
-    auto copier = std::make_unique<ClusterCopier>(
-        task_path, host_id, default_database, local_context, log);
-    copier->setSafeMode(is_safe_mode);
-    copier->setCopyFaultProbability(copy_fault_probability);
-    copier->setMoveFaultProbability(move_fault_probability);
-    copier->setMaxTableTries(max_table_tries);
-    copier->setMaxShardPartitionTries(max_shard_partition_tries);
-    copier->setMaxShardPartitionPieceTriesForAlter(max_shard_partition_piece_tries_for_alter);
-    copier->setRetryDelayMs(retry_delay_ms);
-    copier->setExperimentalUseSampleOffset(experimental_use_sample_offset);
-
-    auto task_file = config().getString("task-file", "");
-    if (!task_file.empty())
-        copier->uploadTaskDescription(task_path, task_file, config().getBool("task-upload-force", false));
-
-    zkutil::validateZooKeeperConfig(config());
-
-    copier->init();
-    copier->process(ConnectionTimeouts::getTCPTimeoutsWithoutFailover(context->getSettingsRef()));
-
-    /// Reset ZooKeeper before removing ClusterCopier.
-    /// Otherwise zookeeper watch can call callback which use already removed ClusterCopier object.
-    context->resetZooKeeper();
-}
-
-
-int ClusterCopierApp::main(const std::vector<std::string> &)
-{
-    if (is_help)
-        return 0;
-
-    try
-    {
-        mainImpl();
-    }
-    catch (...)
-    {
-        tryLogCurrentException(&Poco::Logger::root(), __PRETTY_FUNCTION__);
-        auto code = getCurrentExceptionCode();
-
-        return (code) ? code : -1;
-    }
-
-    return 0;
-}
-
-
-}
-
-#pragma GCC diagnostic ignored "-Wunused-function"
-#pragma GCC diagnostic ignored "-Wmissing-declarations"
-
-int mainEntryClickHouseClusterCopier(int argc, char ** argv)
-{
-    try
-    {
-        DB::ClusterCopierApp app;
-        return app.run(argc, argv);
-    }
-    catch (...)
-    {
-        std::cerr << DB::getCurrentExceptionMessage(true) << "\n";
-        auto code = DB::getCurrentExceptionCode();
-
-        return (code) ? code : -1;
-    }
-}
diff --git a/programs/copier/ClusterCopierApp.h b/programs/copier/ClusterCopierApp.h
deleted file mode 100644
index 0ddc232381e..00000000000
--- a/programs/copier/ClusterCopierApp.h
+++ /dev/null
@@ -1,99 +0,0 @@
-#pragma once
-
-#include <Poco/Util/ServerApplication.h>
-#include <Daemon/BaseDaemon.h>
-
-#include "ClusterCopier.h"
-
-/* clickhouse cluster copier util
- * Copies tables data from one cluster to new tables of other (possibly the same) cluster in distributed fault-tolerant manner.
- *
- * See overview in the docs: docs/en/utils/clickhouse-copier.md
- *
- * Implementation details:
- *
- * cluster-copier workers pull each partition of each shard of the source cluster and push it to the destination cluster through
- * Distributed table (to perform data resharding). So, worker job is a partition of a source shard.
- * A job has three states: Active, Finished and Abandoned. Abandoned means that worker died and did not finish the job.
- *
- * If an error occurred during the copying (a worker failed or a worker did not finish the INSERT), then the whole partition (on
- * all destination servers) should be dropped and refilled. So, copying entity is a partition of all destination shards.
- * If a failure is detected a special /is_dirty node is created in ZooKeeper signalling that other workers copying the same partition
- * should stop, after a refilling procedure should start.
- *
- * ZooKeeper task node has the following structure:
- *  /task/path_root                     - path passed in --task-path parameter
- *      /description                    - contains user-defined XML config of the task
- *      /task_active_workers            - contains ephemeral nodes of all currently active workers, used to implement max_workers limitation
- *          /server_fqdn#PID_timestamp  - cluster-copier worker ID
- *          ...
- *      /tables             - directory with table tasks
- *      /cluster.db.table1  - directory of table_hits task
- *          /partition1     - directory for partition1
- *              /shards     - directory for source cluster shards
- *                  /1      - worker job for the first shard of partition1 of table test.hits
- *                            Contains info about current status (Active or Finished) and worker ID.
- *                  /2
- *                  ...
- *              /partition_active_workers
- *                  /1      - for each job in /shards a corresponding ephemeral node created in /partition_active_workers
- *                            It is used to detect Abandoned jobs (if there is Active node in /shards and there is no node in
- *                            /partition_active_workers).
- *                            Also, it is used to track active workers in the partition (when we need to refill the partition we do
- *                            not DROP PARTITION while there are active workers)
- *                  /2
- *                  ...
- *              /is_dirty   - the node is set if some worker detected that an error occurred (the INSERT is failed or an Abandoned node is
- *                            detected). If the node appeared workers in this partition should stop and start cleaning and refilling
- *                            partition procedure.
- *                            During this procedure a single 'cleaner' worker is selected. The worker waits for stopping all partition
- *                            workers, removes /shards node, executes DROP PARTITION on each destination node and removes /is_dirty node.
- *                  /cleaner- An ephemeral node used to select 'cleaner' worker. Contains ID of the worker.
- *      /cluster.db.table2
- *          ...
- */
-
-namespace DB
-{
-
-class ClusterCopierApp : public BaseDaemon
-{
-public:
-
-    void initialize(Poco::Util::Application & self) override;
-
-    void handleHelp(const std::string &, const std::string &);
-
-    void defineOptions(Poco::Util::OptionSet & options) override;
-
-    int main(const std::vector<std::string> &) override;
-
-private:
-
-    using Base = BaseDaemon;
-
-    void mainImpl();
-
-    std::string config_xml_path;
-    std::string task_path;
-    std::string log_level = "info";
-    bool is_safe_mode = false;
-    bool is_status_mode = false;
-    double copy_fault_probability = 0.0;
-    double move_fault_probability = 0.0;
-    bool is_help = false;
-
-    UInt64 max_table_tries = 3;
-    UInt64 max_shard_partition_tries = 3;
-    UInt64 max_shard_partition_piece_tries_for_alter = 10;
-    std::chrono::milliseconds retry_delay_ms{1000};
-
-    bool experimental_use_sample_offset{false};
-
-    std::string base_dir;
-    std::string process_path;
-    std::string process_id;
-    std::string host_id;
-};
-
-}
diff --git a/programs/copier/ClusterPartition.h b/programs/copier/ClusterPartition.h
deleted file mode 100644
index 22063989e22..00000000000
--- a/programs/copier/ClusterPartition.h
+++ /dev/null
@@ -1,22 +0,0 @@
-#pragma once
-
-#include <base/types.h>
-#include <map>
-
-namespace DB
-{
-
-/// Contains info about all shards that contain a partition
-struct ClusterPartition
-{
-    double elapsed_time_seconds = 0;
-    UInt64 bytes_copied = 0;
-    UInt64 rows_copied = 0;
-    UInt64 blocks_copied = 0;
-
-    UInt64 total_tries = 0;
-};
-
-using ClusterPartitions = std::map<String, ClusterPartition, std::greater<>>;
-
-}
diff --git a/programs/copier/Internals.cpp b/programs/copier/Internals.cpp
deleted file mode 100644
index dcd199c6b38..00000000000
--- a/programs/copier/Internals.cpp
+++ /dev/null
@@ -1,280 +0,0 @@
-#include "Internals.h"
-#include <Parsers/ASTFunction.h>
-#include <Parsers/ASTIdentifier.h>
-#include <Processors/Executors/PullingPipelineExecutor.h>
-#include <Processors/Transforms/SquashingChunksTransform.h>
-#include <Storages/MergeTree/MergeTreeData.h>
-#include <Storages/extractKeyExpressionList.h>
-#include <QueryPipeline/QueryPipelineBuilder.h>
-
-namespace DB
-{
-namespace ErrorCodes
-{
-    extern const int BAD_ARGUMENTS;
-}
-
-using ConfigurationPtr = Poco::AutoPtr<Poco::Util::AbstractConfiguration>;
-
-ConfigurationPtr getConfigurationFromXMLString(const std::string & xml_data)
-{
-    std::stringstream ss(xml_data);         // STYLE_CHECK_ALLOW_STD_STRING_STREAM
-    Poco::XML::InputSource input_source{ss};
-    return {new Poco::Util::XMLConfiguration{&input_source}};
-}
-
-String getQuotedTable(const String & database, const String & table)
-{
-    if (database.empty())
-        return backQuoteIfNeed(table);
-
-    return backQuoteIfNeed(database) + "." + backQuoteIfNeed(table);
-}
-
-String getQuotedTable(const DatabaseAndTableName & db_and_table)
-{
-    return getQuotedTable(db_and_table.first, db_and_table.second);
-}
-
-
-// Creates AST representing 'ENGINE = Distributed(cluster, db, table, [sharding_key])
-std::shared_ptr<ASTStorage> createASTStorageDistributed(
-        const String & cluster_name, const String & database, const String & table,
-        const ASTPtr & sharding_key_ast)
-{
-    auto args = std::make_shared<ASTExpressionList>();
-    args->children.emplace_back(std::make_shared<ASTLiteral>(cluster_name));
-    args->children.emplace_back(std::make_shared<ASTIdentifier>(database));
-    args->children.emplace_back(std::make_shared<ASTIdentifier>(table));
-    if (sharding_key_ast)
-        args->children.emplace_back(sharding_key_ast);
-
-    auto engine = std::make_shared<ASTFunction>();
-    engine->name = "Distributed";
-    engine->arguments = args;
-
-    auto storage = std::make_shared<ASTStorage>();
-    storage->set(storage->engine, engine);
-
-    return storage;
-}
-
-
-Block getBlockWithAllStreamData(QueryPipelineBuilder builder)
-{
-    builder.addTransform(std::make_shared<SquashingChunksTransform>(
-        builder.getHeader(),
-        std::numeric_limits<size_t>::max(),
-        std::numeric_limits<size_t>::max()));
-
-    auto cur_pipeline = QueryPipelineBuilder::getPipeline(std::move(builder));
-    Block block;
-    PullingPipelineExecutor executor(cur_pipeline);
-    executor.pull(block);
-
-    return block;
-}
-
-bool isExtendedDefinitionStorage(const ASTPtr & storage_ast)
-{
-    const auto & storage = storage_ast->as<ASTStorage &>();
-    return storage.partition_by || storage.order_by || storage.sample_by;
-}
-
-ASTPtr extractPartitionKey(const ASTPtr & storage_ast)
-{
-    String storage_str = queryToString(storage_ast);
-
-    const auto & storage = storage_ast->as<ASTStorage &>();
-    const auto & engine = storage.engine->as<ASTFunction &>();
-
-    if (!endsWith(engine.name, "MergeTree"))
-    {
-        throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unsupported engine was specified in {}, only *MergeTree engines are supported", storage_str);
-    }
-
-    if (isExtendedDefinitionStorage(storage_ast))
-    {
-        if (storage.partition_by)
-            return storage.partition_by->clone();
-
-        static const char * all = "all";
-        return std::make_shared<ASTLiteral>(Field(all, strlen(all)));
-    }
-    else
-    {
-        bool is_replicated = startsWith(engine.name, "Replicated");
-        size_t min_args = is_replicated ? 3 : 1;
-
-        if (!engine.arguments)
-            throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected arguments in {}", storage_str);
-
-        ASTPtr arguments_ast = engine.arguments->clone();
-        ASTs & arguments = arguments_ast->children;
-
-        if (arguments.size() < min_args)
-            throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected at least {} arguments in {}", min_args, storage_str);
-
-        ASTPtr & month_arg = is_replicated ? arguments[2] : arguments[1];
-        return makeASTFunction("toYYYYMM", month_arg->clone());
-    }
-}
-
-ASTPtr extractPrimaryKey(const ASTPtr & storage_ast)
-{
-    String storage_str = queryToString(storage_ast);
-
-    const auto & storage = storage_ast->as<ASTStorage &>();
-    const auto & engine = storage.engine->as<ASTFunction &>();
-
-    if (!endsWith(engine.name, "MergeTree"))
-    {
-        throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unsupported engine was specified in {}, only *MergeTree engines are supported", storage_str);
-    }
-
-    if (!isExtendedDefinitionStorage(storage_ast))
-    {
-        throw Exception(ErrorCodes::BAD_ARGUMENTS, "Is not extended deginition storage {} Will be fixed later.", storage_str);
-    }
-
-    if (storage.primary_key)
-        return storage.primary_key->clone();
-
-    return nullptr;
-}
-
-
-ASTPtr extractOrderBy(const ASTPtr & storage_ast)
-{
-    String storage_str = queryToString(storage_ast);
-
-    const auto & storage = storage_ast->as<ASTStorage &>();
-    const auto & engine = storage.engine->as<ASTFunction &>();
-
-    if (!endsWith(engine.name, "MergeTree"))
-    {
-        throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unsupported engine was specified in {}, only *MergeTree engines are supported", storage_str);
-    }
-
-    if (!isExtendedDefinitionStorage(storage_ast))
-    {
-        throw Exception(ErrorCodes::BAD_ARGUMENTS, "Is not extended deginition storage {} Will be fixed later.", storage_str);
-    }
-
-    if (storage.order_by)
-        return storage.order_by->clone();
-
-    throw Exception(ErrorCodes::BAD_ARGUMENTS, "ORDER BY cannot be empty");
-}
-
-/// Wraps only identifiers with backticks.
-std::string wrapIdentifiersWithBackticks(const ASTPtr & root)
-{
-    if (auto identifier = std::dynamic_pointer_cast<ASTIdentifier>(root))
-        return backQuote(identifier->name());
-
-    if (auto function = std::dynamic_pointer_cast<ASTFunction>(root))
-        return function->name + '(' + wrapIdentifiersWithBackticks(function->arguments) + ')';
-
-    if (auto expression_list = std::dynamic_pointer_cast<ASTExpressionList>(root))
-    {
-        Names function_arguments(expression_list->children.size());
-        for (size_t i = 0; i < expression_list->children.size(); ++i)
-            function_arguments[i] = wrapIdentifiersWithBackticks(expression_list->children[0]);
-        return boost::algorithm::join(function_arguments, ", ");
-    }
-
-    throw Exception(ErrorCodes::BAD_ARGUMENTS, "Primary key could be represented only as columns or functions from columns.");
-}
-
-
-Names extractPrimaryKeyColumnNames(const ASTPtr & storage_ast)
-{
-    const auto sorting_key_ast = extractOrderBy(storage_ast);
-    const auto primary_key_ast = extractPrimaryKey(storage_ast);
-
-    const auto sorting_key_expr_list = extractKeyExpressionList(sorting_key_ast);
-    const auto primary_key_expr_list = primary_key_ast
-                           ? extractKeyExpressionList(primary_key_ast) : sorting_key_expr_list->clone();
-
-    /// Maybe we have to handle VersionedCollapsing engine separately. But in our case in looks pointless.
-
-    size_t primary_key_size = primary_key_expr_list->children.size();
-    size_t sorting_key_size = sorting_key_expr_list->children.size();
-
-    if (primary_key_size > sorting_key_size)
-        throw Exception(ErrorCodes::BAD_ARGUMENTS, "Primary key must be a prefix of the sorting key, but its length: "
-                        "{} is greater than the sorting key length: {}",
-                        primary_key_size, sorting_key_size);
-
-    Names primary_key_columns;
-    NameSet primary_key_columns_set;
-
-    for (size_t i = 0; i < sorting_key_size; ++i)
-    {
-        /// Column name could be represented as a f_1(f_2(...f_n(column_name))).
-        /// Each f_i could take one or more parameters.
-        /// We will wrap identifiers with backticks to allow non-standard identifier names.
-        String sorting_key_column = sorting_key_expr_list->children[i]->getColumnName();
-
-        if (i < primary_key_size)
-        {
-            String pk_column = primary_key_expr_list->children[i]->getColumnName();
-            if (pk_column != sorting_key_column)
-                throw Exception(ErrorCodes::BAD_ARGUMENTS,
-                                "Primary key must be a prefix of the sorting key, "
-                                "but the column in the position {} is {}, not {}", i, sorting_key_column, pk_column);
-
-            if (!primary_key_columns_set.emplace(pk_column).second)
-                throw Exception(ErrorCodes::BAD_ARGUMENTS, "Primary key contains duplicate columns");
-
-            primary_key_columns.push_back(wrapIdentifiersWithBackticks(primary_key_expr_list->children[i]));
-        }
-    }
-
-    return primary_key_columns;
-}
-
-bool isReplicatedTableEngine(const ASTPtr & storage_ast)
-{
-    const auto & storage = storage_ast->as<ASTStorage &>();
-    const auto & engine = storage.engine->as<ASTFunction &>();
-
-    if (!endsWith(engine.name, "MergeTree"))
-    {
-        String storage_str = queryToString(storage_ast);
-        throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unsupported engine was specified in {}, only *MergeTree engines are supported", storage_str);
-    }
-
-    return startsWith(engine.name, "Replicated");
-}
-
-ShardPriority getReplicasPriority(const Cluster::Addresses & replicas, const std::string & local_hostname, UInt8 random)
-{
-    ShardPriority res;
-
-    if (replicas.empty())
-        return res;
-
-    res.is_remote = 1;
-    for (const auto & replica : replicas)
-    {
-        if (isLocalAddress(DNSResolver::instance().resolveHostAllInOriginOrder(replica.host_name).front()))
-        {
-            res.is_remote = 0;
-            break;
-        }
-    }
-
-    res.hostname_difference = std::numeric_limits<size_t>::max();
-    for (const auto & replica : replicas)
-    {
-        size_t difference = getHostNamePrefixDistance(local_hostname, replica.host_name);
-        res.hostname_difference = std::min(difference, res.hostname_difference);
-    }
-
-    res.random = random;
-    return res;
-}
-
-}
diff --git a/programs/copier/Internals.h b/programs/copier/Internals.h
deleted file mode 100644
index 27fedd5d9e8..00000000000
--- a/programs/copier/Internals.h
+++ /dev/null
@@ -1,198 +0,0 @@
-#pragma once
-
-#include <chrono>
-#include <optional>
-#include <Poco/Util/XMLConfiguration.h>
-#include <Poco/Logger.h>
-#include <Poco/ConsoleChannel.h>
-#include <Poco/FormattingChannel.h>
-#include <Poco/PatternFormatter.h>
-#include <Poco/UUIDGenerator.h>
-#include <Poco/Process.h>
-#include <Poco/FileChannel.h>
-#include <Poco/SplitterChannel.h>
-#include <Poco/Util/HelpFormatter.h>
-#include <boost/algorithm/string.hpp>
-#include <Common/logger_useful.h>
-#include <Common/ThreadPool.h>
-#include <Common/Exception.h>
-#include <Common/ZooKeeper/ZooKeeper.h>
-#include <Common/ZooKeeper/KeeperException.h>
-#include <base/getFQDNOrHostName.h>
-#include <Common/isLocalAddress.h>
-#include <Common/typeid_cast.h>
-#include <Common/ClickHouseRevision.h>
-#include <Common/formatReadable.h>
-#include <Common/DNSResolver.h>
-#include <Common/CurrentThread.h>
-#include <Common/getNumberOfPhysicalCPUCores.h>
-#include <Common/ThreadStatus.h>
-#include <Client/Connection.h>
-#include <Interpreters/Context.h>
-#include <Interpreters/Cluster.h>
-#include <Interpreters/InterpreterFactory.h>
-#include <Interpreters/InterpreterExistsQuery.h>
-#include <Interpreters/InterpreterShowCreateQuery.h>
-#include <Interpreters/InterpreterDropQuery.h>
-#include <Interpreters/InterpreterCreateQuery.h>
-#include <Columns/ColumnString.h>
-#include <Columns/ColumnsNumber.h>
-#include <DataTypes/DataTypeString.h>
-#include <DataTypes/NestedUtils.h>
-#include <Parsers/ParserCreateQuery.h>
-#include <Parsers/parseQuery.h>
-#include <Parsers/ParserQuery.h>
-#include <Parsers/ASTCreateQuery.h>
-#include <Parsers/queryToString.h>
-#include <Parsers/ASTDropQuery.h>
-#include <Parsers/ASTLiteral.h>
-#include <Parsers/ASTExpressionList.h>
-#include <Formats/FormatSettings.h>
-#include <QueryPipeline/RemoteQueryExecutor.h>
-#include <IO/ConnectionTimeouts.h>
-#include <IO/Operators.h>
-#include <IO/ReadBufferFromString.h>
-#include <IO/ReadBufferFromFile.h>
-#include <Functions/registerFunctions.h>
-#include <TableFunctions/registerTableFunctions.h>
-#include <AggregateFunctions/registerAggregateFunctions.h>
-#include <Storages/registerStorages.h>
-#include <Storages/StorageDistributed.h>
-#include <Dictionaries/registerDictionaries.h>
-#include <Disks/registerDisks.h>
-#include <Databases/DatabaseMemory.h>
-
-#include "Aliases.h"
-
-namespace DB
-{
-
-namespace ErrorCodes
-{
-    extern const int LOGICAL_ERROR;
-}
-
-
-ConfigurationPtr getConfigurationFromXMLString(const std::string & xml_data);
-
-String getQuotedTable(const String & database, const String & table);
-
-String getQuotedTable(const DatabaseAndTableName & db_and_table);
-
-
-enum class TaskState
-{
-    Started = 0,
-    Finished,
-    Unknown
-};
-
-/// Used to mark status of shard partition tasks
-struct TaskStateWithOwner
-{
-    TaskStateWithOwner() = default;
-
-    TaskStateWithOwner(TaskState state_, const String & owner_) : state(state_), owner(owner_) {}
-
-    TaskState state{TaskState::Unknown};
-    String owner;
-
-    static String getData(TaskState state, const String &owner)
-    {
-        return TaskStateWithOwner(state, owner).toString();
-    }
-
-    String toString() const
-    {
-        WriteBufferFromOwnString wb;
-        wb << static_cast<UInt32>(state) << "\n" << escape << owner;
-        return wb.str();
-    }
-
-    static TaskStateWithOwner fromString(const String & data)
-    {
-        ReadBufferFromString rb(data);
-        TaskStateWithOwner res;
-        UInt32 state;
-
-        rb >> state >> "\n" >> escape >> res.owner;
-
-        if (state >= static_cast<int>(TaskState::Unknown))
-            throw Exception(ErrorCodes::LOGICAL_ERROR, "Unknown state {}", data);
-
-        res.state = static_cast<TaskState>(state);
-        return res;
-    }
-};
-
-
-struct ShardPriority
-{
-    UInt8 is_remote = 1;
-    size_t hostname_difference = 0;
-    UInt8 random = 0;
-
-    static bool greaterPriority(const ShardPriority & current, const ShardPriority & other)
-    {
-        return std::forward_as_tuple(current.is_remote, current.hostname_difference, current.random)
-               < std::forward_as_tuple(other.is_remote, other.hostname_difference, other.random);
-    }
-};
-
-/// Execution status of a task.
-/// Is used for: partition copying task status, partition piece copying task status, partition moving task status.
-enum class TaskStatus
-{
-    Active,
-    Finished,
-    Error,
-};
-
-struct MultiTransactionInfo
-{
-    int32_t code;
-    Coordination::Requests requests;
-    Coordination::Responses responses;
-};
-
-// Creates AST representing 'ENGINE = Distributed(cluster, db, table, [sharding_key])
-std::shared_ptr<ASTStorage> createASTStorageDistributed(
-        const String & cluster_name, const String & database, const String & table,
-        const ASTPtr & sharding_key_ast = nullptr);
-
-Block getBlockWithAllStreamData(QueryPipelineBuilder builder);
-
-bool isExtendedDefinitionStorage(const ASTPtr & storage_ast);
-
-ASTPtr extractPartitionKey(const ASTPtr & storage_ast);
-
-/*
-* Choosing a Primary Key that Differs from the Sorting Key
-* It is possible to specify a primary key (an expression with values that are written in the index file for each mark)
-* that is different from the sorting key (an expression for sorting the rows in data parts).
-* In this case the primary key expression tuple must be a prefix of the sorting key expression tuple.
-* This feature is helpful when using the SummingMergeTree and AggregatingMergeTree table engines.
-* In a common case when using these engines, the table has two types of columns: dimensions and measures.
-* Typical queries aggregate values of measure columns with arbitrary GROUP BY and filtering by dimensions.
-* Because SummingMergeTree and AggregatingMergeTree aggregate rows with the same value of the sorting key,
-* it is natural to add all dimensions to it. As a result, the key expression consists of a long list of columns
-* and this list must be frequently updated with newly added dimensions.
-* In this case it makes sense to leave only a few columns in the primary key that will provide efficient
-* range scans and add the remaining dimension columns to the sorting key tuple.
-* ALTER of the sorting key is a lightweight operation because when a new column is simultaneously added t
-* o the table and to the sorting key, existing data parts don't need to be changed.
-* Since the old sorting key is a prefix of the new sorting key and there is no data in the newly added column,
-* the data is sorted by both the old and new sorting keys at the moment of table modification.
-*
-* */
-ASTPtr extractPrimaryKey(const ASTPtr & storage_ast);
-
-ASTPtr extractOrderBy(const ASTPtr & storage_ast);
-
-Names extractPrimaryKeyColumnNames(const ASTPtr & storage_ast);
-
-bool isReplicatedTableEngine(const ASTPtr & storage_ast);
-
-ShardPriority getReplicasPriority(const Cluster::Addresses & replicas, const std::string & local_hostname, UInt8 random);
-
-}
diff --git a/programs/copier/ShardPartition.cpp b/programs/copier/ShardPartition.cpp
deleted file mode 100644
index 4c962fc807d..00000000000
--- a/programs/copier/ShardPartition.cpp
+++ /dev/null
@@ -1,70 +0,0 @@
-#include "ShardPartition.h"
-
-#include "TaskShard.h"
-#include "TaskTable.h"
-
-namespace DB
-{
-
-ShardPartition::ShardPartition(TaskShard & parent, String name_quoted_, size_t number_of_splits)
-    : task_shard(parent)
-    , name(std::move(name_quoted_))
-{
-    pieces.reserve(number_of_splits);
-}
-
-String ShardPartition::getPartitionCleanStartPath() const
-{
-    return getPartitionPath() + "/clean_start";
-}
-
-String ShardPartition::getPartitionPieceCleanStartPath(size_t current_piece_number) const
-{
-    assert(current_piece_number < task_shard.task_table.number_of_splits);
-    return getPartitionPiecePath(current_piece_number) + "/clean_start";
-}
-
-String ShardPartition::getPartitionPath() const
-{
-    return task_shard.task_table.getPartitionPath(name);
-}
-
-String ShardPartition::getPartitionPiecePath(size_t current_piece_number) const
-{
-    assert(current_piece_number < task_shard.task_table.number_of_splits);
-    return task_shard.task_table.getPartitionPiecePath(name, current_piece_number);
-}
-
-String ShardPartition::getShardStatusPath() const
-{
-    // schema: /<root...>/tables/<table>/<partition>/shards/<shard>
-    // e.g. /root/table_test.hits/201701/shards/1
-    return getPartitionShardsPath() + "/" + toString(task_shard.numberInCluster());
-}
-
-String ShardPartition::getPartitionShardsPath() const
-{
-    return getPartitionPath() + "/shards";
-}
-
-String ShardPartition::getPartitionActiveWorkersPath() const
-{
-    return getPartitionPath() + "/partition_active_workers";
-}
-
-String ShardPartition::getActiveWorkerPath() const
-{
-    return getPartitionActiveWorkersPath() + "/" + toString(task_shard.numberInCluster());
-}
-
-String ShardPartition::getCommonPartitionIsDirtyPath() const
-{
-    return getPartitionPath() + "/is_dirty";
-}
-
-String ShardPartition::getCommonPartitionIsCleanedPath() const
-{
-    return getCommonPartitionIsDirtyPath() + "/cleaned";
-}
-
-}
diff --git a/programs/copier/ShardPartition.h b/programs/copier/ShardPartition.h
deleted file mode 100644
index 2457213733c..00000000000
--- a/programs/copier/ShardPartition.h
+++ /dev/null
@@ -1,54 +0,0 @@
-#pragma once
-
-#include "ShardPartitionPiece.h"
-
-#include <base/types.h>
-
-#include <map>
-
-namespace DB
-{
-
-struct TaskShard;
-
-/// Just destination partition of a shard
-/// I don't know what this comment means.
-/// In short, when we discovered what shards contain currently processing partition,
-/// This class describes a partition (name) that is stored on the shard (parent).
-struct ShardPartition
-{
-    ShardPartition(TaskShard &parent, String name_quoted_, size_t number_of_splits = 10);
-
-    String getPartitionPath() const;
-
-    String getPartitionPiecePath(size_t current_piece_number) const;
-
-    String getPartitionCleanStartPath() const;
-
-    String getPartitionPieceCleanStartPath(size_t current_piece_number) const;
-
-    String getCommonPartitionIsDirtyPath() const;
-
-    String getCommonPartitionIsCleanedPath() const;
-
-    String getPartitionActiveWorkersPath() const;
-
-    String getActiveWorkerPath() const;
-
-    String getPartitionShardsPath() const;
-
-    String getShardStatusPath() const;
-
-    /// What partition pieces are present in current shard.
-    /// FYI: Piece is a part of partition which has modulo equals to concrete constant (less than number_of_splits obliously)
-    /// For example SELECT ... from ... WHERE partition=current_partition AND cityHash64(*) == const;
-    /// Absent pieces have field is_absent_piece equals to true.
-    PartitionPieces pieces;
-
-    TaskShard & task_shard;
-    String name;
-};
-
-using TasksPartition = std::map<String, ShardPartition, std::greater<>>;
-
-}
diff --git a/programs/copier/ShardPartitionPiece.cpp b/programs/copier/ShardPartitionPiece.cpp
deleted file mode 100644
index 36d1621e012..00000000000
--- a/programs/copier/ShardPartitionPiece.cpp
+++ /dev/null
@@ -1,64 +0,0 @@
-#include "ShardPartitionPiece.h"
-
-#include "ShardPartition.h"
-#include "TaskShard.h"
-
-#include <IO/WriteHelpers.h>
-
-namespace DB
-{
-
-ShardPartitionPiece::ShardPartitionPiece(ShardPartition & parent, size_t current_piece_number_, bool is_present_piece_)
-    : is_absent_piece(!is_present_piece_)
-    , current_piece_number(current_piece_number_)
-    , shard_partition(parent)
-{
-}
-
-String ShardPartitionPiece::getPartitionPiecePath() const
-{
-    return shard_partition.getPartitionPath() + "/piece_" + toString(current_piece_number);
-}
-
-String ShardPartitionPiece::getPartitionPieceCleanStartPath() const
-{
-    return getPartitionPiecePath() + "/clean_start";
-}
-
-String ShardPartitionPiece::getPartitionPieceIsDirtyPath() const
-{
-    return getPartitionPiecePath() + "/is_dirty";
-}
-
-String ShardPartitionPiece::getPartitionPieceIsCleanedPath() const
-{
-    return getPartitionPieceIsDirtyPath() + "/cleaned";
-}
-
-String ShardPartitionPiece::getPartitionPieceActiveWorkersPath() const
-{
-    return getPartitionPiecePath() + "/partition_piece_active_workers";
-}
-
-String ShardPartitionPiece::getActiveWorkerPath() const
-{
-    return getPartitionPieceActiveWorkersPath() + "/" + toString(shard_partition.task_shard.numberInCluster());
-}
-
-/// On what shards do we have current partition.
-String ShardPartitionPiece::getPartitionPieceShardsPath() const
-{
-    return getPartitionPiecePath() + "/shards";
-}
-
-String ShardPartitionPiece::getShardStatusPath() const
-{
-    return getPartitionPieceShardsPath() + "/" + toString(shard_partition.task_shard.numberInCluster());
-}
-
-String ShardPartitionPiece::getPartitionPieceCleanerPath() const
-{
-    return getPartitionPieceIsDirtyPath() + "/cleaner";
-}
-
-}
diff --git a/programs/copier/ShardPartitionPiece.h b/programs/copier/ShardPartitionPiece.h
deleted file mode 100644
index 453364c0fc8..00000000000
--- a/programs/copier/ShardPartitionPiece.h
+++ /dev/null
@@ -1,43 +0,0 @@
-#pragma once
-
-#include <base/types.h>
-
-#include <vector>
-
-namespace DB
-{
-
-struct ShardPartition;
-
-struct ShardPartitionPiece
-{
-    ShardPartitionPiece(ShardPartition & parent, size_t current_piece_number_, bool is_present_piece_);
-
-    String getPartitionPiecePath() const;
-
-    String getPartitionPieceCleanStartPath() const;
-
-    String getPartitionPieceIsDirtyPath() const;
-
-    String getPartitionPieceIsCleanedPath() const;
-
-    String getPartitionPieceActiveWorkersPath() const;
-
-    String getActiveWorkerPath() const ;
-
-    /// On what shards do we have current partition.
-    String getPartitionPieceShardsPath() const;
-
-    String getShardStatusPath() const;
-
-    String getPartitionPieceCleanerPath() const;
-
-    bool is_absent_piece;
-    const size_t current_piece_number;
-
-    ShardPartition & shard_partition;
-};
-
-using PartitionPieces = std::vector<ShardPartitionPiece>;
-
-}
diff --git a/programs/copier/StatusAccumulator.cpp b/programs/copier/StatusAccumulator.cpp
deleted file mode 100644
index 77adeac708c..00000000000
--- a/programs/copier/StatusAccumulator.cpp
+++ /dev/null
@@ -1,48 +0,0 @@
-#include "StatusAccumulator.h"
-
-#include <Poco/JSON/Parser.h>
-#include <Poco/JSON/JSON.h>
-#include <Poco/JSON/Object.h>
-#include <Poco/JSON/Stringifier.h>
-
-#include <iostream>
-
-namespace DB
-{
-
-StatusAccumulator::MapPtr StatusAccumulator::fromJSON(String state_json)
-{
-    Poco::JSON::Parser parser;
-    auto state = parser.parse(state_json).extract<Poco::JSON::Object::Ptr>();
-    MapPtr result_ptr = std::make_shared<Map>();
-    for (const auto & table_name : state->getNames())
-    {
-        auto table_status_json = state->getValue<String>(table_name);
-        auto table_status = parser.parse(table_status_json).extract<Poco::JSON::Object::Ptr>();
-        /// Map entry will be created if it is absent
-        auto & map_table_status = (*result_ptr)[table_name];
-        map_table_status.all_partitions_count += table_status->getValue<size_t>("all_partitions_count");
-        map_table_status.processed_partitions_count += table_status->getValue<size_t>("processed_partitions_count");
-    }
-    return result_ptr;
-}
-
-String StatusAccumulator::serializeToJSON(MapPtr statuses)
-{
-    Poco::JSON::Object result_json;
-    for (const auto & [table_name, table_status] : *statuses)
-    {
-        Poco::JSON::Object status_json;
-        status_json.set("all_partitions_count", table_status.all_partitions_count);
-        status_json.set("processed_partitions_count", table_status.processed_partitions_count);
-
-        result_json.set(table_name, status_json);
-    }
-    std::ostringstream oss;     // STYLE_CHECK_ALLOW_STD_STRING_STREAM
-    oss.exceptions(std::ios::failbit);
-    Poco::JSON::Stringifier::stringify(result_json, oss);
-    auto result = oss.str();
-    return result;
-}
-
-}
diff --git a/programs/copier/StatusAccumulator.h b/programs/copier/StatusAccumulator.h
deleted file mode 100644
index d420b611602..00000000000
--- a/programs/copier/StatusAccumulator.h
+++ /dev/null
@@ -1,27 +0,0 @@
-#pragma once
-
-#include <base/types.h>
-
-#include <memory>
-#include <unordered_map>
-
-namespace DB
-{
-
-class StatusAccumulator
-{
-public:
-    struct TableStatus
-    {
-        size_t all_partitions_count;
-        size_t processed_partitions_count;
-    };
-
-    using Map = std::unordered_map<String, TableStatus>;
-    using MapPtr = std::shared_ptr<Map>;
-
-    static MapPtr fromJSON(String state_json);
-    static String serializeToJSON(MapPtr statuses);
-};
-
-}
diff --git a/programs/copier/TaskCluster.cpp b/programs/copier/TaskCluster.cpp
deleted file mode 100644
index 0fb06616e50..00000000000
--- a/programs/copier/TaskCluster.cpp
+++ /dev/null
@@ -1,74 +0,0 @@
-#include "TaskCluster.h"
-
-namespace DB
-{
-
-namespace ErrorCodes
-{
-    extern const int BAD_ARGUMENTS;
-}
-
-TaskCluster::TaskCluster(const String & task_zookeeper_path_, const String & default_local_database_)
-    : task_zookeeper_path(task_zookeeper_path_)
-    , default_local_database(default_local_database_)
-{}
-
-void DB::TaskCluster::loadTasks(const Poco::Util::AbstractConfiguration & config, const String & base_key)
-{
-    String prefix = base_key.empty() ? "" : base_key + ".";
-
-    clusters_prefix = prefix + "remote_servers";
-    if (!config.has(clusters_prefix))
-        throw Exception(ErrorCodes::BAD_ARGUMENTS, "You should specify list of clusters in {}", clusters_prefix);
-
-    Poco::Util::AbstractConfiguration::Keys tables_keys;
-    config.keys(prefix + "tables", tables_keys);
-
-    for (const auto & table_key : tables_keys)
-    {
-        table_tasks.emplace_back(*this, config, prefix + "tables", table_key);
-    }
-}
-
-void DB::TaskCluster::reloadSettings(const Poco::Util::AbstractConfiguration & config, const String & base_key)
-{
-    String prefix = base_key.empty() ? "" : base_key + ".";
-
-    max_workers = config.getUInt64(prefix + "max_workers");
-
-    settings_common = Settings();
-    if (config.has(prefix + "settings"))
-        settings_common.loadSettingsFromConfig(prefix + "settings", config);
-
-    settings_common.prefer_localhost_replica = false;
-
-    settings_pull = settings_common;
-    if (config.has(prefix + "settings_pull"))
-        settings_pull.loadSettingsFromConfig(prefix + "settings_pull", config);
-
-    settings_push = settings_common;
-    if (config.has(prefix + "settings_push"))
-        settings_push.loadSettingsFromConfig(prefix + "settings_push", config);
-
-    auto set_default_value = [] (auto && setting, auto && default_value)
-    {
-        setting = setting.changed ? setting.value : default_value;
-    };
-
-    /// Override important settings
-    settings_pull.readonly = 1;
-    settings_pull.prefer_localhost_replica = false;
-    settings_push.distributed_foreground_insert = true;
-    settings_push.prefer_localhost_replica = false;
-
-    set_default_value(settings_pull.load_balancing, LoadBalancing::NEAREST_HOSTNAME);
-    set_default_value(settings_pull.max_threads, 1);
-    set_default_value(settings_pull.max_block_size, 8192UL);
-    set_default_value(settings_pull.preferred_block_size_bytes, 0);
-
-    set_default_value(settings_push.distributed_background_insert_timeout, 0);
-    set_default_value(settings_push.alter_sync, 2);
-}
-
-}
-
diff --git a/programs/copier/TaskCluster.h b/programs/copier/TaskCluster.h
deleted file mode 100644
index a7f8bc3baca..00000000000
--- a/programs/copier/TaskCluster.h
+++ /dev/null
@@ -1,51 +0,0 @@
-#pragma once
-
-#include "TaskTable.h"
-
-#include <Core/Settings.h>
-#include <base/types.h>
-
-#include <Poco/Util/AbstractConfiguration.h>
-
-#include <pcg_random.hpp>
-
-namespace DB
-{
-
-struct TaskCluster
-{
-    TaskCluster(const String & task_zookeeper_path_, const String & default_local_database_);
-
-    void loadTasks(const Poco::Util::AbstractConfiguration & config, const String & base_key = "");
-
-    /// Set (or update) settings and max_workers param
-    void reloadSettings(const Poco::Util::AbstractConfiguration & config, const String & base_key = "");
-
-    /// Base node for all tasks. Its structure:
-    ///  workers/ - directory with active workers (amount of them is less or equal max_workers)
-    ///  description - node with task configuration
-    ///  table_table1/ - directories with per-partition copying status
-    String task_zookeeper_path;
-
-    /// Database used to create temporary Distributed tables
-    String default_local_database;
-
-    /// Limits number of simultaneous workers
-    UInt64 max_workers = 0;
-
-    /// Base settings for pull and push
-    Settings settings_common;
-    /// Settings used to fetch data
-    Settings settings_pull;
-    /// Settings used to insert data
-    Settings settings_push;
-
-    String clusters_prefix;
-
-    /// Subtasks
-    TasksTable table_tasks;
-
-    pcg64 random_engine;
-};
-
-}
diff --git a/programs/copier/TaskShard.cpp b/programs/copier/TaskShard.cpp
deleted file mode 100644
index d156f451a84..00000000000
--- a/programs/copier/TaskShard.cpp
+++ /dev/null
@@ -1,37 +0,0 @@
-#include "TaskShard.h"
-
-#include "TaskTable.h"
-
-namespace DB
-{
-
-TaskShard::TaskShard(TaskTable & parent, const Cluster::ShardInfo & info_)
-    : task_table(parent)
-    , info(info_)
-{
-    list_of_split_tables_on_shard.assign(task_table.number_of_splits, DatabaseAndTableName());
-}
-
-UInt32 TaskShard::numberInCluster() const
-{
-    return info.shard_num;
-}
-
-UInt32 TaskShard::indexInCluster() const
-{
-    return info.shard_num - 1;
-}
-
-String DB::TaskShard::getDescription() const
-{
-    return fmt::format("N{} (having a replica {}, pull table {} of cluster {}",
-                       numberInCluster(), getHostNameExample(), getQuotedTable(task_table.table_pull), task_table.cluster_pull_name);
-}
-
-String DB::TaskShard::getHostNameExample() const
-{
-    const auto & replicas = task_table.cluster_pull->getShardsAddresses().at(indexInCluster());
-    return replicas.at(0).readableString();
-}
-
-}
diff --git a/programs/copier/TaskShard.h b/programs/copier/TaskShard.h
deleted file mode 100644
index 05d652077ea..00000000000
--- a/programs/copier/TaskShard.h
+++ /dev/null
@@ -1,56 +0,0 @@
-#pragma once
-
-#include "Aliases.h"
-#include "Internals.h"
-#include "ClusterPartition.h"
-#include "ShardPartition.h"
-
-
-namespace DB
-{
-
-struct TaskTable;
-
-struct TaskShard
-{
-    TaskShard(TaskTable & parent, const Cluster::ShardInfo & info_);
-
-    TaskTable & task_table;
-
-    Cluster::ShardInfo info;
-
-    UInt32 numberInCluster() const;
-
-    UInt32 indexInCluster() const;
-
-    String getDescription() const;
-
-    String getHostNameExample() const;
-
-    /// Used to sort clusters by their proximity
-    ShardPriority priority;
-
-    /// Column with unique destination partitions (computed from engine_push_partition_key expr.) in the shard
-    ColumnWithTypeAndName partition_key_column;
-
-    /// There is a task for each destination partition
-    TasksPartition partition_tasks;
-
-    /// Which partitions have been checked for existence
-    /// If some partition from this lists is exists, it is in partition_tasks
-    std::set<String> checked_partitions;
-
-    /// Last CREATE TABLE query of the table of the shard
-    ASTPtr current_pull_table_create_query;
-    ASTPtr current_push_table_create_query;
-
-    /// Internal distributed tables
-    DatabaseAndTableName table_read_shard;
-    DatabaseAndTableName main_table_split_shard;
-    ListOfDatabasesAndTableNames list_of_split_tables_on_shard;
-};
-
-using TaskShardPtr = std::shared_ptr<TaskShard>;
-using TasksShard = std::vector<TaskShardPtr>;
-
-}
diff --git a/programs/copier/TaskTable.cpp b/programs/copier/TaskTable.cpp
deleted file mode 100644
index d055ceb4c7b..00000000000
--- a/programs/copier/TaskTable.cpp
+++ /dev/null
@@ -1,222 +0,0 @@
-#include "TaskTable.h"
-
-#include "ClusterPartition.h"
-#include "TaskCluster.h"
-
-#include <Parsers/ASTFunction.h>
-#include <Common/escapeForFileName.h>
-
-#include <boost/algorithm/string/join.hpp>
-
-
-namespace DB
-{
-namespace ErrorCodes
-{
-    extern const int UNKNOWN_ELEMENT_IN_CONFIG;
-    extern const int LOGICAL_ERROR;
-}
-
-TaskTable::TaskTable(TaskCluster & parent, const Poco::Util::AbstractConfiguration & config,
-                     const String & prefix_, const String & table_key)
-        : task_cluster(parent)
-{
-    String table_prefix = prefix_ + "." + table_key + ".";
-
-    name_in_config = table_key;
-
-    number_of_splits = config.getUInt64(table_prefix + "number_of_splits", 3);
-
-    allow_to_copy_alias_and_materialized_columns = config.getBool(table_prefix + "allow_to_copy_alias_and_materialized_columns", false);
-    allow_to_drop_target_partitions = config.getBool(table_prefix + "allow_to_drop_target_partitions", false);
-
-    cluster_pull_name = config.getString(table_prefix + "cluster_pull");
-    cluster_push_name = config.getString(table_prefix + "cluster_push");
-
-    table_pull.first = config.getString(table_prefix + "database_pull");
-    table_pull.second = config.getString(table_prefix + "table_pull");
-
-    table_push.first = config.getString(table_prefix + "database_push");
-    table_push.second = config.getString(table_prefix + "table_push");
-
-    /// Used as node name in ZooKeeper
-    table_id = escapeForFileName(cluster_push_name)
-               + "." + escapeForFileName(table_push.first)
-               + "." + escapeForFileName(table_push.second);
-
-    engine_push_str = config.getString(table_prefix + "engine", "rand()");
-
-    {
-        ParserStorage parser_storage{ParserStorage::TABLE_ENGINE};
-        engine_push_ast = parseQuery(parser_storage, engine_push_str, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
-        engine_push_partition_key_ast = extractPartitionKey(engine_push_ast);
-        primary_key_comma_separated = boost::algorithm::join(extractPrimaryKeyColumnNames(engine_push_ast), ", ");
-        is_replicated_table = isReplicatedTableEngine(engine_push_ast);
-    }
-
-    sharding_key_str = config.getString(table_prefix + "sharding_key");
-
-    auxiliary_engine_split_asts.reserve(number_of_splits);
-    {
-        ParserExpressionWithOptionalAlias parser_expression(false);
-        sharding_key_ast = parseQuery(parser_expression, sharding_key_str, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
-        main_engine_split_ast = createASTStorageDistributed(cluster_push_name, table_push.first, table_push.second,
-                                                            sharding_key_ast);
-
-        for (const auto piece_number : collections::range(0, number_of_splits))
-        {
-            auxiliary_engine_split_asts.emplace_back
-                    (
-                            createASTStorageDistributed(cluster_push_name, table_push.first,
-                                                        table_push.second + "_piece_" + toString(piece_number), sharding_key_ast)
-                    );
-        }
-    }
-
-    where_condition_str = config.getString(table_prefix + "where_condition", "");
-    if (!where_condition_str.empty())
-    {
-        ParserExpressionWithOptionalAlias parser_expression(false);
-        where_condition_ast = parseQuery(parser_expression, where_condition_str, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
-
-        // Will use canonical expression form
-        where_condition_str = queryToString(where_condition_ast);
-    }
-
-    String enabled_partitions_prefix = table_prefix + "enabled_partitions";
-    has_enabled_partitions = config.has(enabled_partitions_prefix);
-
-    if (has_enabled_partitions)
-    {
-        Strings keys;
-        config.keys(enabled_partitions_prefix, keys);
-
-        if (keys.empty())
-        {
-            /// Parse list of partition from space-separated string
-            String partitions_str = config.getString(table_prefix + "enabled_partitions");
-            boost::trim_if(partitions_str, isWhitespaceASCII);
-            boost::split(enabled_partitions, partitions_str, isWhitespaceASCII, boost::token_compress_on);
-        }
-        else
-        {
-            /// Parse sequence of <partition>...</partition>
-            for (const String &key : keys)
-            {
-                if (!startsWith(key, "partition"))
-                    throw Exception(ErrorCodes::UNKNOWN_ELEMENT_IN_CONFIG, "Unknown key {} in {}", key, enabled_partitions_prefix);
-
-                enabled_partitions.emplace_back(config.getString(enabled_partitions_prefix + "." + key));
-            }
-        }
-
-        std::copy(enabled_partitions.begin(), enabled_partitions.end(), std::inserter(enabled_partitions_set, enabled_partitions_set.begin()));
-    }
-}
-
-
-String TaskTable::getPartitionPath(const String & partition_name) const
-{
-    return task_cluster.task_zookeeper_path             // root
-           + "/tables/" + table_id                      // tables/dst_cluster.merge.hits
-           + "/" + escapeForFileName(partition_name);   // 201701
-}
-
-String TaskTable::getPartitionAttachIsActivePath(const String & partition_name) const
-{
-    return getPartitionPath(partition_name) + "/attach_active";
-}
-
-String TaskTable::getPartitionAttachIsDonePath(const String & partition_name) const
-{
-    return getPartitionPath(partition_name) + "/attach_is_done";
-}
-
-String TaskTable::getPartitionPiecePath(const String & partition_name, size_t piece_number) const
-{
-    assert(piece_number < number_of_splits);
-    return getPartitionPath(partition_name) + "/piece_" + toString(piece_number);  // 1...number_of_splits
-}
-
-String TaskTable::getCertainPartitionIsDirtyPath(const String &partition_name) const
-{
-    return getPartitionPath(partition_name) + "/is_dirty";
-}
-
-String TaskTable::getCertainPartitionPieceIsDirtyPath(const String & partition_name, const size_t piece_number) const
-{
-    return getPartitionPiecePath(partition_name, piece_number) + "/is_dirty";
-}
-
-String TaskTable::getCertainPartitionIsCleanedPath(const String & partition_name) const
-{
-    return getCertainPartitionIsDirtyPath(partition_name) + "/cleaned";
-}
-
-String TaskTable::getCertainPartitionPieceIsCleanedPath(const String & partition_name, const size_t piece_number) const
-{
-    return getCertainPartitionPieceIsDirtyPath(partition_name, piece_number) + "/cleaned";
-}
-
-String TaskTable::getCertainPartitionTaskStatusPath(const String & partition_name) const
-{
-    return getPartitionPath(partition_name) + "/shards";
-}
-
-String TaskTable::getCertainPartitionPieceTaskStatusPath(const String & partition_name, const size_t piece_number) const
-{
-    return getPartitionPiecePath(partition_name, piece_number) + "/shards";
-}
-
-bool TaskTable::isReplicatedTable() const
-{
-    return is_replicated_table;
-}
-
-String TaskTable::getStatusAllPartitionCount() const
-{
-    return task_cluster.task_zookeeper_path + "/status/all_partitions_count";
-}
-
-String TaskTable::getStatusProcessedPartitionsCount() const
-{
-    return task_cluster.task_zookeeper_path + "/status/processed_partitions_count";
-}
-
-ASTPtr TaskTable::rewriteReplicatedCreateQueryToPlain() const
-{
-    ASTPtr prev_engine_push_ast = engine_push_ast->clone();
-
-    auto & new_storage_ast = prev_engine_push_ast->as<ASTStorage &>();
-    auto & new_engine_ast = new_storage_ast.engine->as<ASTFunction &>();
-
-    /// Remove "Replicated" from name
-    new_engine_ast.name = new_engine_ast.name.substr(10);
-
-    if (new_engine_ast.arguments)
-    {
-        auto & replicated_table_arguments = new_engine_ast.arguments->children;
-
-
-        /// In some cases of Atomic database engine usage ReplicatedMergeTree tables
-        /// could be created without arguments.
-        if (!replicated_table_arguments.empty())
-        {
-            /// Delete first two arguments of Replicated...MergeTree() table.
-            replicated_table_arguments.erase(replicated_table_arguments.begin());
-            replicated_table_arguments.erase(replicated_table_arguments.begin());
-        }
-    }
-
-    return new_storage_ast.clone();
-}
-
-ClusterPartition & TaskTable::getClusterPartition(const String & partition_name)
-{
-    auto it = cluster_partitions.find(partition_name);
-    if (it == cluster_partitions.end())
-        throw Exception(ErrorCodes::LOGICAL_ERROR, "There are no cluster partition {} in {}", partition_name, table_id);
-    return it->second;
-}
-
-}
diff --git a/programs/copier/TaskTable.h b/programs/copier/TaskTable.h
deleted file mode 100644
index 2bb7f078bc6..00000000000
--- a/programs/copier/TaskTable.h
+++ /dev/null
@@ -1,173 +0,0 @@
-#pragma once
-
-#include "Aliases.h"
-#include "TaskShard.h"
-
-
-namespace DB
-{
-
-struct ClusterPartition;
-struct TaskCluster;
-
-struct TaskTable
-{
-    TaskTable(TaskCluster & parent, const Poco::Util::AbstractConfiguration & config, const String & prefix, const String & table_key);
-
-    TaskCluster & task_cluster;
-
-    /// These functions used in checkPartitionIsDone() or checkPartitionPieceIsDone()
-    /// They are implemented here not to call task_table.tasks_shard[partition_name].second.pieces[current_piece_number] etc.
-
-    String getPartitionPath(const String & partition_name) const;
-
-    String getPartitionAttachIsActivePath(const String & partition_name) const;
-
-    String getPartitionAttachIsDonePath(const String & partition_name) const;
-
-    String getPartitionPiecePath(const String & partition_name, size_t piece_number) const;
-
-    String getCertainPartitionIsDirtyPath(const String & partition_name) const;
-
-    String getCertainPartitionPieceIsDirtyPath(const String & partition_name, size_t piece_number) const;
-
-    String getCertainPartitionIsCleanedPath(const String & partition_name) const;
-
-    String getCertainPartitionPieceIsCleanedPath(const String & partition_name, size_t piece_number) const;
-
-    String getCertainPartitionTaskStatusPath(const String & partition_name) const;
-
-    String getCertainPartitionPieceTaskStatusPath(const String & partition_name, size_t piece_number) const;
-
-    bool isReplicatedTable() const;
-
-    /// These nodes are used for check-status option
-    String getStatusAllPartitionCount() const;
-    String getStatusProcessedPartitionsCount() const;
-
-    /// Partitions will be split into number-of-splits pieces.
-    /// Each piece will be copied independently. (10 by default)
-    size_t number_of_splits;
-
-    bool allow_to_copy_alias_and_materialized_columns{false};
-    bool allow_to_drop_target_partitions{false};
-
-    String name_in_config;
-
-    /// Used as task ID
-    String table_id;
-
-    /// Column names in primary key
-    String primary_key_comma_separated;
-
-    /// Source cluster and table
-    String cluster_pull_name;
-    DatabaseAndTableName table_pull;
-
-    /// Destination cluster and table
-    String cluster_push_name;
-    DatabaseAndTableName table_push;
-
-    /// Storage of destination table
-    /// (tables that are stored on each shard of target cluster)
-    String engine_push_str;
-    ASTPtr engine_push_ast;
-    ASTPtr engine_push_partition_key_ast;
-
-    /// First argument of Replicated...MergeTree()
-    String engine_push_zk_path;
-    bool is_replicated_table;
-
-    ASTPtr rewriteReplicatedCreateQueryToPlain() const;
-
-    /*
-     * A Distributed table definition used to split data
-     * Distributed table will be created on each shard of default
-     * cluster to perform data copying and resharding
-     * */
-    String sharding_key_str;
-    ASTPtr sharding_key_ast;
-    ASTPtr main_engine_split_ast;
-
-    /*
-     * To copy partition piece form one cluster to another we have to use Distributed table.
-     * In case of usage separate table (engine_push) for each partition piece,
-     * we have to use many Distributed tables.
-     * */
-    ASTs auxiliary_engine_split_asts;
-
-    /// Additional WHERE expression to filter input data
-    String where_condition_str;
-    ASTPtr where_condition_ast;
-
-    /// Resolved clusters
-    ClusterPtr cluster_pull;
-    ClusterPtr cluster_push;
-
-    /// Filter partitions that should be copied
-    bool has_enabled_partitions = false;
-    Strings enabled_partitions;
-    NameSet enabled_partitions_set;
-
-    /**
-     * Prioritized list of shards
-     * all_shards contains information about all shards in the table.
-     * So we have to check whether particular shard have current partition or not while processing.
-     */
-    TasksShard all_shards;
-    TasksShard local_shards;
-
-    /// All partitions of the current table.
-    ClusterPartitions cluster_partitions;
-    NameSet finished_cluster_partitions;
-
-    /// Partition names to process in user-specified order
-    Strings ordered_partition_names;
-
-    ClusterPartition & getClusterPartition(const String & partition_name);
-
-    Stopwatch watch;
-    UInt64 bytes_copied = 0;
-    UInt64 rows_copied = 0;
-
-    template <typename RandomEngine>
-    void initShards(RandomEngine &&random_engine);
-};
-
-using TasksTable = std::list<TaskTable>;
-
-
-template<typename RandomEngine>
-inline void TaskTable::initShards(RandomEngine && random_engine)
-{
-    const String & fqdn_name = getFQDNOrHostName();
-    std::uniform_int_distribution<uint8_t> get_urand(0, std::numeric_limits<UInt8>::max());
-
-    // Compute the priority
-    for (const auto & shard_info : cluster_pull->getShardsInfo())
-    {
-        TaskShardPtr task_shard = std::make_shared<TaskShard>(*this, shard_info);
-        const auto & replicas = cluster_pull->getShardsAddresses().at(task_shard->indexInCluster());
-        task_shard->priority = getReplicasPriority(replicas, fqdn_name, get_urand(random_engine));
-
-        all_shards.emplace_back(task_shard);
-    }
-
-    // Sort by priority
-    std::sort(all_shards.begin(), all_shards.end(),
-              [](const TaskShardPtr & lhs, const TaskShardPtr & rhs)
-              {
-                  return ShardPriority::greaterPriority(lhs->priority, rhs->priority);
-              });
-
-    // Cut local shards
-    auto it_first_remote = std::lower_bound(all_shards.begin(), all_shards.end(), 1,
-                                            [](const TaskShardPtr & lhs, UInt8 is_remote)
-                                            {
-                                                return lhs->priority.is_remote < is_remote;
-                                            });
-
-    local_shards.assign(all_shards.begin(), it_first_remote);
-}
-
-}
diff --git a/programs/copier/ZooKeeperStaff.h b/programs/copier/ZooKeeperStaff.h
deleted file mode 100644
index c15db73f060..00000000000
--- a/programs/copier/ZooKeeperStaff.h
+++ /dev/null
@@ -1,221 +0,0 @@
-#pragma once
-
-/** Allows to compare two incremental counters of type UInt32 in presence of possible overflow.
-  * We assume that we compare values that are not too far away.
-  * For example, when we increment 0xFFFFFFFF, we get 0. So, 0xFFFFFFFF is less than 0.
-  */
-class WrappingUInt32
-{
-public:
-    UInt32 value;
-
-    explicit WrappingUInt32(UInt32 _value)
-            : value(_value)
-    {}
-
-    bool operator<(const WrappingUInt32 & other) const
-    {
-        return value != other.value && *this <= other;
-    }
-
-    bool operator<=(const WrappingUInt32 & other) const
-    {
-        const UInt32 HALF = static_cast<UInt32>(1) << 31;
-        return (value <= other.value && other.value - value < HALF)
-               || (value > other.value && value - other.value > HALF);
-    }
-
-    bool operator==(const WrappingUInt32 & other) const
-    {
-        return value == other.value;
-    }
-};
-
-/** Conforming Zxid definition.
-  * cf. https://github.com/apache/zookeeper/blob/631d1b284f0edb1c4f6b0fb221bf2428aec71aaa/zookeeper-docs/src/main/resources/markdown/zookeeperInternals.md#guarantees-properties-and-definitions
-  *
-  * But it is better to read this: https://zookeeper.apache.org/doc/r3.1.2/zookeeperProgrammers.html
-  *
-  * Actually here is the definition of Zxid.
-  * Every change to the ZooKeeper state receives a stamp in the form of a zxid (ZooKeeper Transaction Id).
-  * This exposes the total ordering of all changes to ZooKeeper. Each change will have a unique zxid
-  * and if zxid1 is smaller than zxid2 then zxid1 happened before zxid2.
-  */
-class Zxid
-{
-public:
-    WrappingUInt32 epoch;
-    WrappingUInt32 counter;
-    explicit Zxid(UInt64 _zxid)
-            : epoch(static_cast<UInt32>(_zxid >> 32))
-            , counter(static_cast<UInt32>(_zxid))
-    {}
-
-    bool operator<=(const Zxid & other) const
-    {
-        return (epoch < other.epoch)
-               || (epoch == other.epoch && counter <= other.counter);
-    }
-
-    bool operator==(const Zxid & other) const
-    {
-        return epoch == other.epoch && counter == other.counter;
-    }
-};
-
-/* When multiple ClusterCopiers discover that the target partition is not empty,
- * they will attempt to clean up this partition before proceeding to copying.
- *
- * Instead of purging is_dirty, the history of cleaning work is preserved and partition hygiene is established
- * based on a happens-before relation between the events.
- * This relation is encoded by LogicalClock based on the mzxid of the is_dirty ZNode and is_dirty/cleaned.
- * The fact of the partition hygiene is encoded by CleanStateClock.
- *
- * For you to know what mzxid means:
- *
- * ZooKeeper Stat Structure:
- * The Stat structure for each znode in ZooKeeper is made up of the following fields:
- *
- * -- czxid
- * The zxid of the change that caused this znode to be created.
- *
- * -- mzxid
- * The zxid of the change that last modified this znode.
- *
- * -- ctime
- * The time in milliseconds from epoch when this znode was created.
- *
- * -- mtime
- * The time in milliseconds from epoch when this znode was last modified.
- *
- * -- version
- * The number of changes to the data of this znode.
- *
- * -- cversion
- * The number of changes to the children of this znode.
- *
- * -- aversion
- * The number of changes to the ACL of this znode.
- *
- * -- ephemeralOwner
- * The session id of the owner of this znode if the znode is an ephemeral node.
- * If it is not an ephemeral node, it will be zero.
- *
- * -- dataLength
- * The length of the data field of this znode.
- *
- * -- numChildren
- * The number of children of this znode.
- * */
-
-class LogicalClock
-{
-public:
-    std::optional<Zxid> zxid;
-
-    LogicalClock() = default;
-
-    explicit LogicalClock(UInt64 _zxid)
-            : zxid(_zxid)
-    {}
-
-    bool hasHappened() const
-    {
-        return bool(zxid);
-    }
-
-    /// happens-before relation with a reasonable time bound
-    bool happensBefore(const LogicalClock & other) const
-    {
-        return !zxid
-               || (other.zxid && *zxid <= *other.zxid);
-    }
-
-    bool operator<=(const LogicalClock & other) const
-    {
-        return happensBefore(other);
-    }
-
-    /// strict equality check
-    bool operator==(const LogicalClock & other) const
-    {
-        return zxid == other.zxid;
-    }
-};
-
-
-class CleanStateClock
-{
-public:
-    LogicalClock discovery_zxid;
-    std::optional<UInt32> discovery_version;
-
-    LogicalClock clean_state_zxid;
-    std::optional<UInt32> clean_state_version;
-
-    std::shared_ptr<std::atomic_bool> stale;
-
-    bool is_clean() const
-    {
-        return !is_stale()
-            && (!discovery_zxid.hasHappened() || (clean_state_zxid.hasHappened() && discovery_zxid <= clean_state_zxid));
-    }
-
-    bool is_stale() const
-    {
-        return stale->load();
-    }
-
-    CleanStateClock(
-            const zkutil::ZooKeeperPtr & zookeeper,
-            const String & discovery_path,
-            const String & clean_state_path)
-            : stale(std::make_shared<std::atomic_bool>(false))
-    {
-        Coordination::Stat stat{};
-        String _some_data;
-        auto watch_callback =
-                [my_stale = stale] (const Coordination::WatchResponse & rsp)
-                {
-                    auto logger = getLogger("ClusterCopier");
-                    if (rsp.error == Coordination::Error::ZOK)
-                    {
-                        switch (rsp.type) /// NOLINT(bugprone-switch-missing-default-case)
-                        {
-                            case Coordination::CREATED:
-                                LOG_DEBUG(logger, "CleanStateClock change: CREATED, at {}", rsp.path);
-                                my_stale->store(true);
-                                break;
-                            case Coordination::CHANGED:
-                                LOG_DEBUG(logger, "CleanStateClock change: CHANGED, at {}", rsp.path);
-                                my_stale->store(true);
-                        }
-                    }
-                };
-        if (zookeeper->tryGetWatch(discovery_path, _some_data, &stat, watch_callback))
-        {
-            discovery_zxid = LogicalClock(stat.mzxid);
-            discovery_version = stat.version;
-        }
-        if (zookeeper->tryGetWatch(clean_state_path, _some_data, &stat, watch_callback))
-        {
-            clean_state_zxid = LogicalClock(stat.mzxid);
-            clean_state_version = stat.version;
-        }
-    }
-
-    bool operator==(const CleanStateClock & other) const
-    {
-        return !is_stale()
-               && !other.is_stale()
-               && discovery_zxid == other.discovery_zxid
-               && discovery_version == other.discovery_version
-               && clean_state_zxid == other.clean_state_zxid
-               && clean_state_version == other.clean_state_version;
-    }
-
-    bool operator!=(const CleanStateClock & other) const
-    {
-        return !(*this == other);
-    }
-};
diff --git a/programs/copier/clickhouse-copier.cpp b/programs/copier/clickhouse-copier.cpp
deleted file mode 100644
index 4dabb01775b..00000000000
--- a/programs/copier/clickhouse-copier.cpp
+++ /dev/null
@@ -1 +0,0 @@
-int mainEntryClickHouseClusterCopier(int argc, char ** argv);
diff --git a/programs/diagnostics/testdata/configs/xml/config.xml b/programs/diagnostics/testdata/configs/xml/config.xml
index ae09d207091..eb7c70cf498 100644
--- a/programs/diagnostics/testdata/configs/xml/config.xml
+++ b/programs/diagnostics/testdata/configs/xml/config.xml
@@ -94,7 +94,7 @@
     <http_port>8123</http_port>
 
     <!-- Port for interaction by native protocol with:
-         - clickhouse-client and other native ClickHouse tools (clickhouse-benchmark, clickhouse-copier);
+         - clickhouse-client and other native ClickHouse tools (clickhouse-benchmark);
          - clickhouse-server with other clickhouse-servers for distributed query processing;
          - ClickHouse drivers and applications supporting native protocol
          (this protocol is also informally called as "the TCP protocol");
diff --git a/programs/diagnostics/testdata/configs/yaml/config.yaml b/programs/diagnostics/testdata/configs/yaml/config.yaml
index 354065a8a9b..e577a99e675 100644
--- a/programs/diagnostics/testdata/configs/yaml/config.yaml
+++ b/programs/diagnostics/testdata/configs/yaml/config.yaml
@@ -56,7 +56,7 @@ logger:
 http_port: 8123
 
 # Port for interaction by native protocol with:
-# - clickhouse-client and other native ClickHouse tools (clickhouse-benchmark, clickhouse-copier);
+# - clickhouse-client and other native ClickHouse tools (clickhouse-benchmark);
 # - clickhouse-server with other clickhouse-servers for distributed query processing;
 # - ClickHouse drivers and applications supporting native protocol
 # (this protocol is also informally called as "the TCP protocol");
diff --git a/programs/diagnostics/testdata/configs/yandex_xml/config.xml b/programs/diagnostics/testdata/configs/yandex_xml/config.xml
index 181f52a92f8..40d1fa34b1a 100644
--- a/programs/diagnostics/testdata/configs/yandex_xml/config.xml
+++ b/programs/diagnostics/testdata/configs/yandex_xml/config.xml
@@ -94,7 +94,7 @@
     <http_port>8123</http_port>
 
     <!-- Port for interaction by native protocol with:
-         - clickhouse-client and other native ClickHouse tools (clickhouse-benchmark, clickhouse-copier);
+         - clickhouse-client and other native ClickHouse tools (clickhouse-benchmark);
          - clickhouse-server with other clickhouse-servers for distributed query processing;
          - ClickHouse drivers and applications supporting native protocol
          (this protocol is also informally called as "the TCP protocol");
diff --git a/programs/install/Install.cpp b/programs/install/Install.cpp
index c3d2c61d6d0..0ff0faff624 100644
--- a/programs/install/Install.cpp
+++ b/programs/install/Install.cpp
@@ -433,7 +433,6 @@ int mainEntryClickHouseInstall(int argc, char ** argv)
             "clickhouse-client",
             "clickhouse-local",
             "clickhouse-benchmark",
-            "clickhouse-copier",
             "clickhouse-obfuscator",
             "clickhouse-git-import",
             "clickhouse-compressor",
diff --git a/programs/main.cpp b/programs/main.cpp
index 3896b3819a8..c5f1b30f60e 100644
--- a/programs/main.cpp
+++ b/programs/main.cpp
@@ -31,7 +31,6 @@ int mainEntryClickHouseBenchmark(int argc, char ** argv);
 int mainEntryClickHouseExtractFromConfig(int argc, char ** argv);
 int mainEntryClickHouseCompressor(int argc, char ** argv);
 int mainEntryClickHouseFormat(int argc, char ** argv);
-int mainEntryClickHouseClusterCopier(int argc, char ** argv);
 int mainEntryClickHouseObfuscator(int argc, char ** argv);
 int mainEntryClickHouseGitImport(int argc, char ** argv);
 int mainEntryClickHouseStaticFilesDiskUploader(int argc, char ** argv);
@@ -80,7 +79,6 @@ std::pair<std::string_view, MainFunc> clickhouse_applications[] =
     {"extract-from-config", mainEntryClickHouseExtractFromConfig},
     {"compressor", mainEntryClickHouseCompressor},
     {"format", mainEntryClickHouseFormat},
-    {"copier", mainEntryClickHouseClusterCopier},
     {"obfuscator", mainEntryClickHouseObfuscator},
     {"git-import", mainEntryClickHouseGitImport},
     {"static-files-disk-uploader", mainEntryClickHouseStaticFilesDiskUploader},
diff --git a/programs/server/config.xml b/programs/server/config.xml
index d19f6d77f30..ea3ead47c32 100644
--- a/programs/server/config.xml
+++ b/programs/server/config.xml
@@ -135,7 +135,7 @@
     <http_port>8123</http_port>
 
     <!-- Port for interaction by native protocol with:
-         - clickhouse-client and other native ClickHouse tools (clickhouse-benchmark, clickhouse-copier);
+         - clickhouse-client and other native ClickHouse tools (clickhouse-benchmark);
          - clickhouse-server with other clickhouse-servers for distributed query processing;
          - ClickHouse drivers and applications supporting native protocol
          (this protocol is also informally called as "the TCP protocol");
diff --git a/programs/server/config.yaml.example b/programs/server/config.yaml.example
index b472b6f4a45..9fc188e97aa 100644
--- a/programs/server/config.yaml.example
+++ b/programs/server/config.yaml.example
@@ -55,7 +55,7 @@ logger:
 http_port: 8123
 
 # Port for interaction by native protocol with:
-# - clickhouse-client and other native ClickHouse tools (clickhouse-benchmark, clickhouse-copier);
+# - clickhouse-client and other native ClickHouse tools (clickhouse-benchmark);
 # - clickhouse-server with other clickhouse-servers for distributed query processing;
 # - ClickHouse drivers and applications supporting native protocol
 # (this protocol is also informally called as "the TCP protocol");
diff --git a/src/Storages/StorageDistributed.h b/src/Storages/StorageDistributed.h
index cf5b78305a5..323646ab911 100644
--- a/src/Storages/StorageDistributed.h
+++ b/src/Storages/StorageDistributed.h
@@ -154,7 +154,6 @@ public:
     /// Used by InterpreterSystemQuery
     void flushClusterNodesAllData(ContextPtr context);
 
-    /// Used by ClusterCopier
     size_t getShardCount() const;
 
     bool initializeDiskOnConfigChange(const std::set<String> & new_added_disks) override;
diff --git a/tests/integration/README.md b/tests/integration/README.md
index 1b5a0ee8994..ac01c43769e 100644
--- a/tests/integration/README.md
+++ b/tests/integration/README.md
@@ -76,7 +76,7 @@ docker pull clickhouse/integration-tests-runner
 Notes:
 * If you want to run integration tests without `sudo` you have to add your user to docker group `sudo usermod -aG docker $USER`. [More information](https://docs.docker.com/install/linux/linux-postinstall/) about docker configuration.
 * If you already had run these tests without `./runner` script you may have problems with pytest cache. It can be removed with `rm -r __pycache__ .pytest_cache/`.
-* Some tests maybe require a lot of resources (CPU, RAM, etc.). Better not try large tests like `test_cluster_copier` or `test_distributed_ddl*` on your laptop.
+* Some tests maybe require a lot of resources (CPU, RAM, etc.). Better not try large tests like `test_distributed_ddl*` on your laptop.
 
 You can run tests via `./runner` script and pass pytest arguments as last arg:
 ```
diff --git a/tests/integration/test_cluster_copier/__init__.py b/tests/integration/test_cluster_copier/__init__.py
deleted file mode 100644
index e69de29bb2d..00000000000
diff --git a/tests/integration/test_cluster_copier/configs/conf.d/clusters.xml b/tests/integration/test_cluster_copier/configs/conf.d/clusters.xml
deleted file mode 100644
index 07363e6e43a..00000000000
--- a/tests/integration/test_cluster_copier/configs/conf.d/clusters.xml
+++ /dev/null
@@ -1,73 +0,0 @@
-<clickhouse>
-    <remote_servers>
-        <cluster0>
-            <shard>
-                <internal_replication>true</internal_replication>
-                <replica>
-                    <host>s0_0_0</host>
-                    <port>9000</port>
-                </replica>
-                <replica>
-                    <host>s0_0_1</host>
-                    <port>9000</port>
-                </replica>
-            </shard>
-            <shard>
-                <internal_replication>true</internal_replication>
-                <replica>
-                    <host>s0_1_0</host>
-                    <port>9000</port>
-                </replica>
-            </shard>
-        </cluster0>
-        <cluster1>
-            <shard>
-                <internal_replication>true</internal_replication>
-                <replica>
-                    <host>s1_0_0</host>
-                    <port>9000</port>
-                </replica>
-                <replica>
-                    <host>s1_0_1</host>
-                    <port>9000</port>
-                </replica>
-            </shard>
-            <shard>
-                <internal_replication>true</internal_replication>
-                <replica>
-                    <host>s1_1_0</host>
-                    <port>9000</port>
-                </replica>
-            </shard>
-        </cluster1>
-        <shard_0_0>
-            <shard>
-                <internal_replication>true</internal_replication>
-                <replica>
-                    <host>s0_0_0</host>
-                    <port>9000</port>
-                </replica>
-                <replica>
-                    <host>s0_0_1</host>
-                    <port>9000</port>
-                </replica>
-            </shard>
-        </shard_0_0>
-        <source_trivial_cluster>
-            <shard>
-                <replica>
-                    <host>s0_0_0</host>
-                    <port>9000</port>
-                </replica>
-            </shard>
-        </source_trivial_cluster>
-        <destination_trivial_cluster>
-            <shard>
-                <replica>
-                    <host>s1_0_0</host>
-                    <port>9000</port>
-                </replica>
-            </shard>
-        </destination_trivial_cluster>
-    </remote_servers>
-</clickhouse>
diff --git a/tests/integration/test_cluster_copier/configs/conf.d/clusters_trivial.xml b/tests/integration/test_cluster_copier/configs/conf.d/clusters_trivial.xml
deleted file mode 100644
index c91ec627580..00000000000
--- a/tests/integration/test_cluster_copier/configs/conf.d/clusters_trivial.xml
+++ /dev/null
@@ -1,20 +0,0 @@
-<clickhouse>
-    <remote_servers>
-        <source_trivial_cluster>
-            <shard>
-                <replica>
-                    <host>first_trivial</host>
-                    <port>9000</port>
-                </replica>
-            </shard>
-        </source_trivial_cluster>
-        <destination_trivial_cluster>
-            <shard>
-                <replica>
-                    <host>second_trivial</host>
-                    <port>9000</port>
-                </replica>
-            </shard>
-        </destination_trivial_cluster>
-    </remote_servers>
-</clickhouse>
diff --git a/tests/integration/test_cluster_copier/configs/conf.d/ddl.xml b/tests/integration/test_cluster_copier/configs/conf.d/ddl.xml
deleted file mode 100644
index 64fa32335ab..00000000000
--- a/tests/integration/test_cluster_copier/configs/conf.d/ddl.xml
+++ /dev/null
@@ -1,5 +0,0 @@
-<clickhouse>
-    <distributed_ddl>
-        <path>/clickhouse/task_queue/ddl</path>
-    </distributed_ddl>
-</clickhouse>
\ No newline at end of file
diff --git a/tests/integration/test_cluster_copier/configs/conf.d/query_log.xml b/tests/integration/test_cluster_copier/configs/conf.d/query_log.xml
deleted file mode 100644
index 1efc65bdb28..00000000000
--- a/tests/integration/test_cluster_copier/configs/conf.d/query_log.xml
+++ /dev/null
@@ -1,14 +0,0 @@
-<clickhouse>
-    <!-- Query log. Used only for queries with setting log_queries = 1. -->
-    <query_log>
-        <!-- What table to insert data. If table is not exist, it will be created.
-             When query log structure is changed after system update,
-              then old table will be renamed and new table will be created automatically.
-        -->
-        <database>system</database>
-        <table>query_log</table>
-
-        <!-- Interval of flushing data. -->
-        <flush_interval_milliseconds>1000</flush_interval_milliseconds>
-    </query_log>
-</clickhouse>
\ No newline at end of file
diff --git a/tests/integration/test_cluster_copier/configs/config-copier.xml b/tests/integration/test_cluster_copier/configs/config-copier.xml
deleted file mode 100644
index 590b1892f8d..00000000000
--- a/tests/integration/test_cluster_copier/configs/config-copier.xml
+++ /dev/null
@@ -1,11 +0,0 @@
-<clickhouse>
-    <logger>
-        <level>information</level>
-        <log>/var/log/clickhouse-server/copier/log.log</log>
-        <errorlog>/var/log/clickhouse-server/copier/log.err.log</errorlog>
-        <size>1000M</size>
-        <count>10</count>
-        <stderr>/var/log/clickhouse-server/copier/stderr.log</stderr>
-        <stdout>/var/log/clickhouse-server/copier/stdout.log</stdout>
-    </logger>
-</clickhouse>
diff --git a/tests/integration/test_cluster_copier/configs/users.xml b/tests/integration/test_cluster_copier/configs/users.xml
deleted file mode 100644
index b463dfc81e7..00000000000
--- a/tests/integration/test_cluster_copier/configs/users.xml
+++ /dev/null
@@ -1,34 +0,0 @@
-<clickhouse>
-    <profiles>
-        <default>
-            <log_queries>1</log_queries>
-            <!-- Just to test settings_pull -->
-            <max_rows_in_distinct>5</max_rows_in_distinct>
-            <allow_deprecated_syntax_for_merge_tree>1</allow_deprecated_syntax_for_merge_tree>
-        </default>
-    </profiles>
-
-    <users>
-        <default>
-            <password></password>
-            <networks incl="networks" replace="replace">
-                <ip>::/0</ip>
-            </networks>
-            <profile>default</profile>
-            <quota>default</quota>
-        </default>
-        <dbuser>
-            <password>12345678</password>
-            <networks incl="networks" replace="replace">
-                <ip>::/0</ip>
-            </networks>
-            <profile>default</profile>
-            <quota>default</quota>
-        </dbuser>
-    </users>
-
-    <quotas>
-        <default>
-        </default>
-    </quotas>
-</clickhouse>
diff --git a/tests/integration/test_cluster_copier/configs_three_nodes/conf.d/clusters.xml b/tests/integration/test_cluster_copier/configs_three_nodes/conf.d/clusters.xml
deleted file mode 100644
index 9de7b57de27..00000000000
--- a/tests/integration/test_cluster_copier/configs_three_nodes/conf.d/clusters.xml
+++ /dev/null
@@ -1,27 +0,0 @@
-<clickhouse>
-    <remote_servers>
-        <events>
-             <shard>
-                 <internal_replication>false</internal_replication>
-                 <replica>
-                     <host>first</host>
-                     <port>9000</port>
-                 </replica>
-             </shard>
-             <shard>
-                 <internal_replication>false</internal_replication>
-                 <replica>
-                     <host>second</host>
-                     <port>9000</port>
-                 </replica>
-             </shard>
-             <shard>
-                 <internal_replication>false</internal_replication>
-                 <replica>
-                     <host>third</host>
-                     <port>9000</port>
-                 </replica>
-             </shard>
-         </events>
-    </remote_servers>
-</clickhouse>
diff --git a/tests/integration/test_cluster_copier/configs_three_nodes/conf.d/ddl.xml b/tests/integration/test_cluster_copier/configs_three_nodes/conf.d/ddl.xml
deleted file mode 100644
index 64fa32335ab..00000000000
--- a/tests/integration/test_cluster_copier/configs_three_nodes/conf.d/ddl.xml
+++ /dev/null
@@ -1,5 +0,0 @@
-<clickhouse>
-    <distributed_ddl>
-        <path>/clickhouse/task_queue/ddl</path>
-    </distributed_ddl>
-</clickhouse>
\ No newline at end of file
diff --git a/tests/integration/test_cluster_copier/configs_three_nodes/config-copier.xml b/tests/integration/test_cluster_copier/configs_three_nodes/config-copier.xml
deleted file mode 100644
index d0cab0fafb7..00000000000
--- a/tests/integration/test_cluster_copier/configs_three_nodes/config-copier.xml
+++ /dev/null
@@ -1,27 +0,0 @@
-<clickhouse>
-    <logger>
-        <level>information</level>
-        <log>/var/log/clickhouse-server/copier/log.log</log>
-        <errorlog>/var/log/clickhouse-server/copier/log.err.log</errorlog>
-        <size>1000M</size>
-        <count>10</count>
-        <stderr>/var/log/clickhouse-server/copier/stderr.log</stderr>
-        <stdout>/var/log/clickhouse-server/copier/stdout.log</stdout>
-    </logger>
-
-    <zookeeper>
-        <node index="1">
-            <host>zoo1</host>
-            <port>2181</port>
-        </node>
-        <node index="2">
-            <host>zoo2</host>
-            <port>2181</port>
-        </node>
-            <node index="3">
-            <host>zoo3</host>
-            <port>2181</port>
-        </node>
-        <session_timeout_ms>2000</session_timeout_ms>
-    </zookeeper>
-</clickhouse>
diff --git a/tests/integration/test_cluster_copier/configs_three_nodes/users.xml b/tests/integration/test_cluster_copier/configs_three_nodes/users.xml
deleted file mode 100644
index badaf46a5ca..00000000000
--- a/tests/integration/test_cluster_copier/configs_three_nodes/users.xml
+++ /dev/null
@@ -1,32 +0,0 @@
-<clickhouse>
-    <profiles>
-        <default>
-            <log_queries>1</log_queries>
-            <allow_deprecated_syntax_for_merge_tree>1</allow_deprecated_syntax_for_merge_tree>
-        </default>
-    </profiles>
-
-    <users>
-        <default>
-            <password></password>
-            <networks incl="networks" replace="replace">
-                <ip>::/0</ip>
-            </networks>
-            <profile>default</profile>
-            <quota>default</quota>
-        </default>
-        <dbuser>
-            <password>12345678</password>
-            <networks incl="networks" replace="replace">
-                <ip>::/0</ip>
-            </networks>
-            <profile>default</profile>
-            <quota>default</quota>
-        </dbuser>
-    </users>
-
-    <quotas>
-        <default>
-        </default>
-    </quotas>
-</clickhouse>
diff --git a/tests/integration/test_cluster_copier/configs_two_nodes/conf.d/clusters.xml b/tests/integration/test_cluster_copier/configs_two_nodes/conf.d/clusters.xml
deleted file mode 100644
index 38d88308631..00000000000
--- a/tests/integration/test_cluster_copier/configs_two_nodes/conf.d/clusters.xml
+++ /dev/null
@@ -1,22 +0,0 @@
-<clickhouse>
-    <remote_servers>
-        <source>
-             <shard>
-                 <internal_replication>false</internal_replication>
-                 <replica>
-                     <host>first_of_two</host>
-                     <port>9000</port>
-                 </replica>
-             </shard>
-        </source>
-        <destination>
-             <shard>
-                 <internal_replication>false</internal_replication>
-                 <replica>
-                     <host>second_of_two</host>
-                     <port>9000</port>
-                 </replica>
-             </shard>
-        </destination>
-    </remote_servers>
-</clickhouse>
diff --git a/tests/integration/test_cluster_copier/configs_two_nodes/conf.d/ddl.xml b/tests/integration/test_cluster_copier/configs_two_nodes/conf.d/ddl.xml
deleted file mode 100644
index 64fa32335ab..00000000000
--- a/tests/integration/test_cluster_copier/configs_two_nodes/conf.d/ddl.xml
+++ /dev/null
@@ -1,5 +0,0 @@
-<clickhouse>
-    <distributed_ddl>
-        <path>/clickhouse/task_queue/ddl</path>
-    </distributed_ddl>
-</clickhouse>
\ No newline at end of file
diff --git a/tests/integration/test_cluster_copier/configs_two_nodes/conf.d/storage_configuration.xml b/tests/integration/test_cluster_copier/configs_two_nodes/conf.d/storage_configuration.xml
deleted file mode 100644
index 8306f40ad6a..00000000000
--- a/tests/integration/test_cluster_copier/configs_two_nodes/conf.d/storage_configuration.xml
+++ /dev/null
@@ -1,34 +0,0 @@
-<clickhouse>
-
-<storage_configuration>
-    <disks>
-        <default>
-        </default>
-        <jbod1>
-            <path>/jbod1/</path>
-        </jbod1>
-        <jbod2>
-            <path>/jbod2/</path>
-        </jbod2>
-        <external>
-            <path>/external/</path>
-        </external>
-    </disks>
-
-    <policies>
-        <external_with_jbods>
-            <volumes>
-                <external>
-                    <disk>external</disk>
-                </external>
-                <main>
-                    <disk>jbod1</disk>
-                    <disk>jbod2</disk>
-                </main>
-            </volumes>
-        </external_with_jbods>
-    </policies>
-
-</storage_configuration>
-
-</clickhouse>
diff --git a/tests/integration/test_cluster_copier/configs_two_nodes/config-copier.xml b/tests/integration/test_cluster_copier/configs_two_nodes/config-copier.xml
deleted file mode 100644
index 55bd24816ae..00000000000
--- a/tests/integration/test_cluster_copier/configs_two_nodes/config-copier.xml
+++ /dev/null
@@ -1,19 +0,0 @@
-<clickhouse>
-    <logger>
-        <level>information</level>
-        <log>/var/log/clickhouse-server/copier/log.log</log>
-        <errorlog>/var/log/clickhouse-server/copier/log.err.log</errorlog>
-        <size>1000M</size>
-        <count>10</count>
-        <stderr>/var/log/clickhouse-server/copier/stderr.log</stderr>
-        <stdout>/var/log/clickhouse-server/copier/stdout.log</stdout>
-    </logger>
-
-    <zookeeper>
-        <node index="1">
-            <host>zoo1</host>
-            <port>2181</port>
-        </node>
-        <session_timeout_ms>2000</session_timeout_ms>
-    </zookeeper>
-</clickhouse>
diff --git a/tests/integration/test_cluster_copier/configs_two_nodes/users.xml b/tests/integration/test_cluster_copier/configs_two_nodes/users.xml
deleted file mode 100644
index badaf46a5ca..00000000000
--- a/tests/integration/test_cluster_copier/configs_two_nodes/users.xml
+++ /dev/null
@@ -1,32 +0,0 @@
-<clickhouse>
-    <profiles>
-        <default>
-            <log_queries>1</log_queries>
-            <allow_deprecated_syntax_for_merge_tree>1</allow_deprecated_syntax_for_merge_tree>
-        </default>
-    </profiles>
-
-    <users>
-        <default>
-            <password></password>
-            <networks incl="networks" replace="replace">
-                <ip>::/0</ip>
-            </networks>
-            <profile>default</profile>
-            <quota>default</quota>
-        </default>
-        <dbuser>
-            <password>12345678</password>
-            <networks incl="networks" replace="replace">
-                <ip>::/0</ip>
-            </networks>
-            <profile>default</profile>
-            <quota>default</quota>
-        </dbuser>
-    </users>
-
-    <quotas>
-        <default>
-        </default>
-    </quotas>
-</clickhouse>
diff --git a/tests/integration/test_cluster_copier/task0_description.xml b/tests/integration/test_cluster_copier/task0_description.xml
deleted file mode 100644
index 8d74d0bdde0..00000000000
--- a/tests/integration/test_cluster_copier/task0_description.xml
+++ /dev/null
@@ -1,95 +0,0 @@
-<clickhouse>
-    <!-- How many simualteneous workers are posssible -->
-    <max_workers>3</max_workers>
-
-    <!-- Common setting for pull and push operations -->
-    <settings>
-        <connect_timeout>1</connect_timeout>
-    </settings>
-
-    <!-- Setting used to fetch data -->
-    <settings_pull>
-        <max_rows_in_distinct>0</max_rows_in_distinct>
-    </settings_pull>
-
-    <!-- Setting used to insert data -->
-    <settings_push>
-    </settings_push>
-
-    <!-- Tasks -->
-    <tables>
-        <hits>
-            <cluster_pull>cluster0</cluster_pull>
-            <database_pull>default</database_pull>
-            <table_pull>hits</table_pull>
-
-            <cluster_push>cluster1</cluster_push>
-            <database_push>default</database_push>
-            <table_push>hits</table_push>
-
-            <number_of_splits>2</number_of_splits>
-
-            <enabled_partitions>3 4 5 6 1 2 0   </enabled_partitions>
-
-            <!-- Engine of destination tables -->
-            <engine>ENGINE=ReplicatedMergeTree PARTITION BY d % 3 ORDER BY (d, sipHash64(d)) SAMPLE BY sipHash64(d) SETTINGS index_granularity = 16</engine>
-
-            <!-- Which sarding key to use while copying -->
-            <sharding_key>d + 1</sharding_key>
-
-            <!-- Optional expression that filter copying data -->
-            <where_condition>d - d = 0</where_condition>
-        </hits>
-    </tables>
-
-    <!-- Configuration of clusters -->
-    <remote_servers>
-    <cluster0>
-        <shard>
-            <internal_replication>true</internal_replication>
-            <replica>
-                <host>s0_0_0</host>
-                <port>9000</port>
-            </replica>
-            <replica>
-                <host>s0_0_1</host>
-                <port>9000</port>
-            </replica>
-        </shard>
-        <shard>
-            <internal_replication>true</internal_replication>
-            <replica>
-                <host>s0_1_0</host>
-                <port>9000</port>
-            </replica>
-        </shard>
-    </cluster0>
-
-    <cluster1>
-        <shard>
-            <internal_replication>true</internal_replication>
-            <replica>
-                <host>s1_0_0</host>
-                <port>9000</port>
-            </replica>
-            <replica>
-                <host>s1_0_1</host>
-                <port>9000</port>
-            </replica>
-        </shard>
-        <shard>
-            <internal_replication>true</internal_replication>
-            <replica>
-                <host>s1_1_0</host>
-                <port>9000</port>
-            </replica>
-            <replica>
-                <!-- Died replica -->
-                <host>255.255.255.255</host>
-                <port>9000</port>
-            </replica>
-        </shard>
-    </cluster1>
-    </remote_servers>
-
-</clickhouse>
diff --git a/tests/integration/test_cluster_copier/task_drop_target_partition.xml b/tests/integration/test_cluster_copier/task_drop_target_partition.xml
deleted file mode 100644
index dc8e6452243..00000000000
--- a/tests/integration/test_cluster_copier/task_drop_target_partition.xml
+++ /dev/null
@@ -1,41 +0,0 @@
-<clickhouse>
-<remote_servers>
-        <source>
-             <shard>
-                 <internal_replication>false</internal_replication>
-                 <replica>
-                     <host>first_of_two</host>
-                     <port>9000</port>
-                 </replica>
-             </shard>
-        </source>
-        <destination>
-             <shard>
-                 <internal_replication>false</internal_replication>
-                 <replica>
-                     <host>second_of_two</host>
-                     <port>9000</port>
-                 </replica>
-             </shard>
-        </destination>
-    </remote_servers>
-
-   <max_workers>2</max_workers>
-
-   <tables>
-     <table_events>
-         <cluster_pull>source</cluster_pull>
-         <database_pull>db_drop_target_partition</database_pull>
-         <table_pull>source</table_pull>
-
-         <cluster_push>destination</cluster_push>
-         <database_push>db_drop_target_partition</database_push>
-         <table_push>destination</table_push>
-
-         <allow_to_drop_target_partitions>true</allow_to_drop_target_partitions>
-
-         <engine>ENGINE = MergeTree() PARTITION BY toYYYYMMDD(Column3) ORDER BY (Column3, Column2, Column1)</engine>
-         <sharding_key>rand()</sharding_key>
-     </table_events>
-   </tables>
- </clickhouse>
diff --git a/tests/integration/test_cluster_copier/task_month_to_week_description.xml b/tests/integration/test_cluster_copier/task_month_to_week_description.xml
deleted file mode 100644
index bc290ca397f..00000000000
--- a/tests/integration/test_cluster_copier/task_month_to_week_description.xml
+++ /dev/null
@@ -1,99 +0,0 @@
-<clickhouse>
-    <!-- How many simualteneous workers are posssible -->
-    <max_workers>4</max_workers>
-
-    <!-- Common setting for pull and push operations -->
-    <settings>
-        <connect_timeout>1</connect_timeout>
-        <max_block_size>2</max_block_size>
-    </settings>
-
-    <settings_pull>
-        <max_rows_in_distinct>0</max_rows_in_distinct>
-    </settings_pull>
-
-    <!-- Tasks -->
-    <tables>
-        <AB>
-            <cluster_pull>cluster0</cluster_pull>
-            <database_pull>default</database_pull>
-            <table_pull>a</table_pull>
-
-            <cluster_push>cluster1</cluster_push>
-            <database_push>default</database_push>
-            <table_push>b</table_push>
-
-            <!--
-            <enabled_partitions>
-                <partition></partition>
-            </enabled_partitions>
-            -->
-
-            <number_of_splits>2</number_of_splits>
-
-            <!-- Engine of destination tables -->
-            <engine>ENGINE=
-                ReplicatedMergeTree
-                PARTITION BY toMonday(date)
-                ORDER BY d
-            </engine>
-
-            <!-- Which sarding key to use while copying -->
-            <sharding_key>jumpConsistentHash(intHash64(d), 2)</sharding_key>
-
-            <!-- Optional expression that filter copying data -->
-            <!-- <where_condition></where_condition> -->
-        </AB>
-    </tables>
-
-    <!-- Configuration of clusters -->
-    <remote_servers>
-    <cluster0>
-        <shard>
-            <internal_replication>true</internal_replication>
-            <replica>
-                <host>s0_0_0</host>
-                <port>9000</port>
-            </replica>
-            <replica>
-                <host>s0_0_1</host>
-                <port>9000</port>
-            </replica>
-        </shard>
-        <shard>
-            <internal_replication>true</internal_replication>
-            <replica>
-                <host>s0_1_0</host>
-                <port>9000</port>
-            </replica>
-        </shard>
-    </cluster0>
-
-    <cluster1>
-        <shard>
-            <internal_replication>true</internal_replication>
-            <replica>
-                <host>s1_0_0</host>
-                <port>9000</port>
-            </replica>
-            <replica>
-                <host>s1_0_1</host>
-                <port>9000</port>
-            </replica>
-        </shard>
-        <shard>
-            <internal_replication>true</internal_replication>
-            <replica>
-                <host>s1_1_0</host>
-                <port>9000</port>
-            </replica>
-            <replica>
-                <!-- Died replica -->
-                <host>255.255.255.255</host>
-                <port>9000</port>
-            </replica>
-        </shard>
-    </cluster1>
-    </remote_servers>
-
-</clickhouse>
diff --git a/tests/integration/test_cluster_copier/task_no_arg.xml b/tests/integration/test_cluster_copier/task_no_arg.xml
deleted file mode 100644
index 262ff073537..00000000000
--- a/tests/integration/test_cluster_copier/task_no_arg.xml
+++ /dev/null
@@ -1,39 +0,0 @@
-<clickhouse>
-    <remote_servers>   
-       <source_cluster>
-          <shard>
-              <weight>1</weight>
-              <replica>
-                  <host>s0_0_0</host>
-                  <port>9000</port>
-              </replica>
-          </shard>
-      </source_cluster>     
-      <default_cluster>
-      
-        <shard>
-          <weight>1</weight>
-          <replica>
-              <host>s1_1_0</host>
-              <port>9000</port>
-          </replica>
-        </shard>
-            
-      </default_cluster>     
-    </remote_servers>
-    <max_workers>1</max_workers>
-
-    <tables>
-      <table_crm_fin_account>
-        <cluster_pull>source_cluster</cluster_pull>
-        <database_pull>default</database_pull>
-        <table_pull>copier_test1</table_pull>
-
-        <cluster_push>default_cluster</cluster_push>
-        <database_push>default</database_push>
-        <table_push>copier_test1_1</table_push>
-        <engine>ENGINE = MergeTree PARTITION BY date ORDER BY (date, sipHash64(date)) SAMPLE BY sipHash64(date)</engine>
-        <sharding_key>rand()</sharding_key>
-      </table_crm_fin_account>
-    </tables>
-</clickhouse>
diff --git a/tests/integration/test_cluster_copier/task_no_index.xml b/tests/integration/test_cluster_copier/task_no_index.xml
deleted file mode 100644
index 265f99e21a6..00000000000
--- a/tests/integration/test_cluster_copier/task_no_index.xml
+++ /dev/null
@@ -1,109 +0,0 @@
-<clickhouse>
-    <remote_servers>
-        <source_cluster>
-            <shard>
-                <internal_replication>false</internal_replication>
-                    <replica>
-                        <host>s0_0_0</host>
-                        <port>9000</port>
-                    </replica>
-            </shard>
-        </source_cluster>
-
-        <destination_cluster>
-               <shard>
-                <internal_replication>false</internal_replication>
-                    <replica>
-                        <host>s1_1_0</host>
-                        <port>9000</port>
-                    </replica>
-               </shard>
-        </destination_cluster>
-    </remote_servers>
-
-    <!-- How many simultaneously active workers are possible. If you run more workers superfluous workers will sleep. -->
-    <max_workers>2</max_workers>
-
-    <!-- Setting used to fetch (pull) data from source cluster tables -->
-    <settings_pull>
-        <readonly>1</readonly>
-    </settings_pull>
-
-    <!-- Setting used to insert (push) data to destination cluster tables -->
-    <settings_push>
-        <readonly>0</readonly>
-    </settings_push>
-
-    <!-- Common setting for fetch (pull) and insert (push) operations. Also, copier process context uses it.
-         They are overlaid by <settings_pull/> and <settings_push/> respectively. -->
-    <settings>
-        <connect_timeout>3</connect_timeout>
-        <!-- Sync insert is set forcibly, leave it here just in case. -->
-        <distributed_foreground_insert>1</distributed_foreground_insert>
-    </settings>
-
-    <!-- Copying tasks description.
-         You could specify several table task in the same task description (in the same ZooKeeper node), they will be performed
-         sequentially.
-    -->
-    <tables>
-        <!-- A table task, copies one table. -->
-        <table_hits>
-            <!-- Source cluster name (from <remote_servers/> section) and tables in it that should be copied -->
-            <cluster_pull>source_cluster</cluster_pull>
-            <database_pull>default</database_pull>
-            <table_pull>ontime</table_pull>
-        <!-- <table_pull>onetime</table_pull> -->
-
-            <!-- Destination cluster name and tables in which the data should be inserted -->
-            <cluster_push>destination_cluster</cluster_push>
-            <database_push>default</database_push>
-            <table_push>ontime22</table_push>
-        <!-- <table_pull>onetime</table_pull> -->
-
-            <!-- Engine of destination tables.
-                 If destination tables have not be created, workers create them using columns definition from source tables and engine
-                 definition from here.
-
-                 NOTE: If the first worker starts insert data and detects that destination partition is not empty then the partition will
-                 be dropped and refilled, take it into account if you already have some data in destination tables. You could directly
-                 specify partitions that should be copied in <enabled_partitions/>, they should be in quoted format like partition column of
-                 system.parts table.
-        -->
-
-
-            <engine>
-    ENGINE = MergeTree() PARTITION BY Year ORDER BY (Year, FlightDate) SETTINGS index_granularity=8192
-        </engine>
-
-            <!-- Sharding key used to insert data to destination cluster -->
-
-        <sharding_key>jumpConsistentHash(intHash64(Year), 2)</sharding_key>
-
-            <!-- Optional expression that filter data while pull them from source servers -->
-        <!--  <where_condition>CounterID != 0</where_condition> -->
-
-            <!-- This section specifies partitions that should be copied, other partition will be ignored.
-                 Partition names should have the same format as
-                 partition column of system.parts table (i.e. a quoted text).
-                 Since partition key of source and destination cluster could be different,
-                 these partition names specify destination partitions.
-
-                 NOTE: In spite of this section is optional (if it is not specified, all partitions will be copied),
-                 it is strictly recommended to specify them explicitly.
-                 If you already have some ready partitions on destination cluster they
-                 will be removed at the start of the copying since they will be interpeted
-                 as unfinished data from the previous copying!!!
-        -->
-            <enabled_partitions>
-                <partition>2017</partition>
-            </enabled_partitions>
-        </table_hits>
-
-        <!-- Next table to copy. It is not copied until previous table is copying. -->
-    <!-- </table_visits>
-
-        </table_visits>
-    -->
-    </tables>
-</clickhouse>
diff --git a/tests/integration/test_cluster_copier/task_non_partitioned_table.xml b/tests/integration/test_cluster_copier/task_non_partitioned_table.xml
deleted file mode 100644
index d5424b95f45..00000000000
--- a/tests/integration/test_cluster_copier/task_non_partitioned_table.xml
+++ /dev/null
@@ -1,39 +0,0 @@
-<clickhouse>
-    <remote_servers>   
-       <source_cluster>
-          <shard>
-              <weight>1</weight>
-              <replica>
-                  <host>s0_0_0</host>
-                  <port>9000</port>
-              </replica>
-          </shard>
-      </source_cluster>     
-      <default_cluster>
-      
-        <shard>
-          <weight>1</weight>
-          <replica>
-              <host>s1_1_0</host>
-              <port>9000</port>
-          </replica>
-        </shard>
-            
-      </default_cluster>     
-    </remote_servers>
-    <max_workers>1</max_workers>
-
-    <tables>
-      <table_copier_test1>
-        <cluster_pull>source_cluster</cluster_pull>
-        <database_pull>default</database_pull>
-        <table_pull>copier_test1</table_pull>
-
-        <cluster_push>default_cluster</cluster_push>
-        <database_push>default</database_push>
-        <table_push>copier_test1_1</table_push>
-        <engine>ENGINE = MergeTree ORDER BY date SETTINGS index_granularity = 8192</engine>
-        <sharding_key>rand()</sharding_key>
-      </table_copier_test1>
-    </tables>
-</clickhouse>
diff --git a/tests/integration/test_cluster_copier/task_self_copy.xml b/tests/integration/test_cluster_copier/task_self_copy.xml
deleted file mode 100644
index 21d577bc397..00000000000
--- a/tests/integration/test_cluster_copier/task_self_copy.xml
+++ /dev/null
@@ -1,63 +0,0 @@
-<clickhouse>
-    <tcp_port_secure>9440</tcp_port_secure>
-    <remote_servers>
-        <source_cluster>
-            <shard>
-                <internal_replication>false</internal_replication>
-                <replica>
-                    <host>s0_0_0</host>
-                    <port>9000</port>
-                    <user>dbuser</user>
-                    <password>12345678</password>
-                    <secure>0</secure>
-                </replica>
-            </shard>
-        </source_cluster>
-
-        <destination_cluster>
-            <shard>
-                <internal_replication>false</internal_replication>
-                <replica>
-                    <host>s0_0_0</host>
-                    <port>9000</port>
-                    <user>dbuser</user>
-                    <password>12345678</password>
-                    <secure>0</secure>
-                </replica>
-            </shard>
-        </destination_cluster>
-    </remote_servers>
-
-    <max_workers>2</max_workers>
-
-    <settings_pull>
-        <readonly>1</readonly>
-    </settings_pull>
-
-    <settings_push>
-        <readonly>0</readonly>
-    </settings_push>
-
-    <settings>
-        <connect_timeout>3</connect_timeout>
-        <distributed_foreground_insert>1</distributed_foreground_insert>
-    </settings>
-
-    <tables>
-        <table1>
-            <cluster_pull>source_cluster</cluster_pull>
-    	    <database_pull>db1</database_pull>
-            <table_pull>source_table</table_pull>
-
-            <cluster_push>destination_cluster</cluster_push>
-            <database_push>db2</database_push>
-            <table_push>destination_table</table_push>
-
-            <engine>
-	            ENGINE = MergeTree PARTITION BY a ORDER BY a SETTINGS index_granularity = 8192
-	        </engine>
-
-    	    <sharding_key>rand()</sharding_key>
-        </table1>
-    </tables>
-</clickhouse>
\ No newline at end of file
diff --git a/tests/integration/test_cluster_copier/task_skip_index.xml b/tests/integration/test_cluster_copier/task_skip_index.xml
deleted file mode 100644
index b04cec963d4..00000000000
--- a/tests/integration/test_cluster_copier/task_skip_index.xml
+++ /dev/null
@@ -1,39 +0,0 @@
-<clickhouse>
-<remote_servers>
-        <source>
-             <shard>
-                 <internal_replication>false</internal_replication>
-                 <replica>
-                     <host>first_of_two</host>
-                     <port>9000</port>
-                 </replica>
-             </shard>
-        </source>
-        <destination>
-             <shard>
-                 <internal_replication>false</internal_replication>
-                 <replica>
-                     <host>second_of_two</host>
-                     <port>9000</port>
-                 </replica>
-             </shard>
-        </destination>
-    </remote_servers>
-
-   <max_workers>2</max_workers>
-
-   <tables>
-     <table_events>
-         <cluster_pull>source</cluster_pull>
-         <database_pull>db_skip_index</database_pull>
-         <table_pull>source</table_pull>
-
-         <cluster_push>destination</cluster_push>
-         <database_push>db_skip_index</database_push>
-         <table_push>destination</table_push>
-
-         <engine>ENGINE = MergeTree() PARTITION BY toYYYYMMDD(Column3) ORDER BY (Column3, Column2, Column1)</engine>
-         <sharding_key>rand()</sharding_key>
-     </table_events>
-   </tables>
- </clickhouse>
diff --git a/tests/integration/test_cluster_copier/task_taxi_data.xml b/tests/integration/test_cluster_copier/task_taxi_data.xml
deleted file mode 100644
index 94fa5087338..00000000000
--- a/tests/integration/test_cluster_copier/task_taxi_data.xml
+++ /dev/null
@@ -1,42 +0,0 @@
-<clickhouse>
-   <remote_servers>
-         <events>
-             <shard>
-                 <internal_replication>false</internal_replication>
-                 <replica>
-                     <host>first</host>
-                     <port>9000</port>
-                 </replica>
-             </shard>
-             <shard>
-                 <internal_replication>false</internal_replication>
-                 <replica>
-                     <host>second</host>
-                     <port>9000</port>
-                 </replica>
-             </shard>
-             <shard>
-                 <internal_replication>false</internal_replication>
-                 <replica>
-                     <host>third</host>
-                     <port>9000</port>
-                 </replica>
-             </shard>
-         </events>
-   </remote_servers>
-
-   <max_workers>2</max_workers>
-
-   <tables>
-     <table_events>
-         <cluster_pull>events</cluster_pull>
-         <database_pull>dailyhistory</database_pull>
-         <table_pull>yellow_tripdata_staging</table_pull>
-         <cluster_push>events</cluster_push>
-         <database_push>monthlyhistory</database_push>
-         <table_push>yellow_tripdata_staging</table_push>
-         <engine>Engine=ReplacingMergeTree() PRIMARY KEY (tpep_pickup_datetime, id) ORDER BY (tpep_pickup_datetime, id) PARTITION BY (pickup_location_id, toYYYYMM(tpep_pickup_datetime))</engine>
-         <sharding_key>sipHash64(id) % 3</sharding_key>
-     </table_events>
-   </tables>
- </clickhouse>
\ No newline at end of file
diff --git a/tests/integration/test_cluster_copier/task_test_block_size.xml b/tests/integration/test_cluster_copier/task_test_block_size.xml
deleted file mode 100644
index bf29c7e1832..00000000000
--- a/tests/integration/test_cluster_copier/task_test_block_size.xml
+++ /dev/null
@@ -1,101 +0,0 @@
-<clickhouse>
-    <!-- How many simualteneous workers are posssible -->
-    <max_workers>1</max_workers>
-
-    <!-- Common setting for pull and push operations -->
-    <settings>
-        <connect_timeout>1</connect_timeout>
-    </settings>
-
-    <settings_pull>
-    </settings_pull>
-
-    <!-- Tasks -->
-    <tables>
-        <test_block_size>
-            <cluster_pull>shard_0_0</cluster_pull>
-            <database_pull>default</database_pull>
-            <table_pull>test_block_size</table_pull>
-
-            <cluster_push>cluster1</cluster_push>
-            <database_push>default</database_push>
-            <table_push>test_block_size</table_push>
-
-            <enabled_partitions>
-                <partition>'1970-01-01'</partition>
-            </enabled_partitions>
-
-            <!-- Engine of destination tables -->
-            <engine>ENGINE=
-                ReplicatedMergeTree
-                ORDER BY d PARTITION BY partition
-            </engine>
-
-            <!-- Which sarding key to use while copying -->
-            <sharding_key>jumpConsistentHash(intHash64(d), 2)</sharding_key>
-
-            <!-- Optional expression that filter copying data -->
-            <!-- <where_condition></where_condition> -->
-        </test_block_size>
-    </tables>
-
-    <!-- Configuration of clusters -->
-    <remote_servers>
-    <cluster0>
-        <shard>
-            <internal_replication>true</internal_replication>
-            <replica>
-                <host>s0_0_0</host>
-                <port>9000</port>
-            </replica>
-            <replica>
-                <host>s0_0_1</host>
-                <port>9000</port>
-            </replica>
-        </shard>
-        <shard>
-            <internal_replication>true</internal_replication>
-            <replica>
-                <host>s0_1_0</host>
-                <port>9000</port>
-            </replica>
-        </shard>
-    </cluster0>
-
-    <cluster1>
-        <shard>
-            <internal_replication>true</internal_replication>
-            <replica>
-                <host>s1_0_0</host>
-                <port>9000</port>
-            </replica>
-            <replica>
-                <host>s1_0_1</host>
-                <port>9000</port>
-            </replica>
-        </shard>
-        <shard>
-            <internal_replication>true</internal_replication>
-            <replica>
-                <host>s1_1_0</host>
-                <port>9000</port>
-            </replica>
-        </shard>
-    </cluster1>
-
-    <shard_0_0>
-        <shard>
-            <internal_replication>true</internal_replication>
-            <replica>
-                <host>s0_0_0</host>
-                <port>9000</port>
-            </replica>
-            <replica>
-                <host>s0_0_1</host>
-                <port>9000</port>
-            </replica>
-        </shard>
-    </shard_0_0>
-    </remote_servers>
-
-</clickhouse>
diff --git a/tests/integration/test_cluster_copier/task_trivial.xml b/tests/integration/test_cluster_copier/task_trivial.xml
deleted file mode 100644
index a3b8bc03888..00000000000
--- a/tests/integration/test_cluster_copier/task_trivial.xml
+++ /dev/null
@@ -1,63 +0,0 @@
-<clickhouse>
-    <!-- How many simualteneous workers are posssible -->
-    <max_workers>3</max_workers>
-
-    <!-- Common setting for pull and push operations -->
-    <settings>
-        <connect_timeout>1</connect_timeout>
-    </settings>
-
-    <!-- Setting used to fetch data -->
-    <settings_pull>
-        <max_rows_in_distinct>0</max_rows_in_distinct>
-    </settings_pull>
-
-    <!-- Setting used to insert data -->
-    <settings_push>
-    </settings_push>
-
-    <!-- Tasks -->
-    <tables>
-        <hits>
-            <cluster_pull>source_trivial_cluster</cluster_pull>
-            <database_pull>default</database_pull>
-            <table_pull>trivial</table_pull>
-
-            <cluster_push>destination_trivial_cluster</cluster_push>
-            <database_push>default</database_push>
-            <table_push>trivial</table_push>
-
-            <!-- Engine of destination tables -->
-            <engine>ENGINE=ReplicatedMergeTree('/clickhouse/tables/cluster{cluster}/{shard}/hits', '{replica}') PARTITION BY d % 5 ORDER BY (d, sipHash64(d)) SAMPLE BY sipHash64(d) SETTINGS index_granularity = 16</engine>
-
-            <!-- Which sarding key to use while copying -->
-            <sharding_key>d + 1</sharding_key>
-
-            <!-- Optional expression that filter copying data -->
-            <where_condition>d - d = 0</where_condition>
-        </hits>
-    </tables>
-
-    <!-- Configuration of clusters -->
-    <remote_servers>
-        <source_trivial_cluster>
-            <shard>
-                <replica>
-                    <host>first_trivial</host>
-                    <port>9000</port>
-                </replica>
-            </shard>
-        </source_trivial_cluster>
-
-
-        <destination_trivial_cluster>
-            <shard>
-                <replica>
-                    <host>second_trivial</host>
-                    <port>9000</port>
-                </replica>
-            </shard>
-        </destination_trivial_cluster>
-    </remote_servers>
-
-</clickhouse>
diff --git a/tests/integration/test_cluster_copier/task_trivial_without_arguments.xml b/tests/integration/test_cluster_copier/task_trivial_without_arguments.xml
deleted file mode 100644
index 0197dee0181..00000000000
--- a/tests/integration/test_cluster_copier/task_trivial_without_arguments.xml
+++ /dev/null
@@ -1,63 +0,0 @@
-<clickhouse>
-    <!-- How many simualteneous workers are posssible -->
-    <max_workers>3</max_workers>
-
-    <!-- Common setting for pull and push operations -->
-    <settings>
-        <connect_timeout>1</connect_timeout>
-    </settings>
-
-    <!-- Setting used to fetch data -->
-    <settings_pull>
-        <max_rows_in_distinct>0</max_rows_in_distinct>
-    </settings_pull>
-
-    <!-- Setting used to insert data -->
-    <settings_push>
-    </settings_push>
-
-    <!-- Tasks -->
-    <tables>
-        <hits>
-            <cluster_pull>source_trivial_cluster</cluster_pull>
-            <database_pull>default</database_pull>
-            <table_pull>trivial_without_arguments</table_pull>
-
-            <cluster_push>destination_trivial_cluster</cluster_push>
-            <database_push>default</database_push>
-            <table_push>trivial_without_arguments</table_push>
-
-            <!-- Engine of destination tables -->
-            <engine>ENGINE=ReplicatedMergeTree() PARTITION BY d % 5 ORDER BY (d, sipHash64(d)) SAMPLE BY sipHash64(d) SETTINGS index_granularity = 16</engine>
-
-            <!-- Which sarding key to use while copying -->
-            <sharding_key>d + 1</sharding_key>
-
-            <!-- Optional expression that filter copying data -->
-            <where_condition>d - d = 0</where_condition>
-        </hits>
-    </tables>
-
-    <!-- Configuration of clusters -->
-    <remote_servers>
-        <source_trivial_cluster>
-            <shard>
-                <replica>
-                    <host>first_trivial</host>
-                    <port>9000</port>
-                </replica>
-            </shard>
-        </source_trivial_cluster>
-
-
-        <destination_trivial_cluster>
-            <shard>
-                <replica>
-                    <host>second_trivial</host>
-                    <port>9000</port>
-                </replica>
-            </shard>
-        </destination_trivial_cluster>
-    </remote_servers>
-
-</clickhouse>
diff --git a/tests/integration/test_cluster_copier/task_ttl_columns.xml b/tests/integration/test_cluster_copier/task_ttl_columns.xml
deleted file mode 100644
index 2069c509c87..00000000000
--- a/tests/integration/test_cluster_copier/task_ttl_columns.xml
+++ /dev/null
@@ -1,39 +0,0 @@
-<clickhouse>
-<remote_servers>
-        <source>
-             <shard>
-                 <internal_replication>false</internal_replication>
-                 <replica>
-                     <host>first_of_two</host>
-                     <port>9000</port>
-                 </replica>
-             </shard>
-        </source>
-        <destination>
-             <shard>
-                 <internal_replication>false</internal_replication>
-                 <replica>
-                     <host>second_of_two</host>
-                     <port>9000</port>
-                 </replica>
-             </shard>
-        </destination>
-    </remote_servers>
-
-   <max_workers>2</max_workers>
-
-   <tables>
-     <table_events>
-         <cluster_pull>source</cluster_pull>
-         <database_pull>db_ttl_columns</database_pull>
-         <table_pull>source</table_pull>
-
-         <cluster_push>destination</cluster_push>
-         <database_push>db_ttl_columns</database_push>
-         <table_push>destination</table_push>
-
-         <engine>ENGINE = MergeTree() PARTITION BY toYYYYMMDD(Column3) ORDER BY (Column3, Column2, Column1)</engine>
-         <sharding_key>rand()</sharding_key>
-     </table_events>
-   </tables>
- </clickhouse>
diff --git a/tests/integration/test_cluster_copier/task_ttl_move_to_volume.xml b/tests/integration/test_cluster_copier/task_ttl_move_to_volume.xml
deleted file mode 100644
index 2a51fa7a66d..00000000000
--- a/tests/integration/test_cluster_copier/task_ttl_move_to_volume.xml
+++ /dev/null
@@ -1,39 +0,0 @@
-<clickhouse>
-<remote_servers>
-        <source>
-             <shard>
-                 <internal_replication>false</internal_replication>
-                 <replica>
-                     <host>first_of_two</host>
-                     <port>9000</port>
-                 </replica>
-             </shard>
-        </source>
-        <destination>
-             <shard>
-                 <internal_replication>false</internal_replication>
-                 <replica>
-                     <host>second_of_two</host>
-                     <port>9000</port>
-                 </replica>
-             </shard>
-        </destination>
-    </remote_servers>
-
-   <max_workers>2</max_workers>
-
-   <tables>
-     <table_events>
-         <cluster_pull>source</cluster_pull>
-         <database_pull>db_move_to_volume</database_pull>
-         <table_pull>source</table_pull>
-
-         <cluster_push>destination</cluster_push>
-         <database_push>db_move_to_volume</database_push>
-         <table_push>destination</table_push>
-
-         <engine>ENGINE = MergeTree() PARTITION BY toYYYYMMDD(Column3) ORDER BY (Column3, Column2, Column1) TTL Column3 + INTERVAL 1 MONTH TO VOLUME 'external' SETTINGS storage_policy = 'external_with_jbods'</engine>
-         <sharding_key>rand()</sharding_key>
-     </table_events>
-   </tables>
- </clickhouse>
diff --git a/tests/integration/test_cluster_copier/task_with_different_schema.xml b/tests/integration/test_cluster_copier/task_with_different_schema.xml
deleted file mode 100644
index e1e6ee4dc42..00000000000
--- a/tests/integration/test_cluster_copier/task_with_different_schema.xml
+++ /dev/null
@@ -1,39 +0,0 @@
-<clickhouse>
-<remote_servers>
-        <source>
-             <shard>
-                 <internal_replication>false</internal_replication>
-                 <replica>
-                     <host>first_of_two</host>
-                     <port>9000</port>
-                 </replica>
-             </shard>
-        </source>
-        <destination>
-             <shard>
-                 <internal_replication>false</internal_replication>
-                 <replica>
-                     <host>second_of_two</host>
-                     <port>9000</port>
-                 </replica>
-             </shard>
-        </destination>
-    </remote_servers>
-
-   <max_workers>2</max_workers>
-
-   <tables>
-     <table_events>
-         <cluster_pull>source</cluster_pull>
-         <database_pull>db_different_schema</database_pull>
-         <table_pull>source</table_pull>
-
-         <cluster_push>destination</cluster_push>
-         <database_push>db_different_schema</database_push>
-         <table_push>destination</table_push>
-
-         <engine>ENGINE = MergeTree() PARTITION BY toYYYYMMDD(Column3) ORDER BY (Column9, Column1, Column2, Column3, Column4)</engine>
-         <sharding_key>rand()</sharding_key>
-     </table_events>
-   </tables>
- </clickhouse>
diff --git a/tests/integration/test_cluster_copier/test.py b/tests/integration/test_cluster_copier/test.py
deleted file mode 100644
index be71fc21e33..00000000000
--- a/tests/integration/test_cluster_copier/test.py
+++ /dev/null
@@ -1,653 +0,0 @@
-import os
-import random
-import sys
-import time
-import kazoo
-import pytest
-import string
-import random
-from contextlib import contextmanager
-from helpers.cluster import ClickHouseCluster
-from helpers.test_tools import TSV
-
-import docker
-
-CURRENT_TEST_DIR = os.path.dirname(os.path.abspath(__file__))
-sys.path.insert(0, os.path.dirname(CURRENT_TEST_DIR))
-
-COPYING_FAIL_PROBABILITY = 0.2
-MOVING_FAIL_PROBABILITY = 0.2
-
-cluster = ClickHouseCluster(__file__)
-
-
-def generateRandomString(count):
-    return "".join(
-        random.choice(string.ascii_uppercase + string.digits) for _ in range(count)
-    )
-
-
-def check_all_hosts_sucesfully_executed(tsv_content, num_hosts):
-    M = TSV.toMat(tsv_content)
-    hosts = [(l[0], l[1]) for l in M]  # (host, port)
-    codes = [l[2] for l in M]
-
-    assert len(hosts) == num_hosts and len(set(hosts)) == num_hosts, "\n" + tsv_content
-    assert len(set(codes)) == 1, "\n" + tsv_content
-    assert codes[0] == "0", "\n" + tsv_content
-
-
-def ddl_check_query(instance, query, num_hosts=3):
-    contents = instance.query(query)
-    check_all_hosts_sucesfully_executed(contents, num_hosts)
-    return contents
-
-
-@pytest.fixture(scope="module")
-def started_cluster():
-    global cluster
-    try:
-        clusters_schema = {
-            "0": {"0": ["0", "1"], "1": ["0"]},
-            "1": {"0": ["0", "1"], "1": ["0"]},
-        }
-
-        for cluster_name, shards in clusters_schema.items():
-            for shard_name, replicas in shards.items():
-                for replica_name in replicas:
-                    name = "s{}_{}_{}".format(cluster_name, shard_name, replica_name)
-                    cluster.add_instance(
-                        name,
-                        main_configs=[
-                            "configs/conf.d/query_log.xml",
-                            "configs/conf.d/ddl.xml",
-                            "configs/conf.d/clusters.xml",
-                        ],
-                        user_configs=["configs/users.xml"],
-                        macros={
-                            "cluster": cluster_name,
-                            "shard": shard_name,
-                            "replica": replica_name,
-                        },
-                        with_zookeeper=True,
-                    )
-
-        cluster.start()
-        yield cluster
-
-    finally:
-        cluster.shutdown()
-
-
-class Task1:
-    def __init__(self, cluster):
-        self.cluster = cluster
-        self.zk_task_path = "/clickhouse-copier/task_simple_" + generateRandomString(10)
-        self.container_task_file = "/task0_description.xml"
-
-        for instance_name, _ in cluster.instances.items():
-            instance = cluster.instances[instance_name]
-            instance.copy_file_to_container(
-                os.path.join(CURRENT_TEST_DIR, "./task0_description.xml"),
-                self.container_task_file,
-            )
-            print(
-                "Copied task file to container of '{}' instance. Path {}".format(
-                    instance_name, self.container_task_file
-                )
-            )
-
-    def start(self):
-        instance = cluster.instances["s0_0_0"]
-
-        for cluster_num in ["0", "1"]:
-            ddl_check_query(
-                instance,
-                "DROP DATABASE IF EXISTS default ON CLUSTER cluster{} SYNC".format(
-                    cluster_num
-                ),
-            )
-            ddl_check_query(
-                instance,
-                "CREATE DATABASE default ON CLUSTER cluster{} ".format(cluster_num),
-            )
-
-        ddl_check_query(
-            instance,
-            "CREATE TABLE hits ON CLUSTER cluster0 (d UInt64, d1 UInt64 MATERIALIZED d+1) "
-            + "ENGINE=ReplicatedMergeTree "
-            + "PARTITION BY d % 3 ORDER BY (d, sipHash64(d)) SAMPLE BY sipHash64(d) SETTINGS index_granularity = 16",
-        )
-        ddl_check_query(
-            instance,
-            "CREATE TABLE hits_all ON CLUSTER cluster0 (d UInt64) ENGINE=Distributed(cluster0, default, hits, d)",
-        )
-        ddl_check_query(
-            instance,
-            "CREATE TABLE hits_all ON CLUSTER cluster1 (d UInt64) ENGINE=Distributed(cluster1, default, hits, d + 1)",
-        )
-        instance.query(
-            "INSERT INTO hits_all SELECT * FROM system.numbers LIMIT 1002",
-            settings={"distributed_foreground_insert": 1},
-        )
-
-    def check(self):
-        assert (
-            self.cluster.instances["s0_0_0"]
-            .query("SELECT count() FROM hits_all")
-            .strip()
-            == "1002"
-        )
-        assert (
-            self.cluster.instances["s1_0_0"]
-            .query("SELECT count() FROM hits_all")
-            .strip()
-            == "1002"
-        )
-
-        assert (
-            self.cluster.instances["s1_0_0"]
-            .query("SELECT DISTINCT d % 2 FROM hits")
-            .strip()
-            == "1"
-        )
-        assert (
-            self.cluster.instances["s1_1_0"]
-            .query("SELECT DISTINCT d % 2 FROM hits")
-            .strip()
-            == "0"
-        )
-
-        instance = self.cluster.instances["s0_0_0"]
-        ddl_check_query(instance, "DROP TABLE hits_all ON CLUSTER cluster0")
-        ddl_check_query(instance, "DROP TABLE hits_all ON CLUSTER cluster1")
-        ddl_check_query(instance, "DROP TABLE hits ON CLUSTER cluster0")
-        ddl_check_query(instance, "DROP TABLE hits ON CLUSTER cluster1")
-
-
-class Task2:
-    def __init__(self, cluster, unique_zk_path):
-        self.cluster = cluster
-        self.zk_task_path = (
-            "/clickhouse-copier/task_month_to_week_partition_" + generateRandomString(5)
-        )
-        self.unique_zk_path = generateRandomString(10)
-        self.container_task_file = "/task_month_to_week_description.xml"
-
-        for instance_name, _ in cluster.instances.items():
-            instance = cluster.instances[instance_name]
-            instance.copy_file_to_container(
-                os.path.join(CURRENT_TEST_DIR, "./task_month_to_week_description.xml"),
-                self.container_task_file,
-            )
-            print(
-                "Copied task file to container of '{}' instance. Path {}".format(
-                    instance_name, self.container_task_file
-                )
-            )
-
-    def start(self):
-        instance = cluster.instances["s0_0_0"]
-
-        for cluster_num in ["0", "1"]:
-            ddl_check_query(
-                instance,
-                "DROP DATABASE IF EXISTS default ON CLUSTER cluster{}".format(
-                    cluster_num
-                ),
-            )
-            ddl_check_query(
-                instance,
-                "CREATE DATABASE IF NOT EXISTS default ON CLUSTER cluster{}".format(
-                    cluster_num
-                ),
-            )
-
-        ddl_check_query(
-            instance,
-            "CREATE TABLE a ON CLUSTER cluster0 (date Date, d UInt64, d1 UInt64 ALIAS d+1) "
-            "ENGINE=ReplicatedMergeTree('/clickhouse/tables/cluster_{cluster}/{shard}/"
-            + self.unique_zk_path
-            + "', "
-            "'{replica}', date, intHash64(d), (date, intHash64(d)), 8192)",
-        )
-        ddl_check_query(
-            instance,
-            "CREATE TABLE a_all ON CLUSTER cluster0 (date Date, d UInt64) ENGINE=Distributed(cluster0, default, a, d)",
-        )
-
-        instance.query(
-            "INSERT INTO a_all SELECT toDate(17581 + number) AS date, number AS d FROM system.numbers LIMIT 85",
-            settings={"distributed_foreground_insert": 1},
-        )
-
-    def check(self):
-        assert TSV(
-            self.cluster.instances["s0_0_0"].query(
-                "SELECT count() FROM cluster(cluster0, default, a)"
-            )
-        ) == TSV("85\n")
-        assert TSV(
-            self.cluster.instances["s1_0_0"].query(
-                "SELECT count(), uniqExact(date) FROM cluster(cluster1, default, b)"
-            )
-        ) == TSV("85\t85\n")
-
-        assert TSV(
-            self.cluster.instances["s1_0_0"].query(
-                "SELECT DISTINCT jumpConsistentHash(intHash64(d), 2) FROM b"
-            )
-        ) == TSV("0\n")
-        assert TSV(
-            self.cluster.instances["s1_1_0"].query(
-                "SELECT DISTINCT jumpConsistentHash(intHash64(d), 2) FROM b"
-            )
-        ) == TSV("1\n")
-
-        assert TSV(
-            self.cluster.instances["s1_0_0"].query(
-                "SELECT uniqExact(partition) IN (12, 13) FROM system.parts WHERE active AND database='default' AND table='b'"
-            )
-        ) == TSV("1\n")
-        assert TSV(
-            self.cluster.instances["s1_1_0"].query(
-                "SELECT uniqExact(partition) IN (12, 13) FROM system.parts WHERE active AND database='default' AND table='b'"
-            )
-        ) == TSV("1\n")
-
-        instance = cluster.instances["s0_0_0"]
-        ddl_check_query(instance, "DROP TABLE a ON CLUSTER cluster0")
-        ddl_check_query(instance, "DROP TABLE b ON CLUSTER cluster1")
-
-
-class Task_test_block_size:
-    def __init__(self, cluster):
-        self.cluster = cluster
-        self.zk_task_path = (
-            "/clickhouse-copier/task_test_block_size_" + generateRandomString(5)
-        )
-        self.rows = 1000000
-        self.container_task_file = "/task_test_block_size.xml"
-
-        for instance_name, _ in cluster.instances.items():
-            instance = cluster.instances[instance_name]
-            instance.copy_file_to_container(
-                os.path.join(CURRENT_TEST_DIR, "./task_test_block_size.xml"),
-                self.container_task_file,
-            )
-            print(
-                "Copied task file to container of '{}' instance. Path {}".format(
-                    instance_name, self.container_task_file
-                )
-            )
-
-    def start(self):
-        instance = cluster.instances["s0_0_0"]
-
-        ddl_check_query(
-            instance,
-            """
-            CREATE TABLE test_block_size ON CLUSTER shard_0_0 (partition Date, d UInt64)
-            ENGINE=ReplicatedMergeTree
-            ORDER BY (d, sipHash64(d)) SAMPLE BY sipHash64(d)""",
-            2,
-        )
-
-        instance.query(
-            "INSERT INTO test_block_size SELECT toDate(0) AS partition, number as d FROM system.numbers LIMIT {}".format(
-                self.rows
-            )
-        )
-
-    def check(self):
-        assert TSV(
-            self.cluster.instances["s1_0_0"].query(
-                "SELECT count() FROM cluster(cluster1, default, test_block_size)"
-            )
-        ) == TSV("{}\n".format(self.rows))
-
-        instance = cluster.instances["s0_0_0"]
-        ddl_check_query(instance, "DROP TABLE test_block_size ON CLUSTER shard_0_0", 2)
-        ddl_check_query(instance, "DROP TABLE test_block_size ON CLUSTER cluster1")
-
-
-class Task_no_index:
-    def __init__(self, cluster):
-        self.cluster = cluster
-        self.zk_task_path = "/clickhouse-copier/task_no_index_" + generateRandomString(
-            5
-        )
-        self.rows = 1000000
-        self.container_task_file = "/task_no_index.xml"
-
-        for instance_name, _ in cluster.instances.items():
-            instance = cluster.instances[instance_name]
-            instance.copy_file_to_container(
-                os.path.join(CURRENT_TEST_DIR, "./task_no_index.xml"),
-                self.container_task_file,
-            )
-            print(
-                "Copied task file to container of '{}' instance. Path {}".format(
-                    instance_name, self.container_task_file
-                )
-            )
-
-    def start(self):
-        instance = cluster.instances["s0_0_0"]
-        instance.query("DROP TABLE IF EXISTS ontime SYNC")
-        instance.query(
-            "create table IF NOT EXISTS ontime (Year UInt16, FlightDate String) ENGINE = Memory"
-        )
-        instance.query(
-            "insert into ontime values (2016, 'test6'), (2017, 'test7'), (2018, 'test8')"
-        )
-
-    def check(self):
-        assert TSV(
-            self.cluster.instances["s1_1_0"].query("SELECT Year FROM ontime22")
-        ) == TSV("2017\n")
-        instance = cluster.instances["s0_0_0"]
-        instance.query("DROP TABLE ontime")
-        instance = cluster.instances["s1_1_0"]
-        instance.query("DROP TABLE ontime22")
-
-
-class Task_no_arg:
-    def __init__(self, cluster):
-        self.cluster = cluster
-        self.zk_task_path = "/clickhouse-copier/task_no_arg"
-        self.rows = 1000000
-        self.container_task_file = "/task_no_arg.xml"
-
-        for instance_name, _ in cluster.instances.items():
-            instance = cluster.instances[instance_name]
-            instance.copy_file_to_container(
-                os.path.join(CURRENT_TEST_DIR, "./task_no_arg.xml"),
-                self.container_task_file,
-            )
-            print(
-                "Copied task file to container of '{}' instance. Path {}".format(
-                    instance_name, self.container_task_file
-                )
-            )
-
-    def start(self):
-        instance = cluster.instances["s0_0_0"]
-        instance.query("DROP TABLE IF EXISTS copier_test1 SYNC")
-        instance.query(
-            "create table if not exists copier_test1 (date Date, id UInt32) engine = MergeTree PARTITION BY date ORDER BY date SETTINGS index_granularity = 8192"
-        )
-        instance.query("insert into copier_test1 values ('2016-01-01', 10);")
-
-    def check(self):
-        assert TSV(
-            self.cluster.instances["s1_1_0"].query("SELECT date FROM copier_test1_1")
-        ) == TSV("2016-01-01\n")
-        instance = cluster.instances["s0_0_0"]
-        instance.query("DROP TABLE copier_test1 SYNC")
-        instance = cluster.instances["s1_1_0"]
-        instance.query("DROP TABLE copier_test1_1 SYNC")
-
-
-class Task_non_partitioned_table:
-    def __init__(self, cluster):
-        self.cluster = cluster
-        self.zk_task_path = "/clickhouse-copier/task_non_partitoned_table"
-        self.rows = 1000000
-        self.container_task_file = "/task_non_partitioned_table.xml"
-
-        for instance_name, _ in cluster.instances.items():
-            instance = cluster.instances[instance_name]
-            instance.copy_file_to_container(
-                os.path.join(CURRENT_TEST_DIR, "./task_non_partitioned_table.xml"),
-                self.container_task_file,
-            )
-            print(
-                "Copied task file to container of '{}' instance. Path {}".format(
-                    instance_name, self.container_task_file
-                )
-            )
-
-    def start(self):
-        instance = cluster.instances["s0_0_0"]
-        instance.query("DROP TABLE IF EXISTS copier_test1 SYNC")
-        instance.query(
-            "create table copier_test1 (date Date, id UInt32) engine = MergeTree ORDER BY date SETTINGS index_granularity = 8192"
-        )
-        instance.query("insert into copier_test1 values ('2016-01-01', 10);")
-
-    def check(self):
-        assert TSV(
-            self.cluster.instances["s1_1_0"].query("SELECT date FROM copier_test1_1")
-        ) == TSV("2016-01-01\n")
-        instance = cluster.instances["s0_0_0"]
-        instance.query("DROP TABLE copier_test1")
-        instance = cluster.instances["s1_1_0"]
-        instance.query("DROP TABLE copier_test1_1")
-
-
-class Task_self_copy:
-    def __init__(self, cluster):
-        self.cluster = cluster
-        self.zk_task_path = "/clickhouse-copier/task_self_copy"
-        self.container_task_file = "/task_self_copy.xml"
-
-        for instance_name, _ in cluster.instances.items():
-            instance = cluster.instances[instance_name]
-            instance.copy_file_to_container(
-                os.path.join(CURRENT_TEST_DIR, "./task_self_copy.xml"),
-                self.container_task_file,
-            )
-            print(
-                "Copied task file to container of '{}' instance. Path {}".format(
-                    instance_name, self.container_task_file
-                )
-            )
-
-    def start(self):
-        instance = cluster.instances["s0_0_0"]
-        instance.query("DROP DATABASE IF EXISTS db1 SYNC")
-        instance.query("DROP DATABASE IF EXISTS db2 SYNC")
-        instance.query("CREATE DATABASE IF NOT EXISTS db1;")
-        instance.query(
-            "CREATE TABLE IF NOT EXISTS db1.source_table (`a` Int8, `b` String, `c` Int8) ENGINE = MergeTree PARTITION BY a ORDER BY a SETTINGS index_granularity = 8192"
-        )
-        instance.query("CREATE DATABASE IF NOT EXISTS db2;")
-        instance.query(
-            "CREATE TABLE IF NOT EXISTS db2.destination_table (`a` Int8, `b` String, `c` Int8) ENGINE = MergeTree PARTITION BY a ORDER BY a SETTINGS index_granularity = 8192"
-        )
-        instance.query("INSERT INTO db1.source_table VALUES (1, 'ClickHouse', 1);")
-        instance.query("INSERT INTO db1.source_table VALUES (2, 'Copier', 2);")
-
-    def check(self):
-        instance = cluster.instances["s0_0_0"]
-        assert TSV(
-            instance.query("SELECT * FROM db2.destination_table ORDER BY a")
-        ) == TSV(instance.query("SELECT * FROM db1.source_table ORDER BY a"))
-        instance = cluster.instances["s0_0_0"]
-        instance.query("DROP DATABASE IF EXISTS db1 SYNC")
-        instance.query("DROP DATABASE IF EXISTS db2 SYNC")
-
-
-def execute_task(started_cluster, task, cmd_options):
-    task.start()
-
-    zk = started_cluster.get_kazoo_client("zoo1")
-    print("Use ZooKeeper server: {}:{}".format(zk.hosts[0][0], zk.hosts[0][1]))
-
-    try:
-        zk.delete("/clickhouse-copier", recursive=True)
-    except kazoo.exceptions.NoNodeError:
-        print("No node /clickhouse-copier. It is Ok in first test.")
-
-    # Run cluster-copier processes on each node
-    docker_api = started_cluster.docker_client.api
-    copiers_exec_ids = []
-
-    cmd = [
-        "/usr/bin/clickhouse",
-        "copier",
-        "--config",
-        "/etc/clickhouse-server/config-copier.xml",
-        "--task-path",
-        task.zk_task_path,
-        "--task-file",
-        task.container_task_file,
-        "--task-upload-force",
-        "true",
-        "--base-dir",
-        "/var/log/clickhouse-server/copier",
-    ]
-    cmd += cmd_options
-
-    print(cmd)
-
-    copiers = random.sample(list(started_cluster.instances.keys()), 3)
-
-    for instance_name in copiers:
-        instance = started_cluster.instances[instance_name]
-        container = instance.get_docker_handle()
-        instance.copy_file_to_container(
-            os.path.join(CURRENT_TEST_DIR, "configs/config-copier.xml"),
-            "/etc/clickhouse-server/config-copier.xml",
-        )
-        print("Copied copier config to {}".format(instance.name))
-        exec_id = docker_api.exec_create(container.id, cmd, stderr=True)
-        output = docker_api.exec_start(exec_id).decode("utf8")
-        print(output)
-        copiers_exec_ids.append(exec_id)
-        print(
-            "Copier for {} ({}) has started".format(instance.name, instance.ip_address)
-        )
-
-    # Wait for copiers stopping and check their return codes
-    for exec_id, instance_name in zip(copiers_exec_ids, copiers):
-        instance = started_cluster.instances[instance_name]
-        while True:
-            res = docker_api.exec_inspect(exec_id)
-            if not res["Running"]:
-                break
-            time.sleep(0.5)
-
-        assert res["ExitCode"] == 0, "Instance: {} ({}). Info: {}".format(
-            instance.name, instance.ip_address, repr(res)
-        )
-
-    try:
-        task.check()
-    finally:
-        zk.delete(task.zk_task_path, recursive=True)
-
-
-# Tests
-
-
-@pytest.mark.parametrize(("use_sample_offset"), [False, True])
-def test_copy_simple(started_cluster, use_sample_offset):
-    if use_sample_offset:
-        execute_task(
-            started_cluster,
-            Task1(started_cluster),
-            ["--experimental-use-sample-offset", "1"],
-        )
-    else:
-        execute_task(started_cluster, Task1(started_cluster), [])
-
-
-@pytest.mark.parametrize(("use_sample_offset"), [False, True])
-def test_copy_with_recovering(started_cluster, use_sample_offset):
-    if use_sample_offset:
-        execute_task(
-            started_cluster,
-            Task1(started_cluster),
-            [
-                "--copy-fault-probability",
-                str(COPYING_FAIL_PROBABILITY),
-                "--experimental-use-sample-offset",
-                "1",
-                "--max-table-tries",
-                "10",
-            ],
-        )
-    else:
-        execute_task(
-            started_cluster,
-            Task1(started_cluster),
-            [
-                "--copy-fault-probability",
-                str(COPYING_FAIL_PROBABILITY),
-                "--max-table-tries",
-                "10",
-            ],
-        )
-
-
-@pytest.mark.parametrize(("use_sample_offset"), [False, True])
-def test_copy_with_recovering_after_move_faults(started_cluster, use_sample_offset):
-    if use_sample_offset:
-        execute_task(
-            started_cluster,
-            Task1(started_cluster),
-            [
-                "--move-fault-probability",
-                str(MOVING_FAIL_PROBABILITY),
-                "--experimental-use-sample-offset",
-                "1",
-            ],
-        )
-    else:
-        execute_task(
-            started_cluster,
-            Task1(started_cluster),
-            ["--move-fault-probability", str(MOVING_FAIL_PROBABILITY)],
-        )
-
-
-@pytest.mark.timeout(600)
-def test_copy_month_to_week_partition(started_cluster):
-    execute_task(started_cluster, Task2(started_cluster, "test1"), [])
-
-
-@pytest.mark.timeout(600)
-def test_copy_month_to_week_partition_with_recovering(started_cluster):
-    execute_task(
-        started_cluster,
-        Task2(started_cluster, "test2"),
-        [
-            "--copy-fault-probability",
-            str(COPYING_FAIL_PROBABILITY),
-            "--max-table-tries",
-            "10",
-        ],
-    )
-
-
-@pytest.mark.timeout(600)
-def test_copy_month_to_week_partition_with_recovering_after_move_faults(
-    started_cluster,
-):
-    execute_task(
-        started_cluster,
-        Task2(started_cluster, "test3"),
-        ["--move-fault-probability", str(MOVING_FAIL_PROBABILITY)],
-    )
-
-
-def test_block_size(started_cluster):
-    execute_task(started_cluster, Task_test_block_size(started_cluster), [])
-
-
-def test_no_index(started_cluster):
-    execute_task(started_cluster, Task_no_index(started_cluster), [])
-
-
-def test_no_arg(started_cluster):
-    execute_task(started_cluster, Task_no_arg(started_cluster), [])
-
-
-def test_non_partitioned_table(started_cluster):
-    execute_task(started_cluster, Task_non_partitioned_table(started_cluster), [])
-
-
-def test_self_copy(started_cluster):
-    execute_task(started_cluster, Task_self_copy(started_cluster), [])
diff --git a/tests/integration/test_cluster_copier/test_three_nodes.py b/tests/integration/test_cluster_copier/test_three_nodes.py
deleted file mode 100644
index e7d07757adb..00000000000
--- a/tests/integration/test_cluster_copier/test_three_nodes.py
+++ /dev/null
@@ -1,286 +0,0 @@
-import os
-import sys
-import time
-import logging
-import pytest
-
-from helpers.cluster import ClickHouseCluster
-from helpers.test_tools import TSV
-
-import docker
-
-CURRENT_TEST_DIR = os.path.dirname(os.path.abspath(__file__))
-sys.path.insert(0, os.path.dirname(CURRENT_TEST_DIR))
-
-cluster = ClickHouseCluster(__file__)
-
-
-@pytest.fixture(scope="module")
-def started_cluster():
-    global cluster
-    try:
-        for name in ["first", "second", "third"]:
-            cluster.add_instance(
-                name,
-                main_configs=[
-                    "configs_three_nodes/conf.d/clusters.xml",
-                    "configs_three_nodes/conf.d/ddl.xml",
-                ],
-                user_configs=["configs_three_nodes/users.xml"],
-                with_zookeeper=True,
-            )
-
-        cluster.start()
-        yield cluster
-
-    finally:
-        cluster.shutdown()
-
-
-class Task:
-    def __init__(self, cluster):
-        self.cluster = cluster
-        self.zk_task_path = "/clickhouse-copier/task"
-        self.container_task_file = "/task_taxi_data.xml"
-
-        for instance_name, _ in cluster.instances.items():
-            instance = cluster.instances[instance_name]
-            instance.copy_file_to_container(
-                os.path.join(CURRENT_TEST_DIR, "./task_taxi_data.xml"),
-                self.container_task_file,
-            )
-            logging.debug(
-                f"Copied task file to container of '{instance_name}' instance. Path {self.container_task_file}"
-            )
-
-    def start(self):
-        for name in ["first", "second", "third"]:
-            node = cluster.instances[name]
-            node.query("DROP DATABASE IF EXISTS dailyhistory SYNC;")
-            node.query("DROP DATABASE IF EXISTS monthlyhistory SYNC;")
-
-        first = cluster.instances["first"]
-
-        # daily partition database
-        first.query("CREATE DATABASE IF NOT EXISTS dailyhistory on cluster events;")
-        first.query(
-            """CREATE TABLE dailyhistory.yellow_tripdata_staging ON CLUSTER events
-        (
-            id UUID DEFAULT generateUUIDv4(),
-            vendor_id String,
-            tpep_pickup_datetime DateTime('UTC'),
-            tpep_dropoff_datetime DateTime('UTC'),
-            passenger_count Nullable(Float64),
-            trip_distance String,
-            pickup_longitude Float64,
-            pickup_latitude Float64,
-            rate_code_id String,
-            store_and_fwd_flag String,
-            dropoff_longitude Float64,
-            dropoff_latitude Float64,
-            payment_type String,
-            fare_amount String,
-            extra String,
-            mta_tax String,
-            tip_amount String,
-            tolls_amount String,
-            improvement_surcharge String,
-            total_amount String,
-            pickup_location_id String,
-            dropoff_location_id String,
-            congestion_surcharge String,
-            junk1 String,  junk2 String
-        )
-        Engine = ReplacingMergeTree()
-        PRIMARY KEY (tpep_pickup_datetime, id)
-        ORDER BY (tpep_pickup_datetime, id)
-        PARTITION BY (toYYYYMMDD(tpep_pickup_datetime))"""
-        )
-
-        first.query(
-            """CREATE TABLE dailyhistory.yellow_tripdata
-            ON CLUSTER events
-            AS dailyhistory.yellow_tripdata_staging
-            ENGINE = Distributed('events', 'dailyhistory', yellow_tripdata_staging, sipHash64(id) % 3);"""
-        )
-
-        first.query(
-            """INSERT INTO dailyhistory.yellow_tripdata
-            SELECT * FROM generateRandom(
-                'id UUID DEFAULT generateUUIDv4(),
-                vendor_id String,
-                tpep_pickup_datetime DateTime(\\'UTC\\'),
-                tpep_dropoff_datetime DateTime(\\'UTC\\'),
-                passenger_count Nullable(Float64),
-                trip_distance String,
-                pickup_longitude Float64,
-                pickup_latitude Float64,
-                rate_code_id String,
-                store_and_fwd_flag String,
-                dropoff_longitude Float64,
-                dropoff_latitude Float64,
-                payment_type String,
-                fare_amount String,
-                extra String,
-                mta_tax String,
-                tip_amount String,
-                tolls_amount String,
-                improvement_surcharge String,
-                total_amount String,
-                pickup_location_id String,
-                dropoff_location_id String,
-                congestion_surcharge String,
-                junk1 String,
-                junk2 String',
-            1, 10, 2) LIMIT 50;"""
-        )
-
-        # monthly partition database
-        first.query("create database IF NOT EXISTS monthlyhistory on cluster events;")
-        first.query(
-            """CREATE TABLE monthlyhistory.yellow_tripdata_staging ON CLUSTER events
-        (
-            id UUID DEFAULT generateUUIDv4(),
-            vendor_id String,
-            tpep_pickup_datetime DateTime('UTC'),
-            tpep_dropoff_datetime DateTime('UTC'),
-            passenger_count Nullable(Float64),
-            trip_distance String,
-            pickup_longitude Float64,
-            pickup_latitude Float64,
-            rate_code_id String,
-            store_and_fwd_flag String,
-            dropoff_longitude Float64,
-            dropoff_latitude Float64,
-            payment_type String,
-            fare_amount String,
-            extra String,
-            mta_tax String,
-            tip_amount String,
-            tolls_amount String,
-            improvement_surcharge String,
-            total_amount String,
-            pickup_location_id String,
-            dropoff_location_id String,
-            congestion_surcharge String,
-            junk1 String,
-            junk2 String
-        )
-        Engine = ReplacingMergeTree()
-        PRIMARY KEY (tpep_pickup_datetime, id)
-        ORDER BY (tpep_pickup_datetime, id)
-        PARTITION BY (pickup_location_id, toYYYYMM(tpep_pickup_datetime))"""
-        )
-
-        first.query(
-            """CREATE TABLE monthlyhistory.yellow_tripdata
-            ON CLUSTER events
-            AS monthlyhistory.yellow_tripdata_staging
-            ENGINE = Distributed('events', 'monthlyhistory', yellow_tripdata_staging, sipHash64(id) % 3);"""
-        )
-
-    def check(self):
-        first = cluster.instances["first"]
-        a = TSV(first.query("SELECT count() from dailyhistory.yellow_tripdata"))
-        b = TSV(first.query("SELECT count() from monthlyhistory.yellow_tripdata"))
-        assert a == b, "Distributed tables"
-
-        for instance_name, instance in cluster.instances.items():
-            instance = cluster.instances[instance_name]
-            a = instance.query(
-                "SELECT count() from dailyhistory.yellow_tripdata_staging"
-            )
-            b = instance.query(
-                "SELECT count() from monthlyhistory.yellow_tripdata_staging"
-            )
-            assert a == b, "MergeTree tables on each shard"
-
-            a = TSV(
-                instance.query(
-                    "SELECT sipHash64(*) from dailyhistory.yellow_tripdata_staging ORDER BY id"
-                )
-            )
-            b = TSV(
-                instance.query(
-                    "SELECT sipHash64(*) from monthlyhistory.yellow_tripdata_staging ORDER BY id"
-                )
-            )
-
-            assert a == b, "Data on each shard"
-
-        for name in ["first", "second", "third"]:
-            node = cluster.instances[name]
-            node.query("DROP DATABASE IF EXISTS dailyhistory SYNC;")
-            node.query("DROP DATABASE IF EXISTS monthlyhistory SYNC;")
-
-
-def execute_task(started_cluster, task, cmd_options):
-    task.start()
-
-    zk = started_cluster.get_kazoo_client("zoo1")
-    logging.debug("Use ZooKeeper server: {}:{}".format(zk.hosts[0][0], zk.hosts[0][1]))
-
-    # Run cluster-copier processes on each node
-    docker_api = started_cluster.docker_client.api
-    copiers_exec_ids = []
-
-    cmd = [
-        "/usr/bin/clickhouse",
-        "copier",
-        "--config",
-        "/etc/clickhouse-server/config-copier.xml",
-        "--task-path",
-        task.zk_task_path,
-        "--task-file",
-        task.container_task_file,
-        "--task-upload-force",
-        "true",
-        "--base-dir",
-        "/var/log/clickhouse-server/copier",
-    ]
-    cmd += cmd_options
-
-    logging.debug(f"execute_task cmd: {cmd}")
-
-    for instance_name in started_cluster.instances.keys():
-        instance = started_cluster.instances[instance_name]
-        container = instance.get_docker_handle()
-        instance.copy_file_to_container(
-            os.path.join(CURRENT_TEST_DIR, "configs_three_nodes/config-copier.xml"),
-            "/etc/clickhouse-server/config-copier.xml",
-        )
-        logging.info("Copied copier config to {}".format(instance.name))
-        exec_id = docker_api.exec_create(container.id, cmd, stderr=True)
-        output = docker_api.exec_start(exec_id).decode("utf8")
-        logging.info(output)
-        copiers_exec_ids.append(exec_id)
-        logging.info(
-            "Copier for {} ({}) has started".format(instance.name, instance.ip_address)
-        )
-
-    # time.sleep(1000)
-
-    # Wait for copiers stopping and check their return codes
-    for exec_id, instance in zip(
-        copiers_exec_ids, iter(started_cluster.instances.values())
-    ):
-        while True:
-            res = docker_api.exec_inspect(exec_id)
-            if not res["Running"]:
-                break
-            time.sleep(1)
-
-        assert res["ExitCode"] == 0, "Instance: {} ({}). Info: {}".format(
-            instance.name, instance.ip_address, repr(res)
-        )
-
-    try:
-        task.check()
-    finally:
-        zk.delete(task.zk_task_path, recursive=True)
-
-
-# Tests
-@pytest.mark.timeout(600)
-def test(started_cluster):
-    execute_task(started_cluster, Task(started_cluster), [])
diff --git a/tests/integration/test_cluster_copier/test_trivial.py b/tests/integration/test_cluster_copier/test_trivial.py
deleted file mode 100644
index b8060583ef8..00000000000
--- a/tests/integration/test_cluster_copier/test_trivial.py
+++ /dev/null
@@ -1,227 +0,0 @@
-import os
-import sys
-import time
-import random
-import string
-
-from helpers.cluster import ClickHouseCluster
-from helpers.test_tools import TSV
-
-import kazoo
-import pytest
-import docker
-
-
-CURRENT_TEST_DIR = os.path.dirname(os.path.abspath(__file__))
-sys.path.insert(0, os.path.dirname(CURRENT_TEST_DIR))
-
-
-COPYING_FAIL_PROBABILITY = 0.1
-MOVING_FAIL_PROBABILITY = 0.1
-
-cluster = ClickHouseCluster(__file__)
-
-
-def generateRandomString(count):
-    return "".join(
-        random.choice(string.ascii_uppercase + string.digits) for _ in range(count)
-    )
-
-
-@pytest.fixture(scope="module")
-def started_cluster():
-    global cluster
-    try:
-        for name in ["first_trivial", "second_trivial"]:
-            instance = cluster.add_instance(
-                name,
-                main_configs=["configs/conf.d/clusters_trivial.xml"],
-                user_configs=["configs_two_nodes/users.xml"],
-                macros={
-                    "cluster": name,
-                    "shard": "the_only_shard",
-                    "replica": "the_only_replica",
-                },
-                with_zookeeper=True,
-            )
-
-        cluster.start()
-        yield cluster
-
-    finally:
-        cluster.shutdown()
-
-
-class TaskTrivial:
-    def __init__(self, cluster):
-        self.cluster = cluster
-        self.zk_task_path = "/clickhouse-copier/task_trivial"
-        self.copier_task_config = open(
-            os.path.join(CURRENT_TEST_DIR, "task_trivial.xml"), "r"
-        ).read()
-
-    def start(self):
-        source = cluster.instances["first_trivial"]
-        destination = cluster.instances["second_trivial"]
-
-        for node in [source, destination]:
-            node.query("DROP DATABASE IF EXISTS default")
-            node.query("CREATE DATABASE IF NOT EXISTS default")
-
-        source.query(
-            "CREATE TABLE trivial (d UInt64, d1 UInt64 MATERIALIZED d+1)"
-            "ENGINE=ReplicatedMergeTree('/clickhouse/tables/source_trivial_cluster/1/trivial/{}', '1') "
-            "PARTITION BY d % 5 ORDER BY (d, sipHash64(d)) SAMPLE BY sipHash64(d) SETTINGS index_granularity = 16".format(
-                generateRandomString(10)
-            )
-        )
-
-        source.query(
-            "INSERT INTO trivial SELECT * FROM system.numbers LIMIT 1002",
-            settings={"distributed_foreground_insert": 1},
-        )
-
-    def check(self):
-        zk = cluster.get_kazoo_client("zoo1")
-        status_data, _ = zk.get(self.zk_task_path + "/status")
-        assert (
-            status_data
-            == b'{"hits":{"all_partitions_count":5,"processed_partitions_count":5}}'
-        )
-
-        source = cluster.instances["first_trivial"]
-        destination = cluster.instances["second_trivial"]
-
-        assert TSV(source.query("SELECT count() FROM trivial")) == TSV("1002\n")
-        assert TSV(destination.query("SELECT count() FROM trivial")) == TSV("1002\n")
-
-        for node in [source, destination]:
-            node.query("DROP TABLE trivial")
-
-
-class TaskReplicatedWithoutArguments:
-    def __init__(self, cluster):
-        self.cluster = cluster
-        self.zk_task_path = "/clickhouse-copier/task_trivial_without_arguments"
-        self.copier_task_config = open(
-            os.path.join(CURRENT_TEST_DIR, "task_trivial_without_arguments.xml"), "r"
-        ).read()
-
-    def start(self):
-        source = cluster.instances["first_trivial"]
-        destination = cluster.instances["second_trivial"]
-
-        for node in [source, destination]:
-            node.query("DROP DATABASE IF EXISTS default")
-            node.query("CREATE DATABASE IF NOT EXISTS default")
-
-        source.query(
-            "CREATE TABLE trivial_without_arguments ON CLUSTER source_trivial_cluster (d UInt64, d1 UInt64 MATERIALIZED d+1) "
-            "ENGINE=ReplicatedMergeTree() "
-            "PARTITION BY d % 5 ORDER BY (d, sipHash64(d)) SAMPLE BY sipHash64(d) SETTINGS index_granularity = 16"
-        )
-
-        source.query(
-            "INSERT INTO trivial_without_arguments SELECT * FROM system.numbers LIMIT 1002",
-            settings={"distributed_foreground_insert": 1},
-        )
-
-    def check(self):
-        zk = cluster.get_kazoo_client("zoo1")
-        status_data, _ = zk.get(self.zk_task_path + "/status")
-        assert (
-            status_data
-            == b'{"hits":{"all_partitions_count":5,"processed_partitions_count":5}}'
-        )
-
-        source = cluster.instances["first_trivial"]
-        destination = cluster.instances["second_trivial"]
-
-        assert TSV(
-            source.query("SELECT count() FROM trivial_without_arguments")
-        ) == TSV("1002\n")
-        assert TSV(
-            destination.query("SELECT count() FROM trivial_without_arguments")
-        ) == TSV("1002\n")
-
-        for node in [source, destination]:
-            node.query("DROP TABLE trivial_without_arguments")
-
-
-def execute_task(started_cluster, task, cmd_options):
-    task.start()
-
-    zk = started_cluster.get_kazoo_client("zoo1")
-    print("Use ZooKeeper server: {}:{}".format(zk.hosts[0][0], zk.hosts[0][1]))
-
-    try:
-        zk.delete("/clickhouse-copier", recursive=True)
-    except kazoo.exceptions.NoNodeError:
-        print("No node /clickhouse-copier. It is Ok in first test.")
-
-    zk_task_path = task.zk_task_path
-    zk.ensure_path(zk_task_path)
-    zk.create(zk_task_path + "/description", task.copier_task_config.encode())
-
-    # Run cluster-copier processes on each node
-    docker_api = started_cluster.docker_client.api
-    copiers_exec_ids = []
-
-    cmd = [
-        "/usr/bin/clickhouse",
-        "copier",
-        "--config",
-        "/etc/clickhouse-server/config-copier.xml",
-        "--task-path",
-        zk_task_path,
-        "--base-dir",
-        "/var/log/clickhouse-server/copier",
-    ]
-    cmd += cmd_options
-
-    copiers = list(started_cluster.instances.keys())
-
-    for instance_name in copiers:
-        instance = started_cluster.instances[instance_name]
-        container = instance.get_docker_handle()
-        instance.copy_file_to_container(
-            os.path.join(CURRENT_TEST_DIR, "configs/config-copier.xml"),
-            "/etc/clickhouse-server/config-copier.xml",
-        )
-        print("Copied copier config to {}".format(instance.name))
-        exec_id = docker_api.exec_create(container.id, cmd, stderr=True)
-        output = docker_api.exec_start(exec_id).decode("utf8")
-        print(output)
-        copiers_exec_ids.append(exec_id)
-        print(
-            "Copier for {} ({}) has started".format(instance.name, instance.ip_address)
-        )
-
-    # Wait for copiers stopping and check their return codes
-    for exec_id, instance_name in zip(copiers_exec_ids, copiers):
-        instance = started_cluster.instances[instance_name]
-        while True:
-            res = docker_api.exec_inspect(exec_id)
-            if not res["Running"]:
-                break
-            time.sleep(0.5)
-
-        assert res["ExitCode"] == 0, "Instance: {} ({}). Info: {}".format(
-            instance.name, instance.ip_address, repr(res)
-        )
-
-    try:
-        task.check()
-    finally:
-        zk.delete(zk_task_path, recursive=True)
-
-
-# Tests
-
-
-def test_trivial_copy(started_cluster):
-    execute_task(started_cluster, TaskTrivial(started_cluster), [])
-
-
-def test_trivial_without_arguments(started_cluster):
-    execute_task(started_cluster, TaskReplicatedWithoutArguments(started_cluster), [])
diff --git a/tests/integration/test_cluster_copier/test_two_nodes.py b/tests/integration/test_cluster_copier/test_two_nodes.py
deleted file mode 100644
index 1bd3561f24f..00000000000
--- a/tests/integration/test_cluster_copier/test_two_nodes.py
+++ /dev/null
@@ -1,597 +0,0 @@
-import os
-import sys
-import time
-import logging
-import pytest
-
-from helpers.cluster import ClickHouseCluster
-from helpers.test_tools import TSV
-
-import docker
-
-CURRENT_TEST_DIR = os.path.dirname(os.path.abspath(__file__))
-sys.path.insert(0, os.path.dirname(CURRENT_TEST_DIR))
-
-cluster = ClickHouseCluster(__file__)
-
-
-@pytest.fixture(scope="module")
-def started_cluster():
-    global cluster
-    try:
-        for name in ["first_of_two", "second_of_two"]:
-            instance = cluster.add_instance(
-                name,
-                main_configs=[
-                    "configs_two_nodes/conf.d/clusters.xml",
-                    "configs_two_nodes/conf.d/ddl.xml",
-                    "configs_two_nodes/conf.d/storage_configuration.xml",
-                ],
-                user_configs=["configs_two_nodes/users.xml"],
-                with_zookeeper=True,
-            )
-
-        cluster.start()
-
-        for name in ["first_of_two", "second_of_two"]:
-            instance = cluster.instances[name]
-            instance.exec_in_container(["bash", "-c", "mkdir /jbod1"])
-            instance.exec_in_container(["bash", "-c", "mkdir /jbod2"])
-            instance.exec_in_container(["bash", "-c", "mkdir /external"])
-
-        yield cluster
-
-    finally:
-        cluster.shutdown()
-
-
-# Will copy table from `first` node to `second`
-class TaskWithDifferentSchema:
-    def __init__(self, cluster):
-        self.cluster = cluster
-        self.zk_task_path = "/clickhouse-copier/task_with_different_schema"
-        self.container_task_file = "/task_with_different_schema.xml"
-
-        for instance_name, _ in cluster.instances.items():
-            instance = cluster.instances[instance_name]
-            instance.copy_file_to_container(
-                os.path.join(CURRENT_TEST_DIR, "./task_with_different_schema.xml"),
-                self.container_task_file,
-            )
-            print(
-                "Copied task file to container of '{}' instance. Path {}".format(
-                    instance_name, self.container_task_file
-                )
-            )
-
-    def start(self):
-        first = cluster.instances["first_of_two"]
-        second = cluster.instances["second_of_two"]
-
-        first.query("DROP DATABASE IF EXISTS db_different_schema SYNC")
-        second.query("DROP DATABASE IF EXISTS db_different_schema SYNC")
-
-        first.query("CREATE DATABASE IF NOT EXISTS db_different_schema;")
-        first.query(
-            """CREATE TABLE db_different_schema.source
-        (
-            Column1 String,
-            Column2 UInt32,
-            Column3 Date,
-            Column4 DateTime,
-            Column5 UInt16,
-            Column6 String,
-            Column7 String,
-            Column8 String,
-            Column9 String,
-            Column10 String,
-            Column11 String,
-            Column12 Decimal(3, 1),
-            Column13 DateTime,
-            Column14 UInt16
-        )
-        ENGINE = MergeTree()
-        PARTITION BY (toYYYYMMDD(Column3), Column3)
-        PRIMARY KEY (Column1, Column2, Column3, Column4, Column6, Column7, Column8, Column9)
-        ORDER BY (Column1, Column2, Column3, Column4, Column6, Column7, Column8, Column9)
-        SETTINGS index_granularity = 8192"""
-        )
-
-        first.query(
-            """INSERT INTO db_different_schema.source SELECT * FROM generateRandom(
-            'Column1 String, Column2 UInt32, Column3 Date, Column4 DateTime, Column5 UInt16,
-            Column6 String, Column7 String, Column8 String, Column9 String, Column10 String,
-            Column11 String, Column12 Decimal(3, 1), Column13 DateTime, Column14 UInt16', 1, 10, 2) LIMIT 50;"""
-        )
-
-        second.query("CREATE DATABASE IF NOT EXISTS db_different_schema;")
-        second.query(
-            """CREATE TABLE db_different_schema.destination
-        (
-            Column1 LowCardinality(String) CODEC(LZ4),
-            Column2 UInt32 CODEC(LZ4),
-            Column3 Date CODEC(DoubleDelta, LZ4),
-            Column4 DateTime CODEC(DoubleDelta, LZ4),
-            Column5 UInt16 CODEC(LZ4),
-            Column6 LowCardinality(String) CODEC(ZSTD),
-            Column7 LowCardinality(String) CODEC(ZSTD),
-            Column8 LowCardinality(String) CODEC(ZSTD),
-            Column9 LowCardinality(String) CODEC(ZSTD),
-            Column10 String CODEC(ZSTD(6)),
-            Column11 LowCardinality(String) CODEC(LZ4),
-            Column12 Decimal(3,1) CODEC(LZ4),
-            Column13 DateTime CODEC(DoubleDelta, LZ4),
-            Column14 UInt16 CODEC(LZ4)
-        ) ENGINE = MergeTree()
-        PARTITION BY toYYYYMMDD(Column3)
-        ORDER BY (Column9, Column1, Column2, Column3, Column4);"""
-        )
-
-        print("Preparation completed")
-
-    def check(self):
-        first = cluster.instances["first_of_two"]
-        second = cluster.instances["second_of_two"]
-
-        a = first.query("SELECT count() from db_different_schema.source")
-        b = second.query("SELECT count() from db_different_schema.destination")
-        assert a == b, "Count"
-
-        a = TSV(
-            first.query(
-                """SELECT sipHash64(*) from db_different_schema.source
-            ORDER BY (Column1, Column2, Column3, Column4, Column5, Column6, Column7, Column8, Column9, Column10, Column11, Column12, Column13, Column14)"""
-            )
-        )
-        b = TSV(
-            second.query(
-                """SELECT sipHash64(*) from db_different_schema.destination
-            ORDER BY (Column1, Column2, Column3, Column4, Column5, Column6, Column7, Column8, Column9, Column10, Column11, Column12, Column13, Column14)"""
-            )
-        )
-        assert a == b, "Data"
-
-        first.query("DROP DATABASE IF EXISTS db_different_schema SYNC")
-        second.query("DROP DATABASE IF EXISTS db_different_schema SYNC")
-
-
-# Just simple copying, but table schema has TTL on columns
-# Also table will have slightly different schema
-class TaskTTL:
-    def __init__(self, cluster):
-        self.cluster = cluster
-        self.zk_task_path = "/clickhouse-copier/task_ttl_columns"
-        self.container_task_file = "/task_ttl_columns.xml"
-
-        for instance_name, _ in cluster.instances.items():
-            instance = cluster.instances[instance_name]
-            instance.copy_file_to_container(
-                os.path.join(CURRENT_TEST_DIR, "./task_ttl_columns.xml"),
-                self.container_task_file,
-            )
-            print(
-                "Copied task file to container of '{}' instance. Path {}".format(
-                    instance_name, self.container_task_file
-                )
-            )
-
-    def start(self):
-        first = cluster.instances["first_of_two"]
-        second = cluster.instances["second_of_two"]
-
-        first.query("DROP DATABASE IF EXISTS db_ttl_columns SYNC")
-        second.query("DROP DATABASE IF EXISTS db_ttl_columns SYNC")
-
-        first.query("CREATE DATABASE IF NOT EXISTS db_ttl_columns;")
-        first.query(
-            """CREATE TABLE db_ttl_columns.source
-        (
-            Column1 String,
-            Column2 UInt32,
-            Column3 Date,
-            Column4 DateTime,
-            Column5 UInt16,
-            Column6 String TTL now() + INTERVAL 1 MONTH,
-            Column7 Decimal(3, 1) TTL now() + INTERVAL 1 MONTH,
-            Column8 Tuple(Float64, Float64) TTL now() + INTERVAL 1 MONTH
-        )
-        ENGINE = MergeTree()
-        PARTITION BY (toYYYYMMDD(Column3), Column3)
-        PRIMARY KEY (Column1, Column2, Column3)
-        ORDER BY (Column1, Column2, Column3)
-        SETTINGS index_granularity = 8192"""
-        )
-
-        first.query(
-            """INSERT INTO db_ttl_columns.source SELECT * FROM generateRandom(
-            'Column1 String, Column2 UInt32, Column3 Date, Column4 DateTime, Column5 UInt16,
-            Column6 String, Column7 Decimal(3, 1), Column8 Tuple(Float64, Float64)', 1, 10, 2) LIMIT 50;"""
-        )
-
-        second.query("CREATE DATABASE IF NOT EXISTS db_ttl_columns;")
-        second.query(
-            """CREATE TABLE db_ttl_columns.destination
-        (
-            Column1 String,
-            Column2 UInt32,
-            Column3 Date,
-            Column4 DateTime TTL now() + INTERVAL 1 MONTH,
-            Column5 UInt16 TTL now() + INTERVAL 1 MONTH,
-            Column6 String TTL now() + INTERVAL 1 MONTH,
-            Column7 Decimal(3, 1) TTL now() + INTERVAL 1 MONTH,
-            Column8 Tuple(Float64, Float64)
-        ) ENGINE = MergeTree()
-        PARTITION BY toYYYYMMDD(Column3)
-        ORDER BY (Column3, Column2, Column1);"""
-        )
-
-        print("Preparation completed")
-
-    def check(self):
-        first = cluster.instances["first_of_two"]
-        second = cluster.instances["second_of_two"]
-
-        a = first.query("SELECT count() from db_ttl_columns.source")
-        b = second.query("SELECT count() from db_ttl_columns.destination")
-        assert a == b, "Count"
-
-        a = TSV(
-            first.query(
-                """SELECT sipHash64(*) from db_ttl_columns.source
-            ORDER BY (Column1, Column2, Column3, Column4, Column5, Column6, Column7, Column8)"""
-            )
-        )
-        b = TSV(
-            second.query(
-                """SELECT sipHash64(*) from db_ttl_columns.destination
-            ORDER BY (Column1, Column2, Column3, Column4, Column5, Column6, Column7, Column8)"""
-            )
-        )
-        assert a == b, "Data"
-
-        first.query("DROP DATABASE IF EXISTS db_ttl_columns SYNC")
-        second.query("DROP DATABASE IF EXISTS db_ttl_columns SYNC")
-
-
-class TaskSkipIndex:
-    def __init__(self, cluster):
-        self.cluster = cluster
-        self.zk_task_path = "/clickhouse-copier/task_skip_index"
-        self.container_task_file = "/task_skip_index.xml"
-
-        for instance_name, _ in cluster.instances.items():
-            instance = cluster.instances[instance_name]
-            instance.copy_file_to_container(
-                os.path.join(CURRENT_TEST_DIR, "./task_skip_index.xml"),
-                self.container_task_file,
-            )
-            print(
-                "Copied task file to container of '{}' instance. Path {}".format(
-                    instance_name, self.container_task_file
-                )
-            )
-
-    def start(self):
-        first = cluster.instances["first_of_two"]
-        second = cluster.instances["second_of_two"]
-
-        first.query("DROP DATABASE IF EXISTS db_skip_index SYNC")
-        second.query("DROP DATABASE IF EXISTS db_skip_index SYNC")
-
-        first.query("CREATE DATABASE IF NOT EXISTS db_skip_index;")
-        first.query(
-            """CREATE TABLE db_skip_index.source
-        (
-            Column1 UInt64,
-            Column2 Int32,
-            Column3 Date,
-            Column4 DateTime,
-            Column5 String,
-            INDEX a (Column1 * Column2, Column5) TYPE minmax GRANULARITY 3,
-            INDEX b (Column1 * length(Column5)) TYPE set(1000) GRANULARITY 4
-        )
-        ENGINE = MergeTree()
-        PARTITION BY (toYYYYMMDD(Column3), Column3)
-        PRIMARY KEY (Column1, Column2, Column3)
-        ORDER BY (Column1, Column2, Column3)
-        SETTINGS index_granularity = 8192"""
-        )
-
-        first.query(
-            """INSERT INTO db_skip_index.source SELECT * FROM generateRandom(
-            'Column1 UInt64, Column2 Int32, Column3 Date, Column4 DateTime, Column5 String', 1, 10, 2) LIMIT 100;"""
-        )
-
-        second.query("CREATE DATABASE IF NOT EXISTS db_skip_index;")
-        second.query(
-            """CREATE TABLE db_skip_index.destination
-        (
-            Column1 UInt64,
-            Column2 Int32,
-            Column3 Date,
-            Column4 DateTime,
-            Column5 String,
-            INDEX a (Column1 * Column2, Column5) TYPE minmax GRANULARITY 3,
-            INDEX b (Column1 * length(Column5)) TYPE set(1000) GRANULARITY 4
-        ) ENGINE = MergeTree()
-        PARTITION BY toYYYYMMDD(Column3)
-        ORDER BY (Column3, Column2, Column1);"""
-        )
-
-        print("Preparation completed")
-
-    def check(self):
-        first = cluster.instances["first_of_two"]
-        second = cluster.instances["second_of_two"]
-
-        a = first.query("SELECT count() from db_skip_index.source")
-        b = second.query("SELECT count() from db_skip_index.destination")
-        assert a == b, "Count"
-
-        a = TSV(
-            first.query(
-                """SELECT sipHash64(*) from db_skip_index.source
-            ORDER BY (Column1, Column2, Column3, Column4, Column5)"""
-            )
-        )
-        b = TSV(
-            second.query(
-                """SELECT sipHash64(*) from db_skip_index.destination
-            ORDER BY (Column1, Column2, Column3, Column4, Column5)"""
-            )
-        )
-        assert a == b, "Data"
-
-        first.query("DROP DATABASE IF EXISTS db_skip_index SYNC")
-        second.query("DROP DATABASE IF EXISTS db_skip_index SYNC")
-
-
-class TaskTTLMoveToVolume:
-    def __init__(self, cluster):
-        self.cluster = cluster
-        self.zk_task_path = "/clickhouse-copier/task_ttl_move_to_volume"
-        self.container_task_file = "/task_ttl_move_to_volume.xml"
-
-        for instance_name, _ in cluster.instances.items():
-            instance = cluster.instances[instance_name]
-            instance.copy_file_to_container(
-                os.path.join(CURRENT_TEST_DIR, "./task_ttl_move_to_volume.xml"),
-                self.container_task_file,
-            )
-            print(
-                "Copied task file to container of '{}' instance. Path {}".format(
-                    instance_name, self.container_task_file
-                )
-            )
-
-    def start(self):
-        first = cluster.instances["first_of_two"]
-        second = cluster.instances["first_of_two"]
-
-        first.query("DROP DATABASE IF EXISTS db_move_to_volume SYNC")
-        second.query("DROP DATABASE IF EXISTS db_move_to_volume SYNC")
-
-        first.query("CREATE DATABASE IF NOT EXISTS db_move_to_volume;")
-        first.query(
-            """CREATE TABLE db_move_to_volume.source
-        (
-            Column1 UInt64,
-            Column2 Int32,
-            Column3 Date,
-            Column4 DateTime,
-            Column5 String
-        )
-        ENGINE = MergeTree()
-        PARTITION BY (toYYYYMMDD(Column3), Column3)
-        PRIMARY KEY (Column1, Column2, Column3)
-        ORDER BY (Column1, Column2, Column3)
-        TTL Column3 + INTERVAL 1 MONTH TO VOLUME 'external'
-        SETTINGS storage_policy = 'external_with_jbods';"""
-        )
-
-        first.query(
-            """INSERT INTO db_move_to_volume.source SELECT * FROM generateRandom(
-            'Column1 UInt64, Column2 Int32, Column3 Date, Column4 DateTime, Column5 String', 1, 10, 2) LIMIT 100;"""
-        )
-
-        second.query("CREATE DATABASE IF NOT EXISTS db_move_to_volume;")
-        second.query(
-            """CREATE TABLE db_move_to_volume.destination
-        (
-            Column1 UInt64,
-            Column2 Int32,
-            Column3 Date,
-            Column4 DateTime,
-            Column5 String
-        ) ENGINE = MergeTree()
-        PARTITION BY toYYYYMMDD(Column3)
-        ORDER BY (Column3, Column2, Column1)
-        TTL Column3 + INTERVAL 1 MONTH TO VOLUME 'external'
-        SETTINGS storage_policy = 'external_with_jbods';"""
-        )
-
-        print("Preparation completed")
-
-    def check(self):
-        first = cluster.instances["first_of_two"]
-        second = cluster.instances["second_of_two"]
-
-        a = first.query("SELECT count() from db_move_to_volume.source")
-        b = second.query("SELECT count() from db_move_to_volume.destination")
-        assert a == b, "Count"
-
-        a = TSV(
-            first.query(
-                """SELECT sipHash64(*) from db_move_to_volume.source
-            ORDER BY (Column1, Column2, Column3, Column4, Column5)"""
-            )
-        )
-        b = TSV(
-            second.query(
-                """SELECT sipHash64(*) from db_move_to_volume.destination
-            ORDER BY (Column1, Column2, Column3, Column4, Column5)"""
-            )
-        )
-        assert a == b, "Data"
-
-        first.query("DROP DATABASE IF EXISTS db_move_to_volume SYNC")
-        second.query("DROP DATABASE IF EXISTS db_move_to_volume SYNC")
-
-
-class TaskDropTargetPartition:
-    def __init__(self, cluster):
-        self.cluster = cluster
-        self.zk_task_path = "/clickhouse-copier/task_drop_target_partition"
-        self.container_task_file = "/task_drop_target_partition.xml"
-
-        for instance_name, _ in cluster.instances.items():
-            instance = cluster.instances[instance_name]
-            instance.copy_file_to_container(
-                os.path.join(CURRENT_TEST_DIR, "./task_drop_target_partition.xml"),
-                self.container_task_file,
-            )
-            print(
-                "Copied task file to container of '{}' instance. Path {}".format(
-                    instance_name, self.container_task_file
-                )
-            )
-
-    def start(self):
-        first = cluster.instances["first_of_two"]
-        second = cluster.instances["second_of_two"]
-
-        first.query("DROP DATABASE IF EXISTS db_drop_target_partition SYNC")
-        second.query("DROP DATABASE IF EXISTS db_drop_target_partition SYNC")
-
-        first.query("CREATE DATABASE IF NOT EXISTS db_drop_target_partition;")
-        first.query(
-            """CREATE TABLE db_drop_target_partition.source
-        (
-            Column1 UInt64,
-            Column2 Int32,
-            Column3 Date,
-            Column4 DateTime,
-            Column5 String
-        )
-        ENGINE = MergeTree()
-        PARTITION BY (toYYYYMMDD(Column3), Column3)
-        PRIMARY KEY (Column1, Column2, Column3)
-        ORDER BY (Column1, Column2, Column3);"""
-        )
-
-        first.query(
-            """INSERT INTO db_drop_target_partition.source SELECT * FROM generateRandom(
-            'Column1 UInt64, Column2 Int32, Column3 Date, Column4 DateTime, Column5 String', 1, 10, 2) LIMIT 100;"""
-        )
-
-        second.query("CREATE DATABASE IF NOT EXISTS db_drop_target_partition;")
-        second.query(
-            """CREATE TABLE db_drop_target_partition.destination
-        (
-            Column1 UInt64,
-            Column2 Int32,
-            Column3 Date,
-            Column4 DateTime,
-            Column5 String
-        ) ENGINE = MergeTree()
-        PARTITION BY toYYYYMMDD(Column3)
-        ORDER BY (Column3, Column2, Column1);"""
-        )
-
-        # Insert data in target too. It has to be dropped.
-        first.query(
-            """INSERT INTO db_drop_target_partition.destination SELECT * FROM db_drop_target_partition.source;"""
-        )
-
-        print("Preparation completed")
-
-    def check(self):
-        first = cluster.instances["first_of_two"]
-        second = cluster.instances["second_of_two"]
-
-        a = first.query("SELECT count() from db_drop_target_partition.source")
-        b = second.query("SELECT count() from db_drop_target_partition.destination")
-        assert a == b, "Count"
-
-        a = TSV(
-            first.query(
-                """SELECT sipHash64(*) from db_drop_target_partition.source
-            ORDER BY (Column1, Column2, Column3, Column4, Column5)"""
-            )
-        )
-        b = TSV(
-            second.query(
-                """SELECT sipHash64(*) from db_drop_target_partition.destination
-            ORDER BY (Column1, Column2, Column3, Column4, Column5)"""
-            )
-        )
-        assert a == b, "Data"
-
-        first.query("DROP DATABASE IF EXISTS db_drop_target_partition SYNC")
-        second.query("DROP DATABASE IF EXISTS db_drop_target_partition SYNC")
-
-
-def execute_task(started_cluster, task, cmd_options):
-    task.start()
-
-    zk = started_cluster.get_kazoo_client("zoo1")
-    print("Use ZooKeeper server: {}:{}".format(zk.hosts[0][0], zk.hosts[0][1]))
-
-    # Run cluster-copier processes on each node
-    docker_api = started_cluster.docker_client.api
-    copiers_exec_ids = []
-
-    cmd = [
-        "/usr/bin/clickhouse",
-        "copier",
-        "--config",
-        "/etc/clickhouse-server/config-copier.xml",
-        "--task-path",
-        task.zk_task_path,
-        "--task-file",
-        task.container_task_file,
-        "--task-upload-force",
-        "true",
-        "--base-dir",
-        "/var/log/clickhouse-server/copier",
-    ]
-    cmd += cmd_options
-
-    print(cmd)
-
-    for instance_name in started_cluster.instances.keys():
-        instance = started_cluster.instances[instance_name]
-        container = instance.get_docker_handle()
-        instance.copy_file_to_container(
-            os.path.join(CURRENT_TEST_DIR, "configs_two_nodes/config-copier.xml"),
-            "/etc/clickhouse-server/config-copier.xml",
-        )
-        logging.info("Copied copier config to {}".format(instance.name))
-        exec_id = docker_api.exec_create(container.id, cmd, stderr=True)
-        output = docker_api.exec_start(exec_id).decode("utf8")
-        logging.info(output)
-        copiers_exec_ids.append(exec_id)
-        logging.info(
-            "Copier for {} ({}) has started".format(instance.name, instance.ip_address)
-        )
-
-    # time.sleep(1000)
-
-    # Wait for copiers stopping and check their return codes
-    for exec_id, instance in zip(
-        copiers_exec_ids, iter(started_cluster.instances.values())
-    ):
-        while True:
-            res = docker_api.exec_inspect(exec_id)
-            if not res["Running"]:
-                break
-            time.sleep(1)
-
-        assert res["ExitCode"] == 0, "Instance: {} ({}). Info: {}".format(
-            instance.name, instance.ip_address, repr(res)
-        )
-
-    try:
-        task.check()
-    finally:
-        zk.delete(task.zk_task_path, recursive=True)
diff --git a/tests/integration/test_config_xml_full/configs/config.xml b/tests/integration/test_config_xml_full/configs/config.xml
index ac59b3428e8..628e1432350 100644
--- a/tests/integration/test_config_xml_full/configs/config.xml
+++ b/tests/integration/test_config_xml_full/configs/config.xml
@@ -72,7 +72,7 @@
     <http_port>8123</http_port>
 
     <!-- Port for interaction by native protocol with:
-         - clickhouse-client and other native ClickHouse tools (clickhouse-benchmark, clickhouse-copier);
+         - clickhouse-client and other native ClickHouse tools (clickhouse-benchmark);
          - clickhouse-server with other clickhouse-servers for distributed query processing;
          - ClickHouse drivers and applications supporting native protocol
          (this protocol is also informally called as "the TCP protocol");

From 7365da9293adf2b63bef173769e171581d727d82 Mon Sep 17 00:00:00 2001
From: Nikita Mikhaylov <nikitamikhaylov@clickhouse.com>
Date: Thu, 7 Mar 2024 19:30:31 +0000
Subject: [PATCH 082/443] Better

---
 docs/ru/operations/backup.md | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/docs/ru/operations/backup.md b/docs/ru/operations/backup.md
index 50ee6b45e09..44877ff8071 100644
--- a/docs/ru/operations/backup.md
+++ b/docs/ru/operations/backup.md
@@ -24,6 +24,8 @@ sidebar_label: "Резервное копирование данных"
 
 Некоторые локальные файловые системы позволяют делать снимки (например, [ZFS](https://en.wikipedia.org/wiki/ZFS)), но они могут быть не лучшим выбором для обслуживания живых запросов. Возможным решением является создание дополнительных реплик с такой файловой системой и исключение их из [Distributed](../engines/table-engines/special/distributed.md) таблиц, используемых для запросов `SELECT`. Снимки на таких репликах будут недоступны для запросов, изменяющих данные. В качестве бонуса, эти реплики могут иметь особые конфигурации оборудования с большим количеством дисков, подключенных к серверу, что будет экономически эффективным.
 
+Для небольших объёмов данных можно применять `INSERT INTO ... SELECT ...` в удалённые таблицы.
+
 ## Манипуляции с партициями {#manipuliatsii-s-partitsiiami}
 
 ClickHouse позволяет использовать запрос `ALTER TABLE ... FREEZE PARTITION ...` для создания локальной копии партиций таблицы. Это реализуется с помощью жестких ссылок (hardlinks) на каталог `/var/lib/clickhouse/shadow/`, поэтому такая копия обычно не занимает дополнительное место на диске для старых данных. Созданные копии файлов не обрабатываются сервером ClickHouse, поэтому вы можете просто оставить их там: у вас будет простая резервная копия, которая не требует дополнительной внешней системы, однако при аппаратных проблемах вы можете утратить и актуальные данные и сохраненную копию. По этой причине, лучше удаленно скопировать их в другое место, а затем удалить локальную копию. Распределенные файловые системы и хранилища объектов по-прежнему являются хорошими вариантами для этого, однако можно использовать и обычные присоединенные файловые серверы с достаточно большой ёмкостью (в этом случае передача будет происходить через сетевую файловую систему или, возможно, [rsync](https://en.wikipedia.org/wiki/Rsync)).

From 4c3fa4e8642780d2e27ed4f31590345646f0bb26 Mon Sep 17 00:00:00 2001
From: Nikolai Kochetov <nk@clickhouse.com>
Date: Fri, 8 Mar 2024 09:42:49 +0000
Subject: [PATCH 083/443] Use global scalars cache in analyzer.

---
 src/Analyzer/Passes/QueryAnalysisPass.cpp | 47 ++++++++++++++++++++---
 1 file changed, 42 insertions(+), 5 deletions(-)

diff --git a/src/Analyzer/Passes/QueryAnalysisPass.cpp b/src/Analyzer/Passes/QueryAnalysisPass.cpp
index c62641ca05c..e7c48b81fbc 100644
--- a/src/Analyzer/Passes/QueryAnalysisPass.cpp
+++ b/src/Analyzer/Passes/QueryAnalysisPass.cpp
@@ -86,6 +86,7 @@
 namespace ProfileEvents
 {
     extern const Event ScalarSubqueriesGlobalCacheHit;
+    extern const Event ScalarSubqueriesLocalCacheHit;
     extern const Event ScalarSubqueriesCacheMiss;
 }
 
@@ -1444,7 +1445,8 @@ private:
     std::unordered_map<QueryTreeNodePtr, size_t> node_to_tree_size;
 
     /// Global scalar subquery to scalar value map
-    std::unordered_map<QueryTreeNodePtrWithHash, Block> scalar_subquery_to_scalar_value;
+    std::unordered_map<QueryTreeNodePtrWithHash, Block> scalar_subquery_to_scalar_value_local;
+    std::unordered_map<QueryTreeNodePtrWithHash, Block> scalar_subquery_to_scalar_value_global;
 
     const bool only_analyze;
 };
@@ -1951,6 +1953,24 @@ QueryTreeNodePtr QueryAnalyzer::tryGetLambdaFromSQLUserDefinedFunctions(const st
     return result_node;
 }
 
+bool subtreeHasViewSource(const IQueryTreeNode * node, const Context & context)
+{
+    if (!node)
+        return false;
+
+    if (const auto * table_node = node->as<TableNode>())
+    {
+        if (table_node->getStorageID().getFullNameNotQuoted() == context.getViewSource()->getStorageID().getFullNameNotQuoted())
+            return true;
+    }
+
+    for (const auto & child : node->getChildren())
+        if (subtreeHasViewSource(child.get(), context))
+            return true;
+
+    return false;
+}
+
 /// Evaluate scalar subquery and perform constant folding if scalar subquery does not have constant value
 void QueryAnalyzer::evaluateScalarSubqueryIfNeeded(QueryTreeNodePtr & node, IdentifierResolveScope & scope)
 {
@@ -1970,13 +1990,28 @@ void QueryAnalyzer::evaluateScalarSubqueryIfNeeded(QueryTreeNodePtr & node, Iden
     node_without_alias->removeAlias();
 
     QueryTreeNodePtrWithHash node_with_hash(node_without_alias);
-    auto scalar_value_it = scalar_subquery_to_scalar_value.find(node_with_hash);
+    auto str_hash = DB::toString(node_with_hash.hash);
 
-    if (scalar_value_it != scalar_subquery_to_scalar_value.end())
+    bool can_use_global_scalars = !(context->getViewSource() && subtreeHasViewSource(node_without_alias.get(), *context));
+
+    auto & scalars_cache = can_use_global_scalars ? scalar_subquery_to_scalar_value_global : scalar_subquery_to_scalar_value_local;
+    auto scalar_value_it = scalars_cache.find(node_with_hash);
+
+    if (scalar_value_it != scalars_cache.end())
     {
-        ProfileEvents::increment(ProfileEvents::ScalarSubqueriesGlobalCacheHit);
+        if (can_use_global_scalars)
+            ProfileEvents::increment(ProfileEvents::ScalarSubqueriesGlobalCacheHit);
+        else
+            ProfileEvents::increment(ProfileEvents::ScalarSubqueriesLocalCacheHit);
+
         scalar_block = scalar_value_it->second;
     }
+    else if (context->hasQueryContext() && can_use_global_scalars && context->getQueryContext()->hasScalar(str_hash))
+    {
+        scalar_block = context->getQueryContext()->getScalar(str_hash);
+        scalar_subquery_to_scalar_value_global.emplace(node_with_hash, scalar_block);
+        ProfileEvents::increment(ProfileEvents::ScalarSubqueriesGlobalCacheHit);
+    }
     else
     {
         ProfileEvents::increment(ProfileEvents::ScalarSubqueriesCacheMiss);
@@ -2087,7 +2122,9 @@ void QueryAnalyzer::evaluateScalarSubqueryIfNeeded(QueryTreeNodePtr & node, Iden
             }
         }
 
-        scalar_subquery_to_scalar_value.emplace(node_with_hash, scalar_block);
+        scalars_cache.emplace(node_with_hash, scalar_block);
+        if (can_use_global_scalars && context->hasQueryContext())
+            context->getQueryContext()->addScalar(str_hash, scalar_block);
     }
 
     const auto & scalar_column_with_type = scalar_block.safeGetByPosition(0);

From 435e78d01dec0448002d0b7057e4d8b4307aea59 Mon Sep 17 00:00:00 2001
From: Dmitry Novik <mrnovikd@gmail.com>
Date: Fri, 8 Mar 2024 12:33:40 +0100
Subject: [PATCH 084/443] Analyzer: fix
 01244_optimize_distributed_group_by_sharding_key

---
 tests/analyzer_tech_debt.txt                                    | 1 -
 .../01244_optimize_distributed_group_by_sharding_key.sql        | 2 +-
 2 files changed, 1 insertion(+), 2 deletions(-)

diff --git a/tests/analyzer_tech_debt.txt b/tests/analyzer_tech_debt.txt
index dc6284d20c5..f6a662177c8 100644
--- a/tests/analyzer_tech_debt.txt
+++ b/tests/analyzer_tech_debt.txt
@@ -4,7 +4,6 @@
 01062_pm_all_join_with_block_continuation
 01083_expressions_in_engine_arguments
 01155_rename_move_materialized_view
-01244_optimize_distributed_group_by_sharding_key
 01584_distributed_buffer_cannot_find_column
 01624_soft_constraints
 01747_join_view_filter_dictionary
diff --git a/tests/queries/0_stateless/01244_optimize_distributed_group_by_sharding_key.sql b/tests/queries/0_stateless/01244_optimize_distributed_group_by_sharding_key.sql
index 0d24b238d64..e80f41b173c 100644
--- a/tests/queries/0_stateless/01244_optimize_distributed_group_by_sharding_key.sql
+++ b/tests/queries/0_stateless/01244_optimize_distributed_group_by_sharding_key.sql
@@ -42,7 +42,7 @@ select 'GROUP BY number ORDER BY number DESC';
 select count(), * from dist_01247 group by number order by number desc;
 
 select 'GROUP BY toString(number)';
-select count(), * from dist_01247 group by toString(number);
+select count(), any(number) from dist_01247 group by toString(number);
 
 select 'GROUP BY number%2';
 select count(), any(number) from dist_01247 group by number%2;

From cacbd3ce3450e9a2f8915b392ce8ec08f9662282 Mon Sep 17 00:00:00 2001
From: Max Kainov <max.kainov@clickhouse.com>
Date: Fri, 8 Mar 2024 14:13:23 +0000
Subject: [PATCH 085/443] CI: support merge queue event in pr_info

 #do_not_test
---
 tests/ci/pr_info.py | 56 +++++++++++++++++++++++++++++++--------------
 1 file changed, 39 insertions(+), 17 deletions(-)

diff --git a/tests/ci/pr_info.py b/tests/ci/pr_info.py
index aba32d88c0a..6f4b400f7a8 100644
--- a/tests/ci/pr_info.py
+++ b/tests/ci/pr_info.py
@@ -44,11 +44,12 @@ RETRY_SLEEP = 0
 
 
 class EventType:
-    UNKNOWN = 0
-    PUSH = 1
-    PULL_REQUEST = 2
-    SCHEDULE = 3
-    DISPATCH = 4
+    UNKNOWN = "unknown"
+    PUSH = "commits"
+    PULL_REQUEST = "pull_request"
+    SCHEDULE = "schedule"
+    DISPATCH = "dispatch"
+    MERGE_QUEUE = "merge_group"
 
 
 def get_pr_for_commit(sha, ref):
@@ -114,6 +115,12 @@ class PRInfo:
         # release_pr and merged_pr are used for docker images additional cache
         self.release_pr = 0
         self.merged_pr = 0
+        self.labels = set()
+
+        repo_prefix = f"{GITHUB_SERVER_URL}/{GITHUB_REPOSITORY}"
+        self.task_url = GITHUB_RUN_URL
+        self.repo_full_name = GITHUB_REPOSITORY
+
         self.event_type = EventType.UNKNOWN
         ref = github_event.get("ref", "refs/heads/master")
         if ref and ref.startswith("refs/heads/"):
@@ -154,10 +161,6 @@ class PRInfo:
             else:
                 self.sha = github_event["pull_request"]["head"]["sha"]
 
-            repo_prefix = f"{GITHUB_SERVER_URL}/{GITHUB_REPOSITORY}"
-            self.task_url = GITHUB_RUN_URL
-
-            self.repo_full_name = GITHUB_REPOSITORY
             self.commit_html_url = f"{repo_prefix}/commits/{self.sha}"
             self.pr_html_url = f"{repo_prefix}/pull/{self.number}"
 
@@ -176,7 +179,7 @@ class PRInfo:
             self.body = github_event["pull_request"]["body"]
             self.labels = {
                 label["name"] for label in github_event["pull_request"]["labels"]
-            }  # type: Set[str]
+            }
 
             self.user_login = github_event["pull_request"]["user"]["login"]  # type: str
             self.user_orgs = set()  # type: Set[str]
@@ -191,6 +194,28 @@ class PRInfo:
 
             self.diff_urls.append(self.compare_pr_url(github_event["pull_request"]))
 
+        elif (
+            EventType.MERGE_QUEUE in github_event
+        ):  # pull request and other similar events
+            self.event_type = EventType.MERGE_QUEUE
+            # FIXME: need pr? we can parse it from ["head_ref": "refs/heads/gh-readonly-queue/test-merge-queue/pr-6751-4690229995a155e771c52e95fbd446d219c069bf"]
+            self.number = 0
+            self.sha = github_event[EventType.MERGE_QUEUE]["head_sha"]
+            self.base_ref = github_event[EventType.MERGE_QUEUE]["base_ref"]
+            base_sha = github_event[EventType.MERGE_QUEUE]["base_sha"]  # type: str
+            # ClickHouse/ClickHouse
+            self.base_name = github_event["repository"]["full_name"]
+            # any_branch-name - the name of working branch name
+            self.head_ref = github_event[EventType.MERGE_QUEUE]["head_ref"]
+            # UserName/ClickHouse or ClickHouse/ClickHouse
+            self.head_name = self.base_name
+            self.user_login = github_event["sender"]["login"]
+            self.diff_urls.append(
+                github_event["repository"]["compare_url"]
+                .replace("{base}", base_sha)
+                .replace("{head}", self.sha)
+            )
+
         elif "commits" in github_event:
             self.event_type = EventType.PUSH
             # `head_commit` always comes with `commits`
@@ -203,10 +228,8 @@ class PRInfo:
                     logging.error("Failed to convert %s to integer", merged_pr)
             self.sha = github_event["after"]
             pull_request = get_pr_for_commit(self.sha, github_event["ref"])
-            repo_prefix = f"{GITHUB_SERVER_URL}/{GITHUB_REPOSITORY}"
-            self.task_url = GITHUB_RUN_URL
             self.commit_html_url = f"{repo_prefix}/commits/{self.sha}"
-            self.repo_full_name = GITHUB_REPOSITORY
+
             if pull_request is None or pull_request["state"] == "closed":
                 # it's merged PR to master
                 self.number = 0
@@ -272,11 +295,7 @@ class PRInfo:
                 "GITHUB_SHA", "0000000000000000000000000000000000000000"
             )
             self.number = 0
-            self.labels = set()
-            repo_prefix = f"{GITHUB_SERVER_URL}/{GITHUB_REPOSITORY}"
-            self.task_url = GITHUB_RUN_URL
             self.commit_html_url = f"{repo_prefix}/commits/{self.sha}"
-            self.repo_full_name = GITHUB_REPOSITORY
             self.pr_html_url = f"{repo_prefix}/commits/{ref}"
             self.base_ref = ref
             self.base_name = self.repo_full_name
@@ -300,6 +319,9 @@ class PRInfo:
     def is_scheduled(self):
         return self.event_type == EventType.SCHEDULE
 
+    def is_merge_queue(self):
+        return self.event_type == EventType.MERGE_QUEUE
+
     def is_dispatched(self):
         return self.event_type == EventType.DISPATCH
 

From 2196c75dd8ddaeb1d2f18ca7b05fb4ae37550a4b Mon Sep 17 00:00:00 2001
From: johnnymatthews <9611008+johnnymatthews@users.noreply.github.com>
Date: Fri, 8 Mar 2024 11:07:04 -0400
Subject: [PATCH 086/443] Adds substring-UTF8 docs.

---
 .../functions/string-functions.md             | 69 ++++++++++++++++++-
 1 file changed, 67 insertions(+), 2 deletions(-)

diff --git a/docs/en/sql-reference/functions/string-functions.md b/docs/en/sql-reference/functions/string-functions.md
index 3b49e4954ed..f9c3f91a12b 100644
--- a/docs/en/sql-reference/functions/string-functions.md
+++ b/docs/en/sql-reference/functions/string-functions.md
@@ -588,8 +588,41 @@ Result:
 
 ## substringUTF8
 
-Like `substring` but for Unicode code points. Assumes that the string contains valid UTF-8 encoded text. If this assumption is violated, no exception is thrown and the result is undefined.
+Returns the substring of a string `s` which starts at the specified byte index `offset` for Unicode code points. Byte counting starts from `1`. If `offset` is `0`, an empty string is returned. If `offset` is negative, the substring starts `pos` characters from the end of the string, rather than from the beginning. An optional argument `length` specifies the maximum number of bytes the returned substring may have.
 
+Assumes that the string contains valid UTF-8 encoded text. If this assumption is violated, no exception is thrown and the result is undefined.
+
+**Syntax**
+
+```sql
+substringUTF8(s, offset[, length])
+```
+
+**Arguments**
+
+- `s`: The string to calculate a substring from. [String](../../sql-reference/data-types/string.md), [FixedString](../../sql-reference/data-types/fixedstring.md) or [Enum](../../sql-reference/data-types/enum.md)
+- `offset`: The starting position of the substring in `s` . [(U)Int*](../../sql-reference/data-types/int-uint.md).
+- `length`: The maximum length of the substring. [(U)Int*](../../sql-reference/data-types/int-uint.md). Optional.
+
+**Returned value**
+
+A substring of `s` with `length` many bytes, starting at index `offset`.
+
+**Implementation details**
+
+Assumes that the string contains valid UTF-8 encoded text. If this assumption is violated, no exception is thrown and the result is undefined.
+
+**Example**
+
+```sql
+SELECT 'database' AS string, substringUTF8(string, 5), substringUTF8(string, 5, 1)
+```
+
+```response
+┌─string───┬─substringUTF8('database', 5)─┬─substringUTF8('database', 5, 1)─┐
+│ database │ base                         │ b                               │
+└──────────┴──────────────────────────────┴─────────────────────────────────┘
+```
 
 ## substringIndex
 
@@ -624,7 +657,39 @@ Result:
 
 ## substringIndexUTF8
 
-Like `substringIndex` but for Unicode code points. Assumes that the string contains valid UTF-8 encoded text. If this assumption is violated, no exception is thrown and the result is undefined.
+Returns the substring of `s` before `count` occurrences of the delimiter `delim`, specifically for Unicode code points.
+
+Assumes that the string contains valid UTF-8 encoded text. If this assumption is violated, no exception is thrown and the result is undefined.
+
+**Syntax**
+
+```sql
+substringIndexUTF8(s, delim, count)
+```
+
+**Arguments**
+
+- `s`: The string to extract substring from. [String](../../sql-reference/data-types/string.md).
+- `delim`: The character to split. [String](../../sql-reference/data-types/string.md).
+- `count`: The number of occurrences of the delimiter to count before extracting the substring. If count is positive, everything to the left of the final delimiter (counting from the left) is returned. If count is negative, everything to the right of the final delimiter (counting from the right) is returned. [UInt or Int](../data-types/int-uint.md)
+
+**Returned value**
+
+A substring [String](../../sql-reference/data-types/string.md) of `s` before `count` occurrences of `delim`.
+
+**Implementation details**
+
+Assumes that the string contains valid UTF-8 encoded text. If this assumption is violated, no exception is thrown and the result is undefined.
+
+**Example**
+
+```sql
+SELECT substringIndexUTF8('www.clickhouse.com', '.', 2)
+```
+
+```response
+www.clickhouse
+```
 
 ## appendTrailingCharIfAbsent
 

From 0336ef3557b0c3c05ef974a4c6aa6771b3aa0757 Mon Sep 17 00:00:00 2001
From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com>
Date: Fri, 8 Mar 2024 16:14:11 +0100
Subject: [PATCH 087/443] reload CI


From 6f31a77f704b4a0400d433541b67236f6d87a958 Mon Sep 17 00:00:00 2001
From: Nikolai Kochetov <nk@clickhouse.com>
Date: Fri, 8 Mar 2024 15:33:45 +0000
Subject: [PATCH 088/443] Fix test 02174_cte_scalar_cache_mv

---
 src/Analyzer/Passes/QueryAnalysisPass.cpp     |  7 +-
 .../02174_cte_scalar_cache_mv.reference       | 63 ++++++++++++
 .../0_stateless/02174_cte_scalar_cache_mv.sql | 99 +++++++++++++++++++
 3 files changed, 165 insertions(+), 4 deletions(-)

diff --git a/src/Analyzer/Passes/QueryAnalysisPass.cpp b/src/Analyzer/Passes/QueryAnalysisPass.cpp
index e7c48b81fbc..1da5d86edf3 100644
--- a/src/Analyzer/Passes/QueryAnalysisPass.cpp
+++ b/src/Analyzer/Passes/QueryAnalysisPass.cpp
@@ -1992,19 +1992,18 @@ void QueryAnalyzer::evaluateScalarSubqueryIfNeeded(QueryTreeNodePtr & node, Iden
     QueryTreeNodePtrWithHash node_with_hash(node_without_alias);
     auto str_hash = DB::toString(node_with_hash.hash);
 
-    bool can_use_global_scalars = !(context->getViewSource() && subtreeHasViewSource(node_without_alias.get(), *context));
+    bool can_use_global_scalars = !only_analyze && !(context->getViewSource() && subtreeHasViewSource(node_without_alias.get(), *context));
 
     auto & scalars_cache = can_use_global_scalars ? scalar_subquery_to_scalar_value_global : scalar_subquery_to_scalar_value_local;
-    auto scalar_value_it = scalars_cache.find(node_with_hash);
 
-    if (scalar_value_it != scalars_cache.end())
+    if (scalars_cache.contains(node_with_hash))
     {
         if (can_use_global_scalars)
             ProfileEvents::increment(ProfileEvents::ScalarSubqueriesGlobalCacheHit);
         else
             ProfileEvents::increment(ProfileEvents::ScalarSubqueriesLocalCacheHit);
 
-        scalar_block = scalar_value_it->second;
+        scalar_block = scalars_cache.at(node_with_hash);
     }
     else if (context->hasQueryContext() && can_use_global_scalars && context->getQueryContext()->hasScalar(str_hash))
     {
diff --git a/tests/queries/0_stateless/02174_cte_scalar_cache_mv.reference b/tests/queries/0_stateless/02174_cte_scalar_cache_mv.reference
index 8ec3608317f..dcfab092b5c 100644
--- a/tests/queries/0_stateless/02174_cte_scalar_cache_mv.reference
+++ b/tests/queries/0_stateless/02174_cte_scalar_cache_mv.reference
@@ -19,6 +19,48 @@
 94	94	94	94	5
 99	99	99	99	5
 02177_MV	7	80	22
+4	4	4	4	5
+9	9	9	9	5
+14	14	14	14	5
+19	19	19	19	5
+24	24	24	24	5
+29	29	29	29	5
+34	34	34	34	5
+39	39	39	39	5
+44	44	44	44	5
+49	49	49	49	5
+54	54	54	54	5
+59	59	59	59	5
+64	64	64	64	5
+69	69	69	69	5
+74	74	74	74	5
+79	79	79	79	5
+84	84	84	84	5
+89	89	89	89	5
+94	94	94	94	5
+99	99	99	99	5
+02177_MV	0	0	22
+10
+40
+70
+100
+130
+160
+190
+220
+250
+280
+310
+340
+370
+400
+430
+460
+490
+520
+550
+580
+02177_MV_2	0	0	21
 10
 40
 70
@@ -61,3 +103,24 @@
 188
 198
 02177_MV_3	20	0	1
+8
+18
+28
+38
+48
+58
+68
+78
+88
+98
+108
+118
+128
+138
+148
+158
+168
+178
+188
+198
+02177_MV_3	19	0	2
diff --git a/tests/queries/0_stateless/02174_cte_scalar_cache_mv.sql b/tests/queries/0_stateless/02174_cte_scalar_cache_mv.sql
index 742d72fe2b2..ca54b9e1400 100644
--- a/tests/queries/0_stateless/02174_cte_scalar_cache_mv.sql
+++ b/tests/queries/0_stateless/02174_cte_scalar_cache_mv.sql
@@ -14,6 +14,8 @@ CREATE MATERIALIZED VIEW mv1 TO t2 AS
     FROM t1
     LIMIT 5;
 
+set allow_experimental_analyzer = 0;
+
 -- FIRST INSERT
 INSERT INTO t1
 WITH
@@ -58,8 +60,48 @@ WHERE
   AND query LIKE '-- FIRST INSERT\nINSERT INTO t1\n%'
   AND event_date >= yesterday() AND event_time > now() - interval 10 minute;
 
+truncate table t2;
+set allow_experimental_analyzer = 1;
+
+-- FIRST INSERT ANALYZER
+INSERT INTO t1
+WITH
+    (SELECT max(i) FROM t1) AS t1
+SELECT
+       number as i,
+       t1 + t1 + t1 AS j -- Using global cache
+FROM system.numbers
+LIMIT 100
+SETTINGS
+    min_insert_block_size_rows=5,
+    max_insert_block_size=5,
+    min_insert_block_size_rows_for_materialized_views=5,
+    max_block_size=5,
+    max_threads=1;
+
+SELECT k, l, m, n, count()
+FROM t2
+GROUP BY k, l, m, n
+ORDER BY k, l, m, n;
+
+SYSTEM FLUSH LOGS;
+
+SELECT
+    '02177_MV',
+    ProfileEvents['ScalarSubqueriesGlobalCacheHit'] as scalar_cache_global_hit,
+    ProfileEvents['ScalarSubqueriesLocalCacheHit'] as scalar_cache_local_hit,
+    ProfileEvents['ScalarSubqueriesCacheMiss'] as scalar_cache_miss
+FROM system.query_log
+WHERE
+      current_database = currentDatabase()
+  AND type = 'QueryFinish'
+  AND query LIKE '-- FIRST INSERT ANALYZER\nINSERT INTO t1\n%'
+  AND event_date >= yesterday() AND event_time > now() - interval 10 minute;
+
 DROP TABLE mv1;
 
+set allow_experimental_analyzer = 0;
+
 CREATE TABLE t3 (z Int64) ENGINE = Memory;
 CREATE MATERIALIZED VIEW mv2 TO t3 AS
 SELECT
@@ -91,8 +133,36 @@ WHERE
   AND query LIKE '-- SECOND INSERT\nINSERT INTO t1%'
   AND event_date >= yesterday() AND event_time > now() - interval 10 minute;
 
+truncate table t3;
+set allow_experimental_analyzer = 1;
+
+-- SECOND INSERT ANALYZER
+INSERT INTO t1
+SELECT 0 as i, number as j from numbers(100)
+SETTINGS
+    min_insert_block_size_rows=5,
+    max_insert_block_size=5,
+    min_insert_block_size_rows_for_materialized_views=5,
+    max_block_size=5,
+    max_threads=1;
+
+SELECT * FROM t3 ORDER BY z ASC;
+SYSTEM FLUSH LOGS;
+SELECT
+    '02177_MV_2',
+    ProfileEvents['ScalarSubqueriesGlobalCacheHit'] as scalar_cache_global_hit,
+    ProfileEvents['ScalarSubqueriesLocalCacheHit'] as scalar_cache_local_hit,
+    ProfileEvents['ScalarSubqueriesCacheMiss'] as scalar_cache_miss
+FROM system.query_log
+WHERE
+        current_database = currentDatabase()
+  AND type = 'QueryFinish'
+  AND query LIKE '-- SECOND INSERT ANALYZER\nINSERT INTO t1%'
+  AND event_date >= yesterday() AND event_time > now() - interval 10 minute;
+
 DROP TABLE mv2;
 
+set allow_experimental_analyzer = 0;
 
 CREATE TABLE t4 (z Int64) ENGINE = Memory;
 CREATE MATERIALIZED VIEW mv3 TO t4 AS
@@ -126,6 +196,35 @@ WHERE
   AND query LIKE '-- THIRD INSERT\nINSERT INTO t1%'
   AND event_date >= yesterday() AND event_time > now() - interval 10 minute;
 
+truncate table t4;
+set allow_experimental_analyzer = 1;
+
+-- THIRD INSERT ANALYZER
+INSERT INTO t1
+SELECT number as i, number as j from numbers(100)
+    SETTINGS
+    min_insert_block_size_rows=5,
+    max_insert_block_size=5,
+    min_insert_block_size_rows_for_materialized_views=5,
+    max_block_size=5,
+    max_threads=1;
+SYSTEM FLUSH LOGS;
+
+SELECT * FROM t4 ORDER BY z ASC;
+
+SELECT
+    '02177_MV_3',
+    ProfileEvents['ScalarSubqueriesGlobalCacheHit'] as scalar_cache_global_hit,
+    ProfileEvents['ScalarSubqueriesLocalCacheHit'] as scalar_cache_local_hit,
+    ProfileEvents['ScalarSubqueriesCacheMiss'] as scalar_cache_miss
+FROM system.query_log
+WHERE
+        current_database = currentDatabase()
+  AND type = 'QueryFinish'
+  AND query LIKE '-- THIRD INSERT ANALYZER\nINSERT INTO t1%'
+  AND event_date >= yesterday() AND event_time > now() - interval 10 minute;
+
+
 DROP TABLE mv3;
 DROP TABLE t1;
 DROP TABLE t2;

From ceb82cb3bd8579c4ba8accd6e701f4a5cd05d505 Mon Sep 17 00:00:00 2001
From: Nikolai Kochetov <nk@clickhouse.com>
Date: Fri, 8 Mar 2024 15:34:19 +0000
Subject: [PATCH 089/443] Fix test 02174_cte_scalar_cache_mv

---
 tests/analyzer_tech_debt.txt | 1 -
 1 file changed, 1 deletion(-)

diff --git a/tests/analyzer_tech_debt.txt b/tests/analyzer_tech_debt.txt
index dc6284d20c5..e856da394d2 100644
--- a/tests/analyzer_tech_debt.txt
+++ b/tests/analyzer_tech_debt.txt
@@ -11,7 +11,6 @@
 01761_cast_to_enum_nullable
 01925_join_materialized_columns
 01952_optimize_distributed_group_by_sharding_key
-02174_cte_scalar_cache_mv
 02354_annoy
 # Check after constants refactoring
 02901_parallel_replicas_rollup

From 4997f95426786a037328d4ed2bbce2245144eb1f Mon Sep 17 00:00:00 2001
From: Han Fei <hanfei19910905@gmail.com>
Date: Fri, 8 Mar 2024 17:30:58 +0100
Subject: [PATCH 090/443] fix build

---
 programs/keeper/Keeper.cpp                |  3 +++
 programs/server/Server.cpp                |  5 +++++
 src/Common/CgroupsMemoryUsageObserver.cpp |  4 +---
 src/Common/CgroupsMemoryUsageObserver.h   | 10 ++++++++--
 4 files changed, 17 insertions(+), 5 deletions(-)

diff --git a/programs/keeper/Keeper.cpp b/programs/keeper/Keeper.cpp
index 76dd8cb15a5..c2ad81a3227 100644
--- a/programs/keeper/Keeper.cpp
+++ b/programs/keeper/Keeper.cpp
@@ -630,6 +630,9 @@ try
     {
         observer.emplace(std::chrono::seconds(cgroups_memory_observer_wait_time));
         observer->startThread();
+        observer->setOnMemoryLimitUpdate([&](){
+            main_config_reloader->reload();
+        });
     }
 
 
diff --git a/programs/server/Server.cpp b/programs/server/Server.cpp
index ee55cfd1837..c81e9b56e35 100644
--- a/programs/server/Server.cpp
+++ b/programs/server/Server.cpp
@@ -1701,6 +1701,11 @@ try
         throw;
     }
 
+    if (cgroups_memory_usage_observer)
+        cgroups_memory_usage_observer->setOnMemoryLimitUpdate([&](){
+            main_config_reloader->reload();
+        });
+
     /// Reload config in SYSTEM RELOAD CONFIG query.
     global_context->setConfigReloadCallback([&]()
     {
diff --git a/src/Common/CgroupsMemoryUsageObserver.cpp b/src/Common/CgroupsMemoryUsageObserver.cpp
index 5f24c2553b5..3fda51a119f 100644
--- a/src/Common/CgroupsMemoryUsageObserver.cpp
+++ b/src/Common/CgroupsMemoryUsageObserver.cpp
@@ -8,7 +8,6 @@
 #include <IO/ReadBufferFromFile.h>
 #include <IO/ReadBufferFromFileDescriptor.h>
 #include <IO/ReadHelpers.h>
-#include <Interpreters/Context.h>
 #include <base/cgroupsv2.h>
 #include <base/getMemoryAmount.h>
 #include <base/sleep.h>
@@ -291,8 +290,7 @@ void CgroupsMemoryUsageObserver::runThread()
                 last_memory_amount = memory_limit;
                 /// if we find memory amount changes, we just reload config.
                 /// Reloading config will check the memory amount again and calculate soft/hard limit again.
-                auto global_context = getContext()->getGlobalContext();
-                global_context->reloadConfig();
+                on_memory_limit_update();
             }
             std::lock_guard<std::mutex> set_limit_lock(set_limit_mutex);
             if (soft_limit > 0 && hard_limit > 0)
diff --git a/src/Common/CgroupsMemoryUsageObserver.h b/src/Common/CgroupsMemoryUsageObserver.h
index 6edf2e2049d..639433b5016 100644
--- a/src/Common/CgroupsMemoryUsageObserver.h
+++ b/src/Common/CgroupsMemoryUsageObserver.h
@@ -1,7 +1,6 @@
 #pragma once
 
 #include <Common/ThreadPool.h>
-#include <Interpreters/Context_fwd.h>
 
 #include <atomic>
 #include <chrono>
@@ -15,7 +14,7 @@ namespace DB
 /// - When the soft memory limit is hit, drop jemalloc cache.
 /// - When the hard memory limit is hit, update MemoryTracking metric to throw memory exceptions faster.
 #if defined(OS_LINUX)
-class CgroupsMemoryUsageObserver : public WithContext
+class CgroupsMemoryUsageObserver
 {
 public:
     enum class CgroupsVersion
@@ -28,6 +27,10 @@ public:
     ~CgroupsMemoryUsageObserver();
 
     void setLimits(uint64_t hard_limit_, uint64_t soft_limit_);
+    void setOnMemoryLimitUpdate(std::function<void()> on_memory_limit_update_)
+    {
+        on_memory_limit_update = on_memory_limit_update_;
+    }
     void startThread();
 
     size_t getHardLimit() const { return hard_limit; }
@@ -46,6 +49,7 @@ private:
     using CallbackFn = std::function<void(bool)>;
     CallbackFn on_hard_limit;
     CallbackFn on_soft_limit;
+    std::function<void()> on_memory_limit_update;
 
     uint64_t last_usage = 0;
 
@@ -87,8 +91,10 @@ public:
 
     void setLimits(uint64_t, uint64_t) {}
     size_t readMemoryUsage() { return 0; }
+    void startThread();
     size_t getHardLimit() { return 0; }
     size_t getSoftLimit() { return 0; }
+    void setOnMemoryLimitUpdate(std::function<void()>) {}
 };
 #endif
 

From c01a6775d747606b3aae70c9615404720208aeda Mon Sep 17 00:00:00 2001
From: Han Fei <hanfei19910905@gmail.com>
Date: Sat, 9 Mar 2024 02:28:39 +0100
Subject: [PATCH 091/443] fix style

---
 programs/keeper/Keeper.cpp | 3 ++-
 programs/server/Server.cpp | 5 ++++-
 2 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/programs/keeper/Keeper.cpp b/programs/keeper/Keeper.cpp
index c2ad81a3227..31e0b7dc576 100644
--- a/programs/keeper/Keeper.cpp
+++ b/programs/keeper/Keeper.cpp
@@ -630,7 +630,8 @@ try
     {
         observer.emplace(std::chrono::seconds(cgroups_memory_observer_wait_time));
         observer->startThread();
-        observer->setOnMemoryLimitUpdate([&](){
+        observer->setOnMemoryLimitUpdate([&]()
+        {
             main_config_reloader->reload();
         });
     }
diff --git a/programs/server/Server.cpp b/programs/server/Server.cpp
index c81e9b56e35..33a30a1de2c 100644
--- a/programs/server/Server.cpp
+++ b/programs/server/Server.cpp
@@ -1702,9 +1702,12 @@ try
     }
 
     if (cgroups_memory_usage_observer)
-        cgroups_memory_usage_observer->setOnMemoryLimitUpdate([&](){
+    {
+        cgroups_memory_usage_observer->setOnMemoryLimitUpdate([&]()
+        {
             main_config_reloader->reload();
         });
+    }
 
     /// Reload config in SYSTEM RELOAD CONFIG query.
     global_context->setConfigReloadCallback([&]()

From 009c2ea9f6e3c9dcd30a2528f7737cd9059296dd Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 9 Mar 2024 05:27:29 +0100
Subject: [PATCH 092/443] Move a class into an anonymous namespace

---
 src/Disks/IO/ReadBufferFromRemoteFSGather.cpp | 3 ---
 src/Functions/CastOverloadResolver.cpp        | 5 +++++
 2 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp b/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp
index 417f7615dd7..38da4e96ff1 100644
--- a/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp
+++ b/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp
@@ -1,7 +1,5 @@
 #include "ReadBufferFromRemoteFSGather.h"
 
-#include <IO/SeekableReadBuffer.h>
-
 #include <Disks/IO/CachedOnDiskReadBufferFromFile.h>
 #include <Disks/ObjectStorages/Cached/CachedObjectStorage.h>
 #include <Interpreters/Cache/FileCache.h>
@@ -9,7 +7,6 @@
 #include <IO/ReadSettings.h>
 #include <IO/SwapHelper.h>
 #include <Interpreters/FilesystemCacheLog.h>
-#include <base/hex.h>
 #include <Common/logger_useful.h>
 
 using namespace DB;
diff --git a/src/Functions/CastOverloadResolver.cpp b/src/Functions/CastOverloadResolver.cpp
index 7fc46db50f1..a72563212ff 100644
--- a/src/Functions/CastOverloadResolver.cpp
+++ b/src/Functions/CastOverloadResolver.cpp
@@ -12,6 +12,9 @@ namespace ErrorCodes
     extern const int ILLEGAL_TYPE_OF_ARGUMENT;
 }
 
+namespace
+{
+
 /** CastInternal does not preserve nullability of the data type,
   * i.e. CastInternal(toNullable(toInt8(1)) as Int32) will be Int32(1).
   *
@@ -136,6 +139,8 @@ using CastOverloadResolver = CastOverloadResolverImpl<cast_type, false, CastOver
 template <CastType cast_type>
 using CastInternalOverloadResolver = CastOverloadResolverImpl<cast_type, true, CastInternalOverloadName, CastInternalName>;
 
+}
+
 
 FunctionOverloadResolverPtr createInternalCastOverloadResolver(CastType type, std::optional<CastDiagnostic> diagnostic)
 {

From 2be09581ddc0ddfb57134f8a0ea6c33314f8071e Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 9 Mar 2024 05:48:52 +0100
Subject: [PATCH 093/443] Split CastOverloadResolver translation unit

---
 programs/library-bridge/CMakeLists.txt |   3 +
 programs/odbc-bridge/CMakeLists.txt    |   3 +
 src/Functions/CastOverloadResolver.cpp | 139 +------------------------
 3 files changed, 7 insertions(+), 138 deletions(-)

diff --git a/programs/library-bridge/CMakeLists.txt b/programs/library-bridge/CMakeLists.txt
index dd0bf67cb64..f42b574b807 100644
--- a/programs/library-bridge/CMakeLists.txt
+++ b/programs/library-bridge/CMakeLists.txt
@@ -11,6 +11,9 @@ set (CLICKHOUSE_LIBRARY_BRIDGE_SOURCES
     LibraryBridgeHandlers.cpp
     SharedLibrary.cpp
     library-bridge.cpp
+
+    ../../src/Functions/CastOverloadResolverImpl.cpp
+    ../../src/Functions/CastInternalOverloadResolverImpl.cpp
 )
 
 clickhouse_add_executable(clickhouse-library-bridge ${CLICKHOUSE_LIBRARY_BRIDGE_SOURCES})
diff --git a/programs/odbc-bridge/CMakeLists.txt b/programs/odbc-bridge/CMakeLists.txt
index 56373601b95..4e5dbac486e 100644
--- a/programs/odbc-bridge/CMakeLists.txt
+++ b/programs/odbc-bridge/CMakeLists.txt
@@ -13,6 +13,9 @@ set (CLICKHOUSE_ODBC_BRIDGE_SOURCES
     getIdentifierQuote.cpp
     odbc-bridge.cpp
     validateODBCConnectionString.cpp
+
+    ../../src/Functions/CastOverloadResolverImpl.cpp
+    ../../src/Functions/CastInternalOverloadResolverImpl.cpp
 )
 
 clickhouse_add_executable(clickhouse-odbc-bridge ${CLICKHOUSE_ODBC_BRIDGE_SOURCES})
diff --git a/src/Functions/CastOverloadResolver.cpp b/src/Functions/CastOverloadResolver.cpp
index a72563212ff..1c57bcfa979 100644
--- a/src/Functions/CastOverloadResolver.cpp
+++ b/src/Functions/CastOverloadResolver.cpp
@@ -1,147 +1,11 @@
 #include <Functions/FunctionFactory.h>
-#include <Functions/FunctionsConversion.h>
 #include <Functions/CastOverloadResolver.h>
-#include <Interpreters/parseColumnsListForTableFunction.h>
+#include <Functions/CastOverloadResolverImpl.h>
 
 
 namespace DB
 {
 
-namespace ErrorCodes
-{
-    extern const int ILLEGAL_TYPE_OF_ARGUMENT;
-}
-
-namespace
-{
-
-/** CastInternal does not preserve nullability of the data type,
-  * i.e. CastInternal(toNullable(toInt8(1)) as Int32) will be Int32(1).
-  *
-  * Cast preserves nullability according to setting `cast_keep_nullable`,
-  * i.e. Cast(toNullable(toInt8(1)) as Int32) will be Nullable(Int32(1)) if `cast_keep_nullable` == 1.
-  */
-template <CastType cast_type, bool internal, typename CastName, typename FunctionName>
-class CastOverloadResolverImpl : public IFunctionOverloadResolver
-{
-public:
-    using MonotonicityForRange = FunctionCastBase::MonotonicityForRange;
-
-    static constexpr auto name = cast_type == CastType::accurate
-        ? CastName::accurate_cast_name
-        : (cast_type == CastType::accurateOrNull ? CastName::accurate_cast_or_null_name : CastName::cast_name);
-
-    String getName() const override { return name; }
-
-    size_t getNumberOfArguments() const override { return 2; }
-
-    ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {1}; }
-
-    explicit CastOverloadResolverImpl(ContextPtr context_, std::optional<CastDiagnostic> diagnostic_, bool keep_nullable_, const DataTypeValidationSettings & data_type_validation_settings_)
-        : context(context_)
-        , diagnostic(std::move(diagnostic_))
-        , keep_nullable(keep_nullable_)
-        , data_type_validation_settings(data_type_validation_settings_)
-    {
-    }
-
-    static FunctionOverloadResolverPtr create(ContextPtr context)
-    {
-        const auto & settings_ref = context->getSettingsRef();
-
-        if constexpr (internal)
-            return createImpl(context, {}, false /*keep_nullable*/);
-
-        return createImpl(context, {}, settings_ref.cast_keep_nullable, DataTypeValidationSettings(settings_ref));
-    }
-
-    static FunctionOverloadResolverPtr createImpl(ContextPtr context, std::optional<CastDiagnostic> diagnostic = {}, bool keep_nullable = false, const DataTypeValidationSettings & data_type_validation_settings = {})
-    {
-        assert(!internal || !keep_nullable);
-        return std::make_unique<CastOverloadResolverImpl>(context, std::move(diagnostic), keep_nullable, data_type_validation_settings);
-    }
-
-    static FunctionOverloadResolverPtr createImpl(std::optional<CastDiagnostic> diagnostic = {}, bool keep_nullable = false, const DataTypeValidationSettings & data_type_validation_settings = {})
-    {
-        assert(!internal || !keep_nullable);
-        return std::make_unique<CastOverloadResolverImpl>(ContextPtr(), std::move(diagnostic), keep_nullable, data_type_validation_settings);
-    }
-
-protected:
-
-    FunctionBasePtr buildImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & return_type) const override
-    {
-        DataTypes data_types(arguments.size());
-
-        for (size_t i = 0; i < arguments.size(); ++i)
-            data_types[i] = arguments[i].type;
-
-        auto monotonicity = MonotonicityHelper::getMonotonicityInformation(arguments.front().type, return_type.get());
-        return std::make_unique<FunctionCast<FunctionName>>(context, name, std::move(monotonicity), data_types, return_type, diagnostic, cast_type);
-    }
-
-    DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override
-    {
-        const auto & column = arguments.back().column;
-        if (!column)
-            throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Second argument to {} must be a constant string describing type. "
-                "Instead there is non-constant column of type {}", getName(), arguments.back().type->getName());
-
-        const auto * type_col = checkAndGetColumnConst<ColumnString>(column.get());
-        if (!type_col)
-            throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Second argument to {} must be a constant string describing type. "
-                "Instead there is a column with the following structure: {}", getName(), column->dumpStructure());
-
-        DataTypePtr type = DataTypeFactory::instance().get(type_col->getValue<String>());
-        validateDataType(type, data_type_validation_settings);
-
-        if constexpr (cast_type == CastType::accurateOrNull)
-            return makeNullable(type);
-
-        if constexpr (internal)
-            return type;
-
-        if (keep_nullable && arguments.front().type->isNullable() && type->canBeInsideNullable())
-            return makeNullable(type);
-
-        return type;
-    }
-
-    bool useDefaultImplementationForNulls() const override { return false; }
-    bool useDefaultImplementationForNothing() const override { return false; }
-    bool useDefaultImplementationForLowCardinalityColumns() const override { return false; }
-
-private:
-    ContextPtr context;
-    std::optional<CastDiagnostic> diagnostic;
-    bool keep_nullable;
-    DataTypeValidationSettings data_type_validation_settings;
-};
-
-
-struct CastOverloadName
-{
-    static constexpr auto cast_name = "CAST";
-    static constexpr auto accurate_cast_name = "accurateCast";
-    static constexpr auto accurate_cast_or_null_name = "accurateCastOrNull";
-};
-
-struct CastInternalOverloadName
-{
-    static constexpr auto cast_name = "_CAST";
-    static constexpr auto accurate_cast_name = "accurate_Cast";
-    static constexpr auto accurate_cast_or_null_name = "accurate_CastOrNull";
-};
-
-template <CastType cast_type>
-using CastOverloadResolver = CastOverloadResolverImpl<cast_type, false, CastOverloadName, CastName>;
-
-template <CastType cast_type>
-using CastInternalOverloadResolver = CastOverloadResolverImpl<cast_type, true, CastInternalOverloadName, CastInternalName>;
-
-}
-
-
 FunctionOverloadResolverPtr createInternalCastOverloadResolver(CastType type, std::optional<CastDiagnostic> diagnostic)
 {
     switch (type)
@@ -155,7 +19,6 @@ FunctionOverloadResolverPtr createInternalCastOverloadResolver(CastType type, st
     }
 }
 
-
 REGISTER_FUNCTION(CastOverloadResolvers)
 {
     factory.registerFunction<CastInternalOverloadResolver<CastType::nonAccurate>>({}, FunctionFactory::CaseInsensitive);

From c67e8d5d73a6ac1647a023740a859e739d6ab2ee Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 9 Mar 2024 05:50:19 +0100
Subject: [PATCH 094/443] Add files

---
 .../CastInternalOverloadResolverImpl.cpp      |  10 ++
 src/Functions/CastOverloadResolverImpl.cpp    |  10 ++
 src/Functions/CastOverloadResolverImpl.h      | 149 ++++++++++++++++++
 3 files changed, 169 insertions(+)
 create mode 100644 src/Functions/CastInternalOverloadResolverImpl.cpp
 create mode 100644 src/Functions/CastOverloadResolverImpl.cpp
 create mode 100644 src/Functions/CastOverloadResolverImpl.h

diff --git a/src/Functions/CastInternalOverloadResolverImpl.cpp b/src/Functions/CastInternalOverloadResolverImpl.cpp
new file mode 100644
index 00000000000..d8ee0c76fd8
--- /dev/null
+++ b/src/Functions/CastInternalOverloadResolverImpl.cpp
@@ -0,0 +1,10 @@
+#include <Functions/CastOverloadResolverImpl.h>
+
+namespace DB
+{
+
+template class CastOverloadResolverImpl<CastType::nonAccurate, true, CastInternalOverloadName, CastInternalName>;
+template class CastOverloadResolverImpl<CastType::accurate, true, CastInternalOverloadName, CastInternalName>;
+template class CastOverloadResolverImpl<CastType::accurateOrNull, true, CastInternalOverloadName, CastInternalName>;
+
+}
diff --git a/src/Functions/CastOverloadResolverImpl.cpp b/src/Functions/CastOverloadResolverImpl.cpp
new file mode 100644
index 00000000000..d2325db5e0a
--- /dev/null
+++ b/src/Functions/CastOverloadResolverImpl.cpp
@@ -0,0 +1,10 @@
+#include <Functions/CastOverloadResolverImpl.h>
+
+namespace DB
+{
+
+template class CastOverloadResolverImpl<CastType::nonAccurate, false, CastOverloadName, CastName>;
+template class CastOverloadResolverImpl<CastType::accurate, false, CastOverloadName, CastName>;
+template class CastOverloadResolverImpl<CastType::accurateOrNull, false, CastOverloadName, CastName>;
+
+}
diff --git a/src/Functions/CastOverloadResolverImpl.h b/src/Functions/CastOverloadResolverImpl.h
new file mode 100644
index 00000000000..61ccc66fb6c
--- /dev/null
+++ b/src/Functions/CastOverloadResolverImpl.h
@@ -0,0 +1,149 @@
+#pragma once
+
+#include <Functions/FunctionsConversion.h>
+#include <Functions/CastOverloadResolver.h>
+#include <Interpreters/parseColumnsListForTableFunction.h>
+
+
+namespace DB
+{
+
+namespace ErrorCodes
+{
+    extern const int ILLEGAL_TYPE_OF_ARGUMENT;
+}
+
+
+/** CastInternal does not preserve nullability of the data type,
+  * i.e. CastInternal(toNullable(toInt8(1)) as Int32) will be Int32(1).
+  *
+  * Cast preserves nullability according to setting `cast_keep_nullable`,
+  * i.e. Cast(toNullable(toInt8(1)) as Int32) will be Nullable(Int32(1)) if `cast_keep_nullable` == 1.
+  */
+template <CastType cast_type, bool internal, typename CastName, typename FunctionName>
+class CastOverloadResolverImpl : public IFunctionOverloadResolver
+{
+public:
+    using MonotonicityForRange = FunctionCastBase::MonotonicityForRange;
+
+    static constexpr auto name = cast_type == CastType::accurate
+        ? CastName::accurate_cast_name
+        : (cast_type == CastType::accurateOrNull ? CastName::accurate_cast_or_null_name : CastName::cast_name);
+
+    String getName() const override { return name; }
+
+    size_t getNumberOfArguments() const override { return 2; }
+
+    ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {1}; }
+
+    explicit CastOverloadResolverImpl(ContextPtr context_, std::optional<CastDiagnostic> diagnostic_, bool keep_nullable_, const DataTypeValidationSettings & data_type_validation_settings_)
+        : context(context_)
+        , diagnostic(std::move(diagnostic_))
+        , keep_nullable(keep_nullable_)
+        , data_type_validation_settings(data_type_validation_settings_)
+    {
+    }
+
+    static FunctionOverloadResolverPtr create(ContextPtr context)
+    {
+        const auto & settings_ref = context->getSettingsRef();
+
+        if constexpr (internal)
+            return createImpl(context, {}, false /*keep_nullable*/);
+
+        return createImpl(context, {}, settings_ref.cast_keep_nullable, DataTypeValidationSettings(settings_ref));
+    }
+
+    static FunctionOverloadResolverPtr createImpl(ContextPtr context, std::optional<CastDiagnostic> diagnostic = {}, bool keep_nullable = false, const DataTypeValidationSettings & data_type_validation_settings = {})
+    {
+        assert(!internal || !keep_nullable);
+        return std::make_unique<CastOverloadResolverImpl>(context, std::move(diagnostic), keep_nullable, data_type_validation_settings);
+    }
+
+    static FunctionOverloadResolverPtr createImpl(std::optional<CastDiagnostic> diagnostic = {}, bool keep_nullable = false, const DataTypeValidationSettings & data_type_validation_settings = {})
+    {
+        assert(!internal || !keep_nullable);
+        return std::make_unique<CastOverloadResolverImpl>(ContextPtr(), std::move(diagnostic), keep_nullable, data_type_validation_settings);
+    }
+
+protected:
+
+    FunctionBasePtr buildImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & return_type) const override
+    {
+        DataTypes data_types(arguments.size());
+
+        for (size_t i = 0; i < arguments.size(); ++i)
+            data_types[i] = arguments[i].type;
+
+        auto monotonicity = MonotonicityHelper::getMonotonicityInformation(arguments.front().type, return_type.get());
+        return std::make_unique<FunctionCast<FunctionName>>(context, name, std::move(monotonicity), data_types, return_type, diagnostic, cast_type);
+    }
+
+    DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override
+    {
+        const auto & column = arguments.back().column;
+        if (!column)
+            throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Second argument to {} must be a constant string describing type. "
+                "Instead there is non-constant column of type {}", getName(), arguments.back().type->getName());
+
+        const auto * type_col = checkAndGetColumnConst<ColumnString>(column.get());
+        if (!type_col)
+            throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Second argument to {} must be a constant string describing type. "
+                "Instead there is a column with the following structure: {}", getName(), column->dumpStructure());
+
+        DataTypePtr type = DataTypeFactory::instance().get(type_col->getValue<String>());
+        validateDataType(type, data_type_validation_settings);
+
+        if constexpr (cast_type == CastType::accurateOrNull)
+            return makeNullable(type);
+
+        if constexpr (internal)
+            return type;
+
+        if (keep_nullable && arguments.front().type->isNullable() && type->canBeInsideNullable())
+            return makeNullable(type);
+
+        return type;
+    }
+
+    bool useDefaultImplementationForNulls() const override { return false; }
+    bool useDefaultImplementationForNothing() const override { return false; }
+    bool useDefaultImplementationForLowCardinalityColumns() const override { return false; }
+
+private:
+    ContextPtr context;
+    std::optional<CastDiagnostic> diagnostic;
+    bool keep_nullable;
+    DataTypeValidationSettings data_type_validation_settings;
+};
+
+
+struct CastOverloadName
+{
+    static constexpr auto cast_name = "CAST";
+    static constexpr auto accurate_cast_name = "accurateCast";
+    static constexpr auto accurate_cast_or_null_name = "accurateCastOrNull";
+};
+
+struct CastInternalOverloadName
+{
+    static constexpr auto cast_name = "_CAST";
+    static constexpr auto accurate_cast_name = "accurate_Cast";
+    static constexpr auto accurate_cast_or_null_name = "accurate_CastOrNull";
+};
+
+template <CastType cast_type>
+using CastOverloadResolver = CastOverloadResolverImpl<cast_type, false, CastOverloadName, CastName>;
+
+template <CastType cast_type>
+using CastInternalOverloadResolver = CastOverloadResolverImpl<cast_type, true, CastInternalOverloadName, CastInternalName>;
+
+extern template class CastOverloadResolverImpl<CastType::nonAccurate, false, CastOverloadName, CastName>;
+extern template class CastOverloadResolverImpl<CastType::accurate, false, CastOverloadName, CastName>;
+extern template class CastOverloadResolverImpl<CastType::accurateOrNull, false, CastOverloadName, CastName>;
+
+extern template class CastOverloadResolverImpl<CastType::nonAccurate, true, CastInternalOverloadName, CastInternalName>;
+extern template class CastOverloadResolverImpl<CastType::accurate, true, CastInternalOverloadName, CastInternalName>;
+extern template class CastOverloadResolverImpl<CastType::accurateOrNull, true, CastInternalOverloadName, CastInternalName>;
+
+}

From 6d45eecdad4b737584b9f64e9a44c8fd845f17a4 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 9 Mar 2024 06:11:34 +0100
Subject: [PATCH 095/443] Remove garbage

---
 programs/odbc-bridge/CMakeLists.txt           |  3 --
 src/Functions/CMakeLists.txt                  |  2 +
 .../CastInternalOverloadResolverImpl.cpp      |  6 +--
 src/Functions/CastOverloadResolverImpl.cpp    |  6 +--
 src/Functions/CastOverloadResolverImpl.h      | 51 +++++++++----------
 src/Functions/FunctionsConversion.cpp         | 15 ++++++
 src/Functions/FunctionsConversion.h           | 17 +------
 7 files changed, 49 insertions(+), 51 deletions(-)

diff --git a/programs/odbc-bridge/CMakeLists.txt b/programs/odbc-bridge/CMakeLists.txt
index 4e5dbac486e..56373601b95 100644
--- a/programs/odbc-bridge/CMakeLists.txt
+++ b/programs/odbc-bridge/CMakeLists.txt
@@ -13,9 +13,6 @@ set (CLICKHOUSE_ODBC_BRIDGE_SOURCES
     getIdentifierQuote.cpp
     odbc-bridge.cpp
     validateODBCConnectionString.cpp
-
-    ../../src/Functions/CastOverloadResolverImpl.cpp
-    ../../src/Functions/CastInternalOverloadResolverImpl.cpp
 )
 
 clickhouse_add_executable(clickhouse-odbc-bridge ${CLICKHOUSE_ODBC_BRIDGE_SOURCES})
diff --git a/src/Functions/CMakeLists.txt b/src/Functions/CMakeLists.txt
index ac3e3671ae0..f27bcae1fe3 100644
--- a/src/Functions/CMakeLists.txt
+++ b/src/Functions/CMakeLists.txt
@@ -15,6 +15,8 @@ extract_into_parent_list(clickhouse_functions_sources dbms_sources
     checkHyperscanRegexp.cpp
     array/has.cpp
     CastOverloadResolver.cpp
+    CastOverloadResolverImpl.cpp
+    CastInternalOverloadResolverImpl.cpp
 )
 extract_into_parent_list(clickhouse_functions_headers dbms_headers
     IFunction.h
diff --git a/src/Functions/CastInternalOverloadResolverImpl.cpp b/src/Functions/CastInternalOverloadResolverImpl.cpp
index d8ee0c76fd8..8b74f76ca39 100644
--- a/src/Functions/CastInternalOverloadResolverImpl.cpp
+++ b/src/Functions/CastInternalOverloadResolverImpl.cpp
@@ -3,8 +3,8 @@
 namespace DB
 {
 
-template class CastOverloadResolverImpl<CastType::nonAccurate, true, CastInternalOverloadName, CastInternalName>;
-template class CastOverloadResolverImpl<CastType::accurate, true, CastInternalOverloadName, CastInternalName>;
-template class CastOverloadResolverImpl<CastType::accurateOrNull, true, CastInternalOverloadName, CastInternalName>;
+template class CastOverloadResolverImpl<CastType::nonAccurate, true>;
+template class CastOverloadResolverImpl<CastType::accurate, true>;
+template class CastOverloadResolverImpl<CastType::accurateOrNull, true>;
 
 }
diff --git a/src/Functions/CastOverloadResolverImpl.cpp b/src/Functions/CastOverloadResolverImpl.cpp
index d2325db5e0a..a7f7024892e 100644
--- a/src/Functions/CastOverloadResolverImpl.cpp
+++ b/src/Functions/CastOverloadResolverImpl.cpp
@@ -3,8 +3,8 @@
 namespace DB
 {
 
-template class CastOverloadResolverImpl<CastType::nonAccurate, false, CastOverloadName, CastName>;
-template class CastOverloadResolverImpl<CastType::accurate, false, CastOverloadName, CastName>;
-template class CastOverloadResolverImpl<CastType::accurateOrNull, false, CastOverloadName, CastName>;
+template class CastOverloadResolverImpl<CastType::nonAccurate, false>;
+template class CastOverloadResolverImpl<CastType::accurate, false>;
+template class CastOverloadResolverImpl<CastType::accurateOrNull, false>;
 
 }
diff --git a/src/Functions/CastOverloadResolverImpl.h b/src/Functions/CastOverloadResolverImpl.h
index 61ccc66fb6c..36376c5f254 100644
--- a/src/Functions/CastOverloadResolverImpl.h
+++ b/src/Functions/CastOverloadResolverImpl.h
@@ -13,6 +13,15 @@ namespace ErrorCodes
     extern const int ILLEGAL_TYPE_OF_ARGUMENT;
 }
 
+struct CastName
+{
+    static constexpr auto name = "CAST";
+};
+
+struct CastInternalName
+{
+    static constexpr auto name = "_CAST";
+};
 
 /** CastInternal does not preserve nullability of the data type,
   * i.e. CastInternal(toNullable(toInt8(1)) as Int32) will be Int32(1).
@@ -20,15 +29,16 @@ namespace ErrorCodes
   * Cast preserves nullability according to setting `cast_keep_nullable`,
   * i.e. Cast(toNullable(toInt8(1)) as Int32) will be Nullable(Int32(1)) if `cast_keep_nullable` == 1.
   */
-template <CastType cast_type, bool internal, typename CastName, typename FunctionName>
+template <CastType cast_type, bool internal>
 class CastOverloadResolverImpl : public IFunctionOverloadResolver
 {
 public:
     using MonotonicityForRange = FunctionCastBase::MonotonicityForRange;
 
     static constexpr auto name = cast_type == CastType::accurate
-        ? CastName::accurate_cast_name
-        : (cast_type == CastType::accurateOrNull ? CastName::accurate_cast_or_null_name : CastName::cast_name);
+        ? "accurateCast"
+        : (cast_type == CastType::accurateOrNull ? "accurateCastOrNull"
+        : (internal ? "_CAST" : "CAST"));
 
     String getName() const override { return name; }
 
@@ -76,7 +86,9 @@ protected:
             data_types[i] = arguments[i].type;
 
         auto monotonicity = MonotonicityHelper::getMonotonicityInformation(arguments.front().type, return_type.get());
-        return std::make_unique<FunctionCast<FunctionName>>(context, name, std::move(monotonicity), data_types, return_type, diagnostic, cast_type);
+
+        using Function = FunctionCast<std::conditional_t<internal, CastInternalName, CastName>>;
+        return std::make_unique<Function>(context, name, std::move(monotonicity), data_types, return_type, diagnostic, cast_type);
     }
 
     DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override
@@ -118,32 +130,19 @@ private:
 };
 
 
-struct CastOverloadName
-{
-    static constexpr auto cast_name = "CAST";
-    static constexpr auto accurate_cast_name = "accurateCast";
-    static constexpr auto accurate_cast_or_null_name = "accurateCastOrNull";
-};
-
-struct CastInternalOverloadName
-{
-    static constexpr auto cast_name = "_CAST";
-    static constexpr auto accurate_cast_name = "accurate_Cast";
-    static constexpr auto accurate_cast_or_null_name = "accurate_CastOrNull";
-};
+template <CastType cast_type>
+using CastOverloadResolver = CastOverloadResolverImpl<cast_type, false>;
 
 template <CastType cast_type>
-using CastOverloadResolver = CastOverloadResolverImpl<cast_type, false, CastOverloadName, CastName>;
+using CastInternalOverloadResolver = CastOverloadResolverImpl<cast_type, true>;
 
-template <CastType cast_type>
-using CastInternalOverloadResolver = CastOverloadResolverImpl<cast_type, true, CastInternalOverloadName, CastInternalName>;
 
-extern template class CastOverloadResolverImpl<CastType::nonAccurate, false, CastOverloadName, CastName>;
-extern template class CastOverloadResolverImpl<CastType::accurate, false, CastOverloadName, CastName>;
-extern template class CastOverloadResolverImpl<CastType::accurateOrNull, false, CastOverloadName, CastName>;
+extern template class CastOverloadResolverImpl<CastType::nonAccurate, false>;
+extern template class CastOverloadResolverImpl<CastType::accurate, false>;
+extern template class CastOverloadResolverImpl<CastType::accurateOrNull, false>;
 
-extern template class CastOverloadResolverImpl<CastType::nonAccurate, true, CastInternalOverloadName, CastInternalName>;
-extern template class CastOverloadResolverImpl<CastType::accurate, true, CastInternalOverloadName, CastInternalName>;
-extern template class CastOverloadResolverImpl<CastType::accurateOrNull, true, CastInternalOverloadName, CastInternalName>;
+extern template class CastOverloadResolverImpl<CastType::nonAccurate, true>;
+extern template class CastOverloadResolverImpl<CastType::accurate, true>;
+extern template class CastOverloadResolverImpl<CastType::accurateOrNull, true>;
 
 }
diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 01e057e19a1..59455ba51b7 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -5,6 +5,21 @@
 namespace DB
 {
 
+UInt32 extractToDecimalScale(const ColumnWithTypeAndName & named_column)
+{
+    const auto * arg_type = named_column.type.get();
+    bool ok = checkAndGetDataType<DataTypeUInt64>(arg_type)
+        || checkAndGetDataType<DataTypeUInt32>(arg_type)
+        || checkAndGetDataType<DataTypeUInt16>(arg_type)
+        || checkAndGetDataType<DataTypeUInt8>(arg_type);
+    if (!ok)
+        throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type of toDecimal() scale {}", named_column.type->getName());
+
+    Field field;
+    named_column.column->get(0, field);
+    return static_cast<UInt32>(field.get<UInt32>());
+}
+
 REGISTER_FUNCTION(Conversion)
 {
     factory.registerFunction<FunctionToUInt8>();
diff --git a/src/Functions/FunctionsConversion.h b/src/Functions/FunctionsConversion.h
index 1522e76893e..fac74715fa0 100644
--- a/src/Functions/FunctionsConversion.h
+++ b/src/Functions/FunctionsConversion.h
@@ -98,20 +98,7 @@ namespace ErrorCodes
   * toType - conversion in "natural way";
   */
 
-inline UInt32 extractToDecimalScale(const ColumnWithTypeAndName & named_column)
-{
-    const auto * arg_type = named_column.type.get();
-    bool ok = checkAndGetDataType<DataTypeUInt64>(arg_type)
-        || checkAndGetDataType<DataTypeUInt32>(arg_type)
-        || checkAndGetDataType<DataTypeUInt16>(arg_type)
-        || checkAndGetDataType<DataTypeUInt8>(arg_type);
-    if (!ok)
-        throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type of toDecimal() scale {}", named_column.type->getName());
-
-    Field field;
-    named_column.column->get(0, field);
-    return static_cast<UInt32>(field.get<UInt32>());
-}
+UInt32 extractToDecimalScale(const ColumnWithTypeAndName & named_column);
 
 /// Function toUnixTimestamp has exactly the same implementation as toDateTime of String type.
 struct NameToUnixTimestamp { static constexpr auto name = "toUnixTimestamp"; };
@@ -3182,8 +3169,6 @@ private:
     std::optional<CastDiagnostic> diagnostic;
 };
 
-struct CastName { static constexpr auto name = "CAST"; };
-struct CastInternalName { static constexpr auto name = "_CAST"; };
 
 class FunctionCastBase : public IFunctionBase
 {

From ea54ac3cb4842c332529bab14c0e619115911a2c Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 9 Mar 2024 06:12:22 +0100
Subject: [PATCH 096/443] Remove garbage

---
 programs/library-bridge/CMakeLists.txt | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/programs/library-bridge/CMakeLists.txt b/programs/library-bridge/CMakeLists.txt
index f42b574b807..dd0bf67cb64 100644
--- a/programs/library-bridge/CMakeLists.txt
+++ b/programs/library-bridge/CMakeLists.txt
@@ -11,9 +11,6 @@ set (CLICKHOUSE_LIBRARY_BRIDGE_SOURCES
     LibraryBridgeHandlers.cpp
     SharedLibrary.cpp
     library-bridge.cpp
-
-    ../../src/Functions/CastOverloadResolverImpl.cpp
-    ../../src/Functions/CastInternalOverloadResolverImpl.cpp
 )
 
 clickhouse_add_executable(clickhouse-library-bridge ${CLICKHOUSE_LIBRARY_BRIDGE_SOURCES})

From 7983e2b6206c5b78158802430b639e9b060d7b0c Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 9 Mar 2024 06:14:32 +0100
Subject: [PATCH 097/443] Remove garbage

---
 src/Functions/CastOverloadResolver.cpp | 15 +++++++++++++++
 src/Functions/FunctionsConversion.cpp  | 15 ---------------
 2 files changed, 15 insertions(+), 15 deletions(-)

diff --git a/src/Functions/CastOverloadResolver.cpp b/src/Functions/CastOverloadResolver.cpp
index 1c57bcfa979..a343dbb62fe 100644
--- a/src/Functions/CastOverloadResolver.cpp
+++ b/src/Functions/CastOverloadResolver.cpp
@@ -6,6 +6,21 @@
 namespace DB
 {
 
+UInt32 extractToDecimalScale(const ColumnWithTypeAndName & named_column)
+{
+    const auto * arg_type = named_column.type.get();
+    bool ok = checkAndGetDataType<DataTypeUInt64>(arg_type)
+        || checkAndGetDataType<DataTypeUInt32>(arg_type)
+        || checkAndGetDataType<DataTypeUInt16>(arg_type)
+        || checkAndGetDataType<DataTypeUInt8>(arg_type);
+    if (!ok)
+        throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type of toDecimal() scale {}", named_column.type->getName());
+
+    Field field;
+    named_column.column->get(0, field);
+    return static_cast<UInt32>(field.get<UInt32>());
+}
+
 FunctionOverloadResolverPtr createInternalCastOverloadResolver(CastType type, std::optional<CastDiagnostic> diagnostic)
 {
     switch (type)
diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 59455ba51b7..01e057e19a1 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -5,21 +5,6 @@
 namespace DB
 {
 
-UInt32 extractToDecimalScale(const ColumnWithTypeAndName & named_column)
-{
-    const auto * arg_type = named_column.type.get();
-    bool ok = checkAndGetDataType<DataTypeUInt64>(arg_type)
-        || checkAndGetDataType<DataTypeUInt32>(arg_type)
-        || checkAndGetDataType<DataTypeUInt16>(arg_type)
-        || checkAndGetDataType<DataTypeUInt8>(arg_type);
-    if (!ok)
-        throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type of toDecimal() scale {}", named_column.type->getName());
-
-    Field field;
-    named_column.column->get(0, field);
-    return static_cast<UInt32>(field.get<UInt32>());
-}
-
 REGISTER_FUNCTION(Conversion)
 {
     factory.registerFunction<FunctionToUInt8>();

From 0d05b8ccc1f56d774c502f780cbbd98326dd8401 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 9 Mar 2024 06:37:50 +0100
Subject: [PATCH 098/443] Fix style

---
 src/Functions/CastOverloadResolver.cpp | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/src/Functions/CastOverloadResolver.cpp b/src/Functions/CastOverloadResolver.cpp
index a343dbb62fe..6bacc7f4847 100644
--- a/src/Functions/CastOverloadResolver.cpp
+++ b/src/Functions/CastOverloadResolver.cpp
@@ -6,6 +6,11 @@
 namespace DB
 {
 
+namespace ErrorCodes
+{
+    extern const int ILLEGAL_TYPE_OF_ARGUMENT;
+}
+
 UInt32 extractToDecimalScale(const ColumnWithTypeAndName & named_column)
 {
     const auto * arg_type = named_column.type.get();

From dc7f4b39eef4c2a9b47cee7b7197c58fd86c0520 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 9 Mar 2024 07:00:28 +0100
Subject: [PATCH 099/443] Remove garbage

---
 src/Functions/CMakeLists.txt                  |  2 -
 .../CastInternalOverloadResolverImpl.cpp      | 10 ---
 src/Functions/CastOverloadResolver.cpp        | 21 +-----
 src/Functions/CastOverloadResolverImpl.cpp    | 10 ---
 src/Functions/CastOverloadResolverImpl.h      | 73 +++++++------------
 5 files changed, 31 insertions(+), 85 deletions(-)
 delete mode 100644 src/Functions/CastInternalOverloadResolverImpl.cpp
 delete mode 100644 src/Functions/CastOverloadResolverImpl.cpp

diff --git a/src/Functions/CMakeLists.txt b/src/Functions/CMakeLists.txt
index f27bcae1fe3..ac3e3671ae0 100644
--- a/src/Functions/CMakeLists.txt
+++ b/src/Functions/CMakeLists.txt
@@ -15,8 +15,6 @@ extract_into_parent_list(clickhouse_functions_sources dbms_sources
     checkHyperscanRegexp.cpp
     array/has.cpp
     CastOverloadResolver.cpp
-    CastOverloadResolverImpl.cpp
-    CastInternalOverloadResolverImpl.cpp
 )
 extract_into_parent_list(clickhouse_functions_headers dbms_headers
     IFunction.h
diff --git a/src/Functions/CastInternalOverloadResolverImpl.cpp b/src/Functions/CastInternalOverloadResolverImpl.cpp
deleted file mode 100644
index 8b74f76ca39..00000000000
--- a/src/Functions/CastInternalOverloadResolverImpl.cpp
+++ /dev/null
@@ -1,10 +0,0 @@
-#include <Functions/CastOverloadResolverImpl.h>
-
-namespace DB
-{
-
-template class CastOverloadResolverImpl<CastType::nonAccurate, true>;
-template class CastOverloadResolverImpl<CastType::accurate, true>;
-template class CastOverloadResolverImpl<CastType::accurateOrNull, true>;
-
-}
diff --git a/src/Functions/CastOverloadResolver.cpp b/src/Functions/CastOverloadResolver.cpp
index 6bacc7f4847..79b17d3297c 100644
--- a/src/Functions/CastOverloadResolver.cpp
+++ b/src/Functions/CastOverloadResolver.cpp
@@ -26,27 +26,14 @@ UInt32 extractToDecimalScale(const ColumnWithTypeAndName & named_column)
     return static_cast<UInt32>(field.get<UInt32>());
 }
 
-FunctionOverloadResolverPtr createInternalCastOverloadResolver(CastType type, std::optional<CastDiagnostic> diagnostic)
-{
-    switch (type)
-    {
-        case CastType::nonAccurate:
-            return CastInternalOverloadResolver<CastType::nonAccurate>::createImpl(diagnostic);
-        case CastType::accurate:
-            return CastInternalOverloadResolver<CastType::accurate>::createImpl(diagnostic);
-        case CastType::accurateOrNull:
-            return CastInternalOverloadResolver<CastType::accurateOrNull>::createImpl(diagnostic);
-    }
-}
-
 REGISTER_FUNCTION(CastOverloadResolvers)
 {
-    factory.registerFunction<CastInternalOverloadResolver<CastType::nonAccurate>>({}, FunctionFactory::CaseInsensitive);
+    factory.registerFunction("_CAST", [](ContextPtr context){ return CastOverloadResolverImpl::create(context, CastType::nonAccurate, true); }, {}, FunctionFactory::CaseInsensitive);
     /// Note: "internal" (not affected by null preserving setting) versions of accurate cast functions are unneeded.
 
-    factory.registerFunction<CastOverloadResolver<CastType::nonAccurate>>({}, FunctionFactory::CaseInsensitive);
-    factory.registerFunction<CastOverloadResolver<CastType::accurate>>();
-    factory.registerFunction<CastOverloadResolver<CastType::accurateOrNull>>();
+    factory.registerFunction("CAST", [](ContextPtr context){ return CastOverloadResolverImpl::create(context, CastType::nonAccurate, false); }, {}, FunctionFactory::CaseInsensitive);
+    factory.registerFunction("accurateCast", [](ContextPtr context){ return CastOverloadResolverImpl::create(context, CastType::accurate, false); }, {});
+    factory.registerFunction("accurateCastOrNull", [](ContextPtr context){ return CastOverloadResolverImpl::create(context, CastType::accurateOrNull, false); }, {});
 }
 
 }
diff --git a/src/Functions/CastOverloadResolverImpl.cpp b/src/Functions/CastOverloadResolverImpl.cpp
deleted file mode 100644
index a7f7024892e..00000000000
--- a/src/Functions/CastOverloadResolverImpl.cpp
+++ /dev/null
@@ -1,10 +0,0 @@
-#include <Functions/CastOverloadResolverImpl.h>
-
-namespace DB
-{
-
-template class CastOverloadResolverImpl<CastType::nonAccurate, false>;
-template class CastOverloadResolverImpl<CastType::accurate, false>;
-template class CastOverloadResolverImpl<CastType::accurateOrNull, false>;
-
-}
diff --git a/src/Functions/CastOverloadResolverImpl.h b/src/Functions/CastOverloadResolverImpl.h
index 36376c5f254..b3f3e50ebf8 100644
--- a/src/Functions/CastOverloadResolverImpl.h
+++ b/src/Functions/CastOverloadResolverImpl.h
@@ -29,55 +29,48 @@ struct CastInternalName
   * Cast preserves nullability according to setting `cast_keep_nullable`,
   * i.e. Cast(toNullable(toInt8(1)) as Int32) will be Nullable(Int32(1)) if `cast_keep_nullable` == 1.
   */
-template <CastType cast_type, bool internal>
 class CastOverloadResolverImpl : public IFunctionOverloadResolver
 {
 public:
     using MonotonicityForRange = FunctionCastBase::MonotonicityForRange;
 
-    static constexpr auto name = cast_type == CastType::accurate
-        ? "accurateCast"
-        : (cast_type == CastType::accurateOrNull ? "accurateCastOrNull"
-        : (internal ? "_CAST" : "CAST"));
-
-    String getName() const override { return name; }
+    String getName() const override
+    {
+        if (cast_type == CastType::accurate)
+            return "accurateCast";
+        if (cast_type == CastType::accurateOrNull)
+            return "accurateCastOrNull";
+        if (internal)
+            return "_CAST";
+        else
+            return "CAST";
+    }
 
     size_t getNumberOfArguments() const override { return 2; }
 
     ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {1}; }
 
-    explicit CastOverloadResolverImpl(ContextPtr context_, std::optional<CastDiagnostic> diagnostic_, bool keep_nullable_, const DataTypeValidationSettings & data_type_validation_settings_)
+    explicit CastOverloadResolverImpl(ContextPtr context_, CastType cast_type_, bool internal_, std::optional<CastDiagnostic> diagnostic_, bool keep_nullable_, const DataTypeValidationSettings & data_type_validation_settings_)
         : context(context_)
+        , cast_type(cast_type_)
+        , internal(internal_)
         , diagnostic(std::move(diagnostic_))
         , keep_nullable(keep_nullable_)
         , data_type_validation_settings(data_type_validation_settings_)
     {
     }
 
-    static FunctionOverloadResolverPtr create(ContextPtr context)
+    static FunctionOverloadResolverPtr create(ContextPtr context, CastType cast_type, bool internal)
     {
         const auto & settings_ref = context->getSettingsRef();
 
-        if constexpr (internal)
-            return createImpl(context, {}, false /*keep_nullable*/);
-
-        return createImpl(context, {}, settings_ref.cast_keep_nullable, DataTypeValidationSettings(settings_ref));
-    }
-
-    static FunctionOverloadResolverPtr createImpl(ContextPtr context, std::optional<CastDiagnostic> diagnostic = {}, bool keep_nullable = false, const DataTypeValidationSettings & data_type_validation_settings = {})
-    {
-        assert(!internal || !keep_nullable);
-        return std::make_unique<CastOverloadResolverImpl>(context, std::move(diagnostic), keep_nullable, data_type_validation_settings);
-    }
-
-    static FunctionOverloadResolverPtr createImpl(std::optional<CastDiagnostic> diagnostic = {}, bool keep_nullable = false, const DataTypeValidationSettings & data_type_validation_settings = {})
-    {
-        assert(!internal || !keep_nullable);
-        return std::make_unique<CastOverloadResolverImpl>(ContextPtr(), std::move(diagnostic), keep_nullable, data_type_validation_settings);
+        if (internal)
+            return std::make_unique<CastOverloadResolverImpl>(context, cast_type, internal, std::nullopt, false /*keep_nullable*/, DataTypeValidationSettings{});
+        else
+            return std::make_unique<CastOverloadResolverImpl>(context, cast_type, internal, std::nullopt, settings_ref.cast_keep_nullable, DataTypeValidationSettings(settings_ref));
     }
 
 protected:
-
     FunctionBasePtr buildImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & return_type) const override
     {
         DataTypes data_types(arguments.size());
@@ -87,8 +80,10 @@ protected:
 
         auto monotonicity = MonotonicityHelper::getMonotonicityInformation(arguments.front().type, return_type.get());
 
-        using Function = FunctionCast<std::conditional_t<internal, CastInternalName, CastName>>;
-        return std::make_unique<Function>(context, name, std::move(monotonicity), data_types, return_type, diagnostic, cast_type);
+        if (internal)
+            return std::make_unique<FunctionCast<CastInternalName>>(context, CastInternalName::name, std::move(monotonicity), data_types, return_type, diagnostic, cast_type);
+        else
+            return std::make_unique<FunctionCast<CastName>>(context, CastName::name, std::move(monotonicity), data_types, return_type, diagnostic, cast_type);
     }
 
     DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override
@@ -106,10 +101,10 @@ protected:
         DataTypePtr type = DataTypeFactory::instance().get(type_col->getValue<String>());
         validateDataType(type, data_type_validation_settings);
 
-        if constexpr (cast_type == CastType::accurateOrNull)
+        if (cast_type == CastType::accurateOrNull)
             return makeNullable(type);
 
-        if constexpr (internal)
+        if (internal)
             return type;
 
         if (keep_nullable && arguments.front().type->isNullable() && type->canBeInsideNullable())
@@ -124,25 +119,11 @@ protected:
 
 private:
     ContextPtr context;
+    CastType cast_type;
+    bool internal;
     std::optional<CastDiagnostic> diagnostic;
     bool keep_nullable;
     DataTypeValidationSettings data_type_validation_settings;
 };
 
-
-template <CastType cast_type>
-using CastOverloadResolver = CastOverloadResolverImpl<cast_type, false>;
-
-template <CastType cast_type>
-using CastInternalOverloadResolver = CastOverloadResolverImpl<cast_type, true>;
-
-
-extern template class CastOverloadResolverImpl<CastType::nonAccurate, false>;
-extern template class CastOverloadResolverImpl<CastType::accurate, false>;
-extern template class CastOverloadResolverImpl<CastType::accurateOrNull, false>;
-
-extern template class CastOverloadResolverImpl<CastType::nonAccurate, true>;
-extern template class CastOverloadResolverImpl<CastType::accurate, true>;
-extern template class CastOverloadResolverImpl<CastType::accurateOrNull, true>;
-
 }

From 83a5611355fdb7bf6ad95a8ae6b3e60321e00c69 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 9 Mar 2024 07:04:05 +0100
Subject: [PATCH 100/443] Remove garbage

---
 src/Functions/CastOverloadResolver.cpp   | 13 +++++++++----
 src/Functions/CastOverloadResolverImpl.h |  6 +++---
 2 files changed, 12 insertions(+), 7 deletions(-)

diff --git a/src/Functions/CastOverloadResolver.cpp b/src/Functions/CastOverloadResolver.cpp
index 79b17d3297c..898842c8505 100644
--- a/src/Functions/CastOverloadResolver.cpp
+++ b/src/Functions/CastOverloadResolver.cpp
@@ -26,14 +26,19 @@ UInt32 extractToDecimalScale(const ColumnWithTypeAndName & named_column)
     return static_cast<UInt32>(field.get<UInt32>());
 }
 
+FunctionOverloadResolverPtr createInternalCastOverloadResolver(CastType type, std::optional<CastDiagnostic> diagnostic)
+{
+    return CastOverloadResolverImpl::create(ContextPtr{}, type, true, diagnostic);
+}
+
 REGISTER_FUNCTION(CastOverloadResolvers)
 {
-    factory.registerFunction("_CAST", [](ContextPtr context){ return CastOverloadResolverImpl::create(context, CastType::nonAccurate, true); }, {}, FunctionFactory::CaseInsensitive);
+    factory.registerFunction("_CAST", [](ContextPtr context){ return CastOverloadResolverImpl::create(context, CastType::nonAccurate, true, {}); }, {}, FunctionFactory::CaseInsensitive);
     /// Note: "internal" (not affected by null preserving setting) versions of accurate cast functions are unneeded.
 
-    factory.registerFunction("CAST", [](ContextPtr context){ return CastOverloadResolverImpl::create(context, CastType::nonAccurate, false); }, {}, FunctionFactory::CaseInsensitive);
-    factory.registerFunction("accurateCast", [](ContextPtr context){ return CastOverloadResolverImpl::create(context, CastType::accurate, false); }, {});
-    factory.registerFunction("accurateCastOrNull", [](ContextPtr context){ return CastOverloadResolverImpl::create(context, CastType::accurateOrNull, false); }, {});
+    factory.registerFunction("CAST", [](ContextPtr context){ return CastOverloadResolverImpl::create(context, CastType::nonAccurate, false, {}); }, {}, FunctionFactory::CaseInsensitive);
+    factory.registerFunction("accurateCast", [](ContextPtr context){ return CastOverloadResolverImpl::create(context, CastType::accurate, false, {}); }, {});
+    factory.registerFunction("accurateCastOrNull", [](ContextPtr context){ return CastOverloadResolverImpl::create(context, CastType::accurateOrNull, false, {}); }, {});
 }
 
 }
diff --git a/src/Functions/CastOverloadResolverImpl.h b/src/Functions/CastOverloadResolverImpl.h
index b3f3e50ebf8..74b3fd3df3f 100644
--- a/src/Functions/CastOverloadResolverImpl.h
+++ b/src/Functions/CastOverloadResolverImpl.h
@@ -60,14 +60,14 @@ public:
     {
     }
 
-    static FunctionOverloadResolverPtr create(ContextPtr context, CastType cast_type, bool internal)
+    static FunctionOverloadResolverPtr create(ContextPtr context, CastType cast_type, bool internal, std::optional<CastDiagnostic> diagnostic)
     {
         const auto & settings_ref = context->getSettingsRef();
 
         if (internal)
-            return std::make_unique<CastOverloadResolverImpl>(context, cast_type, internal, std::nullopt, false /*keep_nullable*/, DataTypeValidationSettings{});
+            return std::make_unique<CastOverloadResolverImpl>(context, cast_type, internal, diagnostic, false /*keep_nullable*/, DataTypeValidationSettings{});
         else
-            return std::make_unique<CastOverloadResolverImpl>(context, cast_type, internal, std::nullopt, settings_ref.cast_keep_nullable, DataTypeValidationSettings(settings_ref));
+            return std::make_unique<CastOverloadResolverImpl>(context, cast_type, internal, diagnostic, settings_ref.cast_keep_nullable, DataTypeValidationSettings(settings_ref));
     }
 
 protected:

From 574d4863224ed7343db64f976f53442f1d166e4a Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 9 Mar 2024 07:55:59 +0100
Subject: [PATCH 101/443] Something

---
 programs/library-bridge/CMakeLists.txt        |  1 +
 .../library-bridge/createFunctionBaseCast.cpp | 19 +++++++++++++
 programs/odbc-bridge/CMakeLists.txt           |  1 +
 .../odbc-bridge/createFunctionBaseCast.cpp    | 19 +++++++++++++
 src/Functions/CastOverloadResolver.cpp        |  1 +
 src/Functions/CastOverloadResolverImpl.h      | 28 ++++---------------
 src/Functions/FunctionsConversion.cpp         | 23 +++++++++++++++
 src/Functions/FunctionsConversion.h           |  7 +++++
 8 files changed, 77 insertions(+), 22 deletions(-)
 create mode 100644 programs/library-bridge/createFunctionBaseCast.cpp
 create mode 100644 programs/odbc-bridge/createFunctionBaseCast.cpp

diff --git a/programs/library-bridge/CMakeLists.txt b/programs/library-bridge/CMakeLists.txt
index dd0bf67cb64..98d8848502d 100644
--- a/programs/library-bridge/CMakeLists.txt
+++ b/programs/library-bridge/CMakeLists.txt
@@ -11,6 +11,7 @@ set (CLICKHOUSE_LIBRARY_BRIDGE_SOURCES
     LibraryBridgeHandlers.cpp
     SharedLibrary.cpp
     library-bridge.cpp
+    createFunctionBaseCast.cpp
 )
 
 clickhouse_add_executable(clickhouse-library-bridge ${CLICKHOUSE_LIBRARY_BRIDGE_SOURCES})
diff --git a/programs/library-bridge/createFunctionBaseCast.cpp b/programs/library-bridge/createFunctionBaseCast.cpp
new file mode 100644
index 00000000000..473aa1ca81d
--- /dev/null
+++ b/programs/library-bridge/createFunctionBaseCast.cpp
@@ -0,0 +1,19 @@
+#include <Functions/IFunction.h>
+#include <Functions/CastOverloadResolver.h>
+
+
+namespace DB
+{
+
+namespace ErrorCodes
+{
+    extern const int NOT_IMPLEMENTED;
+}
+
+FunctionBasePtr createFunctionBaseCast(
+    ContextPtr, const ColumnsWithTypeAndName &, const DataTypePtr &, std::optional<CastDiagnostic>, CastType)
+{
+    throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Type conversions are not implemented for ODBC Bridge");
+}
+
+}
diff --git a/programs/odbc-bridge/CMakeLists.txt b/programs/odbc-bridge/CMakeLists.txt
index 56373601b95..18cda4d7a04 100644
--- a/programs/odbc-bridge/CMakeLists.txt
+++ b/programs/odbc-bridge/CMakeLists.txt
@@ -13,6 +13,7 @@ set (CLICKHOUSE_ODBC_BRIDGE_SOURCES
     getIdentifierQuote.cpp
     odbc-bridge.cpp
     validateODBCConnectionString.cpp
+    createFunctionBaseCast.cpp
 )
 
 clickhouse_add_executable(clickhouse-odbc-bridge ${CLICKHOUSE_ODBC_BRIDGE_SOURCES})
diff --git a/programs/odbc-bridge/createFunctionBaseCast.cpp b/programs/odbc-bridge/createFunctionBaseCast.cpp
new file mode 100644
index 00000000000..473aa1ca81d
--- /dev/null
+++ b/programs/odbc-bridge/createFunctionBaseCast.cpp
@@ -0,0 +1,19 @@
+#include <Functions/IFunction.h>
+#include <Functions/CastOverloadResolver.h>
+
+
+namespace DB
+{
+
+namespace ErrorCodes
+{
+    extern const int NOT_IMPLEMENTED;
+}
+
+FunctionBasePtr createFunctionBaseCast(
+    ContextPtr, const ColumnsWithTypeAndName &, const DataTypePtr &, std::optional<CastDiagnostic>, CastType)
+{
+    throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Type conversions are not implemented for ODBC Bridge");
+}
+
+}
diff --git a/src/Functions/CastOverloadResolver.cpp b/src/Functions/CastOverloadResolver.cpp
index 898842c8505..6ab6f95410f 100644
--- a/src/Functions/CastOverloadResolver.cpp
+++ b/src/Functions/CastOverloadResolver.cpp
@@ -1,6 +1,7 @@
 #include <Functions/FunctionFactory.h>
 #include <Functions/CastOverloadResolver.h>
 #include <Functions/CastOverloadResolverImpl.h>
+#include <DataTypes/DataTypesNumber.h>
 
 
 namespace DB
diff --git a/src/Functions/CastOverloadResolverImpl.h b/src/Functions/CastOverloadResolverImpl.h
index 74b3fd3df3f..2bb83040163 100644
--- a/src/Functions/CastOverloadResolverImpl.h
+++ b/src/Functions/CastOverloadResolverImpl.h
@@ -1,7 +1,12 @@
 #pragma once
 
 #include <Functions/FunctionsConversion.h>
+#include <Functions/FunctionHelpers.h>
 #include <Functions/CastOverloadResolver.h>
+#include <DataTypes/DataTypeFactory.h>
+#include <DataTypes/DataTypeNullable.h>
+#include <Columns/ColumnString.h>
+#include <Interpreters/Context.h>
 #include <Interpreters/parseColumnsListForTableFunction.h>
 
 
@@ -13,15 +18,6 @@ namespace ErrorCodes
     extern const int ILLEGAL_TYPE_OF_ARGUMENT;
 }
 
-struct CastName
-{
-    static constexpr auto name = "CAST";
-};
-
-struct CastInternalName
-{
-    static constexpr auto name = "_CAST";
-};
 
 /** CastInternal does not preserve nullability of the data type,
   * i.e. CastInternal(toNullable(toInt8(1)) as Int32) will be Int32(1).
@@ -32,8 +28,6 @@ struct CastInternalName
 class CastOverloadResolverImpl : public IFunctionOverloadResolver
 {
 public:
-    using MonotonicityForRange = FunctionCastBase::MonotonicityForRange;
-
     String getName() const override
     {
         if (cast_type == CastType::accurate)
@@ -73,17 +67,7 @@ public:
 protected:
     FunctionBasePtr buildImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & return_type) const override
     {
-        DataTypes data_types(arguments.size());
-
-        for (size_t i = 0; i < arguments.size(); ++i)
-            data_types[i] = arguments[i].type;
-
-        auto monotonicity = MonotonicityHelper::getMonotonicityInformation(arguments.front().type, return_type.get());
-
-        if (internal)
-            return std::make_unique<FunctionCast<CastInternalName>>(context, CastInternalName::name, std::move(monotonicity), data_types, return_type, diagnostic, cast_type);
-        else
-            return std::make_unique<FunctionCast<CastName>>(context, CastName::name, std::move(monotonicity), data_types, return_type, diagnostic, cast_type);
+        return createFunctionBaseCast(context, arguments, return_type, diagnostic, cast_type);
     }
 
     DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override
diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 01e057e19a1..ce9e4fd66e1 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -5,6 +5,29 @@
 namespace DB
 {
 
+namespace
+{
+
+struct CastInternalName { static constexpr auto name = "_CAST"; };
+
+}
+
+FunctionBasePtr createFunctionBaseCast(
+    ContextPtr context
+    , const ColumnsWithTypeAndName & arguments
+    , const DataTypePtr & return_type
+    , std::optional<CastDiagnostic> diagnostic
+    , CastType cast_type)
+{
+    DataTypes data_types(arguments.size());
+
+    for (size_t i = 0; i < arguments.size(); ++i)
+        data_types[i] = arguments[i].type;
+
+    auto monotonicity = MonotonicityHelper::getMonotonicityInformation(arguments.front().type, return_type.get());
+    return std::make_unique<FunctionCast<CastInternalName>>(context, CastInternalName::name, std::move(monotonicity), data_types, return_type, diagnostic, cast_type);
+}
+
 REGISTER_FUNCTION(Conversion)
 {
     factory.registerFunction<FunctionToUInt8>();
diff --git a/src/Functions/FunctionsConversion.h b/src/Functions/FunctionsConversion.h
index fac74715fa0..689b72dc917 100644
--- a/src/Functions/FunctionsConversion.h
+++ b/src/Functions/FunctionsConversion.h
@@ -4972,4 +4972,11 @@ public:
     }
 };
 
+FunctionBasePtr createFunctionBaseCast(
+    ContextPtr context
+    , const ColumnsWithTypeAndName & arguments
+    , const DataTypePtr & return_type
+    , std::optional<CastDiagnostic> diagnostic
+    , CastType cast_type);
+
 }

From feed74e598503eeaa0e6aac44684f19e1fd42361 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 9 Mar 2024 07:59:09 +0100
Subject: [PATCH 102/443] Better

---
 src/Functions/CastOverloadResolver.cpp   | 100 +++++++++++++++++++-
 src/Functions/CastOverloadResolverImpl.h | 113 -----------------------
 2 files changed, 99 insertions(+), 114 deletions(-)
 delete mode 100644 src/Functions/CastOverloadResolverImpl.h

diff --git a/src/Functions/CastOverloadResolver.cpp b/src/Functions/CastOverloadResolver.cpp
index 6ab6f95410f..98debc3e2a4 100644
--- a/src/Functions/CastOverloadResolver.cpp
+++ b/src/Functions/CastOverloadResolver.cpp
@@ -1,7 +1,12 @@
 #include <Functions/FunctionFactory.h>
 #include <Functions/CastOverloadResolver.h>
-#include <Functions/CastOverloadResolverImpl.h>
+#include <Functions/FunctionsConversion.h>
+#include <Functions/FunctionHelpers.h>
+#include <DataTypes/DataTypeFactory.h>
 #include <DataTypes/DataTypesNumber.h>
+#include <DataTypes/DataTypeNullable.h>
+#include <Columns/ColumnString.h>
+#include <Interpreters/parseColumnsListForTableFunction.h>
 
 
 namespace DB
@@ -12,6 +17,99 @@ namespace ErrorCodes
     extern const int ILLEGAL_TYPE_OF_ARGUMENT;
 }
 
+
+/** CastInternal does not preserve nullability of the data type,
+  * i.e. CastInternal(toNullable(toInt8(1)) as Int32) will be Int32(1).
+  *
+  * Cast preserves nullability according to setting `cast_keep_nullable`,
+  * i.e. Cast(toNullable(toInt8(1)) as Int32) will be Nullable(Int32(1)) if `cast_keep_nullable` == 1.
+  */
+class CastOverloadResolverImpl : public IFunctionOverloadResolver
+{
+public:
+    String getName() const override
+    {
+        if (cast_type == CastType::accurate)
+            return "accurateCast";
+        if (cast_type == CastType::accurateOrNull)
+            return "accurateCastOrNull";
+        if (internal)
+            return "_CAST";
+        else
+            return "CAST";
+    }
+
+    size_t getNumberOfArguments() const override { return 2; }
+
+    ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {1}; }
+
+    explicit CastOverloadResolverImpl(ContextPtr context_, CastType cast_type_, bool internal_, std::optional<CastDiagnostic> diagnostic_, bool keep_nullable_, const DataTypeValidationSettings & data_type_validation_settings_)
+        : context(context_)
+        , cast_type(cast_type_)
+        , internal(internal_)
+        , diagnostic(std::move(diagnostic_))
+        , keep_nullable(keep_nullable_)
+        , data_type_validation_settings(data_type_validation_settings_)
+    {
+    }
+
+    static FunctionOverloadResolverPtr create(ContextPtr context, CastType cast_type, bool internal, std::optional<CastDiagnostic> diagnostic)
+    {
+        const auto & settings_ref = context->getSettingsRef();
+
+        if (internal)
+            return std::make_unique<CastOverloadResolverImpl>(context, cast_type, internal, diagnostic, false /*keep_nullable*/, DataTypeValidationSettings{});
+        else
+            return std::make_unique<CastOverloadResolverImpl>(context, cast_type, internal, diagnostic, settings_ref.cast_keep_nullable, DataTypeValidationSettings(settings_ref));
+    }
+
+protected:
+    FunctionBasePtr buildImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & return_type) const override
+    {
+        return createFunctionBaseCast(context, arguments, return_type, diagnostic, cast_type);
+    }
+
+    DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override
+    {
+        const auto & column = arguments.back().column;
+        if (!column)
+            throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Second argument to {} must be a constant string describing type. "
+                "Instead there is non-constant column of type {}", getName(), arguments.back().type->getName());
+
+        const auto * type_col = checkAndGetColumnConst<ColumnString>(column.get());
+        if (!type_col)
+            throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Second argument to {} must be a constant string describing type. "
+                "Instead there is a column with the following structure: {}", getName(), column->dumpStructure());
+
+        DataTypePtr type = DataTypeFactory::instance().get(type_col->getValue<String>());
+        validateDataType(type, data_type_validation_settings);
+
+        if (cast_type == CastType::accurateOrNull)
+            return makeNullable(type);
+
+        if (internal)
+            return type;
+
+        if (keep_nullable && arguments.front().type->isNullable() && type->canBeInsideNullable())
+            return makeNullable(type);
+
+        return type;
+    }
+
+    bool useDefaultImplementationForNulls() const override { return false; }
+    bool useDefaultImplementationForNothing() const override { return false; }
+    bool useDefaultImplementationForLowCardinalityColumns() const override { return false; }
+
+private:
+    ContextPtr context;
+    CastType cast_type;
+    bool internal;
+    std::optional<CastDiagnostic> diagnostic;
+    bool keep_nullable;
+    DataTypeValidationSettings data_type_validation_settings;
+};
+
+
 UInt32 extractToDecimalScale(const ColumnWithTypeAndName & named_column)
 {
     const auto * arg_type = named_column.type.get();
diff --git a/src/Functions/CastOverloadResolverImpl.h b/src/Functions/CastOverloadResolverImpl.h
deleted file mode 100644
index 2bb83040163..00000000000
--- a/src/Functions/CastOverloadResolverImpl.h
+++ /dev/null
@@ -1,113 +0,0 @@
-#pragma once
-
-#include <Functions/FunctionsConversion.h>
-#include <Functions/FunctionHelpers.h>
-#include <Functions/CastOverloadResolver.h>
-#include <DataTypes/DataTypeFactory.h>
-#include <DataTypes/DataTypeNullable.h>
-#include <Columns/ColumnString.h>
-#include <Interpreters/Context.h>
-#include <Interpreters/parseColumnsListForTableFunction.h>
-
-
-namespace DB
-{
-
-namespace ErrorCodes
-{
-    extern const int ILLEGAL_TYPE_OF_ARGUMENT;
-}
-
-
-/** CastInternal does not preserve nullability of the data type,
-  * i.e. CastInternal(toNullable(toInt8(1)) as Int32) will be Int32(1).
-  *
-  * Cast preserves nullability according to setting `cast_keep_nullable`,
-  * i.e. Cast(toNullable(toInt8(1)) as Int32) will be Nullable(Int32(1)) if `cast_keep_nullable` == 1.
-  */
-class CastOverloadResolverImpl : public IFunctionOverloadResolver
-{
-public:
-    String getName() const override
-    {
-        if (cast_type == CastType::accurate)
-            return "accurateCast";
-        if (cast_type == CastType::accurateOrNull)
-            return "accurateCastOrNull";
-        if (internal)
-            return "_CAST";
-        else
-            return "CAST";
-    }
-
-    size_t getNumberOfArguments() const override { return 2; }
-
-    ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {1}; }
-
-    explicit CastOverloadResolverImpl(ContextPtr context_, CastType cast_type_, bool internal_, std::optional<CastDiagnostic> diagnostic_, bool keep_nullable_, const DataTypeValidationSettings & data_type_validation_settings_)
-        : context(context_)
-        , cast_type(cast_type_)
-        , internal(internal_)
-        , diagnostic(std::move(diagnostic_))
-        , keep_nullable(keep_nullable_)
-        , data_type_validation_settings(data_type_validation_settings_)
-    {
-    }
-
-    static FunctionOverloadResolverPtr create(ContextPtr context, CastType cast_type, bool internal, std::optional<CastDiagnostic> diagnostic)
-    {
-        const auto & settings_ref = context->getSettingsRef();
-
-        if (internal)
-            return std::make_unique<CastOverloadResolverImpl>(context, cast_type, internal, diagnostic, false /*keep_nullable*/, DataTypeValidationSettings{});
-        else
-            return std::make_unique<CastOverloadResolverImpl>(context, cast_type, internal, diagnostic, settings_ref.cast_keep_nullable, DataTypeValidationSettings(settings_ref));
-    }
-
-protected:
-    FunctionBasePtr buildImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & return_type) const override
-    {
-        return createFunctionBaseCast(context, arguments, return_type, diagnostic, cast_type);
-    }
-
-    DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override
-    {
-        const auto & column = arguments.back().column;
-        if (!column)
-            throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Second argument to {} must be a constant string describing type. "
-                "Instead there is non-constant column of type {}", getName(), arguments.back().type->getName());
-
-        const auto * type_col = checkAndGetColumnConst<ColumnString>(column.get());
-        if (!type_col)
-            throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Second argument to {} must be a constant string describing type. "
-                "Instead there is a column with the following structure: {}", getName(), column->dumpStructure());
-
-        DataTypePtr type = DataTypeFactory::instance().get(type_col->getValue<String>());
-        validateDataType(type, data_type_validation_settings);
-
-        if (cast_type == CastType::accurateOrNull)
-            return makeNullable(type);
-
-        if (internal)
-            return type;
-
-        if (keep_nullable && arguments.front().type->isNullable() && type->canBeInsideNullable())
-            return makeNullable(type);
-
-        return type;
-    }
-
-    bool useDefaultImplementationForNulls() const override { return false; }
-    bool useDefaultImplementationForNothing() const override { return false; }
-    bool useDefaultImplementationForLowCardinalityColumns() const override { return false; }
-
-private:
-    ContextPtr context;
-    CastType cast_type;
-    bool internal;
-    std::optional<CastDiagnostic> diagnostic;
-    bool keep_nullable;
-    DataTypeValidationSettings data_type_validation_settings;
-};
-
-}

From 061cee257c489eccdcbcda7e694937fa93444d18 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 9 Mar 2024 08:10:59 +0100
Subject: [PATCH 103/443] Remove crap

---
 src/Functions/CastOverloadResolver.cpp |  9 +++-
 src/Functions/FunctionsConversion.cpp  | 59 +++++++++++++++++---
 src/Functions/FunctionsConversion.h    | 75 ++++++--------------------
 src/Functions/concat.cpp               |  2 +-
 4 files changed, 76 insertions(+), 69 deletions(-)

diff --git a/src/Functions/CastOverloadResolver.cpp b/src/Functions/CastOverloadResolver.cpp
index 98debc3e2a4..4a081d684f6 100644
--- a/src/Functions/CastOverloadResolver.cpp
+++ b/src/Functions/CastOverloadResolver.cpp
@@ -1,12 +1,12 @@
 #include <Functions/FunctionFactory.h>
 #include <Functions/CastOverloadResolver.h>
-#include <Functions/FunctionsConversion.h>
 #include <Functions/FunctionHelpers.h>
 #include <DataTypes/DataTypeFactory.h>
 #include <DataTypes/DataTypesNumber.h>
 #include <DataTypes/DataTypeNullable.h>
 #include <Columns/ColumnString.h>
 #include <Interpreters/parseColumnsListForTableFunction.h>
+#include <Interpreters/Context.h>
 
 
 namespace DB
@@ -17,6 +17,13 @@ namespace ErrorCodes
     extern const int ILLEGAL_TYPE_OF_ARGUMENT;
 }
 
+FunctionBasePtr createFunctionBaseCast(
+    ContextPtr context
+    , const ColumnsWithTypeAndName & arguments
+    , const DataTypePtr & return_type
+    , std::optional<CastDiagnostic> diagnostic
+    , CastType cast_type);
+
 
 /** CastInternal does not preserve nullability of the data type,
   * i.e. CastInternal(toNullable(toInt8(1)) as Int32) will be Int32(1).
diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index ce9e4fd66e1..ebb63f1b25d 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -5,13 +5,6 @@
 namespace DB
 {
 
-namespace
-{
-
-struct CastInternalName { static constexpr auto name = "_CAST"; };
-
-}
-
 FunctionBasePtr createFunctionBaseCast(
     ContextPtr context
     , const ColumnsWithTypeAndName & arguments
@@ -25,7 +18,7 @@ FunctionBasePtr createFunctionBaseCast(
         data_types[i] = arguments[i].type;
 
     auto monotonicity = MonotonicityHelper::getMonotonicityInformation(arguments.front().type, return_type.get());
-    return std::make_unique<FunctionCast<CastInternalName>>(context, CastInternalName::name, std::move(monotonicity), data_types, return_type, diagnostic, cast_type);
+    return std::make_unique<FunctionCast>(context, "CAST", std::move(monotonicity), data_types, return_type, diagnostic, cast_type);
 }
 
 REGISTER_FUNCTION(Conversion)
@@ -153,4 +146,54 @@ REGISTER_FUNCTION(Conversion)
     factory.registerFunction<FunctionConvert<DataTypeInterval, NameToIntervalYear, PositiveMonotonicity>>();
 }
 
+
+MonotonicityHelper::MonotonicityForRange MonotonicityHelper::getMonotonicityInformation(const DataTypePtr & from_type, const IDataType * to_type)
+{
+    if (const auto * type = checkAndGetDataType<DataTypeUInt8>(to_type))
+        return monotonicityForType(type);
+    if (const auto * type = checkAndGetDataType<DataTypeUInt16>(to_type))
+        return monotonicityForType(type);
+    if (const auto * type = checkAndGetDataType<DataTypeUInt32>(to_type))
+        return monotonicityForType(type);
+    if (const auto * type = checkAndGetDataType<DataTypeUInt64>(to_type))
+        return monotonicityForType(type);
+    if (const auto * type = checkAndGetDataType<DataTypeUInt128>(to_type))
+        return monotonicityForType(type);
+    if (const auto * type = checkAndGetDataType<DataTypeUInt256>(to_type))
+        return monotonicityForType(type);
+    if (const auto * type = checkAndGetDataType<DataTypeInt8>(to_type))
+        return monotonicityForType(type);
+    if (const auto * type = checkAndGetDataType<DataTypeInt16>(to_type))
+        return monotonicityForType(type);
+    if (const auto * type = checkAndGetDataType<DataTypeInt32>(to_type))
+        return monotonicityForType(type);
+    if (const auto * type = checkAndGetDataType<DataTypeInt64>(to_type))
+        return monotonicityForType(type);
+    if (const auto * type = checkAndGetDataType<DataTypeInt128>(to_type))
+        return monotonicityForType(type);
+    if (const auto * type = checkAndGetDataType<DataTypeInt256>(to_type))
+        return monotonicityForType(type);
+    if (const auto * type = checkAndGetDataType<DataTypeFloat32>(to_type))
+        return monotonicityForType(type);
+    if (const auto * type = checkAndGetDataType<DataTypeFloat64>(to_type))
+        return monotonicityForType(type);
+    if (const auto * type = checkAndGetDataType<DataTypeDate>(to_type))
+        return monotonicityForType(type);
+    if (const auto * type = checkAndGetDataType<DataTypeDate32>(to_type))
+        return monotonicityForType(type);
+    if (const auto * type = checkAndGetDataType<DataTypeDateTime>(to_type))
+        return monotonicityForType(type);
+    if (const auto * type = checkAndGetDataType<DataTypeString>(to_type))
+        return monotonicityForType(type);
+    if (isEnum(from_type))
+    {
+        if (const auto * type = checkAndGetDataType<DataTypeEnum8>(to_type))
+            return monotonicityForType(type);
+        if (const auto * type = checkAndGetDataType<DataTypeEnum16>(to_type))
+            return monotonicityForType(type);
+    }
+    /// other types like Null, FixedString, Array and Tuple have no monotonicity defined
+    return {};
+}
+
 }
diff --git a/src/Functions/FunctionsConversion.h b/src/Functions/FunctionsConversion.h
index 689b72dc917..0c4b7b41b93 100644
--- a/src/Functions/FunctionsConversion.h
+++ b/src/Functions/FunctionsConversion.h
@@ -3170,13 +3170,17 @@ private:
 };
 
 
+struct FunctionCastName
+{
+    static constexpr auto name = "CAST";
+};
+
 class FunctionCastBase : public IFunctionBase
 {
 public:
     using MonotonicityForRange = std::function<Monotonicity(const IDataType &, const Field &, const Field &)>;
 };
 
-template <typename FunctionName>
 class FunctionCast final : public FunctionCastBase
 {
 public:
@@ -3204,7 +3208,7 @@ public:
         try
         {
             return std::make_unique<ExecutableFunctionCast>(
-                    prepareUnpackDictionaries(getArgumentTypes()[0], getResultType()), cast_name, diagnostic);
+                prepareUnpackDictionaries(getArgumentTypes()[0], getResultType()), cast_name, diagnostic);
         }
         catch (Exception & e)
         {
@@ -3278,7 +3282,7 @@ private:
         {
             /// In case when converting to Nullable type, we apply different parsing rule,
             /// that will not throw an exception but return NULL in case of malformed input.
-            FunctionPtr function = FunctionConvertFromString<ToDataType, FunctionName, ConvertFromStringExceptionMode::Null>::create();
+            FunctionPtr function = FunctionConvertFromString<ToDataType, FunctionCastName, ConvertFromStringExceptionMode::Null>::create();
             return createFunctionAdaptor(function, from_type);
         }
         else if (!can_apply_accurate_cast)
@@ -3304,7 +3308,7 @@ private:
                     {
 #define GENERATE_OVERFLOW_MODE_CASE(OVERFLOW_MODE, ADDITIONS) \
             case FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE: \
-                result_column = ConvertImpl<LeftDataType, RightDataType, FunctionName, ConvertDefaultBehaviorTag, FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE>::execute( \
+                result_column = ConvertImpl<LeftDataType, RightDataType, FunctionCastName, ConvertDefaultBehaviorTag, FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE>::execute( \
                 arguments, result_type, input_rows_count, ADDITIONS()); \
                 break;
                         if (wrapper_cast_type == CastType::accurate)
@@ -3334,7 +3338,7 @@ private:
                     {
 #define GENERATE_OVERFLOW_MODE_CASE(OVERFLOW_MODE, ADDITIONS) \
             case FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE: \
-            result_column = ConvertImpl<LeftDataType, RightDataType, FunctionName, ConvertDefaultBehaviorTag, FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE>::template execute<ADDITIONS>( \
+            result_column = ConvertImpl<LeftDataType, RightDataType, FunctionCastName, ConvertDefaultBehaviorTag, FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE>::template execute<ADDITIONS>( \
 arguments, result_type, input_rows_count); \
                 break;
                         if (wrapper_cast_type == CastType::accurate)
@@ -3368,7 +3372,7 @@ arguments, result_type, input_rows_count); \
             {
                 if (wrapper_cast_type == CastType::accurateOrNull)
                 {
-                    auto nullable_column_wrapper = FunctionCast<FunctionName>::createToNullableColumnWrapper();
+                    auto nullable_column_wrapper = FunctionCast::createToNullableColumnWrapper();
                     return nullable_column_wrapper(arguments, result_type, column_nullable, input_rows_count);
                 }
                 else
@@ -3494,7 +3498,7 @@ arguments, result_type, input_rows_count); \
                     {
                         AccurateConvertStrategyAdditions additions;
                         additions.scale = scale;
-                        result_column = ConvertImpl<LeftDataType, RightDataType, FunctionName>::execute(
+                        result_column = ConvertImpl<LeftDataType, RightDataType, FunctionCastName>::execute(
                             arguments, result_type, input_rows_count, additions);
 
                         return true;
@@ -3503,7 +3507,7 @@ arguments, result_type, input_rows_count); \
                     {
                         AccurateOrNullConvertStrategyAdditions additions;
                         additions.scale = scale;
-                        result_column = ConvertImpl<LeftDataType, RightDataType, FunctionName>::execute(
+                        result_column = ConvertImpl<LeftDataType, RightDataType, FunctionCastName>::execute(
                             arguments, result_type, input_rows_count, additions);
 
                         return true;
@@ -3516,14 +3520,14 @@ arguments, result_type, input_rows_count); \
                         /// Consistent with CAST(Nullable(String) AS Nullable(Numbers))
                         /// In case when converting to Nullable type, we apply different parsing rule,
                         /// that will not throw an exception but return NULL in case of malformed input.
-                        result_column = ConvertImpl<LeftDataType, RightDataType, FunctionName, ConvertReturnNullOnErrorTag>::execute(
+                        result_column = ConvertImpl<LeftDataType, RightDataType, FunctionCastName, ConvertReturnNullOnErrorTag>::execute(
                             arguments, result_type, input_rows_count, scale);
 
                         return true;
                     }
                 }
 
-                result_column = ConvertImpl<LeftDataType, RightDataType, FunctionName>::execute(arguments, result_type, input_rows_count, scale);
+                result_column = ConvertImpl<LeftDataType, RightDataType, FunctionCastName>::execute(arguments, result_type, input_rows_count, scale);
 
                 return true;
             });
@@ -3533,7 +3537,7 @@ arguments, result_type, input_rows_count); \
             {
                 if (wrapper_cast_type == CastType::accurateOrNull)
                 {
-                    auto nullable_column_wrapper = FunctionCast<FunctionName>::createToNullableColumnWrapper();
+                    auto nullable_column_wrapper = FunctionCast::createToNullableColumnWrapper();
                     return nullable_column_wrapper(arguments, result_type, column_nullable, input_rows_count);
                 }
                 else
@@ -4922,54 +4926,7 @@ public:
         return FunctionTo<DataType>::Type::Monotonic::get;
     }
 
-    static MonotonicityForRange getMonotonicityInformation(const DataTypePtr & from_type, const IDataType * to_type)
-    {
-        if (const auto * type = checkAndGetDataType<DataTypeUInt8>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeUInt16>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeUInt32>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeUInt64>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeUInt128>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeUInt256>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeInt8>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeInt16>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeInt32>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeInt64>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeInt128>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeInt256>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeFloat32>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeFloat64>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeDate>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeDate32>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeDateTime>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeString>(to_type))
-            return monotonicityForType(type);
-        if (isEnum(from_type))
-        {
-            if (const auto * type = checkAndGetDataType<DataTypeEnum8>(to_type))
-                return monotonicityForType(type);
-            if (const auto * type = checkAndGetDataType<DataTypeEnum16>(to_type))
-                return monotonicityForType(type);
-        }
-        /// other types like Null, FixedString, Array and Tuple have no monotonicity defined
-        return {};
-    }
+    static MonotonicityForRange getMonotonicityInformation(const DataTypePtr & from_type, const IDataType * to_type);
 };
 
 FunctionBasePtr createFunctionBaseCast(
diff --git a/src/Functions/concat.cpp b/src/Functions/concat.cpp
index d68f5256f6d..c75a806559c 100644
--- a/src/Functions/concat.cpp
+++ b/src/Functions/concat.cpp
@@ -1,8 +1,8 @@
 #include <Columns/ColumnString.h>
+#include <Columns/ColumnStringHelpers.h>
 #include <DataTypes/DataTypeString.h>
 #include <Functions/FunctionFactory.h>
 #include <Functions/FunctionHelpers.h>
-#include <Functions/FunctionsConversion.h>
 #include <Functions/GatherUtils/Algorithms.h>
 #include <Functions/GatherUtils/Sinks.h>
 #include <Functions/GatherUtils/Sources.h>

From 804c07156d90c0b4ce6b632f30b5481077d0f5d3 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 9 Mar 2024 08:24:53 +0100
Subject: [PATCH 104/443] Fix garbage

---
 .../Serializations/SerializationObject.cpp    | 20 +++-
 src/Functions/FunctionsConversion.h           | 97 ++++++++-----------
 2 files changed, 57 insertions(+), 60 deletions(-)

diff --git a/src/DataTypes/Serializations/SerializationObject.cpp b/src/DataTypes/Serializations/SerializationObject.cpp
index e6dc16ef5a0..cd186ec2c46 100644
--- a/src/DataTypes/Serializations/SerializationObject.cpp
+++ b/src/DataTypes/Serializations/SerializationObject.cpp
@@ -5,13 +5,11 @@
 #include <DataTypes/DataTypeNullable.h>
 #include <DataTypes/ObjectUtils.h>
 #include <DataTypes/DataTypeFactory.h>
-#include <DataTypes/NestedUtils.h>
 #include <Common/JSONParsers/SimdJSONParser.h>
 #include <Common/JSONParsers/RapidJSONParser.h>
-#include <Common/HashTable/HashSet.h>
 #include <Columns/ColumnObject.h>
+#include <Columns/ColumnTuple.h>
 #include <Columns/ColumnString.h>
-#include <Functions/FunctionsConversion.h>
 
 #include <IO/ReadHelpers.h>
 #include <IO/WriteHelpers.h>
@@ -30,6 +28,7 @@ namespace ErrorCodes
     extern const int CANNOT_READ_ALL_DATA;
     extern const int ARGUMENT_OUT_OF_BOUND;
     extern const int LOGICAL_ERROR;
+    extern const int CANNOT_PARSE_TEXT;
 }
 
 template <typename Parser>
@@ -344,7 +343,20 @@ void SerializationObject<Parser>::deserializeBinaryBulkFromString(
     state.nested_serialization->deserializeBinaryBulkWithMultipleStreams(
         column_string, limit, settings, state.nested_state, cache);
 
-    ConvertImplGenericFromString<ColumnString>::executeImpl(*column_string, column_object, *this, column_string->size());
+    size_t input_rows_count = column_string->size();
+    column_object.reserve(input_rows_count);
+
+    FormatSettings format_settings;
+    for (size_t i = 0; i < input_rows_count; ++i)
+    {
+        const auto & val = column_string->getDataAt(i);
+        ReadBufferFromMemory read_buffer(val.data, val.size);
+        deserializeWholeText(column_object, read_buffer, format_settings);
+
+        if (!read_buffer.eof())
+            throw Exception(ErrorCodes::CANNOT_PARSE_TEXT,
+                "Cannot parse string to column Object. Expected eof");
+    }
 }
 
 template <typename Parser>
diff --git a/src/Functions/FunctionsConversion.h b/src/Functions/FunctionsConversion.h
index 0c4b7b41b93..c21e85fb40e 100644
--- a/src/Functions/FunctionsConversion.h
+++ b/src/Functions/FunctionsConversion.h
@@ -1818,14 +1818,10 @@ struct ConvertImpl<FromDataType, ToDataType, Name, ConvertReturnZeroOnErrorTag,
     : ConvertThroughParsing<FromDataType, ToDataType, Name, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::Normal> {};
 
 /// Generic conversion of any type from String. Used for complex types: Array and Tuple or types with custom serialization.
-template <typename StringColumnType>
 struct ConvertImplGenericFromString
 {
     static ColumnPtr execute(ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable * column_nullable, size_t input_rows_count)
     {
-        static_assert(std::is_same_v<StringColumnType, ColumnString> || std::is_same_v<StringColumnType, ColumnFixedString>,
-                "Can be used only to parse from ColumnString or ColumnFixedString");
-
         const IColumn & column_from = *arguments[0].column;
         const IDataType & data_type_to = *result_type;
         auto res = data_type_to.createColumn();
@@ -1841,63 +1837,52 @@ struct ConvertImplGenericFromString
         IColumn & column_to,
         const ISerialization & serialization_from,
         size_t input_rows_count,
-        const PaddedPODArray<UInt8> * null_map = nullptr,
-        const IDataType * result_type = nullptr)
+        const PaddedPODArray<UInt8> * null_map,
+        const IDataType * result_type)
     {
-        static_assert(std::is_same_v<StringColumnType, ColumnString> || std::is_same_v<StringColumnType, ColumnFixedString>,
-                "Can be used only to parse from ColumnString or ColumnFixedString");
+        column_to.reserve(input_rows_count);
 
-        if (const StringColumnType * col_from_string = checkAndGetColumn<StringColumnType>(&column_from))
+        FormatSettings format_settings;
+        for (size_t i = 0; i < input_rows_count; ++i)
         {
-            column_to.reserve(input_rows_count);
-
-            FormatSettings format_settings;
-            for (size_t i = 0; i < input_rows_count; ++i)
+            if (null_map && (*null_map)[i])
             {
-                if (null_map && (*null_map)[i])
+                column_to.insertDefault();
+                continue;
+            }
+
+            const auto & val = column_from.getDataAt(i);
+            ReadBufferFromMemory read_buffer(val.data, val.size);
+            try
+            {
+                serialization_from.deserializeWholeText(column_to, read_buffer, format_settings);
+            }
+            catch (const Exception & e)
+            {
+                auto * nullable_column = typeid_cast<ColumnNullable *>(&column_to);
+                if (e.code() == ErrorCodes::CANNOT_PARSE_BOOL && nullable_column)
                 {
-                    column_to.insertDefault();
+                    auto & col_nullmap = nullable_column->getNullMapData();
+                    if (col_nullmap.size() != nullable_column->size())
+                        col_nullmap.resize_fill(nullable_column->size());
+                    if (nullable_column->size() == (i + 1))
+                        nullable_column->popBack(1);
+                    nullable_column->insertDefault();
                     continue;
                 }
+                throw;
+            }
 
-                const auto & val = col_from_string->getDataAt(i);
-                ReadBufferFromMemory read_buffer(val.data, val.size);
-                try
-                {
-                    serialization_from.deserializeWholeText(column_to, read_buffer, format_settings);
-                }
-                catch (const Exception & e)
-                {
-                    auto * nullable_column = typeid_cast<ColumnNullable *>(&column_to);
-                    if (e.code() == ErrorCodes::CANNOT_PARSE_BOOL && nullable_column)
-                    {
-                        auto & col_nullmap = nullable_column->getNullMapData();
-                        if (col_nullmap.size() != nullable_column->size())
-                            col_nullmap.resize_fill(nullable_column->size());
-                        if (nullable_column->size() == (i + 1))
-                            nullable_column->popBack(1);
-                        nullable_column->insertDefault();
-                        continue;
-                    }
-                    throw;
-                }
-
-                if (!read_buffer.eof())
-                {
-                    if (result_type)
-                        throwExceptionForIncompletelyParsedValue(read_buffer, *result_type);
-                    else
-                        throw Exception(ErrorCodes::CANNOT_PARSE_TEXT,
-                            "Cannot parse string to column {}. Expected eof", column_to.getName());
-                }
+            if (!read_buffer.eof())
+            {
+                if (result_type)
+                    throwExceptionForIncompletelyParsedValue(read_buffer, *result_type);
+                else
+                    throw Exception(ErrorCodes::CANNOT_PARSE_TEXT,
+                        "Cannot parse string to column {}. Expected eof", column_to.getName());
             }
         }
-        else
-            throw Exception(ErrorCodes::ILLEGAL_COLUMN,
-                "Illegal column {} of first argument of conversion function from string",
-                column_from.getName());
     }
-
 };
 
 
@@ -3392,7 +3377,7 @@ arguments, result_type, input_rows_count); \
     {
         if (checkAndGetDataType<DataTypeString>(from_type.get()))
         {
-            return &ConvertImplGenericFromString<ColumnString>::execute;
+            return &ConvertImplGenericFromString::execute;
         }
 
         return createWrapper<ToDataType>(from_type, to_type, requested_result_is_nullable);
@@ -3555,7 +3540,7 @@ arguments, result_type, input_rows_count); \
         /// Conversion from String through parsing.
         if (checkAndGetDataType<DataTypeString>(from_type_untyped.get()))
         {
-            return &ConvertImplGenericFromString<ColumnString>::execute;
+            return &ConvertImplGenericFromString::execute;
         }
         else if (const auto * agg_type = checkAndGetDataType<DataTypeAggregateFunction>(from_type_untyped.get()))
         {
@@ -3598,7 +3583,7 @@ arguments, result_type, input_rows_count); \
         /// Conversion from String through parsing.
         if (checkAndGetDataType<DataTypeString>(from_type_untyped.get()))
         {
-            return &ConvertImplGenericFromString<ColumnString>::execute;
+            return &ConvertImplGenericFromString::execute;
         }
 
         DataTypePtr from_type_holder;
@@ -3689,7 +3674,7 @@ arguments, result_type, input_rows_count); \
         /// Conversion from String through parsing.
         if (checkAndGetDataType<DataTypeString>(from_type_untyped.get()))
         {
-            return &ConvertImplGenericFromString<ColumnString>::execute;
+            return &ConvertImplGenericFromString::execute;
         }
 
         const auto * from_type = checkAndGetDataType<DataTypeTuple>(from_type_untyped.get());
@@ -4034,7 +4019,7 @@ arguments, result_type, input_rows_count); \
         {
             return [] (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable * nullable_source, size_t input_rows_count)
             {
-                auto res = ConvertImplGenericFromString<ColumnString>::execute(arguments, result_type, nullable_source, input_rows_count)->assumeMutable();
+                auto res = ConvertImplGenericFromString::execute(arguments, result_type, nullable_source, input_rows_count)->assumeMutable();
                 res->finalize();
                 return res;
             };
@@ -4831,7 +4816,7 @@ arguments, result_type, input_rows_count); \
                         auto wrapped_result_type = result_type;
                         if (requested_result_is_nullable)
                             wrapped_result_type = makeNullable(result_type);
-                        return ConvertImplGenericFromString<typename FromDataType::ColumnType>::execute(
+                        return ConvertImplGenericFromString::execute(
                             arguments, wrapped_result_type, column_nullable, input_rows_count);
                     };
                     return true;

From f2c4a5bb94ea319b38caf35ac90a9b9208cdc745 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 9 Mar 2024 08:30:13 +0100
Subject: [PATCH 105/443] Prevent garbage from appearing

---
 src/Functions/FunctionsConversion.cpp | 5008 ++++++++++++++++++++++++-
 src/Functions/FunctionsConversion.h   | 4924 ------------------------
 2 files changed, 4957 insertions(+), 4975 deletions(-)
 delete mode 100644 src/Functions/FunctionsConversion.h

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index ebb63f1b25d..865f7db8e12 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -1,10 +1,4966 @@
+#include <cstddef>
+#include <type_traits>
+
+#include <IO/WriteBufferFromVector.h>
+#include <IO/ReadBufferFromMemory.h>
+#include <IO/Operators.h>
+#include <IO/parseDateTimeBestEffort.h>
+#include <DataTypes/DataTypeFactory.h>
+#include <DataTypes/DataTypesNumber.h>
+#include <DataTypes/DataTypesDecimal.h>
+#include <DataTypes/DataTypeString.h>
+#include <DataTypes/DataTypeFixedString.h>
+#include <DataTypes/DataTypeDate.h>
+#include <DataTypes/DataTypeDate32.h>
+#include <DataTypes/DataTypeDateTime.h>
+#include <DataTypes/DataTypeDateTime64.h>
+#include <DataTypes/DataTypeEnum.h>
+#include <DataTypes/DataTypeArray.h>
+#include <DataTypes/DataTypeTuple.h>
+#include <DataTypes/DataTypeMap.h>
+#include <DataTypes/DataTypeNullable.h>
+#include <DataTypes/DataTypeNothing.h>
+#include <DataTypes/DataTypeUUID.h>
+#include <DataTypes/DataTypeInterval.h>
+#include <DataTypes/DataTypeAggregateFunction.h>
+#include <DataTypes/DataTypeObject.h>
+#include <DataTypes/ObjectUtils.h>
+#include <DataTypes/DataTypeNested.h>
+#include <DataTypes/DataTypeVariant.h>
+#include <DataTypes/Serializations/SerializationDecimal.h>
+#include <Formats/FormatSettings.h>
+#include <Columns/ColumnString.h>
+#include <Columns/ColumnFixedString.h>
+#include <Columns/ColumnConst.h>
+#include <Columns/ColumnAggregateFunction.h>
+#include <Columns/ColumnArray.h>
+#include <Columns/ColumnNullable.h>
+#include <Columns/ColumnTuple.h>
+#include <Columns/ColumnMap.h>
+#include <Columns/ColumnObject.h>
+#include <Columns/ColumnsCommon.h>
+#include <Columns/ColumnVariant.h>
+#include <Columns/ColumnStringHelpers.h>
+#include <Common/assert_cast.h>
+#include <Common/Concepts.h>
+#include <Common/quoteString.h>
+#include <Common/Exception.h>
+#include <Core/AccurateComparison.h>
 #include <Functions/FunctionFactory.h>
-#include <Functions/FunctionsConversion.h>
+#include <Functions/IFunctionAdaptors.h>
+#include <Functions/FunctionHelpers.h>
+#include <Functions/DateTimeTransforms.h>
+#include <Functions/toFixedString.h>
+#include <Functions/TransformDateTime64.h>
+#include <Functions/FunctionsCodingIP.h>
+#include <Functions/CastOverloadResolver.h>
+#include <DataTypes/DataTypeLowCardinality.h>
+#include <Columns/ColumnLowCardinality.h>
+#include <Interpreters/Context.h>
+#include <Common/HashTable/HashMap.h>
+#include <DataTypes/DataTypeIPv4andIPv6.h>
+#include <Common/IPv6ToBinary.h>
+#include <Core/Types.h>
 
 
 namespace DB
 {
 
+namespace ErrorCodes
+{
+    extern const int ATTEMPT_TO_READ_AFTER_EOF;
+    extern const int CANNOT_PARSE_NUMBER;
+    extern const int CANNOT_READ_ARRAY_FROM_TEXT;
+    extern const int CANNOT_PARSE_INPUT_ASSERTION_FAILED;
+    extern const int CANNOT_PARSE_QUOTED_STRING;
+    extern const int CANNOT_PARSE_ESCAPE_SEQUENCE;
+    extern const int CANNOT_PARSE_DATE;
+    extern const int CANNOT_PARSE_DATETIME;
+    extern const int CANNOT_PARSE_TEXT;
+    extern const int CANNOT_PARSE_UUID;
+    extern const int CANNOT_PARSE_IPV4;
+    extern const int CANNOT_PARSE_IPV6;
+    extern const int TOO_FEW_ARGUMENTS_FOR_FUNCTION;
+    extern const int LOGICAL_ERROR;
+    extern const int TYPE_MISMATCH;
+    extern const int CANNOT_CONVERT_TYPE;
+    extern const int ILLEGAL_COLUMN;
+    extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH;
+    extern const int ILLEGAL_TYPE_OF_ARGUMENT;
+    extern const int NOT_IMPLEMENTED;
+    extern const int CANNOT_INSERT_NULL_IN_ORDINARY_COLUMN;
+    extern const int CANNOT_PARSE_BOOL;
+    extern const int VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE;
+}
+
+/** Type conversion functions.
+  * toType - conversion in "natural way";
+  */
+
+UInt32 extractToDecimalScale(const ColumnWithTypeAndName & named_column);
+
+/// Function toUnixTimestamp has exactly the same implementation as toDateTime of String type.
+struct NameToUnixTimestamp { static constexpr auto name = "toUnixTimestamp"; };
+
+struct AccurateConvertStrategyAdditions
+{
+    UInt32 scale { 0 };
+};
+
+struct AccurateOrNullConvertStrategyAdditions
+{
+    UInt32 scale { 0 };
+};
+
+
+struct ConvertDefaultBehaviorTag {};
+struct ConvertReturnNullOnErrorTag {};
+struct ConvertReturnZeroOnErrorTag {};
+
+/** Conversion of number types to each other, enums to numbers, dates and datetimes to numbers and back: done by straight assignment.
+  *  (Date is represented internally as number of days from some day; DateTime - as unix timestamp)
+  */
+template <typename FromDataType, typename ToDataType, typename Name,
+    typename SpecialTag = ConvertDefaultBehaviorTag,
+    FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior = default_date_time_overflow_behavior>
+struct ConvertImpl
+{
+    using FromFieldType = typename FromDataType::FieldType;
+    using ToFieldType = typename ToDataType::FieldType;
+
+    template <typename Additions = void *>
+    static ColumnPtr NO_SANITIZE_UNDEFINED execute(
+        const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type [[maybe_unused]], size_t input_rows_count,
+        Additions additions [[maybe_unused]] = Additions())
+    {
+        const ColumnWithTypeAndName & named_from = arguments[0];
+
+        using ColVecFrom = typename FromDataType::ColumnType;
+        using ColVecTo = typename ToDataType::ColumnType;
+
+        if constexpr ((IsDataTypeDecimal<FromDataType> || IsDataTypeDecimal<ToDataType>)
+            && !(std::is_same_v<DataTypeDateTime64, FromDataType> || std::is_same_v<DataTypeDateTime64, ToDataType>))
+        {
+            if constexpr (!IsDataTypeDecimalOrNumber<FromDataType> || !IsDataTypeDecimalOrNumber<ToDataType>)
+            {
+                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
+                    named_from.column->getName(), Name::name);
+            }
+        }
+
+        if (const ColVecFrom * col_from = checkAndGetColumn<ColVecFrom>(named_from.column.get()))
+        {
+            typename ColVecTo::MutablePtr col_to = nullptr;
+
+            if constexpr (IsDataTypeDecimal<ToDataType>)
+            {
+                UInt32 scale;
+
+                if constexpr (std::is_same_v<Additions, AccurateConvertStrategyAdditions>
+                    || std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
+                {
+                    scale = additions.scale;
+                }
+                else
+                {
+                    scale = additions;
+                }
+
+                col_to = ColVecTo::create(0, scale);
+            }
+            else
+                col_to = ColVecTo::create();
+
+            const auto & vec_from = col_from->getData();
+            auto & vec_to = col_to->getData();
+            vec_to.resize(input_rows_count);
+
+            ColumnUInt8::MutablePtr col_null_map_to;
+            ColumnUInt8::Container * vec_null_map_to [[maybe_unused]] = nullptr;
+            if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
+            {
+                col_null_map_to = ColumnUInt8::create(input_rows_count, false);
+                vec_null_map_to = &col_null_map_to->getData();
+            }
+
+            bool result_is_bool = isBool(result_type);
+            for (size_t i = 0; i < input_rows_count; ++i)
+            {
+                if constexpr (std::is_same_v<ToDataType, DataTypeUInt8>)
+                {
+                    if (result_is_bool)
+                    {
+                        vec_to[i] = vec_from[i] != FromFieldType(0);
+                        continue;
+                    }
+                }
+
+                if constexpr (std::is_same_v<FromDataType, DataTypeUUID> && std::is_same_v<ToDataType, DataTypeUInt128>)
+                {
+                    static_assert(
+                        std::is_same_v<DataTypeUInt128::FieldType, DataTypeUUID::FieldType::UnderlyingType>,
+                        "UInt128 and UUID types must be same");
+
+                    vec_to[i].items[1] = vec_from[i].toUnderType().items[0];
+                    vec_to[i].items[0] = vec_from[i].toUnderType().items[1];
+
+                    continue;
+                }
+
+                if constexpr (std::is_same_v<FromDataType, DataTypeIPv6> && std::is_same_v<ToDataType, DataTypeUInt128>)
+                {
+                    static_assert(
+                        std::is_same_v<DataTypeUInt128::FieldType, DataTypeUUID::FieldType::UnderlyingType>,
+                        "UInt128 and IPv6 types must be same");
+
+                    vec_to[i].items[1] = std::byteswap(vec_from[i].toUnderType().items[0]);
+                    vec_to[i].items[0] = std::byteswap(vec_from[i].toUnderType().items[1]);
+
+                    continue;
+                }
+
+                if constexpr (std::is_same_v<FromDataType, DataTypeUUID> != std::is_same_v<ToDataType, DataTypeUUID>)
+                {
+                    throw Exception(ErrorCodes::NOT_IMPLEMENTED,
+                                    "Conversion between numeric types and UUID is not supported. "
+                                    "Probably the passed UUID is unquoted");
+                }
+                else if constexpr (
+                    (std::is_same_v<FromDataType, DataTypeIPv4> != std::is_same_v<ToDataType, DataTypeIPv4>)
+                    && !(is_any_of<FromDataType, DataTypeUInt8, DataTypeUInt16, DataTypeUInt32, DataTypeUInt64, DataTypeIPv6> || is_any_of<ToDataType, DataTypeUInt32, DataTypeUInt64, DataTypeUInt128, DataTypeUInt256, DataTypeIPv6>)
+                )
+                {
+                    throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Conversion from {} to {} is not supported",
+                                    TypeName<typename FromDataType::FieldType>, TypeName<typename ToDataType::FieldType>);
+                }
+                else if constexpr (std::is_same_v<FromDataType, DataTypeIPv6> != std::is_same_v<ToDataType, DataTypeIPv6> && !(std::is_same_v<ToDataType, DataTypeIPv4> || std::is_same_v<FromDataType, DataTypeIPv4>))
+                {
+                    throw Exception(ErrorCodes::NOT_IMPLEMENTED,
+                                    "Conversion between numeric types and IPv6 is not supported. "
+                                    "Probably the passed IPv6 is unquoted");
+                }
+                else
+                {
+                    if constexpr (IsDataTypeDecimal<FromDataType> || IsDataTypeDecimal<ToDataType>)
+                    {
+                        if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
+                        {
+                            ToFieldType result;
+                            bool convert_result = false;
+
+                            if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeDecimal<ToDataType>)
+                                convert_result = tryConvertDecimals<FromDataType, ToDataType>(vec_from[i], col_from->getScale(), col_to->getScale(), result);
+                            else if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeNumber<ToDataType>)
+                                convert_result = tryConvertFromDecimal<FromDataType, ToDataType>(vec_from[i], col_from->getScale(), result);
+                            else if constexpr (IsDataTypeNumber<FromDataType> && IsDataTypeDecimal<ToDataType>)
+                                convert_result = tryConvertToDecimal<FromDataType, ToDataType>(vec_from[i], col_to->getScale(), result);
+
+                            if (convert_result)
+                                vec_to[i] = result;
+                            else
+                            {
+                                vec_to[i] = static_cast<ToFieldType>(0);
+                                (*vec_null_map_to)[i] = true;
+                            }
+                        }
+                        else
+                        {
+                            if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeDecimal<ToDataType>)
+                                vec_to[i] = convertDecimals<FromDataType, ToDataType>(vec_from[i], col_from->getScale(), col_to->getScale());
+                            else if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeNumber<ToDataType>)
+                                vec_to[i] = convertFromDecimal<FromDataType, ToDataType>(vec_from[i], col_from->getScale());
+                            else if constexpr (IsDataTypeNumber<FromDataType> && IsDataTypeDecimal<ToDataType>)
+                                vec_to[i] = convertToDecimal<FromDataType, ToDataType>(vec_from[i], col_to->getScale());
+                            else
+                                throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Unsupported data type in conversion function");
+                        }
+                    }
+                    else
+                    {
+                        /// If From Data is Nan or Inf and we convert to integer type, throw exception
+                        if constexpr (std::is_floating_point_v<FromFieldType> && !std::is_floating_point_v<ToFieldType>)
+                        {
+                            if (!isFinite(vec_from[i]))
+                            {
+                                if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
+                                {
+                                    vec_to[i] = 0;
+                                    (*vec_null_map_to)[i] = true;
+                                    continue;
+                                }
+                                else
+                                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Unexpected inf or nan to integer conversion");
+                            }
+                        }
+
+                        if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>
+                                || std::is_same_v<Additions, AccurateConvertStrategyAdditions>)
+                        {
+                            bool convert_result = accurate::convertNumeric(vec_from[i], vec_to[i]);
+
+                            if (!convert_result)
+                            {
+                                if (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
+                                {
+                                    vec_to[i] = 0;
+                                    (*vec_null_map_to)[i] = true;
+                                }
+                                else
+                                {
+                                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Value in column {} cannot be safely converted into type {}",
+                                        named_from.column->getName(), result_type->getName());
+                                }
+                            }
+                        }
+                        else
+                        {
+                            if constexpr (std::is_same_v<ToDataType, DataTypeIPv4> && std::is_same_v<FromDataType, DataTypeIPv6>)
+                            {
+                                const uint8_t ip4_cidr[] {0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x00, 0x00, 0x00, 0x00};
+                                const uint8_t * src = reinterpret_cast<const uint8_t *>(&vec_from[i].toUnderType());
+                                if (!matchIPv6Subnet(src, ip4_cidr, 96))
+                                {
+                                    char addr[IPV6_MAX_TEXT_LENGTH + 1] {};
+                                    char * paddr = addr;
+                                    formatIPv6(src, paddr);
+
+                                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "IPv6 {} in column {} is not in IPv4 mapping block", addr, named_from.column->getName());
+                                }
+
+                                uint8_t * dst = reinterpret_cast<uint8_t *>(&vec_to[i].toUnderType());
+                                if constexpr (std::endian::native == std::endian::little)
+                                {
+                                    dst[0] = src[15];
+                                    dst[1] = src[14];
+                                    dst[2] = src[13];
+                                    dst[3] = src[12];
+                                }
+                                else
+                                {
+                                    dst[0] = src[12];
+                                    dst[1] = src[13];
+                                    dst[2] = src[14];
+                                    dst[3] = src[15];
+                                }
+                            }
+                            else if constexpr (std::is_same_v<ToDataType, DataTypeIPv6> && std::is_same_v<FromDataType, DataTypeIPv4>)
+                            {
+                                const uint8_t * src = reinterpret_cast<const uint8_t *>(&vec_from[i].toUnderType());
+                                uint8_t * dst = reinterpret_cast<uint8_t *>(&vec_to[i].toUnderType());
+                                std::memset(dst, '\0', IPV6_BINARY_LENGTH);
+                                dst[10] = dst[11] = 0xff;
+
+                                if constexpr (std::endian::native == std::endian::little)
+                                {
+                                    dst[12] = src[3];
+                                    dst[13] = src[2];
+                                    dst[14] = src[1];
+                                    dst[15] = src[0];
+                                }
+                                else
+                                {
+                                    dst[12] = src[0];
+                                    dst[13] = src[1];
+                                    dst[14] = src[2];
+                                    dst[15] = src[3];
+                                }
+                            }
+                            else if constexpr (std::is_same_v<ToDataType, DataTypeIPv4> && std::is_same_v<FromDataType, DataTypeUInt64>)
+                                vec_to[i] = static_cast<ToFieldType>(static_cast<IPv4::UnderlyingType>(vec_from[i]));
+                            else if constexpr (std::is_same_v<Name, NameToUnixTimestamp> && (std::is_same_v<FromDataType, DataTypeDate> || std::is_same_v<FromDataType, DataTypeDate32>))
+                                vec_to[i] = static_cast<ToFieldType>(vec_from[i] * DATE_SECONDS_PER_DAY);
+                            else
+                                vec_to[i] = static_cast<ToFieldType>(vec_from[i]);
+                        }
+                    }
+                }
+            }
+
+            if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
+                return ColumnNullable::create(std::move(col_to), std::move(col_null_map_to));
+            else
+                return col_to;
+        }
+        else
+            throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
+                named_from.column->getName(), Name::name);
+    }
+};
+
+/** Conversion of DateTime to Date: throw off time component.
+  */
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeDateTime, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeDateTime, DataTypeDate, ToDateImpl<date_time_overflow_behavior>, false> {};
+
+/** Conversion of DateTime to Date32: throw off time component.
+  */
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeDateTime, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeDateTime, DataTypeDate32, ToDate32Impl, false> {};
+
+/** Conversion of Date to DateTime: adding 00:00:00 time component.
+  */
+template <FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior = default_date_time_overflow_behavior>
+struct ToDateTimeImpl
+{
+    static constexpr auto name = "toDateTime";
+
+    static UInt32 execute(UInt16 d, const DateLUTImpl & time_zone)
+    {
+        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+        {
+            if (d > MAX_DATETIME_DAY_NUM) [[unlikely]]
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Day number {} is out of bounds of type DateTime", d);
+        }
+        else if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Saturate)
+        {
+            if (d > MAX_DATETIME_DAY_NUM)
+                d = MAX_DATETIME_DAY_NUM;
+        }
+        return static_cast<UInt32>(time_zone.fromDayNum(DayNum(d)));
+    }
+
+    static UInt32 execute(Int32 d, const DateLUTImpl & time_zone)
+    {
+        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Saturate)
+        {
+            if (d < 0)
+                return 0;
+            else if (d > MAX_DATETIME_DAY_NUM)
+                d = MAX_DATETIME_DAY_NUM;
+        }
+        else if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+        {
+            if (d < 0 || d > MAX_DATETIME_DAY_NUM) [[unlikely]]
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type DateTime", d);
+        }
+        return static_cast<UInt32>(time_zone.fromDayNum(ExtendedDayNum(d)));
+    }
+
+    static UInt32 execute(UInt32 dt, const DateLUTImpl & /*time_zone*/)
+    {
+        return dt;
+    }
+
+    static UInt32 execute(Int64 dt64, const DateLUTImpl & /*time_zone*/)
+    {
+        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Ignore)
+            return static_cast<UInt32>(dt64);
+        else
+        {
+            if (dt64 < 0 || dt64 >= MAX_DATETIME_TIMESTAMP)
+            {
+                if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Saturate)
+                    return dt64 < 0 ? 0 : std::numeric_limits<UInt32>::max();
+                else
+                    throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type DateTime", dt64);
+            }
+            else
+                return static_cast<UInt32>(dt64);
+        }
+    }
+};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeDate, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeDate, DataTypeDateTime, ToDateTimeImpl<date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeDate32, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeDate32, DataTypeDateTime, ToDateTimeImpl<date_time_overflow_behavior>, false> {};
+
+/// Implementation of toDate function.
+
+template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ToDateTransform32Or64
+{
+    static constexpr auto name = "toDate";
+
+    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
+    {
+        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+        {
+            if (from > MAX_DATETIME_TIMESTAMP) [[unlikely]]
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type Date", from);
+        }
+        /// if value is smaller (or equal) than maximum day value for Date, than treat it as day num,
+        /// otherwise treat it as unix timestamp. This is a bit weird, but we leave this behavior.
+        if (from <= DATE_LUT_MAX_DAY_NUM)
+            return from;
+        else
+            return time_zone.toDayNum(std::min(time_t(from), time_t(MAX_DATETIME_TIMESTAMP)));
+    }
+};
+
+/** Conversion of Date32 to Date.
+  */
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeDate32, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeDate32, DataTypeDate, ToDateImpl<date_time_overflow_behavior>, false> {};
+
+template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ToDateTransform32Or64Signed
+{
+    static constexpr auto name = "toDate";
+
+    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
+    {
+        // TODO: decide narrow or extended range based on FromType
+        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+        {
+            if (from < 0 || from > MAX_DATE_TIMESTAMP) [[unlikely]]
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type Date", from);
+        }
+        else
+        {
+            if (from < 0)
+                return 0;
+        }
+        return (from <= DATE_LUT_MAX_DAY_NUM)
+            ? static_cast<ToType>(from)
+            : time_zone.toDayNum(std::min(time_t(from), time_t(MAX_DATE_TIMESTAMP)));
+    }
+};
+
+template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ToDateTransform8Or16Signed
+{
+    static constexpr auto name = "toDate";
+
+    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
+    {
+        if (from < 0)
+        {
+            if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type Date", from);
+            else
+                return 0;
+        }
+        return from;
+    }
+};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeDateTime64, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+        : DateTimeTransformImpl<DataTypeDateTime64, DataTypeDate32, TransformDateTime64<ToDate32Impl>, false> {};
+
+/// Implementation of toDate32 function.
+
+template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ToDate32Transform32Or64
+{
+    static constexpr auto name = "toDate32";
+
+    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
+    {
+        if (from < DATE_LUT_MAX_EXTEND_DAY_NUM)
+            return static_cast<ToType>(from);
+        else
+        {
+            if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+            {
+                if (from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
+                    throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type Date32", from);
+            }
+            return time_zone.toDayNum(std::min(time_t(from), time_t(MAX_DATETIME64_TIMESTAMP)));
+        }
+    }
+};
+
+template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ToDate32Transform32Or64Signed
+{
+    static constexpr auto name = "toDate32";
+
+    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
+    {
+        static const Int32 daynum_min_offset = -static_cast<Int32>(time_zone.getDayNumOffsetEpoch());
+
+        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+        {
+            if (from < daynum_min_offset || from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type Date32", from);
+        }
+
+        if (from < daynum_min_offset)
+            return daynum_min_offset;
+
+        return (from < DATE_LUT_MAX_EXTEND_DAY_NUM)
+            ? static_cast<ToType>(from)
+            : time_zone.toDayNum(std::min(time_t(Int64(from)), time_t(MAX_DATETIME64_TIMESTAMP)));
+    }
+};
+
+template <typename FromType, typename ToType>
+struct ToDate32Transform8Or16Signed
+{
+    static constexpr auto name = "toDate32";
+
+    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
+    {
+        return from;
+    }
+};
+
+/** Special case of converting Int8, Int16, (U)Int32 or (U)Int64 (and also, for convenience,
+  * Float32, Float64) to Date. If the
+  * number is less than 65536, then it is treated as DayNum, and if it's greater or equals to 65536,
+  * then treated as unix timestamp. If the number exceeds UInt32, saturate to MAX_UINT32 then as DayNum.
+  * It's a bit illogical, as we actually have two functions in one.
+  * But allows to support frequent case,
+  *  when user write toDate(UInt32), expecting conversion of unix timestamp to Date.
+  *  (otherwise such usage would be frequent mistake).
+  */
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeUInt32, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeUInt32, DataTypeDate, ToDateTransform32Or64<UInt32, UInt16, default_date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeUInt64, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeUInt64, DataTypeDate, ToDateTransform32Or64<UInt64, UInt16, default_date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeInt8, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeInt8, DataTypeDate, ToDateTransform8Or16Signed<Int8, UInt16, default_date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeInt16, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeInt16, DataTypeDate, ToDateTransform8Or16Signed<Int16, UInt16, default_date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeInt32, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeInt32, DataTypeDate, ToDateTransform32Or64Signed<Int32, UInt16, default_date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeInt64, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeInt64, DataTypeDate, ToDateTransform32Or64Signed<Int64, UInt16, default_date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeFloat32, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeFloat32, DataTypeDate, ToDateTransform32Or64Signed<Float32, UInt16, default_date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeFloat64, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeFloat64, DataTypeDate, ToDateTransform32Or64Signed<Float64, UInt16, default_date_time_overflow_behavior>, false> {};
+
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeUInt32, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeUInt32, DataTypeDate32, ToDate32Transform32Or64<UInt32, Int32, default_date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeUInt64, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeUInt64, DataTypeDate32, ToDate32Transform32Or64<UInt64, Int32, default_date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeInt8, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeInt8, DataTypeDate32, ToDate32Transform8Or16Signed<Int8, Int32>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeInt16, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeInt16, DataTypeDate32, ToDate32Transform8Or16Signed<Int16, Int32>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeInt32, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeInt32, DataTypeDate32, ToDate32Transform32Or64Signed<Int32, Int32, default_date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeInt64, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeInt64, DataTypeDate32, ToDate32Transform32Or64Signed<Int64, Int32, default_date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeFloat32, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeFloat32, DataTypeDate32, ToDate32Transform32Or64Signed<Float32, Int32, default_date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeFloat64, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeFloat64, DataTypeDate32, ToDate32Transform32Or64Signed<Float64, Int32, default_date_time_overflow_behavior>, false> {};
+
+
+template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ToDateTimeTransform64
+{
+    static constexpr auto name = "toDateTime";
+
+    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
+    {
+        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+        {
+            if (from > MAX_DATETIME_TIMESTAMP) [[unlikely]]
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime", from);
+        }
+        return static_cast<ToType>(std::min(time_t(from), time_t(MAX_DATETIME_TIMESTAMP)));
+    }
+};
+
+template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ToDateTimeTransformSigned
+{
+    static constexpr auto name = "toDateTime";
+
+    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
+    {
+        if (from < 0)
+        {
+            if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime", from);
+            else
+                return 0;
+        }
+        return from;
+    }
+};
+
+template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ToDateTimeTransform64Signed
+{
+    static constexpr auto name = "toDateTime";
+
+    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
+    {
+        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+        {
+            if (from < 0 || from > MAX_DATETIME_TIMESTAMP) [[unlikely]]
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime", from);
+        }
+
+        if (from < 0)
+            return 0;
+        return static_cast<ToType>(std::min(time_t(from), time_t(MAX_DATETIME_TIMESTAMP)));
+    }
+};
+
+/// Special case of converting Int8, Int16, Int32 or (U)Int64 (and also, for convenience, Float32, Float64) to DateTime.
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeInt8, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeInt8, DataTypeDateTime, ToDateTimeTransformSigned<Int8, UInt32, default_date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeInt16, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeInt16, DataTypeDateTime, ToDateTimeTransformSigned<Int16, UInt32, default_date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeInt32, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeInt32, DataTypeDateTime, ToDateTimeTransformSigned<Int32, UInt32, default_date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeInt64, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeInt64, DataTypeDateTime, ToDateTimeTransform64Signed<Int64, UInt32, default_date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeUInt64, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeUInt64, DataTypeDateTime, ToDateTimeTransform64<UInt64, UInt32, default_date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeFloat32, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeFloat32, DataTypeDateTime, ToDateTimeTransform64Signed<Float32, UInt32, default_date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeFloat64, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeFloat64, DataTypeDateTime, ToDateTimeTransform64Signed<Float64, UInt32, default_date_time_overflow_behavior>, false> {};
+
+/** Conversion of numeric to DateTime64
+  */
+
+template <typename FromType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ToDateTime64TransformUnsigned
+{
+    static constexpr auto name = "toDateTime64";
+
+    const DateTime64::NativeType scale_multiplier = 1;
+
+    ToDateTime64TransformUnsigned(UInt32 scale = 0) /// NOLINT
+        : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
+    {}
+
+    NO_SANITIZE_UNDEFINED DateTime64::NativeType execute(FromType from, const DateLUTImpl &) const
+    {
+        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+        {
+            if (from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime64", from);
+            else
+                return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(from, 0, scale_multiplier);
+        }
+        else
+            return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(std::min<time_t>(from, MAX_DATETIME64_TIMESTAMP), 0, scale_multiplier);
+    }
+};
+template <typename FromType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ToDateTime64TransformSigned
+{
+    static constexpr auto name = "toDateTime64";
+
+    const DateTime64::NativeType scale_multiplier = 1;
+
+    ToDateTime64TransformSigned(UInt32 scale = 0) /// NOLINT
+        : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
+    {}
+
+    NO_SANITIZE_UNDEFINED DateTime64::NativeType execute(FromType from, const DateLUTImpl &) const
+    {
+        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+        {
+            if (from < MIN_DATETIME64_TIMESTAMP || from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime64", from);
+        }
+        from = static_cast<FromType>(std::max<time_t>(from, MIN_DATETIME64_TIMESTAMP));
+        from = static_cast<FromType>(std::min<time_t>(from, MAX_DATETIME64_TIMESTAMP));
+
+        return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(from, 0, scale_multiplier);
+    }
+};
+template <typename FromDataType, typename FromType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ToDateTime64TransformFloat
+{
+    static constexpr auto name = "toDateTime64";
+
+    const UInt32 scale = 1;
+
+    ToDateTime64TransformFloat(UInt32 scale_ = 0) /// NOLINT
+        : scale(scale_)
+    {}
+
+    NO_SANITIZE_UNDEFINED DateTime64::NativeType execute(FromType from, const DateLUTImpl &) const
+    {
+        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+        {
+            if (from < MIN_DATETIME64_TIMESTAMP || from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime64", from);
+        }
+
+        from = std::max(from, static_cast<FromType>(MIN_DATETIME64_TIMESTAMP));
+        from = std::min(from, static_cast<FromType>(MAX_DATETIME64_TIMESTAMP));
+        return convertToDecimal<FromDataType, DataTypeDateTime64>(from, scale);
+    }
+};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeInt8, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeInt8, DataTypeDateTime64, ToDateTime64TransformSigned<Int8, date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeInt16, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeInt16, DataTypeDateTime64, ToDateTime64TransformSigned<Int16, date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeInt32, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeInt32, DataTypeDateTime64, ToDateTime64TransformSigned<Int32, date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeInt64, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeInt64, DataTypeDateTime64, ToDateTime64TransformSigned<Int64, date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeUInt64, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeUInt64, DataTypeDateTime64, ToDateTime64TransformUnsigned<UInt64, date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeFloat32, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeFloat32, DataTypeDateTime64, ToDateTime64TransformFloat<DataTypeFloat32, Float32, date_time_overflow_behavior>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeFloat64, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeFloat64, DataTypeDateTime64, ToDateTime64TransformFloat<DataTypeFloat64, Float64, date_time_overflow_behavior>, false> {};
+
+
+/** Conversion of DateTime64 to Date or DateTime: discards fractional part.
+ */
+template <typename Transform>
+struct FromDateTime64Transform
+{
+    static constexpr auto name = Transform::name;
+
+    const DateTime64::NativeType scale_multiplier = 1;
+
+    FromDateTime64Transform(UInt32 scale) /// NOLINT
+        : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
+    {}
+
+    auto execute(DateTime64::NativeType dt, const DateLUTImpl & time_zone) const
+    {
+        const auto c = DecimalUtils::splitWithScaleMultiplier(DateTime64(dt), scale_multiplier);
+        return Transform::execute(static_cast<UInt32>(c.whole), time_zone);
+    }
+};
+
+/** Conversion of DateTime64 to Date or DateTime: discards fractional part.
+ */
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeDateTime64, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeDateTime64, DataTypeDate, TransformDateTime64<ToDateImpl<date_time_overflow_behavior>>, false> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeDateTime64, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeDateTime64, DataTypeDateTime, TransformDateTime64<ToDateTimeImpl<date_time_overflow_behavior>>, false> {};
+
+struct ToDateTime64Transform
+{
+    static constexpr auto name = "toDateTime64";
+
+    const DateTime64::NativeType scale_multiplier = 1;
+
+    ToDateTime64Transform(UInt32 scale = 0) /// NOLINT
+        : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
+    {}
+
+    DateTime64::NativeType execute(UInt16 d, const DateLUTImpl & time_zone) const
+    {
+        const auto dt = ToDateTimeImpl<>::execute(d, time_zone);
+        return execute(dt, time_zone);
+    }
+
+    DateTime64::NativeType execute(Int32 d, const DateLUTImpl & time_zone) const
+    {
+        Int64 dt = static_cast<Int64>(time_zone.fromDayNum(ExtendedDayNum(d)));
+        return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(dt, 0, scale_multiplier);
+    }
+
+    DateTime64::NativeType execute(UInt32 dt, const DateLUTImpl & /*time_zone*/) const
+    {
+        return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(dt, 0, scale_multiplier);
+    }
+};
+
+/** Conversion of Date or DateTime to DateTime64: add zero sub-second part.
+  */
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeDate, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeDate, DataTypeDateTime64, ToDateTime64Transform> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeDate32, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeDate32, DataTypeDateTime64, ToDateTime64Transform> {};
+
+template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ConvertImpl<DataTypeDateTime, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : DateTimeTransformImpl<DataTypeDateTime, DataTypeDateTime64, ToDateTime64Transform> {};
+
+
+/** Transformation of numbers, dates, datetimes to strings: through formatting.
+  */
+template <typename DataType>
+struct FormatImpl
+{
+    template <typename ReturnType = void>
+    static ReturnType execute(const typename DataType::FieldType x, WriteBuffer & wb, const DataType *, const DateLUTImpl *)
+    {
+        writeText(x, wb);
+        return ReturnType(true);
+    }
+};
+
+template <>
+struct FormatImpl<DataTypeDate>
+{
+    template <typename ReturnType = void>
+    static ReturnType execute(const DataTypeDate::FieldType x, WriteBuffer & wb, const DataTypeDate *, const DateLUTImpl * time_zone)
+    {
+        writeDateText(DayNum(x), wb, *time_zone);
+        return ReturnType(true);
+    }
+};
+
+template <>
+struct FormatImpl<DataTypeDate32>
+{
+    template <typename ReturnType = void>
+    static ReturnType execute(const DataTypeDate32::FieldType x, WriteBuffer & wb, const DataTypeDate32 *, const DateLUTImpl * time_zone)
+    {
+        writeDateText(ExtendedDayNum(x), wb, *time_zone);
+        return ReturnType(true);
+    }
+};
+
+template <>
+struct FormatImpl<DataTypeDateTime>
+{
+    template <typename ReturnType = void>
+    static ReturnType execute(const DataTypeDateTime::FieldType x, WriteBuffer & wb, const DataTypeDateTime *, const DateLUTImpl * time_zone)
+    {
+        writeDateTimeText(x, wb, *time_zone);
+        return ReturnType(true);
+    }
+};
+
+template <>
+struct FormatImpl<DataTypeDateTime64>
+{
+    template <typename ReturnType = void>
+    static ReturnType execute(const DataTypeDateTime64::FieldType x, WriteBuffer & wb, const DataTypeDateTime64 * type, const DateLUTImpl * time_zone)
+    {
+        writeDateTimeText(DateTime64(x), type->getScale(), wb, *time_zone);
+        return ReturnType(true);
+    }
+};
+
+
+template <typename FieldType>
+struct FormatImpl<DataTypeEnum<FieldType>>
+{
+    template <typename ReturnType = void>
+    static ReturnType execute(const FieldType x, WriteBuffer & wb, const DataTypeEnum<FieldType> * type, const DateLUTImpl *)
+    {
+        static constexpr bool throw_exception = std::is_same_v<ReturnType, void>;
+
+        if constexpr (throw_exception)
+        {
+            writeString(type->getNameForValue(x), wb);
+        }
+        else
+        {
+            StringRef res;
+            bool is_ok = type->getNameForValue(x, res);
+            if (is_ok)
+                writeString(res, wb);
+            return ReturnType(is_ok);
+        }
+    }
+};
+
+template <typename FieldType>
+struct FormatImpl<DataTypeDecimal<FieldType>>
+{
+    template <typename ReturnType = void>
+    static ReturnType execute(const FieldType x, WriteBuffer & wb, const DataTypeDecimal<FieldType> * type, const DateLUTImpl *)
+    {
+        writeText(x, type->getScale(), wb, false);
+        return ReturnType(true);
+    }
+};
+
+
+/// DataTypeEnum<T> to DataType<T> free conversion
+template <typename FieldType, typename Name>
+struct ConvertImpl<DataTypeEnum<FieldType>, DataTypeNumber<FieldType>, Name, ConvertDefaultBehaviorTag>
+{
+    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/)
+    {
+        return arguments[0].column;
+    }
+};
+
+static inline ColumnUInt8::MutablePtr copyNullMap(ColumnPtr col)
+{
+    ColumnUInt8::MutablePtr null_map = nullptr;
+    if (const auto * col_null = checkAndGetColumn<ColumnNullable>(col.get()))
+    {
+        null_map = ColumnUInt8::create();
+        null_map->insertRangeFrom(col_null->getNullMapColumn(), 0, col_null->size());
+    }
+    return null_map;
+}
+
+template <typename FromDataType, typename Name>
+requires (!std::is_same_v<FromDataType, DataTypeString>)
+struct ConvertImpl<FromDataType, DataTypeString, Name, ConvertDefaultBehaviorTag>
+{
+    using FromFieldType = typename FromDataType::FieldType;
+    using ColVecType = ColumnVectorOrDecimal<FromFieldType>;
+
+    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/)
+    {
+        if constexpr (IsDataTypeDateOrDateTime<FromDataType>)
+        {
+            auto datetime_arg = arguments[0];
+
+            const DateLUTImpl * time_zone = nullptr;
+            const ColumnConst * time_zone_column = nullptr;
+
+            if (arguments.size() == 1)
+            {
+                auto non_null_args = createBlockWithNestedColumns(arguments);
+                time_zone = &extractTimeZoneFromFunctionArguments(non_null_args, 1, 0);
+            }
+            else /// When we have a column for timezone
+            {
+                datetime_arg.column = datetime_arg.column->convertToFullColumnIfConst();
+
+                if constexpr (std::is_same_v<FromDataType, DataTypeDate> || std::is_same_v<FromDataType, DataTypeDate32>)
+                    time_zone = &DateLUT::instance();
+                /// For argument of Date or DateTime type, second argument with time zone could be specified.
+                if constexpr (std::is_same_v<FromDataType, DataTypeDateTime> || std::is_same_v<FromDataType, DataTypeDateTime64>)
+                {
+                    if ((time_zone_column = checkAndGetColumnConst<ColumnString>(arguments[1].column.get())))
+                    {
+                        auto non_null_args = createBlockWithNestedColumns(arguments);
+                        time_zone = &extractTimeZoneFromFunctionArguments(non_null_args, 1, 0);
+                    }
+                }
+            }
+            const auto & col_with_type_and_name = columnGetNested(datetime_arg);
+
+            if (const auto col_from = checkAndGetColumn<ColVecType>(col_with_type_and_name.column.get()))
+            {
+                auto col_to = ColumnString::create();
+
+                const typename ColVecType::Container & vec_from = col_from->getData();
+                ColumnString::Chars & data_to = col_to->getChars();
+                ColumnString::Offsets & offsets_to = col_to->getOffsets();
+                size_t size = vec_from.size();
+
+                if constexpr (std::is_same_v<FromDataType, DataTypeDate>)
+                    data_to.resize(size * (strlen("YYYY-MM-DD") + 1));
+                else if constexpr (std::is_same_v<FromDataType, DataTypeDate32>)
+                    data_to.resize(size * (strlen("YYYY-MM-DD") + 1));
+                else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime>)
+                    data_to.resize(size * (strlen("YYYY-MM-DD hh:mm:ss") + 1));
+                else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime64>)
+                    data_to.resize(size * (strlen("YYYY-MM-DD hh:mm:ss.") + col_from->getScale() + 1));
+                else
+                    data_to.resize(size * 3);   /// Arbitrary
+
+                offsets_to.resize(size);
+
+                WriteBufferFromVector<ColumnString::Chars> write_buffer(data_to);
+                const auto & type = static_cast<const FromDataType &>(*col_with_type_and_name.type);
+
+                ColumnUInt8::MutablePtr null_map = copyNullMap(datetime_arg.column);
+
+                if (!null_map && arguments.size() > 1)
+                    null_map = copyNullMap(arguments[1].column->convertToFullColumnIfConst());
+
+                if (null_map)
+                {
+                    for (size_t i = 0; i < size; ++i)
+                    {
+                        if (!time_zone_column && arguments.size() > 1)
+                        {
+                            if (!arguments[1].column.get()->getDataAt(i).toString().empty())
+                                time_zone = &DateLUT::instance(arguments[1].column.get()->getDataAt(i).toString());
+                            else
+                                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Provided time zone must be non-empty");
+                        }
+                        bool is_ok = FormatImpl<FromDataType>::template execute<bool>(vec_from[i], write_buffer, &type, time_zone);
+                        null_map->getData()[i] |= !is_ok;
+                        writeChar(0, write_buffer);
+                        offsets_to[i] = write_buffer.count();
+                    }
+                }
+                else
+                {
+                    for (size_t i = 0; i < size; ++i)
+                    {
+                        if (!time_zone_column && arguments.size() > 1)
+                        {
+                            if (!arguments[1].column.get()->getDataAt(i).toString().empty())
+                                time_zone = &DateLUT::instance(arguments[1].column.get()->getDataAt(i).toString());
+                            else
+                                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Provided time zone must be non-empty");
+                        }
+                        FormatImpl<FromDataType>::template execute<void>(vec_from[i], write_buffer, &type, time_zone);
+                        writeChar(0, write_buffer);
+                        offsets_to[i] = write_buffer.count();
+                    }
+                }
+
+                write_buffer.finalize();
+
+                if (null_map)
+                    return ColumnNullable::create(std::move(col_to), std::move(null_map));
+                return col_to;
+            }
+            else
+                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
+                        arguments[0].column->getName(), Name::name);
+        }
+        else
+        {
+            ColumnUInt8::MutablePtr null_map = copyNullMap(arguments[0].column);
+
+            const auto & col_with_type_and_name = columnGetNested(arguments[0]);
+            const auto & type = static_cast<const FromDataType &>(*col_with_type_and_name.type);
+
+            if (const auto col_from = checkAndGetColumn<ColVecType>(col_with_type_and_name.column.get()))
+            {
+                auto col_to = ColumnString::create();
+
+                const typename ColVecType::Container & vec_from = col_from->getData();
+                ColumnString::Chars & data_to = col_to->getChars();
+                ColumnString::Offsets & offsets_to = col_to->getOffsets();
+                size_t size = vec_from.size();
+
+                data_to.resize(size * 3);
+                offsets_to.resize(size);
+
+                WriteBufferFromVector<ColumnString::Chars> write_buffer(data_to);
+
+                if (null_map)
+                {
+                    for (size_t i = 0; i < size; ++i)
+                    {
+                        bool is_ok = FormatImpl<FromDataType>::template execute<bool>(vec_from[i], write_buffer, &type, nullptr);
+                        /// We don't use timezones in this branch
+                        null_map->getData()[i] |= !is_ok;
+                        writeChar(0, write_buffer);
+                        offsets_to[i] = write_buffer.count();
+                    }
+                }
+                else
+                {
+                    for (size_t i = 0; i < size; ++i)
+                    {
+                        FormatImpl<FromDataType>::template execute<void>(vec_from[i], write_buffer, &type, nullptr);
+                        writeChar(0, write_buffer);
+                        offsets_to[i] = write_buffer.count();
+                    }
+                }
+
+                write_buffer.finalize();
+
+                if (null_map)
+                    return ColumnNullable::create(std::move(col_to), std::move(null_map));
+                return col_to;
+            }
+            else
+                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
+                        arguments[0].column->getName(), Name::name);
+        }
+    }
+};
+
+
+/// Generic conversion of any type to String or FixedString via serialization to text.
+template <typename StringColumnType>
+struct ConvertImplGenericToString
+{
+    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t /*input_rows_count*/)
+    {
+        static_assert(std::is_same_v<StringColumnType, ColumnString> || std::is_same_v<StringColumnType, ColumnFixedString>,
+                "Can be used only to serialize to ColumnString or ColumnFixedString");
+
+        ColumnUInt8::MutablePtr null_map = copyNullMap(arguments[0].column);
+
+        const auto & col_with_type_and_name = columnGetNested(arguments[0]);
+        const IDataType & type = *col_with_type_and_name.type;
+        const IColumn & col_from = *col_with_type_and_name.column;
+
+        size_t size = col_from.size();
+        auto col_to = removeNullable(result_type)->createColumn();
+
+        {
+            ColumnStringHelpers::WriteHelper write_helper(
+                    assert_cast<StringColumnType &>(*col_to),
+                    size);
+
+            auto & write_buffer = write_helper.getWriteBuffer();
+
+            FormatSettings format_settings;
+            auto serialization = type.getDefaultSerialization();
+            for (size_t row = 0; row < size; ++row)
+            {
+                serialization->serializeText(col_from, row, write_buffer, format_settings);
+                write_helper.rowWritten();
+            }
+
+            write_helper.finalize();
+        }
+
+        if (result_type->isNullable() && null_map)
+            return ColumnNullable::create(std::move(col_to), std::move(null_map));
+        return col_to;
+    }
+};
+
+/** Conversion of time_t to UInt16, Int32, UInt32
+  */
+template <typename DataType>
+void convertFromTime(typename DataType::FieldType & x, time_t & time)
+{
+    x = time;
+}
+
+template <>
+inline void convertFromTime<DataTypeDate>(DataTypeDate::FieldType & x, time_t & time)
+{
+    if (unlikely(time < 0))
+        x = 0;
+    else if (unlikely(time > 0xFFFF))
+        x = 0xFFFF;
+    else
+        x = time;
+}
+
+template <>
+inline void convertFromTime<DataTypeDate32>(DataTypeDate32::FieldType & x, time_t & time)
+{
+    x = static_cast<UInt32>(time);
+}
+
+template <>
+inline void convertFromTime<DataTypeDateTime>(DataTypeDateTime::FieldType & x, time_t & time)
+{
+    if (unlikely(time < 0))
+        x = 0;
+    else if (unlikely(time > MAX_DATETIME_TIMESTAMP))
+        x = MAX_DATETIME_TIMESTAMP;
+    else
+        x = static_cast<UInt32>(time);
+}
+
+/** Conversion of strings to numbers, dates, datetimes: through parsing.
+  */
+template <typename DataType>
+void parseImpl(typename DataType::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool precise_float_parsing)
+{
+    if constexpr (std::is_floating_point_v<typename DataType::FieldType>)
+    {
+        if (precise_float_parsing)
+            readFloatTextPrecise(x, rb);
+        else
+            readFloatTextFast(x, rb);
+    }
+    else
+        readText(x, rb);
+}
+
+template <>
+inline void parseImpl<DataTypeDate>(DataTypeDate::FieldType & x, ReadBuffer & rb, const DateLUTImpl * time_zone, bool)
+{
+    DayNum tmp(0);
+    readDateText(tmp, rb, *time_zone);
+    x = tmp;
+}
+
+template <>
+inline void parseImpl<DataTypeDate32>(DataTypeDate32::FieldType & x, ReadBuffer & rb, const DateLUTImpl * time_zone, bool)
+{
+    ExtendedDayNum tmp(0);
+    readDateText(tmp, rb, *time_zone);
+    x = tmp;
+}
+
+
+// NOTE: no need of extra overload of DateTime64, since readDateTimeText64 has different signature and that case is explicitly handled in the calling code.
+template <>
+inline void parseImpl<DataTypeDateTime>(DataTypeDateTime::FieldType & x, ReadBuffer & rb, const DateLUTImpl * time_zone, bool)
+{
+    time_t time = 0;
+    readDateTimeText(time, rb, *time_zone);
+    convertFromTime<DataTypeDateTime>(x, time);
+}
+
+template <>
+inline void parseImpl<DataTypeUUID>(DataTypeUUID::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool)
+{
+    UUID tmp;
+    readUUIDText(tmp, rb);
+    x = tmp.toUnderType();
+}
+
+template <>
+inline void parseImpl<DataTypeIPv4>(DataTypeIPv4::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool)
+{
+    IPv4 tmp;
+    readIPv4Text(tmp, rb);
+    x = tmp.toUnderType();
+}
+
+template <>
+inline void parseImpl<DataTypeIPv6>(DataTypeIPv6::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool)
+{
+    IPv6 tmp;
+    readIPv6Text(tmp, rb);
+    x = tmp;
+}
+
+template <typename DataType>
+bool tryParseImpl(typename DataType::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool precise_float_parsing)
+{
+    if constexpr (std::is_floating_point_v<typename DataType::FieldType>)
+    {
+        if (precise_float_parsing)
+            return tryReadFloatTextPrecise(x, rb);
+        else
+            return tryReadFloatTextFast(x, rb);
+    }
+    else /*if constexpr (is_integer_v<typename DataType::FieldType>)*/
+        return tryReadIntText(x, rb);
+}
+
+template <>
+inline bool tryParseImpl<DataTypeDate>(DataTypeDate::FieldType & x, ReadBuffer & rb, const DateLUTImpl * time_zone, bool)
+{
+    DayNum tmp(0);
+    if (!tryReadDateText(tmp, rb, *time_zone))
+        return false;
+    x = tmp;
+    return true;
+}
+
+template <>
+inline bool tryParseImpl<DataTypeDate32>(DataTypeDate32::FieldType & x, ReadBuffer & rb, const DateLUTImpl * time_zone, bool)
+{
+    ExtendedDayNum tmp(0);
+    if (!tryReadDateText(tmp, rb, *time_zone))
+        return false;
+    x = tmp;
+    return true;
+}
+
+template <>
+inline bool tryParseImpl<DataTypeDateTime>(DataTypeDateTime::FieldType & x, ReadBuffer & rb, const DateLUTImpl * time_zone, bool)
+{
+    time_t time = 0;
+    if (!tryReadDateTimeText(time, rb, *time_zone))
+        return false;
+    convertFromTime<DataTypeDateTime>(x, time);
+    return true;
+}
+
+template <>
+inline bool tryParseImpl<DataTypeUUID>(DataTypeUUID::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool)
+{
+    UUID tmp;
+    if (!tryReadUUIDText(tmp, rb))
+        return false;
+
+    x = tmp.toUnderType();
+    return true;
+}
+
+template <>
+inline bool tryParseImpl<DataTypeIPv4>(DataTypeIPv4::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool)
+{
+    IPv4 tmp;
+    if (!tryReadIPv4Text(tmp, rb))
+        return false;
+
+    x = tmp.toUnderType();
+    return true;
+}
+
+template <>
+inline bool tryParseImpl<DataTypeIPv6>(DataTypeIPv6::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool)
+{
+    IPv6 tmp;
+    if (!tryReadIPv6Text(tmp, rb))
+        return false;
+
+    x = tmp;
+    return true;
+}
+
+
+/** Throw exception with verbose message when string value is not parsed completely.
+  */
+[[noreturn]] inline void throwExceptionForIncompletelyParsedValue(ReadBuffer & read_buffer, const IDataType & result_type)
+{
+    WriteBufferFromOwnString message_buf;
+    message_buf << "Cannot parse string " << quote << String(read_buffer.buffer().begin(), read_buffer.buffer().size())
+                << " as " << result_type.getName()
+                << ": syntax error";
+
+    if (read_buffer.offset())
+        message_buf << " at position " << read_buffer.offset()
+                    << " (parsed just " << quote << String(read_buffer.buffer().begin(), read_buffer.offset()) << ")";
+    else
+        message_buf << " at begin of string";
+
+    // Currently there are no functions toIPv{4,6}Or{Null,Zero}
+    if (isNativeNumber(result_type) && !(result_type.getName() == "IPv4" || result_type.getName() == "IPv6"))
+        message_buf << ". Note: there are to" << result_type.getName() << "OrZero and to" << result_type.getName() << "OrNull functions, which returns zero/NULL instead of throwing exception.";
+
+    throw Exception(PreformattedMessage{message_buf.str(), "Cannot parse string {} as {}: syntax error {}"}, ErrorCodes::CANNOT_PARSE_TEXT);
+}
+
+
+enum class ConvertFromStringExceptionMode
+{
+    Throw,  /// Throw exception if value cannot be parsed.
+    Zero,   /// Fill with zero or default if value cannot be parsed.
+    Null    /// Return ColumnNullable with NULLs when value cannot be parsed.
+};
+
+enum class ConvertFromStringParsingMode
+{
+    Normal,
+    BestEffort,  /// Only applicable for DateTime. Will use sophisticated method, that is slower.
+    BestEffortUS
+};
+
+template <typename FromDataType, typename ToDataType, typename Name,
+    ConvertFromStringExceptionMode exception_mode, ConvertFromStringParsingMode parsing_mode>
+struct ConvertThroughParsing
+{
+    static_assert(std::is_same_v<FromDataType, DataTypeString> || std::is_same_v<FromDataType, DataTypeFixedString>,
+        "ConvertThroughParsing is only applicable for String or FixedString data types");
+
+    static constexpr bool to_datetime64 = std::is_same_v<ToDataType, DataTypeDateTime64>;
+
+    static bool isAllRead(ReadBuffer & in)
+    {
+        /// In case of FixedString, skip zero bytes at end.
+        if constexpr (std::is_same_v<FromDataType, DataTypeFixedString>)
+            while (!in.eof() && *in.position() == 0)
+                ++in.position();
+
+        if (in.eof())
+            return true;
+
+        /// Special case, that allows to parse string with DateTime or DateTime64 as Date or Date32.
+        if constexpr (std::is_same_v<ToDataType, DataTypeDate> || std::is_same_v<ToDataType, DataTypeDate32>)
+        {
+            if (!in.eof() && (*in.position() == ' ' || *in.position() == 'T'))
+            {
+                if (in.buffer().size() == strlen("YYYY-MM-DD hh:mm:ss"))
+                    return true;
+
+                if (in.buffer().size() >= strlen("YYYY-MM-DD hh:mm:ss.x")
+                    && in.buffer().begin()[19] == '.')
+                {
+                    in.position() = in.buffer().begin() + 20;
+
+                    while (!in.eof() && isNumericASCII(*in.position()))
+                        ++in.position();
+
+                    if (in.eof())
+                        return true;
+                }
+            }
+        }
+
+        return false;
+    }
+
+    template <typename Additions = void *>
+    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr & res_type, size_t input_rows_count,
+                        Additions additions [[maybe_unused]] = Additions())
+    {
+        using ColVecTo = typename ToDataType::ColumnType;
+
+        const DateLUTImpl * local_time_zone [[maybe_unused]] = nullptr;
+        const DateLUTImpl * utc_time_zone [[maybe_unused]] = nullptr;
+
+        /// For conversion to Date or DateTime type, second argument with time zone could be specified.
+        if constexpr (std::is_same_v<ToDataType, DataTypeDateTime> || to_datetime64)
+        {
+            const auto result_type = removeNullable(res_type);
+            // Time zone is already figured out during result type resolution, no need to do it here.
+            if (const auto dt_col = checkAndGetDataType<ToDataType>(result_type.get()))
+                local_time_zone = &dt_col->getTimeZone();
+            else
+                local_time_zone = &extractTimeZoneFromFunctionArguments(arguments, 1, 0);
+
+            if constexpr (parsing_mode == ConvertFromStringParsingMode::BestEffort || parsing_mode == ConvertFromStringParsingMode::BestEffortUS)
+                utc_time_zone = &DateLUT::instance("UTC");
+        }
+        else if constexpr (std::is_same_v<ToDataType, DataTypeDate> || std::is_same_v<ToDataType, DataTypeDate32>)
+        {
+            // Timezone is more or less dummy when parsing Date/Date32 from string.
+            local_time_zone = &DateLUT::instance();
+            utc_time_zone = &DateLUT::instance("UTC");
+        }
+
+        const IColumn * col_from = arguments[0].column.get();
+        const ColumnString * col_from_string = checkAndGetColumn<ColumnString>(col_from);
+        const ColumnFixedString * col_from_fixed_string = checkAndGetColumn<ColumnFixedString>(col_from);
+
+        if (std::is_same_v<FromDataType, DataTypeString> && !col_from_string)
+            throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
+                col_from->getName(), Name::name);
+
+        if (std::is_same_v<FromDataType, DataTypeFixedString> && !col_from_fixed_string)
+            throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
+                col_from->getName(), Name::name);
+
+        size_t size = input_rows_count;
+        typename ColVecTo::MutablePtr col_to = nullptr;
+
+        if constexpr (IsDataTypeDecimal<ToDataType>)
+        {
+            UInt32 scale = additions;
+            if constexpr (to_datetime64)
+            {
+                ToDataType check_bounds_in_ctor(scale, local_time_zone ? local_time_zone->getTimeZone() : String{});
+            }
+            else
+            {
+                ToDataType check_bounds_in_ctor(ToDataType::maxPrecision(), scale);
+            }
+            col_to = ColVecTo::create(size, scale);
+        }
+        else
+            col_to = ColVecTo::create(size);
+
+        typename ColVecTo::Container & vec_to = col_to->getData();
+
+        ColumnUInt8::MutablePtr col_null_map_to;
+        ColumnUInt8::Container * vec_null_map_to [[maybe_unused]] = nullptr;
+        if constexpr (exception_mode == ConvertFromStringExceptionMode::Null)
+        {
+            col_null_map_to = ColumnUInt8::create(size);
+            vec_null_map_to = &col_null_map_to->getData();
+        }
+
+        const ColumnString::Chars * chars = nullptr;
+        const IColumn::Offsets * offsets = nullptr;
+        size_t fixed_string_size = 0;
+
+        if constexpr (std::is_same_v<FromDataType, DataTypeString>)
+        {
+            chars = &col_from_string->getChars();
+            offsets = &col_from_string->getOffsets();
+        }
+        else
+        {
+            chars = &col_from_fixed_string->getChars();
+            fixed_string_size = col_from_fixed_string->getN();
+        }
+
+        size_t current_offset = 0;
+
+        bool precise_float_parsing = false;
+
+        if (DB::CurrentThread::isInitialized())
+        {
+            const DB::ContextPtr query_context = DB::CurrentThread::get().getQueryContext();
+
+            if (query_context)
+                precise_float_parsing = query_context->getSettingsRef().precise_float_parsing;
+        }
+
+        for (size_t i = 0; i < size; ++i)
+        {
+            size_t next_offset = std::is_same_v<FromDataType, DataTypeString> ? (*offsets)[i] : (current_offset + fixed_string_size);
+            size_t string_size = std::is_same_v<FromDataType, DataTypeString> ? next_offset - current_offset - 1 : fixed_string_size;
+
+            ReadBufferFromMemory read_buffer(&(*chars)[current_offset], string_size);
+
+            if constexpr (exception_mode == ConvertFromStringExceptionMode::Throw)
+            {
+                if constexpr (parsing_mode == ConvertFromStringParsingMode::BestEffort)
+                {
+                    if constexpr (to_datetime64)
+                    {
+                        DateTime64 res = 0;
+                        parseDateTime64BestEffort(res, col_to->getScale(), read_buffer, *local_time_zone, *utc_time_zone);
+                        vec_to[i] = res;
+                    }
+                    else
+                    {
+                        time_t res;
+                        parseDateTimeBestEffort(res, read_buffer, *local_time_zone, *utc_time_zone);
+                        convertFromTime<ToDataType>(vec_to[i], res);
+                    }
+                }
+                else if constexpr (parsing_mode == ConvertFromStringParsingMode::BestEffortUS)
+                {
+                    if constexpr (to_datetime64)
+                    {
+                        DateTime64 res = 0;
+                        parseDateTime64BestEffortUS(res, col_to->getScale(), read_buffer, *local_time_zone, *utc_time_zone);
+                        vec_to[i] = res;
+                    }
+                    else
+                    {
+                        time_t res;
+                        parseDateTimeBestEffortUS(res, read_buffer, *local_time_zone, *utc_time_zone);
+                        convertFromTime<ToDataType>(vec_to[i], res);
+                    }
+                }
+                else
+                {
+                    if constexpr (to_datetime64)
+                    {
+                        DateTime64 value = 0;
+                        readDateTime64Text(value, col_to->getScale(), read_buffer, *local_time_zone);
+                        vec_to[i] = value;
+                    }
+                    else if constexpr (IsDataTypeDecimal<ToDataType>)
+                    {
+                        SerializationDecimal<typename ToDataType::FieldType>::readText(
+                            vec_to[i], read_buffer, ToDataType::maxPrecision(), col_to->getScale());
+                    }
+                    else
+                    {
+                        /// we want to utilize constexpr condition here, which is not mixable with value comparison
+                        do
+                        {
+                            if constexpr (std::is_same_v<FromDataType, DataTypeFixedString> && std::is_same_v<ToDataType, DataTypeIPv6>)
+                            {
+                                if (fixed_string_size == IPV6_BINARY_LENGTH)
+                                {
+                                    readBinary(vec_to[i], read_buffer);
+                                    break;
+                                }
+                            }
+                            parseImpl<ToDataType>(vec_to[i], read_buffer, local_time_zone, precise_float_parsing);
+                        } while (false);
+                    }
+                }
+
+                if (!isAllRead(read_buffer))
+                    throwExceptionForIncompletelyParsedValue(read_buffer, *res_type);
+            }
+            else
+            {
+                bool parsed;
+
+                if constexpr (parsing_mode == ConvertFromStringParsingMode::BestEffort)
+                {
+                    if constexpr (to_datetime64)
+                    {
+                        DateTime64 res = 0;
+                        parsed = tryParseDateTime64BestEffort(res, col_to->getScale(), read_buffer, *local_time_zone, *utc_time_zone);
+                        vec_to[i] = res;
+                    }
+                    else
+                    {
+                        time_t res;
+                        parsed = tryParseDateTimeBestEffort(res, read_buffer, *local_time_zone, *utc_time_zone);
+                        convertFromTime<ToDataType>(vec_to[i],res);
+                    }
+                }
+                else if constexpr (parsing_mode == ConvertFromStringParsingMode::BestEffortUS)
+                {
+                    if constexpr (to_datetime64)
+                    {
+                        DateTime64 res = 0;
+                        parsed = tryParseDateTime64BestEffortUS(res, col_to->getScale(), read_buffer, *local_time_zone, *utc_time_zone);
+                        vec_to[i] = res;
+                    }
+                    else
+                    {
+                        time_t res;
+                        parsed = tryParseDateTimeBestEffortUS(res, read_buffer, *local_time_zone, *utc_time_zone);
+                        convertFromTime<ToDataType>(vec_to[i],res);
+                    }
+                }
+                else
+                {
+                    if constexpr (to_datetime64)
+                    {
+                        DateTime64 value = 0;
+                        parsed = tryReadDateTime64Text(value, col_to->getScale(), read_buffer, *local_time_zone);
+                        vec_to[i] = value;
+                    }
+                    else if constexpr (IsDataTypeDecimal<ToDataType>)
+                    {
+                        parsed = SerializationDecimal<typename ToDataType::FieldType>::tryReadText(
+                            vec_to[i], read_buffer, ToDataType::maxPrecision(), col_to->getScale());
+                    }
+                    else
+                    {
+                        /// we want to utilize constexpr condition here, which is not mixable with value comparison
+                        do
+                        {
+                            if constexpr (std::is_same_v<FromDataType, DataTypeFixedString> && std::is_same_v<ToDataType, DataTypeIPv6>)
+                            {
+                                if (fixed_string_size == IPV6_BINARY_LENGTH)
+                                {
+                                    readBinary(vec_to[i], read_buffer);
+                                    parsed = true;
+                                    break;
+                                }
+                            }
+
+                            parsed = tryParseImpl<ToDataType>(vec_to[i], read_buffer, local_time_zone, precise_float_parsing);
+                        } while (false);
+                    }
+                }
+
+                if (!isAllRead(read_buffer))
+                    parsed = false;
+
+                if (!parsed)
+                {
+                    if constexpr (std::is_same_v<ToDataType, DataTypeDate32>)
+                    {
+                        vec_to[i] = -static_cast<Int32>(DateLUT::instance().getDayNumOffsetEpoch());
+                    }
+                    else
+                    {
+                        vec_to[i] = static_cast<typename ToDataType::FieldType>(0);
+                    }
+                }
+
+                if constexpr (exception_mode == ConvertFromStringExceptionMode::Null)
+                    (*vec_null_map_to)[i] = !parsed;
+            }
+
+            current_offset = next_offset;
+        }
+
+        if constexpr (exception_mode == ConvertFromStringExceptionMode::Null)
+            return ColumnNullable::create(std::move(col_to), std::move(col_null_map_to));
+        else
+            return col_to;
+    }
+};
+
+
+template <typename ToDataType, typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+requires (!std::is_same_v<ToDataType, DataTypeString>)
+struct ConvertImpl<DataTypeString, ToDataType, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : ConvertThroughParsing<DataTypeString, ToDataType, Name, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::Normal> {};
+
+template <typename ToDataType, typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+requires (!std::is_same_v<ToDataType, DataTypeFixedString>)
+struct ConvertImpl<DataTypeFixedString, ToDataType, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
+    : ConvertThroughParsing<DataTypeFixedString, ToDataType, Name, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::Normal> {};
+
+template <typename ToDataType, typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+requires (!std::is_same_v<ToDataType, DataTypeString>)
+struct ConvertImpl<DataTypeString, ToDataType, Name, ConvertReturnNullOnErrorTag, date_time_overflow_behavior>
+    : ConvertThroughParsing<DataTypeString, ToDataType, Name, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::Normal> {};
+
+template <typename ToDataType, typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+requires (!std::is_same_v<ToDataType, DataTypeFixedString>)
+struct ConvertImpl<DataTypeFixedString, ToDataType, Name, ConvertReturnNullOnErrorTag, date_time_overflow_behavior>
+    : ConvertThroughParsing<DataTypeFixedString, ToDataType, Name, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::Normal> {};
+
+template <typename FromDataType, typename ToDataType, typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+requires (is_any_of<FromDataType, DataTypeString, DataTypeFixedString> && is_any_of<ToDataType, DataTypeIPv4, DataTypeIPv6>)
+struct ConvertImpl<FromDataType, ToDataType, Name, ConvertReturnZeroOnErrorTag, date_time_overflow_behavior>
+    : ConvertThroughParsing<FromDataType, ToDataType, Name, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::Normal> {};
+
+/// Generic conversion of any type from String. Used for complex types: Array and Tuple or types with custom serialization.
+struct ConvertImplGenericFromString
+{
+    static ColumnPtr execute(ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable * column_nullable, size_t input_rows_count)
+    {
+        const IColumn & column_from = *arguments[0].column;
+        const IDataType & data_type_to = *result_type;
+        auto res = data_type_to.createColumn();
+        auto serialization = data_type_to.getDefaultSerialization();
+        const auto * null_map = column_nullable ? &column_nullable->getNullMapData() : nullptr;
+
+        executeImpl(column_from, *res, *serialization, input_rows_count, null_map, result_type.get());
+        return res;
+    }
+
+    static void executeImpl(
+        const IColumn & column_from,
+        IColumn & column_to,
+        const ISerialization & serialization_from,
+        size_t input_rows_count,
+        const PaddedPODArray<UInt8> * null_map,
+        const IDataType * result_type)
+    {
+        column_to.reserve(input_rows_count);
+
+        FormatSettings format_settings;
+        for (size_t i = 0; i < input_rows_count; ++i)
+        {
+            if (null_map && (*null_map)[i])
+            {
+                column_to.insertDefault();
+                continue;
+            }
+
+            const auto & val = column_from.getDataAt(i);
+            ReadBufferFromMemory read_buffer(val.data, val.size);
+            try
+            {
+                serialization_from.deserializeWholeText(column_to, read_buffer, format_settings);
+            }
+            catch (const Exception & e)
+            {
+                auto * nullable_column = typeid_cast<ColumnNullable *>(&column_to);
+                if (e.code() == ErrorCodes::CANNOT_PARSE_BOOL && nullable_column)
+                {
+                    auto & col_nullmap = nullable_column->getNullMapData();
+                    if (col_nullmap.size() != nullable_column->size())
+                        col_nullmap.resize_fill(nullable_column->size());
+                    if (nullable_column->size() == (i + 1))
+                        nullable_column->popBack(1);
+                    nullable_column->insertDefault();
+                    continue;
+                }
+                throw;
+            }
+
+            if (!read_buffer.eof())
+            {
+                if (result_type)
+                    throwExceptionForIncompletelyParsedValue(read_buffer, *result_type);
+                else
+                    throw Exception(ErrorCodes::CANNOT_PARSE_TEXT,
+                        "Cannot parse string to column {}. Expected eof", column_to.getName());
+            }
+        }
+    }
+};
+
+
+template <>
+struct ConvertImpl<DataTypeString, DataTypeUInt32, NameToUnixTimestamp, ConvertDefaultBehaviorTag>
+    : ConvertImpl<DataTypeString, DataTypeDateTime, NameToUnixTimestamp, ConvertDefaultBehaviorTag> {};
+
+template <>
+struct ConvertImpl<DataTypeString, DataTypeUInt32, NameToUnixTimestamp, ConvertReturnNullOnErrorTag>
+    : ConvertImpl<DataTypeString, DataTypeDateTime, NameToUnixTimestamp, ConvertReturnNullOnErrorTag> {};
+
+/** If types are identical, just take reference to column.
+  */
+template <typename T, typename Name>
+requires (!T::is_parametric)
+struct ConvertImpl<T, T, Name, ConvertDefaultBehaviorTag>
+{
+    template <typename Additions = void *>
+    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/,
+        Additions additions [[maybe_unused]] = Additions())
+    {
+        return arguments[0].column;
+    }
+};
+
+template <typename Name>
+struct ConvertImpl<DataTypeUInt8, DataTypeUInt8, Name, ConvertDefaultBehaviorTag>
+{
+    template <typename Additions = void *>
+    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/,
+                             Additions additions [[maybe_unused]] = Additions())
+    {
+
+        return arguments[0].column;
+    }
+};
+
+
+/** Conversion from FixedString to String.
+  * Cutting sequences of zero bytes from end of strings.
+  */
+template <typename Name>
+struct ConvertImpl<DataTypeFixedString, DataTypeString, Name, ConvertDefaultBehaviorTag>
+{
+    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr & return_type, size_t /*input_rows_count*/)
+    {
+        ColumnUInt8::MutablePtr null_map = copyNullMap(arguments[0].column);
+        const auto & nested =  columnGetNested(arguments[0]);
+        if (const ColumnFixedString * col_from = checkAndGetColumn<ColumnFixedString>(nested.column.get()))
+        {
+            auto col_to = ColumnString::create();
+
+            const ColumnFixedString::Chars & data_from = col_from->getChars();
+            ColumnString::Chars & data_to = col_to->getChars();
+            ColumnString::Offsets & offsets_to = col_to->getOffsets();
+            size_t size = col_from->size();
+            size_t n = col_from->getN();
+            data_to.resize(size * (n + 1)); /// + 1 - zero terminator
+            offsets_to.resize(size);
+
+            size_t offset_from = 0;
+            size_t offset_to = 0;
+            for (size_t i = 0; i < size; ++i)
+            {
+                if (!null_map || !null_map->getData()[i])
+                {
+                    size_t bytes_to_copy = n;
+                    while (bytes_to_copy > 0 && data_from[offset_from + bytes_to_copy - 1] == 0)
+                        --bytes_to_copy;
+
+                    memcpy(&data_to[offset_to], &data_from[offset_from], bytes_to_copy);
+                    offset_to += bytes_to_copy;
+                }
+                data_to[offset_to] = 0;
+                ++offset_to;
+                offsets_to[i] = offset_to;
+                offset_from += n;
+            }
+
+            data_to.resize(offset_to);
+            if (return_type->isNullable() && null_map)
+                return ColumnNullable::create(std::move(col_to), std::move(null_map));
+            return col_to;
+        }
+        else
+            throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
+                    arguments[0].column->getName(), Name::name);
+    }
+};
+
+
+/// Declared early because used below.
+struct NameToDate { static constexpr auto name = "toDate"; };
+struct NameToDate32 { static constexpr auto name = "toDate32"; };
+struct NameToDateTime { static constexpr auto name = "toDateTime"; };
+struct NameToDateTime32 { static constexpr auto name = "toDateTime32"; };
+struct NameToDateTime64 { static constexpr auto name = "toDateTime64"; };
+struct NameToString { static constexpr auto name = "toString"; };
+struct NameToDecimal32 { static constexpr auto name = "toDecimal32"; };
+struct NameToDecimal64 { static constexpr auto name = "toDecimal64"; };
+struct NameToDecimal128 { static constexpr auto name = "toDecimal128"; };
+struct NameToDecimal256 { static constexpr auto name = "toDecimal256"; };
+
+
+#define DEFINE_NAME_TO_INTERVAL(INTERVAL_KIND) \
+    struct NameToInterval ## INTERVAL_KIND \
+    { \
+        static constexpr auto name = "toInterval" #INTERVAL_KIND; \
+        static constexpr auto kind = IntervalKind::Kind::INTERVAL_KIND; \
+    };
+
+DEFINE_NAME_TO_INTERVAL(Nanosecond)
+DEFINE_NAME_TO_INTERVAL(Microsecond)
+DEFINE_NAME_TO_INTERVAL(Millisecond)
+DEFINE_NAME_TO_INTERVAL(Second)
+DEFINE_NAME_TO_INTERVAL(Minute)
+DEFINE_NAME_TO_INTERVAL(Hour)
+DEFINE_NAME_TO_INTERVAL(Day)
+DEFINE_NAME_TO_INTERVAL(Week)
+DEFINE_NAME_TO_INTERVAL(Month)
+DEFINE_NAME_TO_INTERVAL(Quarter)
+DEFINE_NAME_TO_INTERVAL(Year)
+
+#undef DEFINE_NAME_TO_INTERVAL
+
+struct NameParseDateTimeBestEffort;
+struct NameParseDateTimeBestEffortOrZero;
+struct NameParseDateTimeBestEffortOrNull;
+
+template<typename Name, typename ToDataType>
+static inline bool isDateTime64(const ColumnsWithTypeAndName & arguments)
+{
+    if constexpr (std::is_same_v<ToDataType, DataTypeDateTime64>)
+        return true;
+    else if constexpr (std::is_same_v<Name, NameToDateTime> || std::is_same_v<Name, NameParseDateTimeBestEffort>
+        || std::is_same_v<Name, NameParseDateTimeBestEffortOrZero> || std::is_same_v<Name, NameParseDateTimeBestEffortOrNull>)
+    {
+        return (arguments.size() == 2 && isUInt(arguments[1].type)) || arguments.size() == 3;
+    }
+
+    return false;
+}
+
+template <typename ToDataType, typename Name, typename MonotonicityImpl>
+class FunctionConvert : public IFunction
+{
+public:
+    using Monotonic = MonotonicityImpl;
+
+    static constexpr auto name = Name::name;
+    static constexpr bool to_decimal =
+        std::is_same_v<Name, NameToDecimal32> || std::is_same_v<Name, NameToDecimal64>
+         || std::is_same_v<Name, NameToDecimal128> || std::is_same_v<Name, NameToDecimal256>;
+
+    static constexpr bool to_datetime64 = std::is_same_v<ToDataType, DataTypeDateTime64>;
+
+    static constexpr bool to_string_or_fixed_string = std::is_same_v<ToDataType, DataTypeFixedString> ||
+                                                      std::is_same_v<ToDataType, DataTypeString>;
+
+    static constexpr bool to_date_or_datetime = std::is_same_v<ToDataType, DataTypeDate> ||
+                                                std::is_same_v<ToDataType, DataTypeDate32> ||
+                                                std::is_same_v<ToDataType, DataTypeDateTime>;
+
+    static FunctionPtr create(ContextPtr context) { return std::make_shared<FunctionConvert>(context); }
+    static FunctionPtr create() { return std::make_shared<FunctionConvert>(); }
+
+    FunctionConvert() = default;
+    explicit FunctionConvert(ContextPtr context_) : context(context_) {}
+
+    String getName() const override
+    {
+        return name;
+    }
+
+    bool isVariadic() const override { return true; }
+    size_t getNumberOfArguments() const override { return 0; }
+    bool isInjective(const ColumnsWithTypeAndName &) const override { return std::is_same_v<Name, NameToString>; }
+    bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & arguments) const override
+    {
+        /// TODO: We can make more optimizations here.
+        return !(to_date_or_datetime && isNumber(*arguments[0].type));
+    }
+
+    using DefaultReturnTypeGetter = std::function<DataTypePtr(const ColumnsWithTypeAndName &)>;
+    static DataTypePtr getReturnTypeDefaultImplementationForNulls(const ColumnsWithTypeAndName & arguments, const DefaultReturnTypeGetter & getter)
+    {
+        NullPresence null_presence = getNullPresense(arguments);
+
+        if (null_presence.has_null_constant)
+        {
+            return makeNullable(std::make_shared<DataTypeNothing>());
+        }
+        if (null_presence.has_nullable)
+        {
+            auto nested_columns = Block(createBlockWithNestedColumns(arguments));
+            auto return_type = getter(ColumnsWithTypeAndName(nested_columns.begin(), nested_columns.end()));
+            return makeNullable(return_type);
+        }
+
+        return getter(arguments);
+    }
+
+    DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override
+    {
+        auto getter = [&] (const auto & args) { return getReturnTypeImplRemovedNullable(args); };
+        auto res = getReturnTypeDefaultImplementationForNulls(arguments, getter);
+        to_nullable = res->isNullable();
+        checked_return_type = true;
+        return res;
+    }
+
+    DataTypePtr getReturnTypeImplRemovedNullable(const ColumnsWithTypeAndName & arguments) const
+    {
+        FunctionArgumentDescriptors mandatory_args = {{"Value", nullptr, nullptr, nullptr}};
+        FunctionArgumentDescriptors optional_args;
+
+        if constexpr (to_decimal)
+        {
+            mandatory_args.push_back({"scale", static_cast<FunctionArgumentDescriptor::TypeValidator>(&isNativeInteger), &isColumnConst, "const Integer"});
+        }
+
+        if (!to_decimal && isDateTime64<Name, ToDataType>(arguments))
+        {
+            mandatory_args.push_back({"scale", static_cast<FunctionArgumentDescriptor::TypeValidator>(&isNativeInteger), &isColumnConst, "const Integer"});
+        }
+
+        // toString(DateTime or DateTime64, [timezone: String])
+        if ((std::is_same_v<Name, NameToString> && !arguments.empty() && (isDateTime64(arguments[0].type) || isDateTime(arguments[0].type)))
+            // toUnixTimestamp(value[, timezone : String])
+            || std::is_same_v<Name, NameToUnixTimestamp>
+            // toDate(value[, timezone : String])
+            || std::is_same_v<ToDataType, DataTypeDate> // TODO: shall we allow timestamp argument for toDate? DateTime knows nothing about timezones and this argument is ignored below.
+            // toDate32(value[, timezone : String])
+            || std::is_same_v<ToDataType, DataTypeDate32>
+            // toDateTime(value[, timezone: String])
+            || std::is_same_v<ToDataType, DataTypeDateTime>
+            // toDateTime64(value, scale : Integer[, timezone: String])
+            || std::is_same_v<ToDataType, DataTypeDateTime64>)
+        {
+            optional_args.push_back({"timezone", static_cast<FunctionArgumentDescriptor::TypeValidator>(&isString), nullptr, "String"});
+        }
+
+        validateFunctionArgumentTypes(*this, arguments, mandatory_args, optional_args);
+
+        if constexpr (std::is_same_v<ToDataType, DataTypeInterval>)
+        {
+            return std::make_shared<DataTypeInterval>(Name::kind);
+        }
+        else if constexpr (to_decimal)
+        {
+            UInt64 scale = extractToDecimalScale(arguments[1]);
+
+            if constexpr (std::is_same_v<Name, NameToDecimal32>)
+                return createDecimalMaxPrecision<Decimal32>(scale);
+            else if constexpr (std::is_same_v<Name, NameToDecimal64>)
+                return createDecimalMaxPrecision<Decimal64>(scale);
+            else if constexpr (std::is_same_v<Name, NameToDecimal128>)
+                return createDecimalMaxPrecision<Decimal128>(scale);
+            else if constexpr (std::is_same_v<Name, NameToDecimal256>)
+                return createDecimalMaxPrecision<Decimal256>(scale);
+
+            throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected branch in code of conversion function: it is a bug.");
+        }
+        else
+        {
+            // Optional second argument with time zone for DateTime.
+            UInt8 timezone_arg_position = 1;
+            UInt32 scale [[maybe_unused]] = DataTypeDateTime64::default_scale;
+
+            // DateTime64 requires more arguments: scale and timezone. Since timezone is optional, scale should be first.
+            if (isDateTime64<Name, ToDataType>(arguments))
+            {
+                timezone_arg_position += 1;
+                scale = static_cast<UInt32>(arguments[1].column->get64(0));
+
+                if (to_datetime64 || scale != 0) /// toDateTime('xxxx-xx-xx xx:xx:xx', 0) return DateTime
+                    return std::make_shared<DataTypeDateTime64>(scale,
+                        extractTimeZoneNameFromFunctionArguments(arguments, timezone_arg_position, 0, false));
+
+                return std::make_shared<DataTypeDateTime>(extractTimeZoneNameFromFunctionArguments(arguments, timezone_arg_position, 0, false));
+            }
+
+            if constexpr (std::is_same_v<ToDataType, DataTypeDateTime>)
+                return std::make_shared<DataTypeDateTime>(extractTimeZoneNameFromFunctionArguments(arguments, timezone_arg_position, 0, false));
+            else if constexpr (std::is_same_v<ToDataType, DataTypeDateTime64>)
+                throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected branch in code of conversion function: it is a bug.");
+            else
+                return std::make_shared<ToDataType>();
+        }
+    }
+
+    /// Function actually uses default implementation for nulls,
+    /// but we need to know if return type is Nullable or not,
+    /// so we use checked_return_type only to intercept the first call to getReturnTypeImpl(...).
+    bool useDefaultImplementationForNulls() const override
+    {
+        bool to_nullable_string = to_nullable && std::is_same_v<ToDataType, DataTypeString>;
+        return checked_return_type && !to_nullable_string;
+    }
+
+    bool useDefaultImplementationForConstants() const override { return true; }
+    ColumnNumbers getArgumentsThatAreAlwaysConstant() const override
+    {
+        if constexpr (std::is_same_v<ToDataType, DataTypeString>)
+            return {};
+        else if constexpr (std::is_same_v<ToDataType, DataTypeDateTime64>)
+            return {2};
+        return {1};
+    }
+    bool canBeExecutedOnDefaultArguments() const override { return false; }
+
+    ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const override
+    {
+        try
+        {
+            return executeInternal(arguments, result_type, input_rows_count);
+        }
+        catch (Exception & e)
+        {
+            /// More convenient error message.
+            if (e.code() == ErrorCodes::ATTEMPT_TO_READ_AFTER_EOF)
+            {
+                e.addMessage("Cannot parse "
+                    + result_type->getName() + " from "
+                    + arguments[0].type->getName()
+                    + ", because value is too short");
+            }
+            else if (e.code() == ErrorCodes::CANNOT_PARSE_NUMBER
+                || e.code() == ErrorCodes::CANNOT_READ_ARRAY_FROM_TEXT
+                || e.code() == ErrorCodes::CANNOT_PARSE_INPUT_ASSERTION_FAILED
+                || e.code() == ErrorCodes::CANNOT_PARSE_QUOTED_STRING
+                || e.code() == ErrorCodes::CANNOT_PARSE_ESCAPE_SEQUENCE
+                || e.code() == ErrorCodes::CANNOT_PARSE_DATE
+                || e.code() == ErrorCodes::CANNOT_PARSE_DATETIME
+                || e.code() == ErrorCodes::CANNOT_PARSE_UUID
+                || e.code() == ErrorCodes::CANNOT_PARSE_IPV4
+                || e.code() == ErrorCodes::CANNOT_PARSE_IPV6)
+            {
+                e.addMessage("Cannot parse "
+                    + result_type->getName() + " from "
+                    + arguments[0].type->getName());
+            }
+
+            throw;
+        }
+    }
+
+    bool hasInformationAboutMonotonicity() const override
+    {
+        return Monotonic::has();
+    }
+
+    Monotonicity getMonotonicityForRange(const IDataType & type, const Field & left, const Field & right) const override
+    {
+        return Monotonic::get(type, left, right);
+    }
+
+private:
+    ContextPtr context;
+    mutable bool checked_return_type = false;
+    mutable bool to_nullable = false;
+
+    ColumnPtr executeInternal(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const
+    {
+        if (arguments.empty())
+            throw Exception(ErrorCodes::TOO_FEW_ARGUMENTS_FOR_FUNCTION, "Function {} expects at least 1 argument", getName());
+
+        if (result_type->onlyNull())
+            return result_type->createColumnConstWithDefaultValue(input_rows_count);
+
+        const DataTypePtr from_type = removeNullable(arguments[0].type);
+        ColumnPtr result_column;
+
+        [[maybe_unused]] FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior = default_date_time_overflow_behavior;
+
+        if (context)
+            date_time_overflow_behavior = context->getSettingsRef().date_time_overflow_behavior.value;
+
+        auto call = [&](const auto & types, const auto & tag) -> bool
+        {
+            using Types = std::decay_t<decltype(types)>;
+            using LeftDataType = typename Types::LeftType;
+            using RightDataType = typename Types::RightType;
+            using SpecialTag = std::decay_t<decltype(tag)>;
+
+            if constexpr (IsDataTypeDecimal<RightDataType>)
+            {
+                if constexpr (std::is_same_v<RightDataType, DataTypeDateTime64>)
+                {
+                    /// Account for optional timezone argument.
+                    if (arguments.size() != 2 && arguments.size() != 3)
+                        throw Exception(ErrorCodes::TOO_FEW_ARGUMENTS_FOR_FUNCTION, "Function {} expects 2 or 3 arguments for DataTypeDateTime64.", getName());
+                }
+                else if (arguments.size() != 2)
+                {
+                    throw Exception(ErrorCodes::TOO_FEW_ARGUMENTS_FOR_FUNCTION, "Function {} expects 2 arguments for Decimal.", getName());
+                }
+
+                const ColumnWithTypeAndName & scale_column = arguments[1];
+                UInt32 scale = extractToDecimalScale(scale_column);
+
+                switch (date_time_overflow_behavior)
+                {
+                    case FormatSettings::DateTimeOverflowBehavior::Throw:
+                        result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag, FormatSettings::DateTimeOverflowBehavior::Throw>::execute(arguments, result_type, input_rows_count, scale);
+                        break;
+                    case FormatSettings::DateTimeOverflowBehavior::Ignore:
+                        result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag, FormatSettings::DateTimeOverflowBehavior::Ignore>::execute(arguments, result_type, input_rows_count, scale);
+                        break;
+                    case FormatSettings::DateTimeOverflowBehavior::Saturate:
+                        result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag, FormatSettings::DateTimeOverflowBehavior::Saturate>::execute(arguments, result_type, input_rows_count, scale);
+                        break;
+                }
+
+            }
+            else if constexpr (IsDataTypeDateOrDateTime<RightDataType> && std::is_same_v<LeftDataType, DataTypeDateTime64>)
+            {
+                const auto * dt64 = assert_cast<const DataTypeDateTime64 *>(arguments[0].type.get());
+                switch (date_time_overflow_behavior)
+                {
+                    case FormatSettings::DateTimeOverflowBehavior::Throw:
+                        result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag, FormatSettings::DateTimeOverflowBehavior::Throw>::execute(arguments, result_type, input_rows_count, dt64->getScale());
+                        break;
+                    case FormatSettings::DateTimeOverflowBehavior::Ignore:
+                        result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag, FormatSettings::DateTimeOverflowBehavior::Ignore>::execute(arguments, result_type, input_rows_count, dt64->getScale());
+                        break;
+                    case FormatSettings::DateTimeOverflowBehavior::Saturate:
+                        result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag, FormatSettings::DateTimeOverflowBehavior::Saturate>::execute(arguments, result_type, input_rows_count, dt64->getScale());
+                        break;
+                }
+            }
+#define GENERATE_OVERFLOW_MODE_CASE(OVERFLOW_MODE) \
+            case FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE: \
+                result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag, FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE>::execute( \
+                arguments, result_type, input_rows_count); \
+                break;
+
+            else if constexpr (IsDataTypeDecimalOrNumber<LeftDataType> && IsDataTypeDecimalOrNumber<RightDataType>)
+            {
+                using LeftT = typename LeftDataType::FieldType;
+                using RightT = typename RightDataType::FieldType;
+
+                static constexpr bool bad_left =
+                    is_decimal<LeftT> || std::is_floating_point_v<LeftT> || is_big_int_v<LeftT> || is_signed_v<LeftT>;
+                static constexpr bool bad_right =
+                    is_decimal<RightT> || std::is_floating_point_v<RightT> || is_big_int_v<RightT> || is_signed_v<RightT>;
+
+                /// Disallow int vs UUID conversion (but support int vs UInt128 conversion)
+                if constexpr ((bad_left && std::is_same_v<RightDataType, DataTypeUUID>) ||
+                              (bad_right && std::is_same_v<LeftDataType, DataTypeUUID>))
+                {
+                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Wrong UUID conversion");
+                }
+                else
+                {
+                    switch (date_time_overflow_behavior)
+                    {
+                        GENERATE_OVERFLOW_MODE_CASE(Throw)
+                        GENERATE_OVERFLOW_MODE_CASE(Ignore)
+                        GENERATE_OVERFLOW_MODE_CASE(Saturate)
+                    }
+                }
+            }
+            else if constexpr ((IsDataTypeNumber<LeftDataType> || IsDataTypeDateOrDateTime<LeftDataType>)
+                               && IsDataTypeDateOrDateTime<RightDataType>)
+            {
+                switch (date_time_overflow_behavior)
+                {
+                    GENERATE_OVERFLOW_MODE_CASE(Throw)
+                    GENERATE_OVERFLOW_MODE_CASE(Ignore)
+                    GENERATE_OVERFLOW_MODE_CASE(Saturate)
+                }
+            }
+#undef GENERATE_OVERFLOW_MODE_CASE
+            else
+                  result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag>::execute(arguments, result_type, input_rows_count);
+
+            return true;
+        };
+
+        if (isDateTime64<Name, ToDataType>(arguments))
+        {
+            /// For toDateTime('xxxx-xx-xx xx:xx:xx.00', 2[, 'timezone']) we need to it convert to DateTime64
+            const ColumnWithTypeAndName & scale_column = arguments[1];
+            UInt32 scale = extractToDecimalScale(scale_column);
+
+            if (to_datetime64 || scale != 0) /// When scale = 0, the data type is DateTime otherwise the data type is DateTime64
+            {
+                if (!callOnIndexAndDataType<DataTypeDateTime64>(from_type->getTypeId(), call, ConvertDefaultBehaviorTag{}))
+                    throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of argument of function {}",
+                                    arguments[0].type->getName(), getName());
+
+                return result_column;
+            }
+        }
+
+        if constexpr (std::is_same_v<ToDataType, DataTypeString>)
+        {
+            if (from_type->getCustomSerialization())
+                return ConvertImplGenericToString<ColumnString>::execute(arguments, result_type, input_rows_count);
+        }
+
+        bool done = false;
+        if constexpr (to_string_or_fixed_string)
+        {
+            done = callOnIndexAndDataType<ToDataType>(from_type->getTypeId(), call, ConvertDefaultBehaviorTag{});
+        }
+        else
+        {
+            bool cast_ipv4_ipv6_default_on_conversion_error = false;
+            if constexpr (is_any_of<ToDataType, DataTypeIPv4, DataTypeIPv6>)
+                if (context && (cast_ipv4_ipv6_default_on_conversion_error = context->getSettingsRef().cast_ipv4_ipv6_default_on_conversion_error))
+                    done = callOnIndexAndDataType<ToDataType>(from_type->getTypeId(), call, ConvertReturnZeroOnErrorTag{});
+
+            if (!cast_ipv4_ipv6_default_on_conversion_error)
+            {
+                /// We should use ConvertFromStringExceptionMode::Null mode when converting from String (or FixedString)
+                /// to Nullable type, to avoid 'value is too short' error on attempt to parse empty string from NULL values.
+                if (to_nullable && WhichDataType(from_type).isStringOrFixedString())
+                    done = callOnIndexAndDataType<ToDataType>(from_type->getTypeId(), call, ConvertReturnNullOnErrorTag{});
+                else
+                    done = callOnIndexAndDataType<ToDataType>(from_type->getTypeId(), call, ConvertDefaultBehaviorTag{});
+            }
+        }
+
+        if (!done)
+        {
+            /// Generic conversion of any type to String.
+            if (std::is_same_v<ToDataType, DataTypeString>)
+            {
+                return ConvertImplGenericToString<ColumnString>::execute(arguments, result_type, input_rows_count);
+            }
+            else
+                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of argument of function {}",
+                    arguments[0].type->getName(), getName());
+        }
+
+        return result_column;
+    }
+};
+
+
+/** Function toTOrZero (where T is number of date or datetime type):
+  *  try to convert from String to type T through parsing,
+  *  if cannot parse, return default value instead of throwing exception.
+  * Function toTOrNull will return Nullable type with NULL when cannot parse.
+  * NOTE Also need to implement tryToUnixTimestamp with timezone.
+  */
+template <typename ToDataType, typename Name,
+    ConvertFromStringExceptionMode exception_mode,
+    ConvertFromStringParsingMode parsing_mode = ConvertFromStringParsingMode::Normal>
+class FunctionConvertFromString : public IFunction
+{
+public:
+    static constexpr auto name = Name::name;
+    static constexpr bool to_decimal =
+        std::is_same_v<ToDataType, DataTypeDecimal<Decimal32>> ||
+        std::is_same_v<ToDataType, DataTypeDecimal<Decimal64>> ||
+        std::is_same_v<ToDataType, DataTypeDecimal<Decimal128>> ||
+        std::is_same_v<ToDataType, DataTypeDecimal<Decimal256>>;
+
+    static constexpr bool to_datetime64 = std::is_same_v<ToDataType, DataTypeDateTime64>;
+
+    static FunctionPtr create(ContextPtr) { return std::make_shared<FunctionConvertFromString>(); }
+    static FunctionPtr create() { return std::make_shared<FunctionConvertFromString>(); }
+
+    String getName() const override
+    {
+        return name;
+    }
+
+    bool isVariadic() const override { return true; }
+    bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return true; }
+    size_t getNumberOfArguments() const override { return 0; }
+
+    bool useDefaultImplementationForConstants() const override { return true; }
+    bool canBeExecutedOnDefaultArguments() const override { return false; }
+
+    ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {1}; }
+
+    DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override
+    {
+        DataTypePtr res;
+
+        if (isDateTime64<Name, ToDataType>(arguments))
+        {
+            validateFunctionArgumentTypes(*this, arguments,
+                FunctionArgumentDescriptors{{"string", static_cast<FunctionArgumentDescriptor::TypeValidator>(&isStringOrFixedString), nullptr, "String or FixedString"}},
+                // optional
+                FunctionArgumentDescriptors{
+                    {"precision", static_cast<FunctionArgumentDescriptor::TypeValidator>(&isUInt8), isColumnConst, "const UInt8"},
+                    {"timezone", static_cast<FunctionArgumentDescriptor::TypeValidator>(&isStringOrFixedString), isColumnConst, "const String or FixedString"},
+                });
+
+            UInt64 scale = to_datetime64 ? DataTypeDateTime64::default_scale : 0;
+            if (arguments.size() > 1)
+                scale = extractToDecimalScale(arguments[1]);
+            const auto timezone = extractTimeZoneNameFromFunctionArguments(arguments, 2, 0, false);
+
+            res = scale == 0 ? res = std::make_shared<DataTypeDateTime>(timezone) : std::make_shared<DataTypeDateTime64>(scale, timezone);
+        }
+        else
+        {
+            if ((arguments.size() != 1 && arguments.size() != 2) || (to_decimal && arguments.size() != 2))
+                throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH,
+                    "Number of arguments for function {} doesn't match: passed {}, should be 1 or 2. "
+                    "Second argument only make sense for DateTime (time zone, optional) and Decimal (scale).",
+                    getName(), arguments.size());
+
+            if (!isStringOrFixedString(arguments[0].type))
+            {
+                if (this->getName().find("OrZero") != std::string::npos ||
+                    this->getName().find("OrNull") != std::string::npos)
+                    throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of first argument of function {}. "
+                            "Conversion functions with postfix 'OrZero' or 'OrNull' should take String argument",
+                            arguments[0].type->getName(), getName());
+                else
+                    throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of first argument of function {}",
+                            arguments[0].type->getName(), getName());
+            }
+
+            if (arguments.size() == 2)
+            {
+                if constexpr (std::is_same_v<ToDataType, DataTypeDateTime>)
+                {
+                    if (!isString(arguments[1].type))
+                        throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of 2nd argument of function {}",
+                            arguments[1].type->getName(), getName());
+                }
+                else if constexpr (to_decimal)
+                {
+                    if (!isInteger(arguments[1].type))
+                        throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of 2nd argument of function {}",
+                            arguments[1].type->getName(), getName());
+                    if (!arguments[1].column)
+                        throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Second argument for function {} must be constant", getName());
+                }
+                else
+                {
+                    throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH,
+                        "Number of arguments for function {} doesn't match: passed {}, should be 1. "
+                        "Second argument makes sense only for DateTime and Decimal.",
+                        getName(), arguments.size());
+                }
+            }
+
+            if constexpr (std::is_same_v<ToDataType, DataTypeDateTime>)
+                res = std::make_shared<DataTypeDateTime>(extractTimeZoneNameFromFunctionArguments(arguments, 1, 0, false));
+            else if constexpr (std::is_same_v<ToDataType, DataTypeDateTime64>)
+                throw Exception(ErrorCodes::LOGICAL_ERROR, "MaterializedMySQL is a bug.");
+            else if constexpr (to_decimal)
+            {
+                UInt64 scale = extractToDecimalScale(arguments[1]);
+                res = createDecimalMaxPrecision<typename ToDataType::FieldType>(scale);
+                if (!res)
+                    throw Exception(ErrorCodes::LOGICAL_ERROR, "Something wrong with toDecimalNNOrZero() or toDecimalNNOrNull()");
+            }
+            else
+                res = std::make_shared<ToDataType>();
+        }
+
+        if constexpr (exception_mode == ConvertFromStringExceptionMode::Null)
+            res = std::make_shared<DataTypeNullable>(res);
+
+        return res;
+    }
+
+    template <typename ConvertToDataType>
+    ColumnPtr executeInternal(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count, UInt32 scale = 0) const
+    {
+        const IDataType * from_type = arguments[0].type.get();
+
+        if (checkAndGetDataType<DataTypeString>(from_type))
+        {
+            return ConvertThroughParsing<DataTypeString, ConvertToDataType, Name, exception_mode, parsing_mode>::execute(
+                arguments, result_type, input_rows_count, scale);
+        }
+        else if (checkAndGetDataType<DataTypeFixedString>(from_type))
+        {
+            return ConvertThroughParsing<DataTypeFixedString, ConvertToDataType, Name, exception_mode, parsing_mode>::execute(
+                arguments, result_type, input_rows_count, scale);
+        }
+
+        return nullptr;
+    }
+
+    ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const override
+    {
+        ColumnPtr result_column;
+
+        if constexpr (to_decimal)
+            result_column = executeInternal<ToDataType>(arguments, result_type, input_rows_count,
+                assert_cast<const ToDataType &>(*removeNullable(result_type)).getScale());
+        else
+        {
+            if (isDateTime64<Name, ToDataType>(arguments))
+            {
+                UInt64 scale = to_datetime64 ? DataTypeDateTime64::default_scale : 0;
+                if (arguments.size() > 1)
+                    scale = extractToDecimalScale(arguments[1]);
+
+                if (scale == 0)
+                    result_column = executeInternal<DataTypeDateTime>(arguments, result_type, input_rows_count);
+                else
+                {
+                    result_column = executeInternal<DataTypeDateTime64>(arguments, result_type, input_rows_count, static_cast<UInt32>(scale));
+                }
+            }
+            else
+            {
+                result_column = executeInternal<ToDataType>(arguments, result_type, input_rows_count);
+            }
+        }
+
+        if (!result_column)
+            throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of argument of function {}. "
+                "Only String or FixedString argument is accepted for try-conversion function. For other arguments, "
+                "use function without 'orZero' or 'orNull'.", arguments[0].type->getName(), getName());
+
+        return result_column;
+    }
+};
+
+
+/// Monotonicity.
+
+struct PositiveMonotonicity
+{
+    static bool has() { return true; }
+    static IFunction::Monotonicity get(const IDataType &, const Field &, const Field &)
+    {
+        return { .is_monotonic = true };
+    }
+};
+
+struct UnknownMonotonicity
+{
+    static bool has() { return false; }
+    static IFunction::Monotonicity get(const IDataType &, const Field &, const Field &)
+    {
+        return { };
+    }
+};
+
+template <typename T>
+struct ToNumberMonotonicity
+{
+    static bool has() { return true; }
+
+    static UInt64 divideByRangeOfType(UInt64 x)
+    {
+        if constexpr (sizeof(T) < sizeof(UInt64))
+            return x >> (sizeof(T) * 8);
+        else
+            return 0;
+    }
+
+    static IFunction::Monotonicity get(const IDataType & type, const Field & left, const Field & right)
+    {
+        if (!type.isValueRepresentedByNumber())
+            return {};
+
+        /// If type is same, the conversion is always monotonic.
+        /// (Enum has separate case, because it is different data type)
+        if (checkAndGetDataType<DataTypeNumber<T>>(&type) ||
+            checkAndGetDataType<DataTypeEnum<T>>(&type))
+            return { .is_monotonic = true, .is_always_monotonic = true };
+
+        /// Float cases.
+
+        /// When converting to Float, the conversion is always monotonic.
+        if constexpr (std::is_floating_point_v<T>)
+            return { .is_monotonic = true, .is_always_monotonic = true };
+
+        const auto * low_cardinality = typeid_cast<const DataTypeLowCardinality *>(&type);
+        const IDataType * low_cardinality_dictionary_type = nullptr;
+        if (low_cardinality)
+            low_cardinality_dictionary_type = low_cardinality->getDictionaryType().get();
+
+        WhichDataType which_type(type);
+        WhichDataType which_inner_type = low_cardinality
+            ? WhichDataType(low_cardinality_dictionary_type)
+            : WhichDataType(type);
+
+        /// If converting from Float, for monotonicity, arguments must fit in range of result type.
+        if (which_inner_type.isFloat())
+        {
+            if (left.isNull() || right.isNull())
+                return {};
+
+            Float64 left_float = left.get<Float64>();
+            Float64 right_float = right.get<Float64>();
+
+            if (left_float >= static_cast<Float64>(std::numeric_limits<T>::min())
+                && left_float <= static_cast<Float64>(std::numeric_limits<T>::max())
+                && right_float >= static_cast<Float64>(std::numeric_limits<T>::min())
+                && right_float <= static_cast<Float64>(std::numeric_limits<T>::max()))
+                return { .is_monotonic = true };
+
+            return {};
+        }
+
+        /// Integer cases.
+
+        /// Only support types represented by native integers.
+        /// It can be extended to big integers, decimals and DateTime64 later.
+        /// By the way, NULLs are representing unbounded ranges.
+        if (!((left.isNull() || left.getType() == Field::Types::UInt64 || left.getType() == Field::Types::Int64)
+            && (right.isNull() || right.getType() == Field::Types::UInt64 || right.getType() == Field::Types::Int64)))
+            return {};
+
+        const bool from_is_unsigned = type.isValueRepresentedByUnsignedInteger();
+        const bool to_is_unsigned = is_unsigned_v<T>;
+
+        const size_t size_of_from = type.getSizeOfValueInMemory();
+        const size_t size_of_to = sizeof(T);
+
+        const bool left_in_first_half = left.isNull()
+            ? from_is_unsigned
+            : (left.get<Int64>() >= 0);
+
+        const bool right_in_first_half = right.isNull()
+            ? !from_is_unsigned
+            : (right.get<Int64>() >= 0);
+
+        /// Size of type is the same.
+        if (size_of_from == size_of_to)
+        {
+            if (from_is_unsigned == to_is_unsigned)
+                return { .is_monotonic = true, .is_always_monotonic = true };
+
+            if (left_in_first_half == right_in_first_half)
+                return { .is_monotonic = true };
+
+            return {};
+        }
+
+        /// Size of type is expanded.
+        if (size_of_from < size_of_to)
+        {
+            if (from_is_unsigned == to_is_unsigned)
+                return { .is_monotonic = true, .is_always_monotonic = true };
+
+            if (!to_is_unsigned)
+                return { .is_monotonic = true, .is_always_monotonic = true };
+
+            /// signed -> unsigned. If arguments from the same half, then function is monotonic.
+            if (left_in_first_half == right_in_first_half)
+                return { .is_monotonic = true };
+
+            return {};
+        }
+
+        /// Size of type is shrunk.
+        if (size_of_from > size_of_to)
+        {
+            /// Function cannot be monotonic on unbounded ranges.
+            if (left.isNull() || right.isNull())
+                return {};
+
+            /// Function cannot be monotonic when left and right are not on the same ranges.
+            if (divideByRangeOfType(left.get<UInt64>()) != divideByRangeOfType(right.get<UInt64>()))
+                return {};
+
+            if (to_is_unsigned)
+                return { .is_monotonic = true };
+            else
+            {
+                // If To is signed, it's possible that the signedness is different after conversion. So we check it explicitly.
+                const bool is_monotonic = (T(left.get<UInt64>()) >= 0) == (T(right.get<UInt64>()) >= 0);
+
+                return { .is_monotonic = is_monotonic };
+            }
+        }
+
+        UNREACHABLE();
+    }
+};
+
+struct ToDateMonotonicity
+{
+    static bool has() { return true; }
+
+    static IFunction::Monotonicity get(const IDataType & type, const Field & left, const Field & right)
+    {
+        auto which = WhichDataType(type);
+        if (which.isDateOrDate32() || which.isDateTime() || which.isDateTime64() || which.isInt8() || which.isInt16() || which.isUInt8()
+            || which.isUInt16())
+        {
+            return {.is_monotonic = true, .is_always_monotonic = true};
+        }
+        else if (
+            ((left.getType() == Field::Types::UInt64 || left.isNull()) && (right.getType() == Field::Types::UInt64 || right.isNull())
+             && ((left.isNull() || left.get<UInt64>() < 0xFFFF) && (right.isNull() || right.get<UInt64>() >= 0xFFFF)))
+            || ((left.getType() == Field::Types::Int64 || left.isNull()) && (right.getType() == Field::Types::Int64 || right.isNull())
+                && ((left.isNull() || left.get<Int64>() < 0xFFFF) && (right.isNull() || right.get<Int64>() >= 0xFFFF)))
+            || ((
+                (left.getType() == Field::Types::Float64 || left.isNull())
+                && (right.getType() == Field::Types::Float64 || right.isNull())
+                && ((left.isNull() || left.get<Float64>() < 0xFFFF) && (right.isNull() || right.get<Float64>() >= 0xFFFF))))
+            || !isNativeNumber(type))
+        {
+            return {};
+        }
+        else
+        {
+            return {.is_monotonic = true, .is_always_monotonic = true};
+        }
+    }
+};
+
+struct ToDateTimeMonotonicity
+{
+    static bool has() { return true; }
+
+    static IFunction::Monotonicity get(const IDataType & type, const Field &, const Field &)
+    {
+        if (type.isValueRepresentedByNumber())
+            return {.is_monotonic = true, .is_always_monotonic = true};
+        else
+            return {};
+    }
+};
+
+/** The monotonicity for the `toString` function is mainly determined for test purposes.
+  * It is doubtful that anyone is looking to optimize queries with conditions `toString(CounterID) = 34`.
+  */
+struct ToStringMonotonicity
+{
+    static bool has() { return true; }
+
+    static IFunction::Monotonicity get(const IDataType & type, const Field & left, const Field & right)
+    {
+        IFunction::Monotonicity positive{ .is_monotonic = true };
+        IFunction::Monotonicity not_monotonic;
+
+        const auto * type_ptr = &type;
+        if (const auto * low_cardinality_type = checkAndGetDataType<DataTypeLowCardinality>(type_ptr))
+            type_ptr = low_cardinality_type->getDictionaryType().get();
+
+        /// Order on enum values (which is the order on integers) is completely arbitrary in respect to the order on strings.
+        if (WhichDataType(type).isEnum())
+            return not_monotonic;
+
+        /// `toString` function is monotonous if the argument is Date or Date32 or DateTime or String, or non-negative numbers with the same number of symbols.
+        if (checkDataTypes<DataTypeDate, DataTypeDate32, DataTypeDateTime, DataTypeString>(type_ptr))
+            return positive;
+
+        if (left.isNull() || right.isNull())
+            return {};
+
+        if (left.getType() == Field::Types::UInt64
+            && right.getType() == Field::Types::UInt64)
+        {
+            return (left.get<Int64>() == 0 && right.get<Int64>() == 0)
+                || (floor(log10(left.get<UInt64>())) == floor(log10(right.get<UInt64>())))
+                ? positive : not_monotonic;
+        }
+
+        if (left.getType() == Field::Types::Int64
+            && right.getType() == Field::Types::Int64)
+        {
+            return (left.get<Int64>() == 0 && right.get<Int64>() == 0)
+                || (left.get<Int64>() > 0 && right.get<Int64>() > 0 && floor(log10(left.get<Int64>())) == floor(log10(right.get<Int64>())))
+                ? positive : not_monotonic;
+        }
+
+        return not_monotonic;
+    }
+};
+
+
+struct NameToUInt8 { static constexpr auto name = "toUInt8"; };
+struct NameToUInt16 { static constexpr auto name = "toUInt16"; };
+struct NameToUInt32 { static constexpr auto name = "toUInt32"; };
+struct NameToUInt64 { static constexpr auto name = "toUInt64"; };
+struct NameToUInt128 { static constexpr auto name = "toUInt128"; };
+struct NameToUInt256 { static constexpr auto name = "toUInt256"; };
+struct NameToInt8 { static constexpr auto name = "toInt8"; };
+struct NameToInt16 { static constexpr auto name = "toInt16"; };
+struct NameToInt32 { static constexpr auto name = "toInt32"; };
+struct NameToInt64 { static constexpr auto name = "toInt64"; };
+struct NameToInt128 { static constexpr auto name = "toInt128"; };
+struct NameToInt256 { static constexpr auto name = "toInt256"; };
+struct NameToFloat32 { static constexpr auto name = "toFloat32"; };
+struct NameToFloat64 { static constexpr auto name = "toFloat64"; };
+struct NameToUUID { static constexpr auto name = "toUUID"; };
+struct NameToIPv4 { static constexpr auto name = "toIPv4"; };
+struct NameToIPv6 { static constexpr auto name = "toIPv6"; };
+
+using FunctionToUInt8 = FunctionConvert<DataTypeUInt8, NameToUInt8, ToNumberMonotonicity<UInt8>>;
+using FunctionToUInt16 = FunctionConvert<DataTypeUInt16, NameToUInt16, ToNumberMonotonicity<UInt16>>;
+using FunctionToUInt32 = FunctionConvert<DataTypeUInt32, NameToUInt32, ToNumberMonotonicity<UInt32>>;
+using FunctionToUInt64 = FunctionConvert<DataTypeUInt64, NameToUInt64, ToNumberMonotonicity<UInt64>>;
+using FunctionToUInt128 = FunctionConvert<DataTypeUInt128, NameToUInt128, ToNumberMonotonicity<UInt128>>;
+using FunctionToUInt256 = FunctionConvert<DataTypeUInt256, NameToUInt256, ToNumberMonotonicity<UInt256>>;
+using FunctionToInt8 = FunctionConvert<DataTypeInt8, NameToInt8, ToNumberMonotonicity<Int8>>;
+using FunctionToInt16 = FunctionConvert<DataTypeInt16, NameToInt16, ToNumberMonotonicity<Int16>>;
+using FunctionToInt32 = FunctionConvert<DataTypeInt32, NameToInt32, ToNumberMonotonicity<Int32>>;
+using FunctionToInt64 = FunctionConvert<DataTypeInt64, NameToInt64, ToNumberMonotonicity<Int64>>;
+using FunctionToInt128 = FunctionConvert<DataTypeInt128, NameToInt128, ToNumberMonotonicity<Int128>>;
+using FunctionToInt256 = FunctionConvert<DataTypeInt256, NameToInt256, ToNumberMonotonicity<Int256>>;
+using FunctionToFloat32 = FunctionConvert<DataTypeFloat32, NameToFloat32, ToNumberMonotonicity<Float32>>;
+using FunctionToFloat64 = FunctionConvert<DataTypeFloat64, NameToFloat64, ToNumberMonotonicity<Float64>>;
+
+using FunctionToDate = FunctionConvert<DataTypeDate, NameToDate, ToDateMonotonicity>;
+
+using FunctionToDate32 = FunctionConvert<DataTypeDate32, NameToDate32, ToDateMonotonicity>;
+
+using FunctionToDateTime = FunctionConvert<DataTypeDateTime, NameToDateTime, ToDateTimeMonotonicity>;
+
+using FunctionToDateTime32 = FunctionConvert<DataTypeDateTime, NameToDateTime32, ToDateTimeMonotonicity>;
+
+using FunctionToDateTime64 = FunctionConvert<DataTypeDateTime64, NameToDateTime64, ToDateTimeMonotonicity>;
+
+using FunctionToUUID = FunctionConvert<DataTypeUUID, NameToUUID, ToNumberMonotonicity<UInt128>>;
+using FunctionToIPv4 = FunctionConvert<DataTypeIPv4, NameToIPv4, ToNumberMonotonicity<UInt32>>;
+using FunctionToIPv6 = FunctionConvert<DataTypeIPv6, NameToIPv6, ToNumberMonotonicity<UInt128>>;
+using FunctionToString = FunctionConvert<DataTypeString, NameToString, ToStringMonotonicity>;
+using FunctionToUnixTimestamp = FunctionConvert<DataTypeUInt32, NameToUnixTimestamp, ToNumberMonotonicity<UInt32>>;
+using FunctionToDecimal32 = FunctionConvert<DataTypeDecimal<Decimal32>, NameToDecimal32, UnknownMonotonicity>;
+using FunctionToDecimal64 = FunctionConvert<DataTypeDecimal<Decimal64>, NameToDecimal64, UnknownMonotonicity>;
+using FunctionToDecimal128 = FunctionConvert<DataTypeDecimal<Decimal128>, NameToDecimal128, UnknownMonotonicity>;
+using FunctionToDecimal256 = FunctionConvert<DataTypeDecimal<Decimal256>, NameToDecimal256, UnknownMonotonicity>;
+
+template <typename DataType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior = default_date_time_overflow_behavior> struct FunctionTo;
+
+template <> struct FunctionTo<DataTypeUInt8> { using Type = FunctionToUInt8; };
+template <> struct FunctionTo<DataTypeUInt16> { using Type = FunctionToUInt16; };
+template <> struct FunctionTo<DataTypeUInt32> { using Type = FunctionToUInt32; };
+template <> struct FunctionTo<DataTypeUInt64> { using Type = FunctionToUInt64; };
+template <> struct FunctionTo<DataTypeUInt128> { using Type = FunctionToUInt128; };
+template <> struct FunctionTo<DataTypeUInt256> { using Type = FunctionToUInt256; };
+template <> struct FunctionTo<DataTypeInt8> { using Type = FunctionToInt8; };
+template <> struct FunctionTo<DataTypeInt16> { using Type = FunctionToInt16; };
+template <> struct FunctionTo<DataTypeInt32> { using Type = FunctionToInt32; };
+template <> struct FunctionTo<DataTypeInt64> { using Type = FunctionToInt64; };
+template <> struct FunctionTo<DataTypeInt128> { using Type = FunctionToInt128; };
+template <> struct FunctionTo<DataTypeInt256> { using Type = FunctionToInt256; };
+template <> struct FunctionTo<DataTypeFloat32> { using Type = FunctionToFloat32; };
+template <> struct FunctionTo<DataTypeFloat64> { using Type = FunctionToFloat64; };
+
+template <FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct FunctionTo<DataTypeDate, date_time_overflow_behavior> { using Type = FunctionToDate; };
+
+template <FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct FunctionTo<DataTypeDate32, date_time_overflow_behavior> { using Type = FunctionToDate32; };
+
+template <FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct FunctionTo<DataTypeDateTime, date_time_overflow_behavior> { using Type = FunctionToDateTime; };
+
+template <FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct FunctionTo<DataTypeDateTime64, date_time_overflow_behavior> { using Type = FunctionToDateTime64; };
+
+template <> struct FunctionTo<DataTypeUUID> { using Type = FunctionToUUID; };
+template <> struct FunctionTo<DataTypeIPv4> { using Type = FunctionToIPv4; };
+template <> struct FunctionTo<DataTypeIPv6> { using Type = FunctionToIPv6; };
+template <> struct FunctionTo<DataTypeString> { using Type = FunctionToString; };
+template <> struct FunctionTo<DataTypeFixedString> { using Type = FunctionToFixedString; };
+template <> struct FunctionTo<DataTypeDecimal<Decimal32>> { using Type = FunctionToDecimal32; };
+template <> struct FunctionTo<DataTypeDecimal<Decimal64>> { using Type = FunctionToDecimal64; };
+template <> struct FunctionTo<DataTypeDecimal<Decimal128>> { using Type = FunctionToDecimal128; };
+template <> struct FunctionTo<DataTypeDecimal<Decimal256>> { using Type = FunctionToDecimal256; };
+
+template <typename FieldType> struct FunctionTo<DataTypeEnum<FieldType>>
+    : FunctionTo<DataTypeNumber<FieldType>>
+{
+};
+
+struct NameToUInt8OrZero { static constexpr auto name = "toUInt8OrZero"; };
+struct NameToUInt16OrZero { static constexpr auto name = "toUInt16OrZero"; };
+struct NameToUInt32OrZero { static constexpr auto name = "toUInt32OrZero"; };
+struct NameToUInt64OrZero { static constexpr auto name = "toUInt64OrZero"; };
+struct NameToUInt128OrZero { static constexpr auto name = "toUInt128OrZero"; };
+struct NameToUInt256OrZero { static constexpr auto name = "toUInt256OrZero"; };
+struct NameToInt8OrZero { static constexpr auto name = "toInt8OrZero"; };
+struct NameToInt16OrZero { static constexpr auto name = "toInt16OrZero"; };
+struct NameToInt32OrZero { static constexpr auto name = "toInt32OrZero"; };
+struct NameToInt64OrZero { static constexpr auto name = "toInt64OrZero"; };
+struct NameToInt128OrZero { static constexpr auto name = "toInt128OrZero"; };
+struct NameToInt256OrZero { static constexpr auto name = "toInt256OrZero"; };
+struct NameToFloat32OrZero { static constexpr auto name = "toFloat32OrZero"; };
+struct NameToFloat64OrZero { static constexpr auto name = "toFloat64OrZero"; };
+struct NameToDateOrZero { static constexpr auto name = "toDateOrZero"; };
+struct NameToDate32OrZero { static constexpr auto name = "toDate32OrZero"; };
+struct NameToDateTimeOrZero { static constexpr auto name = "toDateTimeOrZero"; };
+struct NameToDateTime64OrZero { static constexpr auto name = "toDateTime64OrZero"; };
+struct NameToDecimal32OrZero { static constexpr auto name = "toDecimal32OrZero"; };
+struct NameToDecimal64OrZero { static constexpr auto name = "toDecimal64OrZero"; };
+struct NameToDecimal128OrZero { static constexpr auto name = "toDecimal128OrZero"; };
+struct NameToDecimal256OrZero { static constexpr auto name = "toDecimal256OrZero"; };
+struct NameToUUIDOrZero { static constexpr auto name = "toUUIDOrZero"; };
+struct NameToIPv4OrZero { static constexpr auto name = "toIPv4OrZero"; };
+struct NameToIPv6OrZero { static constexpr auto name = "toIPv6OrZero"; };
+
+using FunctionToUInt8OrZero = FunctionConvertFromString<DataTypeUInt8, NameToUInt8OrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToUInt16OrZero = FunctionConvertFromString<DataTypeUInt16, NameToUInt16OrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToUInt32OrZero = FunctionConvertFromString<DataTypeUInt32, NameToUInt32OrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToUInt64OrZero = FunctionConvertFromString<DataTypeUInt64, NameToUInt64OrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToUInt128OrZero = FunctionConvertFromString<DataTypeUInt128, NameToUInt128OrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToUInt256OrZero = FunctionConvertFromString<DataTypeUInt256, NameToUInt256OrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToInt8OrZero = FunctionConvertFromString<DataTypeInt8, NameToInt8OrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToInt16OrZero = FunctionConvertFromString<DataTypeInt16, NameToInt16OrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToInt32OrZero = FunctionConvertFromString<DataTypeInt32, NameToInt32OrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToInt64OrZero = FunctionConvertFromString<DataTypeInt64, NameToInt64OrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToInt128OrZero = FunctionConvertFromString<DataTypeInt128, NameToInt128OrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToInt256OrZero = FunctionConvertFromString<DataTypeInt256, NameToInt256OrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToFloat32OrZero = FunctionConvertFromString<DataTypeFloat32, NameToFloat32OrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToFloat64OrZero = FunctionConvertFromString<DataTypeFloat64, NameToFloat64OrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToDateOrZero = FunctionConvertFromString<DataTypeDate, NameToDateOrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToDate32OrZero = FunctionConvertFromString<DataTypeDate32, NameToDate32OrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToDateTimeOrZero = FunctionConvertFromString<DataTypeDateTime, NameToDateTimeOrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToDateTime64OrZero = FunctionConvertFromString<DataTypeDateTime64, NameToDateTime64OrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToDecimal32OrZero = FunctionConvertFromString<DataTypeDecimal<Decimal32>, NameToDecimal32OrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToDecimal64OrZero = FunctionConvertFromString<DataTypeDecimal<Decimal64>, NameToDecimal64OrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToDecimal128OrZero = FunctionConvertFromString<DataTypeDecimal<Decimal128>, NameToDecimal128OrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToDecimal256OrZero = FunctionConvertFromString<DataTypeDecimal<Decimal256>, NameToDecimal256OrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToUUIDOrZero = FunctionConvertFromString<DataTypeUUID, NameToUUIDOrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToIPv4OrZero = FunctionConvertFromString<DataTypeIPv4, NameToIPv4OrZero, ConvertFromStringExceptionMode::Zero>;
+using FunctionToIPv6OrZero = FunctionConvertFromString<DataTypeIPv6, NameToIPv6OrZero, ConvertFromStringExceptionMode::Zero>;
+
+struct NameToUInt8OrNull { static constexpr auto name = "toUInt8OrNull"; };
+struct NameToUInt16OrNull { static constexpr auto name = "toUInt16OrNull"; };
+struct NameToUInt32OrNull { static constexpr auto name = "toUInt32OrNull"; };
+struct NameToUInt64OrNull { static constexpr auto name = "toUInt64OrNull"; };
+struct NameToUInt128OrNull { static constexpr auto name = "toUInt128OrNull"; };
+struct NameToUInt256OrNull { static constexpr auto name = "toUInt256OrNull"; };
+struct NameToInt8OrNull { static constexpr auto name = "toInt8OrNull"; };
+struct NameToInt16OrNull { static constexpr auto name = "toInt16OrNull"; };
+struct NameToInt32OrNull { static constexpr auto name = "toInt32OrNull"; };
+struct NameToInt64OrNull { static constexpr auto name = "toInt64OrNull"; };
+struct NameToInt128OrNull { static constexpr auto name = "toInt128OrNull"; };
+struct NameToInt256OrNull { static constexpr auto name = "toInt256OrNull"; };
+struct NameToFloat32OrNull { static constexpr auto name = "toFloat32OrNull"; };
+struct NameToFloat64OrNull { static constexpr auto name = "toFloat64OrNull"; };
+struct NameToDateOrNull { static constexpr auto name = "toDateOrNull"; };
+struct NameToDate32OrNull { static constexpr auto name = "toDate32OrNull"; };
+struct NameToDateTimeOrNull { static constexpr auto name = "toDateTimeOrNull"; };
+struct NameToDateTime64OrNull { static constexpr auto name = "toDateTime64OrNull"; };
+struct NameToDecimal32OrNull { static constexpr auto name = "toDecimal32OrNull"; };
+struct NameToDecimal64OrNull { static constexpr auto name = "toDecimal64OrNull"; };
+struct NameToDecimal128OrNull { static constexpr auto name = "toDecimal128OrNull"; };
+struct NameToDecimal256OrNull { static constexpr auto name = "toDecimal256OrNull"; };
+struct NameToUUIDOrNull { static constexpr auto name = "toUUIDOrNull"; };
+struct NameToIPv4OrNull { static constexpr auto name = "toIPv4OrNull"; };
+struct NameToIPv6OrNull { static constexpr auto name = "toIPv6OrNull"; };
+
+using FunctionToUInt8OrNull = FunctionConvertFromString<DataTypeUInt8, NameToUInt8OrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToUInt16OrNull = FunctionConvertFromString<DataTypeUInt16, NameToUInt16OrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToUInt32OrNull = FunctionConvertFromString<DataTypeUInt32, NameToUInt32OrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToUInt64OrNull = FunctionConvertFromString<DataTypeUInt64, NameToUInt64OrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToUInt128OrNull = FunctionConvertFromString<DataTypeUInt128, NameToUInt128OrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToUInt256OrNull = FunctionConvertFromString<DataTypeUInt256, NameToUInt256OrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToInt8OrNull = FunctionConvertFromString<DataTypeInt8, NameToInt8OrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToInt16OrNull = FunctionConvertFromString<DataTypeInt16, NameToInt16OrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToInt32OrNull = FunctionConvertFromString<DataTypeInt32, NameToInt32OrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToInt64OrNull = FunctionConvertFromString<DataTypeInt64, NameToInt64OrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToInt128OrNull = FunctionConvertFromString<DataTypeInt128, NameToInt128OrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToInt256OrNull = FunctionConvertFromString<DataTypeInt256, NameToInt256OrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToFloat32OrNull = FunctionConvertFromString<DataTypeFloat32, NameToFloat32OrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToFloat64OrNull = FunctionConvertFromString<DataTypeFloat64, NameToFloat64OrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToDateOrNull = FunctionConvertFromString<DataTypeDate, NameToDateOrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToDate32OrNull = FunctionConvertFromString<DataTypeDate32, NameToDate32OrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToDateTimeOrNull = FunctionConvertFromString<DataTypeDateTime, NameToDateTimeOrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToDateTime64OrNull = FunctionConvertFromString<DataTypeDateTime64, NameToDateTime64OrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToDecimal32OrNull = FunctionConvertFromString<DataTypeDecimal<Decimal32>, NameToDecimal32OrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToDecimal64OrNull = FunctionConvertFromString<DataTypeDecimal<Decimal64>, NameToDecimal64OrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToDecimal128OrNull = FunctionConvertFromString<DataTypeDecimal<Decimal128>, NameToDecimal128OrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToDecimal256OrNull = FunctionConvertFromString<DataTypeDecimal<Decimal256>, NameToDecimal256OrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToUUIDOrNull = FunctionConvertFromString<DataTypeUUID, NameToUUIDOrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToIPv4OrNull = FunctionConvertFromString<DataTypeIPv4, NameToIPv4OrNull, ConvertFromStringExceptionMode::Null>;
+using FunctionToIPv6OrNull = FunctionConvertFromString<DataTypeIPv6, NameToIPv6OrNull, ConvertFromStringExceptionMode::Null>;
+
+struct NameParseDateTimeBestEffort { static constexpr auto name = "parseDateTimeBestEffort"; };
+struct NameParseDateTimeBestEffortOrZero { static constexpr auto name = "parseDateTimeBestEffortOrZero"; };
+struct NameParseDateTimeBestEffortOrNull { static constexpr auto name = "parseDateTimeBestEffortOrNull"; };
+struct NameParseDateTimeBestEffortUS { static constexpr auto name = "parseDateTimeBestEffortUS"; };
+struct NameParseDateTimeBestEffortUSOrZero { static constexpr auto name = "parseDateTimeBestEffortUSOrZero"; };
+struct NameParseDateTimeBestEffortUSOrNull { static constexpr auto name = "parseDateTimeBestEffortUSOrNull"; };
+struct NameParseDateTime32BestEffort { static constexpr auto name = "parseDateTime32BestEffort"; };
+struct NameParseDateTime32BestEffortOrZero { static constexpr auto name = "parseDateTime32BestEffortOrZero"; };
+struct NameParseDateTime32BestEffortOrNull { static constexpr auto name = "parseDateTime32BestEffortOrNull"; };
+struct NameParseDateTime64BestEffort { static constexpr auto name = "parseDateTime64BestEffort"; };
+struct NameParseDateTime64BestEffortOrZero { static constexpr auto name = "parseDateTime64BestEffortOrZero"; };
+struct NameParseDateTime64BestEffortOrNull { static constexpr auto name = "parseDateTime64BestEffortOrNull"; };
+struct NameParseDateTime64BestEffortUS { static constexpr auto name = "parseDateTime64BestEffortUS"; };
+struct NameParseDateTime64BestEffortUSOrZero { static constexpr auto name = "parseDateTime64BestEffortUSOrZero"; };
+struct NameParseDateTime64BestEffortUSOrNull { static constexpr auto name = "parseDateTime64BestEffortUSOrNull"; };
+
+
+using FunctionParseDateTimeBestEffort = FunctionConvertFromString<
+    DataTypeDateTime, NameParseDateTimeBestEffort, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::BestEffort>;
+using FunctionParseDateTimeBestEffortOrZero = FunctionConvertFromString<
+    DataTypeDateTime, NameParseDateTimeBestEffortOrZero, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::BestEffort>;
+using FunctionParseDateTimeBestEffortOrNull = FunctionConvertFromString<
+    DataTypeDateTime, NameParseDateTimeBestEffortOrNull, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::BestEffort>;
+
+using FunctionParseDateTimeBestEffortUS = FunctionConvertFromString<
+    DataTypeDateTime, NameParseDateTimeBestEffortUS, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::BestEffortUS>;
+using FunctionParseDateTimeBestEffortUSOrZero = FunctionConvertFromString<
+    DataTypeDateTime, NameParseDateTimeBestEffortUSOrZero, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::BestEffortUS>;
+using FunctionParseDateTimeBestEffortUSOrNull = FunctionConvertFromString<
+    DataTypeDateTime, NameParseDateTimeBestEffortUSOrNull, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::BestEffortUS>;
+
+using FunctionParseDateTime32BestEffort = FunctionConvertFromString<
+    DataTypeDateTime, NameParseDateTime32BestEffort, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::BestEffort>;
+using FunctionParseDateTime32BestEffortOrZero = FunctionConvertFromString<
+    DataTypeDateTime, NameParseDateTime32BestEffortOrZero, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::BestEffort>;
+using FunctionParseDateTime32BestEffortOrNull = FunctionConvertFromString<
+    DataTypeDateTime, NameParseDateTime32BestEffortOrNull, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::BestEffort>;
+
+using FunctionParseDateTime64BestEffort = FunctionConvertFromString<
+    DataTypeDateTime64, NameParseDateTime64BestEffort, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::BestEffort>;
+using FunctionParseDateTime64BestEffortOrZero = FunctionConvertFromString<
+    DataTypeDateTime64, NameParseDateTime64BestEffortOrZero, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::BestEffort>;
+using FunctionParseDateTime64BestEffortOrNull = FunctionConvertFromString<
+    DataTypeDateTime64, NameParseDateTime64BestEffortOrNull, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::BestEffort>;
+
+using FunctionParseDateTime64BestEffortUS = FunctionConvertFromString<
+    DataTypeDateTime64, NameParseDateTime64BestEffortUS, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::BestEffortUS>;
+using FunctionParseDateTime64BestEffortUSOrZero = FunctionConvertFromString<
+    DataTypeDateTime64, NameParseDateTime64BestEffortUSOrZero, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::BestEffortUS>;
+using FunctionParseDateTime64BestEffortUSOrNull = FunctionConvertFromString<
+    DataTypeDateTime64, NameParseDateTime64BestEffortUSOrNull, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::BestEffortUS>;
+
+
+class ExecutableFunctionCast : public IExecutableFunction
+{
+public:
+    using WrapperType = std::function<ColumnPtr(ColumnsWithTypeAndName &, const DataTypePtr &, const ColumnNullable *, size_t)>;
+
+    explicit ExecutableFunctionCast(
+            WrapperType && wrapper_function_, const char * name_, std::optional<CastDiagnostic> diagnostic_)
+            : wrapper_function(std::move(wrapper_function_)), name(name_), diagnostic(std::move(diagnostic_)) {}
+
+    String getName() const override { return name; }
+
+protected:
+    ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const override
+    {
+        /// drop second argument, pass others
+        ColumnsWithTypeAndName new_arguments{arguments.front()};
+        if (arguments.size() > 2)
+            new_arguments.insert(std::end(new_arguments), std::next(std::begin(arguments), 2), std::end(arguments));
+
+        try
+        {
+            return wrapper_function(new_arguments, result_type, nullptr, input_rows_count);
+        }
+        catch (Exception & e)
+        {
+            if (diagnostic)
+                e.addMessage("while converting source column " + backQuoteIfNeed(diagnostic->column_from) +
+                             " to destination column " + backQuoteIfNeed(diagnostic->column_to));
+            throw;
+        }
+    }
+
+    bool useDefaultImplementationForNulls() const override { return false; }
+    /// CAST(Nothing, T) -> T
+    bool useDefaultImplementationForNothing() const override { return false; }
+    bool useDefaultImplementationForConstants() const override { return true; }
+    bool useDefaultImplementationForLowCardinalityColumns() const override { return false; }
+    ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {1}; }
+
+private:
+    WrapperType wrapper_function;
+    const char * name;
+    std::optional<CastDiagnostic> diagnostic;
+};
+
+
+struct FunctionCastName
+{
+    static constexpr auto name = "CAST";
+};
+
+class FunctionCastBase : public IFunctionBase
+{
+public:
+    using MonotonicityForRange = std::function<Monotonicity(const IDataType &, const Field &, const Field &)>;
+};
+
+class FunctionCast final : public FunctionCastBase
+{
+public:
+    using WrapperType = std::function<ColumnPtr(ColumnsWithTypeAndName &, const DataTypePtr &, const ColumnNullable *, size_t)>;
+
+    FunctionCast(ContextPtr context_
+            , const char * cast_name_
+            , MonotonicityForRange && monotonicity_for_range_
+            , const DataTypes & argument_types_
+            , const DataTypePtr & return_type_
+            , std::optional<CastDiagnostic> diagnostic_
+            , CastType cast_type_)
+        : cast_name(cast_name_), monotonicity_for_range(std::move(monotonicity_for_range_))
+        , argument_types(argument_types_), return_type(return_type_), diagnostic(std::move(diagnostic_))
+        , cast_type(cast_type_)
+        , context(context_)
+    {
+    }
+
+    const DataTypes & getArgumentTypes() const override { return argument_types; }
+    const DataTypePtr & getResultType() const override { return return_type; }
+
+    ExecutableFunctionPtr prepare(const ColumnsWithTypeAndName & /*sample_columns*/) const override
+    {
+        try
+        {
+            return std::make_unique<ExecutableFunctionCast>(
+                prepareUnpackDictionaries(getArgumentTypes()[0], getResultType()), cast_name, diagnostic);
+        }
+        catch (Exception & e)
+        {
+            if (diagnostic)
+                e.addMessage("while converting source column " + backQuoteIfNeed(diagnostic->column_from) +
+                             " to destination column " + backQuoteIfNeed(diagnostic->column_to));
+            throw;
+        }
+    }
+
+    String getName() const override { return cast_name; }
+
+    bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return true; }
+
+    bool hasInformationAboutMonotonicity() const override
+    {
+        return static_cast<bool>(monotonicity_for_range);
+    }
+
+    Monotonicity getMonotonicityForRange(const IDataType & type, const Field & left, const Field & right) const override
+    {
+        return monotonicity_for_range(type, left, right);
+    }
+
+private:
+
+    const char * cast_name;
+    MonotonicityForRange monotonicity_for_range;
+
+    DataTypes argument_types;
+    DataTypePtr return_type;
+
+    std::optional<CastDiagnostic> diagnostic;
+    CastType cast_type;
+    ContextPtr context;
+
+    static WrapperType createFunctionAdaptor(FunctionPtr function, const DataTypePtr & from_type)
+    {
+        auto function_adaptor = std::make_unique<FunctionToOverloadResolverAdaptor>(function)->build({ColumnWithTypeAndName{nullptr, from_type, ""}});
+
+        return [function_adaptor]
+            (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count)
+        {
+            return function_adaptor->execute(arguments, result_type, input_rows_count);
+        };
+    }
+
+    static WrapperType createToNullableColumnWrapper()
+    {
+        return [] (ColumnsWithTypeAndName &, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count)
+        {
+            ColumnPtr res = result_type->createColumn();
+            ColumnUInt8::Ptr col_null_map_to = ColumnUInt8::create(input_rows_count, true);
+            return ColumnNullable::create(res->cloneResized(input_rows_count), std::move(col_null_map_to));
+        };
+    }
+
+    template <typename ToDataType>
+    WrapperType createWrapper(const DataTypePtr & from_type, const ToDataType * const to_type, bool requested_result_is_nullable) const
+    {
+        TypeIndex from_type_index = from_type->getTypeId();
+        WhichDataType which(from_type_index);
+        bool can_apply_accurate_cast = (cast_type == CastType::accurate || cast_type == CastType::accurateOrNull)
+            && (which.isInt() || which.isUInt() || which.isFloat());
+
+        FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior = default_date_time_overflow_behavior;
+        if (context)
+            date_time_overflow_behavior = context->getSettingsRef().date_time_overflow_behavior;
+
+        if (requested_result_is_nullable && checkAndGetDataType<DataTypeString>(from_type.get()))
+        {
+            /// In case when converting to Nullable type, we apply different parsing rule,
+            /// that will not throw an exception but return NULL in case of malformed input.
+            FunctionPtr function = FunctionConvertFromString<ToDataType, FunctionCastName, ConvertFromStringExceptionMode::Null>::create();
+            return createFunctionAdaptor(function, from_type);
+        }
+        else if (!can_apply_accurate_cast)
+        {
+            FunctionPtr function = FunctionTo<ToDataType>::Type::create(context);
+            return createFunctionAdaptor(function, from_type);
+        }
+
+        auto wrapper_cast_type = cast_type;
+
+        return [wrapper_cast_type, from_type_index, to_type, date_time_overflow_behavior]
+            (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *column_nullable, size_t input_rows_count)
+        {
+            ColumnPtr result_column;
+            auto res = callOnIndexAndDataType<ToDataType>(from_type_index, [&](const auto & types) -> bool {
+                using Types = std::decay_t<decltype(types)>;
+                using LeftDataType = typename Types::LeftType;
+                using RightDataType = typename Types::RightType;
+
+                if constexpr (IsDataTypeNumber<LeftDataType>)
+                {
+                    if constexpr (IsDataTypeNumber<RightDataType>)
+                    {
+#define GENERATE_OVERFLOW_MODE_CASE(OVERFLOW_MODE, ADDITIONS) \
+            case FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE: \
+                result_column = ConvertImpl<LeftDataType, RightDataType, FunctionCastName, ConvertDefaultBehaviorTag, FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE>::execute( \
+                arguments, result_type, input_rows_count, ADDITIONS()); \
+                break;
+                        if (wrapper_cast_type == CastType::accurate)
+                        {
+                            switch (date_time_overflow_behavior)
+                            {
+                                GENERATE_OVERFLOW_MODE_CASE(Throw, AccurateConvertStrategyAdditions)
+                                GENERATE_OVERFLOW_MODE_CASE(Ignore, AccurateConvertStrategyAdditions)
+                                GENERATE_OVERFLOW_MODE_CASE(Saturate, AccurateConvertStrategyAdditions)
+                            }
+                        }
+                        else
+                        {
+                            switch (date_time_overflow_behavior)
+                            {
+                                GENERATE_OVERFLOW_MODE_CASE(Throw, AccurateOrNullConvertStrategyAdditions)
+                                GENERATE_OVERFLOW_MODE_CASE(Ignore, AccurateOrNullConvertStrategyAdditions)
+                                GENERATE_OVERFLOW_MODE_CASE(Saturate, AccurateOrNullConvertStrategyAdditions)
+                            }
+                        }
+#undef GENERATE_OVERFLOW_MODE_CASE
+
+                        return true;
+                    }
+
+                    if constexpr (std::is_same_v<RightDataType, DataTypeDate> || std::is_same_v<RightDataType, DataTypeDateTime>)
+                    {
+#define GENERATE_OVERFLOW_MODE_CASE(OVERFLOW_MODE, ADDITIONS) \
+            case FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE: \
+            result_column = ConvertImpl<LeftDataType, RightDataType, FunctionCastName, ConvertDefaultBehaviorTag, FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE>::template execute<ADDITIONS>( \
+arguments, result_type, input_rows_count); \
+                break;
+                        if (wrapper_cast_type == CastType::accurate)
+                        {
+                            switch (date_time_overflow_behavior)
+                            {
+                                GENERATE_OVERFLOW_MODE_CASE(Throw, DateTimeAccurateConvertStrategyAdditions)
+                                GENERATE_OVERFLOW_MODE_CASE(Ignore, DateTimeAccurateConvertStrategyAdditions)
+                                GENERATE_OVERFLOW_MODE_CASE(Saturate, DateTimeAccurateConvertStrategyAdditions)
+                            }
+                        }
+                        else
+                        {
+                            switch (date_time_overflow_behavior)
+                            {
+                                GENERATE_OVERFLOW_MODE_CASE(Throw, DateTimeAccurateOrNullConvertStrategyAdditions)
+                                GENERATE_OVERFLOW_MODE_CASE(Ignore, DateTimeAccurateOrNullConvertStrategyAdditions)
+                                GENERATE_OVERFLOW_MODE_CASE(Saturate, DateTimeAccurateOrNullConvertStrategyAdditions)
+                            }
+                        }
+#undef GENERATE_OVERFLOW_MODE_CASE
+                        return true;
+                    }
+                }
+
+                return false;
+            });
+
+            /// Additionally check if callOnIndexAndDataType wasn't called at all.
+            if (!res)
+            {
+                if (wrapper_cast_type == CastType::accurateOrNull)
+                {
+                    auto nullable_column_wrapper = FunctionCast::createToNullableColumnWrapper();
+                    return nullable_column_wrapper(arguments, result_type, column_nullable, input_rows_count);
+                }
+                else
+                {
+                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE,
+                        "Conversion from {} to {} is not supported",
+                        from_type_index, to_type->getName());
+                }
+            }
+
+            return result_column;
+        };
+    }
+
+    template <typename ToDataType>
+    WrapperType createBoolWrapper(const DataTypePtr & from_type, const ToDataType * const to_type, bool requested_result_is_nullable) const
+    {
+        if (checkAndGetDataType<DataTypeString>(from_type.get()))
+        {
+            return &ConvertImplGenericFromString::execute;
+        }
+
+        return createWrapper<ToDataType>(from_type, to_type, requested_result_is_nullable);
+    }
+
+    WrapperType createUInt8ToBoolWrapper(const DataTypePtr from_type, const DataTypePtr to_type) const
+    {
+        return [from_type, to_type] (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable *, size_t /*input_rows_count*/) -> ColumnPtr
+        {
+            /// Special case when we convert UInt8 column to Bool column.
+            /// both columns have type UInt8, but we shouldn't use identity wrapper,
+            /// because Bool column can contain only 0 and 1.
+            auto res_column = to_type->createColumn();
+            const auto & data_from = checkAndGetColumn<ColumnUInt8>(arguments[0].column.get())->getData();
+            auto & data_to = assert_cast<ColumnUInt8 *>(res_column.get())->getData();
+            data_to.resize(data_from.size());
+            for (size_t i = 0; i != data_from.size(); ++i)
+                data_to[i] = static_cast<bool>(data_from[i]);
+            return res_column;
+        };
+    }
+
+    static WrapperType createStringWrapper(const DataTypePtr & from_type)
+    {
+        FunctionPtr function = FunctionToString::create();
+        return createFunctionAdaptor(function, from_type);
+    }
+
+    WrapperType createFixedStringWrapper(const DataTypePtr & from_type, const size_t N) const
+    {
+        if (!isStringOrFixedString(from_type))
+            throw Exception(ErrorCodes::NOT_IMPLEMENTED, "CAST AS FixedString is only implemented for types String and FixedString");
+
+        bool exception_mode_null = cast_type == CastType::accurateOrNull;
+        return [exception_mode_null, N] (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable *, size_t /*input_rows_count*/)
+        {
+            if (exception_mode_null)
+                return FunctionToFixedString::executeForN<ConvertToFixedStringExceptionMode::Null>(arguments, N);
+            else
+                return FunctionToFixedString::executeForN<ConvertToFixedStringExceptionMode::Throw>(arguments, N);
+        };
+    }
+
+#define GENERATE_INTERVAL_CASE(INTERVAL_KIND) \
+            case IntervalKind::Kind::INTERVAL_KIND: \
+                return createFunctionAdaptor(FunctionConvert<DataTypeInterval, NameToInterval##INTERVAL_KIND, PositiveMonotonicity>::create(), from_type);
+
+    static WrapperType createIntervalWrapper(const DataTypePtr & from_type, IntervalKind kind)
+    {
+        switch (kind)
+        {
+            GENERATE_INTERVAL_CASE(Nanosecond)
+            GENERATE_INTERVAL_CASE(Microsecond)
+            GENERATE_INTERVAL_CASE(Millisecond)
+            GENERATE_INTERVAL_CASE(Second)
+            GENERATE_INTERVAL_CASE(Minute)
+            GENERATE_INTERVAL_CASE(Hour)
+            GENERATE_INTERVAL_CASE(Day)
+            GENERATE_INTERVAL_CASE(Week)
+            GENERATE_INTERVAL_CASE(Month)
+            GENERATE_INTERVAL_CASE(Quarter)
+            GENERATE_INTERVAL_CASE(Year)
+        }
+        throw Exception{ErrorCodes::CANNOT_CONVERT_TYPE, "Conversion to unexpected IntervalKind: {}", kind.toString()};
+    }
+
+#undef GENERATE_INTERVAL_CASE
+
+    template <typename ToDataType>
+    requires IsDataTypeDecimal<ToDataType>
+    WrapperType createDecimalWrapper(const DataTypePtr & from_type, const ToDataType * to_type, bool requested_result_is_nullable) const
+    {
+        TypeIndex type_index = from_type->getTypeId();
+        UInt32 scale = to_type->getScale();
+
+        WhichDataType which(type_index);
+        bool ok = which.isNativeInt() || which.isNativeUInt() || which.isDecimal() || which.isFloat() || which.isDateOrDate32() || which.isDateTime() || which.isDateTime64()
+            || which.isStringOrFixedString();
+        if (!ok)
+        {
+            if (cast_type == CastType::accurateOrNull)
+                return createToNullableColumnWrapper();
+            else
+                throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Conversion from {} to {} is not supported",
+                    from_type->getName(), to_type->getName());
+        }
+
+        auto wrapper_cast_type = cast_type;
+
+        return [wrapper_cast_type, type_index, scale, to_type, requested_result_is_nullable]
+            (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *column_nullable, size_t input_rows_count)
+        {
+            ColumnPtr result_column;
+            auto res = callOnIndexAndDataType<ToDataType>(type_index, [&](const auto & types) -> bool
+            {
+                using Types = std::decay_t<decltype(types)>;
+                using LeftDataType = typename Types::LeftType;
+                using RightDataType = typename Types::RightType;
+
+                if constexpr (IsDataTypeDecimalOrNumber<LeftDataType> && IsDataTypeDecimalOrNumber<RightDataType> && !std::is_same_v<DataTypeDateTime64, RightDataType>)
+                {
+                    if (wrapper_cast_type == CastType::accurate)
+                    {
+                        AccurateConvertStrategyAdditions additions;
+                        additions.scale = scale;
+                        result_column = ConvertImpl<LeftDataType, RightDataType, FunctionCastName>::execute(
+                            arguments, result_type, input_rows_count, additions);
+
+                        return true;
+                    }
+                    else if (wrapper_cast_type == CastType::accurateOrNull)
+                    {
+                        AccurateOrNullConvertStrategyAdditions additions;
+                        additions.scale = scale;
+                        result_column = ConvertImpl<LeftDataType, RightDataType, FunctionCastName>::execute(
+                            arguments, result_type, input_rows_count, additions);
+
+                        return true;
+                    }
+                }
+                else if constexpr (std::is_same_v<LeftDataType, DataTypeString>)
+                {
+                    if (requested_result_is_nullable)
+                    {
+                        /// Consistent with CAST(Nullable(String) AS Nullable(Numbers))
+                        /// In case when converting to Nullable type, we apply different parsing rule,
+                        /// that will not throw an exception but return NULL in case of malformed input.
+                        result_column = ConvertImpl<LeftDataType, RightDataType, FunctionCastName, ConvertReturnNullOnErrorTag>::execute(
+                            arguments, result_type, input_rows_count, scale);
+
+                        return true;
+                    }
+                }
+
+                result_column = ConvertImpl<LeftDataType, RightDataType, FunctionCastName>::execute(arguments, result_type, input_rows_count, scale);
+
+                return true;
+            });
+
+            /// Additionally check if callOnIndexAndDataType wasn't called at all.
+            if (!res)
+            {
+                if (wrapper_cast_type == CastType::accurateOrNull)
+                {
+                    auto nullable_column_wrapper = FunctionCast::createToNullableColumnWrapper();
+                    return nullable_column_wrapper(arguments, result_type, column_nullable, input_rows_count);
+                }
+                else
+                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE,
+                        "Conversion from {} to {} is not supported",
+                        type_index, to_type->getName());
+            }
+
+            return result_column;
+        };
+    }
+
+    WrapperType createAggregateFunctionWrapper(const DataTypePtr & from_type_untyped, const DataTypeAggregateFunction * to_type) const
+    {
+        /// Conversion from String through parsing.
+        if (checkAndGetDataType<DataTypeString>(from_type_untyped.get()))
+        {
+            return &ConvertImplGenericFromString::execute;
+        }
+        else if (const auto * agg_type = checkAndGetDataType<DataTypeAggregateFunction>(from_type_untyped.get()))
+        {
+            if (agg_type->getFunction()->haveSameStateRepresentation(*to_type->getFunction()))
+            {
+                return [function = to_type->getFunction()](
+                           ColumnsWithTypeAndName & arguments,
+                           const DataTypePtr & /* result_type */,
+                           const ColumnNullable * /* nullable_source */,
+                           size_t /*input_rows_count*/) -> ColumnPtr
+                {
+                    const auto & argument_column = arguments.front();
+                    const auto * col_agg = checkAndGetColumn<ColumnAggregateFunction>(argument_column.column.get());
+                    if (col_agg)
+                    {
+                        auto new_col_agg = ColumnAggregateFunction::create(*col_agg);
+                        new_col_agg->set(function);
+                        return new_col_agg;
+                    }
+                    else
+                    {
+                        throw Exception(
+                            ErrorCodes::LOGICAL_ERROR,
+                            "Illegal column {} for function CAST AS AggregateFunction",
+                            argument_column.column->getName());
+                    }
+                };
+            }
+        }
+
+        if (cast_type == CastType::accurateOrNull)
+            return createToNullableColumnWrapper();
+        else
+            throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Conversion from {} to {} is not supported",
+                from_type_untyped->getName(), to_type->getName());
+    }
+
+    WrapperType createArrayWrapper(const DataTypePtr & from_type_untyped, const DataTypeArray & to_type) const
+    {
+        /// Conversion from String through parsing.
+        if (checkAndGetDataType<DataTypeString>(from_type_untyped.get()))
+        {
+            return &ConvertImplGenericFromString::execute;
+        }
+
+        DataTypePtr from_type_holder;
+        const auto * from_type = checkAndGetDataType<DataTypeArray>(from_type_untyped.get());
+        const auto * from_type_map = checkAndGetDataType<DataTypeMap>(from_type_untyped.get());
+
+        /// Convert from Map
+        if (from_type_map)
+        {
+            /// Recreate array of unnamed tuples because otherwise it may work
+            /// unexpectedly while converting to array of named tuples.
+            from_type_holder = from_type_map->getNestedTypeWithUnnamedTuple();
+            from_type = assert_cast<const DataTypeArray *>(from_type_holder.get());
+        }
+
+        if (!from_type)
+        {
+            throw Exception(ErrorCodes::TYPE_MISMATCH,
+                "CAST AS Array can only be performed between same-dimensional Array, Map or String types");
+        }
+
+        DataTypePtr from_nested_type = from_type->getNestedType();
+
+        /// In query SELECT CAST([] AS Array(Array(String))) from type is Array(Nothing)
+        bool from_empty_array = isNothing(from_nested_type);
+
+        if (from_type->getNumberOfDimensions() != to_type.getNumberOfDimensions() && !from_empty_array)
+            throw Exception(ErrorCodes::TYPE_MISMATCH,
+                "CAST AS Array can only be performed between same-dimensional array types");
+
+        const DataTypePtr & to_nested_type = to_type.getNestedType();
+
+        /// Prepare nested type conversion
+        const auto nested_function = prepareUnpackDictionaries(from_nested_type, to_nested_type);
+
+        return [nested_function, from_nested_type, to_nested_type](
+                ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * nullable_source, size_t /*input_rows_count*/) -> ColumnPtr
+        {
+            const auto & argument_column = arguments.front();
+
+            const ColumnArray * col_array = nullptr;
+
+            if (const ColumnMap * col_map = checkAndGetColumn<ColumnMap>(argument_column.column.get()))
+                col_array = &col_map->getNestedColumn();
+            else
+                col_array = checkAndGetColumn<ColumnArray>(argument_column.column.get());
+
+            if (col_array)
+            {
+                /// create columns for converting nested column containing original and result columns
+                ColumnsWithTypeAndName nested_columns{{ col_array->getDataPtr(), from_nested_type, "" }};
+
+                /// convert nested column
+                auto result_column = nested_function(nested_columns, to_nested_type, nullable_source, nested_columns.front().column->size());
+
+                /// set converted nested column to result
+                return ColumnArray::create(result_column, col_array->getOffsetsPtr());
+            }
+            else
+            {
+                throw Exception(ErrorCodes::LOGICAL_ERROR,
+                    "Illegal column {} for function CAST AS Array",
+                    argument_column.column->getName());
+            }
+        };
+    }
+
+    using ElementWrappers = std::vector<WrapperType>;
+
+    ElementWrappers getElementWrappers(const DataTypes & from_element_types, const DataTypes & to_element_types) const
+    {
+        ElementWrappers element_wrappers;
+        element_wrappers.reserve(from_element_types.size());
+
+        /// Create conversion wrapper for each element in tuple
+        for (size_t i = 0; i < from_element_types.size(); ++i)
+        {
+            const DataTypePtr & from_element_type = from_element_types[i];
+            const DataTypePtr & to_element_type = to_element_types[i];
+            element_wrappers.push_back(prepareUnpackDictionaries(from_element_type, to_element_type));
+        }
+
+        return element_wrappers;
+    }
+
+    WrapperType createTupleWrapper(const DataTypePtr & from_type_untyped, const DataTypeTuple * to_type) const
+    {
+        /// Conversion from String through parsing.
+        if (checkAndGetDataType<DataTypeString>(from_type_untyped.get()))
+        {
+            return &ConvertImplGenericFromString::execute;
+        }
+
+        const auto * from_type = checkAndGetDataType<DataTypeTuple>(from_type_untyped.get());
+        if (!from_type)
+            throw Exception(ErrorCodes::TYPE_MISMATCH, "CAST AS Tuple can only be performed between tuple types or from String.\n"
+                            "Left type: {}, right type: {}", from_type_untyped->getName(), to_type->getName());
+
+        const auto & from_element_types = from_type->getElements();
+        const auto & to_element_types = to_type->getElements();
+
+        std::vector<WrapperType> element_wrappers;
+        std::vector<std::optional<size_t>> to_reverse_index;
+
+        /// For named tuples allow conversions for tuples with
+        /// different sets of elements. If element exists in @to_type
+        /// and doesn't exist in @to_type it will be filled by default values.
+        if (from_type->haveExplicitNames() && to_type->haveExplicitNames())
+        {
+            const auto & from_names = from_type->getElementNames();
+            std::unordered_map<String, size_t> from_positions;
+            from_positions.reserve(from_names.size());
+            for (size_t i = 0; i < from_names.size(); ++i)
+                from_positions[from_names[i]] = i;
+
+            const auto & to_names = to_type->getElementNames();
+            element_wrappers.reserve(to_names.size());
+            to_reverse_index.reserve(from_names.size());
+
+            for (size_t i = 0; i < to_names.size(); ++i)
+            {
+                auto it = from_positions.find(to_names[i]);
+                if (it != from_positions.end())
+                {
+                    element_wrappers.emplace_back(prepareUnpackDictionaries(from_element_types[it->second], to_element_types[i]));
+                    to_reverse_index.emplace_back(it->second);
+                }
+                else
+                {
+                    element_wrappers.emplace_back();
+                    to_reverse_index.emplace_back();
+                }
+            }
+        }
+        else
+        {
+            if (from_element_types.size() != to_element_types.size())
+                throw Exception(ErrorCodes::TYPE_MISMATCH, "CAST AS Tuple can only be performed between tuple types "
+                                "with the same number of elements or from String.\nLeft type: {}, right type: {}",
+                                from_type->getName(), to_type->getName());
+
+            element_wrappers = getElementWrappers(from_element_types, to_element_types);
+            to_reverse_index.reserve(to_element_types.size());
+            for (size_t i = 0; i < to_element_types.size(); ++i)
+                to_reverse_index.emplace_back(i);
+        }
+
+        return [element_wrappers, from_element_types, to_element_types, to_reverse_index]
+            (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * nullable_source, size_t input_rows_count) -> ColumnPtr
+        {
+            const auto * col = arguments.front().column.get();
+
+            size_t tuple_size = to_element_types.size();
+            const ColumnTuple & column_tuple = typeid_cast<const ColumnTuple &>(*col);
+
+            Columns converted_columns(tuple_size);
+
+            /// invoke conversion for each element
+            for (size_t i = 0; i < tuple_size; ++i)
+            {
+                if (to_reverse_index[i])
+                {
+                    size_t from_idx = *to_reverse_index[i];
+                    ColumnsWithTypeAndName element = {{column_tuple.getColumns()[from_idx], from_element_types[from_idx], "" }};
+                    converted_columns[i] = element_wrappers[i](element, to_element_types[i], nullable_source, input_rows_count);
+                }
+                else
+                {
+                    converted_columns[i] = to_element_types[i]->createColumn()->cloneResized(input_rows_count);
+                }
+            }
+
+            return ColumnTuple::create(converted_columns);
+        };
+    }
+
+    /// The case of: tuple([key1, key2, ..., key_n], [value1, value2, ..., value_n])
+    WrapperType createTupleToMapWrapper(const DataTypes & from_kv_types, const DataTypes & to_kv_types) const
+    {
+        return [element_wrappers = getElementWrappers(from_kv_types, to_kv_types), from_kv_types, to_kv_types]
+            (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * nullable_source, size_t /*input_rows_count*/) -> ColumnPtr
+        {
+            const auto * col = arguments.front().column.get();
+            const auto & column_tuple = assert_cast<const ColumnTuple &>(*col);
+
+            Columns offsets(2);
+            Columns converted_columns(2);
+            for (size_t i = 0; i < 2; ++i)
+            {
+                const auto & column_array = assert_cast<const ColumnArray &>(column_tuple.getColumn(i));
+                ColumnsWithTypeAndName element = {{column_array.getDataPtr(), from_kv_types[i], ""}};
+                converted_columns[i] = element_wrappers[i](element, to_kv_types[i], nullable_source, (element[0].column)->size());
+                offsets[i] = column_array.getOffsetsPtr();
+            }
+
+            const auto & keys_offsets = assert_cast<const ColumnArray::ColumnOffsets &>(*offsets[0]).getData();
+            const auto & values_offsets = assert_cast<const ColumnArray::ColumnOffsets &>(*offsets[1]).getData();
+            if (keys_offsets != values_offsets)
+                throw Exception(ErrorCodes::TYPE_MISMATCH,
+                    "CAST AS Map can only be performed from tuple of arrays with equal sizes.");
+
+            return ColumnMap::create(converted_columns[0], converted_columns[1], offsets[0]);
+        };
+    }
+
+    WrapperType createMapToMapWrapper(const DataTypes & from_kv_types, const DataTypes & to_kv_types) const
+    {
+        return [element_wrappers = getElementWrappers(from_kv_types, to_kv_types), from_kv_types, to_kv_types]
+            (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * nullable_source, size_t /*input_rows_count*/) -> ColumnPtr
+        {
+            const auto * col = arguments.front().column.get();
+            const auto & column_map = typeid_cast<const ColumnMap &>(*col);
+            const auto & nested_data = column_map.getNestedData();
+
+            Columns converted_columns(2);
+            for (size_t i = 0; i < 2; ++i)
+            {
+                ColumnsWithTypeAndName element = {{nested_data.getColumnPtr(i), from_kv_types[i], ""}};
+                converted_columns[i] = element_wrappers[i](element, to_kv_types[i], nullable_source, (element[0].column)->size());
+            }
+
+            return ColumnMap::create(converted_columns[0], converted_columns[1], column_map.getNestedColumn().getOffsetsPtr());
+        };
+    }
+
+    /// The case of: [(key1, value1), (key2, value2), ...]
+    WrapperType createArrayToMapWrapper(const DataTypes & from_kv_types, const DataTypes & to_kv_types) const
+    {
+        return [element_wrappers = getElementWrappers(from_kv_types, to_kv_types), from_kv_types, to_kv_types]
+            (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * nullable_source, size_t /*input_rows_count*/) -> ColumnPtr
+        {
+            const auto * col = arguments.front().column.get();
+            const auto & column_array = typeid_cast<const ColumnArray &>(*col);
+            const auto & nested_data = typeid_cast<const ColumnTuple &>(column_array.getData());
+
+            Columns converted_columns(2);
+            for (size_t i = 0; i < 2; ++i)
+            {
+                ColumnsWithTypeAndName element = {{nested_data.getColumnPtr(i), from_kv_types[i], ""}};
+                converted_columns[i] = element_wrappers[i](element, to_kv_types[i], nullable_source, (element[0].column)->size());
+            }
+
+            return ColumnMap::create(converted_columns[0], converted_columns[1], column_array.getOffsetsPtr());
+        };
+    }
+
+
+    WrapperType createMapWrapper(const DataTypePtr & from_type_untyped, const DataTypeMap * to_type) const
+    {
+        if (const auto * from_tuple = checkAndGetDataType<DataTypeTuple>(from_type_untyped.get()))
+        {
+            if (from_tuple->getElements().size() != 2)
+                throw Exception(
+                    ErrorCodes::TYPE_MISMATCH,
+                    "CAST AS Map from tuple requires 2 elements. "
+                    "Left type: {}, right type: {}",
+                    from_tuple->getName(),
+                    to_type->getName());
+
+            DataTypes from_kv_types;
+            const auto & to_kv_types = to_type->getKeyValueTypes();
+
+            for (const auto & elem : from_tuple->getElements())
+            {
+                const auto * type_array = checkAndGetDataType<DataTypeArray>(elem.get());
+                if (!type_array)
+                    throw Exception(ErrorCodes::TYPE_MISMATCH,
+                        "CAST AS Map can only be performed from tuples of array. Got: {}", from_tuple->getName());
+
+                from_kv_types.push_back(type_array->getNestedType());
+            }
+
+            return createTupleToMapWrapper(from_kv_types, to_kv_types);
+        }
+        else if (const auto * from_array = typeid_cast<const DataTypeArray *>(from_type_untyped.get()))
+        {
+            const auto * nested_tuple = typeid_cast<const DataTypeTuple *>(from_array->getNestedType().get());
+            if (!nested_tuple || nested_tuple->getElements().size() != 2)
+                throw Exception(
+                    ErrorCodes::TYPE_MISMATCH,
+                    "CAST AS Map from array requires nested tuple of 2 elements. "
+                    "Left type: {}, right type: {}",
+                    from_array->getName(),
+                    to_type->getName());
+
+            return createArrayToMapWrapper(nested_tuple->getElements(), to_type->getKeyValueTypes());
+        }
+        else if (const auto * from_type = checkAndGetDataType<DataTypeMap>(from_type_untyped.get()))
+        {
+            return createMapToMapWrapper(from_type->getKeyValueTypes(), to_type->getKeyValueTypes());
+        }
+        else
+        {
+            throw Exception(ErrorCodes::TYPE_MISMATCH, "Unsupported types to CAST AS Map. "
+                "Left type: {}, right type: {}", from_type_untyped->getName(), to_type->getName());
+        }
+    }
+
+    WrapperType createTupleToObjectWrapper(const DataTypeTuple & from_tuple, bool has_nullable_subcolumns) const
+    {
+        if (!from_tuple.haveExplicitNames())
+            throw Exception(ErrorCodes::TYPE_MISMATCH,
+            "Cast to Object can be performed only from flatten Named Tuple. Got: {}", from_tuple.getName());
+
+        PathsInData paths;
+        DataTypes from_types;
+
+        std::tie(paths, from_types) = flattenTuple(from_tuple.getPtr());
+        auto to_types = from_types;
+
+        for (auto & type : to_types)
+        {
+            if (isTuple(type) || isNested(type))
+                throw Exception(ErrorCodes::TYPE_MISMATCH,
+                    "Cast to Object can be performed only from flatten Named Tuple. Got: {}",
+                    from_tuple.getName());
+
+            type = recursiveRemoveLowCardinality(type);
+        }
+
+        return [element_wrappers = getElementWrappers(from_types, to_types),
+            has_nullable_subcolumns, from_types, to_types, paths]
+            (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * nullable_source, size_t input_rows_count)
+        {
+            size_t tuple_size = to_types.size();
+            auto flattened_column = flattenTuple(arguments.front().column);
+            const auto & column_tuple = assert_cast<const ColumnTuple &>(*flattened_column);
+
+            if (tuple_size != column_tuple.getColumns().size())
+                throw Exception(ErrorCodes::TYPE_MISMATCH,
+                    "Expected tuple with {} subcolumn, but got {} subcolumns",
+                    tuple_size, column_tuple.getColumns().size());
+
+            auto res = ColumnObject::create(has_nullable_subcolumns);
+            for (size_t i = 0; i < tuple_size; ++i)
+            {
+                ColumnsWithTypeAndName element = {{column_tuple.getColumns()[i], from_types[i], "" }};
+                auto converted_column = element_wrappers[i](element, to_types[i], nullable_source, input_rows_count);
+                res->addSubcolumn(paths[i], converted_column->assumeMutable());
+            }
+
+            return res;
+        };
+    }
+
+    WrapperType createMapToObjectWrapper(const DataTypeMap & from_map, bool has_nullable_subcolumns) const
+    {
+        auto key_value_types = from_map.getKeyValueTypes();
+
+        if (!isStringOrFixedString(key_value_types[0]))
+            throw Exception(ErrorCodes::TYPE_MISMATCH,
+                "Cast to Object from Map can be performed only from Map "
+                "with String or FixedString key. Got: {}", from_map.getName());
+
+        const auto & value_type = key_value_types[1];
+        auto to_value_type = value_type;
+
+        if (!has_nullable_subcolumns && value_type->isNullable())
+            to_value_type = removeNullable(value_type);
+
+        if (has_nullable_subcolumns && !value_type->isNullable())
+            to_value_type = makeNullable(value_type);
+
+        DataTypes to_key_value_types{std::make_shared<DataTypeString>(), std::move(to_value_type)};
+        auto element_wrappers = getElementWrappers(key_value_types, to_key_value_types);
+
+        return [has_nullable_subcolumns, element_wrappers, key_value_types, to_key_value_types]
+            (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * nullable_source, size_t) -> ColumnPtr
+        {
+            const auto & column_map = assert_cast<const ColumnMap &>(*arguments.front().column);
+            const auto & offsets = column_map.getNestedColumn().getOffsets();
+            auto key_value_columns = column_map.getNestedData().getColumnsCopy();
+
+            for (size_t i = 0; i < 2; ++i)
+            {
+                ColumnsWithTypeAndName element{{key_value_columns[i], key_value_types[i], ""}};
+                key_value_columns[i] = element_wrappers[i](element, to_key_value_types[i], nullable_source, key_value_columns[i]->size());
+            }
+
+            const auto & key_column_str = assert_cast<const ColumnString &>(*key_value_columns[0]);
+            const auto & value_column = *key_value_columns[1];
+
+            using SubcolumnsMap = HashMap<StringRef, MutableColumnPtr, StringRefHash>;
+            SubcolumnsMap subcolumns;
+
+            for (size_t row = 0; row < offsets.size(); ++row)
+            {
+                for (size_t i = offsets[static_cast<ssize_t>(row) - 1]; i < offsets[row]; ++i)
+                {
+                    auto ref = key_column_str.getDataAt(i);
+
+                    bool inserted;
+                    SubcolumnsMap::LookupResult it;
+                    subcolumns.emplace(ref, it, inserted);
+                    auto & subcolumn = it->getMapped();
+
+                    if (inserted)
+                        subcolumn = value_column.cloneEmpty()->cloneResized(row);
+
+                    /// Map can have duplicated keys. We insert only first one.
+                    if (subcolumn->size() == row)
+                        subcolumn->insertFrom(value_column, i);
+                }
+
+                /// Insert default values for keys missed in current row.
+                for (const auto & [_, subcolumn] : subcolumns)
+                    if (subcolumn->size() == row)
+                        subcolumn->insertDefault();
+            }
+
+            auto column_object = ColumnObject::create(has_nullable_subcolumns);
+            for (auto && [key, subcolumn] : subcolumns)
+            {
+                PathInData path(key.toView());
+                column_object->addSubcolumn(path, std::move(subcolumn));
+            }
+
+            return column_object;
+        };
+    }
+
+    WrapperType createObjectWrapper(const DataTypePtr & from_type, const DataTypeObject * to_type) const
+    {
+        if (const auto * from_tuple = checkAndGetDataType<DataTypeTuple>(from_type.get()))
+        {
+            return createTupleToObjectWrapper(*from_tuple, to_type->hasNullableSubcolumns());
+        }
+        else if (const auto * from_map = checkAndGetDataType<DataTypeMap>(from_type.get()))
+        {
+            return createMapToObjectWrapper(*from_map, to_type->hasNullableSubcolumns());
+        }
+        else if (checkAndGetDataType<DataTypeString>(from_type.get()))
+        {
+            return [] (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable * nullable_source, size_t input_rows_count)
+            {
+                auto res = ConvertImplGenericFromString::execute(arguments, result_type, nullable_source, input_rows_count)->assumeMutable();
+                res->finalize();
+                return res;
+            };
+        }
+        else if (checkAndGetDataType<DataTypeObject>(from_type.get()))
+        {
+            return [is_nullable = to_type->hasNullableSubcolumns()] (ColumnsWithTypeAndName & arguments, const DataTypePtr & , const ColumnNullable * , size_t) -> ColumnPtr
+            {
+                auto & column_object = assert_cast<const ColumnObject &>(*arguments.front().column);
+                auto res = ColumnObject::create(is_nullable);
+                for (size_t i = 0; i < column_object.size(); i++)
+                    res->insert(column_object[i]);
+
+                res->finalize();
+                return res;
+            };
+        }
+
+        throw Exception(ErrorCodes::TYPE_MISMATCH,
+            "Cast to Object can be performed only from flatten named Tuple, Map or String. Got: {}", from_type->getName());
+    }
+
+    WrapperType createVariantToVariantWrapper(const DataTypeVariant & from_variant, const DataTypeVariant & to_variant) const
+    {
+        /// We support only extension of variant type, so, only new types can be added.
+        /// For example: Variant(T1, T2) -> Variant(T1, T2, T3) is supported, but Variant(T1, T2) -> Variant(T1, T3) is not supported.
+        /// We want to extend Variant type for free without rewriting the data, but we sort data types inside Variant during type creation
+        /// (we do it because we want Variant(T1, T2) to be the same as Variant(T2, T1)), but after extension the order of variant types
+        /// (and so their discriminators) can be different. For example: Variant(T1, T3) -> Variant(T1, T2, T3).
+        /// To avoid full rewrite of discriminators column, ColumnVariant supports it's local order of variant columns (and so local
+        /// discriminators) and stores mapping global order -> local order.
+        /// So, to extend Variant with new types for free, we should keep old local order for old variants, append new variants and change
+        /// mapping global order -> local order according to the new global order.
+
+        /// Create map (new variant type) -> (it's global discriminator in new order).
+        const auto & new_variants = to_variant.getVariants();
+        std::unordered_map<String, ColumnVariant::Discriminator> new_variant_types_to_new_global_discriminator;
+        new_variant_types_to_new_global_discriminator.reserve(new_variants.size());
+        for (size_t i = 0; i != new_variants.size(); ++i)
+            new_variant_types_to_new_global_discriminator[new_variants[i]->getName()] = i;
+
+        /// Create set of old variant types.
+        const auto & old_variants = from_variant.getVariants();
+        std::unordered_map<String, ColumnVariant::Discriminator> old_variant_types_to_old_global_discriminator;
+        old_variant_types_to_old_global_discriminator.reserve(old_variants.size());
+        for (size_t i = 0; i != old_variants.size(); ++i)
+            old_variant_types_to_old_global_discriminator[old_variants[i]->getName()] = i;
+
+        /// Check that the set of old variants types is a subset of new variant types and collect new global discriminator for each old global discriminator.
+        std::unordered_map<ColumnVariant::Discriminator, ColumnVariant::Discriminator> old_global_discriminator_to_new;
+        old_global_discriminator_to_new.reserve(old_variants.size());
+        for (const auto & [old_variant_type, old_discriminator] : old_variant_types_to_old_global_discriminator)
+        {
+            auto it = new_variant_types_to_new_global_discriminator.find(old_variant_type);
+            if (it == new_variant_types_to_new_global_discriminator.end())
+                throw Exception(
+                    ErrorCodes::CANNOT_CONVERT_TYPE,
+                    "Cannot convert type {} to {}. Conversion between Variant types is allowed only when new Variant type is an extension "
+                    "of an initial one", from_variant.getName(), to_variant.getName());
+            old_global_discriminator_to_new[old_discriminator] = it->second;
+        }
+
+        /// Collect variant types and their global discriminators that should be added to the old Variant to get the new Variant.
+        std::vector<std::pair<DataTypePtr, ColumnVariant::Discriminator>> variant_types_and_discriminators_to_add;
+        variant_types_and_discriminators_to_add.reserve(new_variants.size() - old_variants.size());
+        for (size_t i = 0; i != new_variants.size(); ++i)
+        {
+            if (!old_variant_types_to_old_global_discriminator.contains(new_variants[i]->getName()))
+                variant_types_and_discriminators_to_add.emplace_back(new_variants[i], i);
+        }
+
+        return [old_global_discriminator_to_new, variant_types_and_discriminators_to_add]
+               (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable *, size_t) -> ColumnPtr
+        {
+            const auto & column_variant = assert_cast<const ColumnVariant &>(*arguments.front().column.get());
+            size_t num_old_variants = column_variant.getNumVariants();
+            Columns new_variant_columns;
+            new_variant_columns.reserve(num_old_variants + variant_types_and_discriminators_to_add.size());
+            std::vector<ColumnVariant::Discriminator> new_local_to_global_discriminators;
+            new_local_to_global_discriminators.reserve(num_old_variants + variant_types_and_discriminators_to_add.size());
+            for (size_t i = 0; i != num_old_variants; ++i)
+            {
+                new_variant_columns.push_back(column_variant.getVariantPtrByLocalDiscriminator(i));
+                new_local_to_global_discriminators.push_back(old_global_discriminator_to_new.at(column_variant.globalDiscriminatorByLocal(i)));
+            }
+
+            for (const auto & [new_variant_type, new_global_discriminator] : variant_types_and_discriminators_to_add)
+            {
+                new_variant_columns.push_back(new_variant_type->createColumn());
+                new_local_to_global_discriminators.push_back(new_global_discriminator);
+            }
+
+            return ColumnVariant::create(column_variant.getLocalDiscriminatorsPtr(), column_variant.getOffsetsPtr(), new_variant_columns, new_local_to_global_discriminators);
+        };
+    }
+
+    WrapperType createVariantToColumnWrapper(const DataTypeVariant & from_variant, const DataTypePtr & to_type) const
+    {
+        const auto & variant_types = from_variant.getVariants();
+        std::vector<WrapperType> variant_wrappers;
+        variant_wrappers.reserve(variant_types.size());
+
+        /// Create conversion wrapper for each variant.
+        for (const auto & variant_type : variant_types)
+            variant_wrappers.push_back(prepareUnpackDictionaries(variant_type, to_type));
+
+        return [variant_wrappers, variant_types, to_type]
+               (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count) -> ColumnPtr
+        {
+            const auto & column_variant = assert_cast<const ColumnVariant &>(*arguments.front().column.get());
+
+            /// First, cast each variant to the result type.
+            std::vector<ColumnPtr> casted_variant_columns;
+            casted_variant_columns.reserve(variant_types.size());
+            for (size_t i = 0; i != variant_types.size(); ++i)
+            {
+                auto variant_col = column_variant.getVariantPtrByLocalDiscriminator(i);
+                ColumnsWithTypeAndName variant = {{variant_col, variant_types[i], "" }};
+                const auto & variant_wrapper = variant_wrappers[column_variant.globalDiscriminatorByLocal(i)];
+                casted_variant_columns.push_back(variant_wrapper(variant, result_type, nullptr, variant_col->size()));
+            }
+
+            /// Second, construct resulting column from casted variant columns according to discriminators.
+            const auto & local_discriminators = column_variant.getLocalDiscriminators();
+            auto res = result_type->createColumn();
+            res->reserve(input_rows_count);
+            for (size_t i = 0; i != input_rows_count; ++i)
+            {
+                auto local_discr = local_discriminators[i];
+                if (local_discr == ColumnVariant::NULL_DISCRIMINATOR)
+                    res->insertDefault();
+                else
+                    res->insertFrom(*casted_variant_columns[local_discr], column_variant.offsetAt(i));
+            }
+
+            return res;
+        };
+    }
+
+    static ColumnPtr createVariantFromDescriptorsAndOneNonEmptyVariant(const DataTypes & variant_types, const ColumnPtr & discriminators, const ColumnPtr & variant, ColumnVariant::Discriminator variant_discr)
+    {
+        Columns variants;
+        variants.reserve(variant_types.size());
+        for (size_t i = 0; i != variant_types.size(); ++i)
+        {
+            if (i == variant_discr)
+                variants.emplace_back(variant);
+            else
+                variants.push_back(variant_types[i]->createColumn());
+        }
+
+        return ColumnVariant::create(discriminators, variants);
+    }
+
+    WrapperType createColumnToVariantWrapper(const DataTypePtr & from_type, const DataTypeVariant & to_variant) const
+    {
+        /// We allow converting NULL to Variant(...) as Variant can store NULLs.
+        if (from_type->onlyNull())
+        {
+            return [](ColumnsWithTypeAndName &, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count) -> ColumnPtr
+            {
+                auto result_column = result_type->createColumn();
+                result_column->insertManyDefaults(input_rows_count);
+                return result_column;
+            };
+        }
+
+        auto variant_discr_opt = to_variant.tryGetVariantDiscriminator(*removeNullableOrLowCardinalityNullable(from_type));
+        if (!variant_discr_opt)
+            throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Cannot convert type {} to {}. Conversion to Variant allowed only for types from this Variant", from_type->getName(), to_variant.getName());
+
+        return [variant_discr = *variant_discr_opt]
+               (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *, size_t) -> ColumnPtr
+        {
+            const auto & result_variant_type = assert_cast<const DataTypeVariant &>(*result_type);
+            const auto & variant_types = result_variant_type.getVariants();
+            if (const ColumnNullable * col_nullable = typeid_cast<const ColumnNullable *>(arguments.front().column.get()))
+            {
+                const auto & column = col_nullable->getNestedColumnPtr();
+                const auto & null_map = col_nullable->getNullMapData();
+                IColumn::Filter filter;
+                filter.reserve(column->size());
+                auto discriminators = ColumnVariant::ColumnDiscriminators::create();
+                auto & discriminators_data = discriminators->getData();
+                discriminators_data.reserve(column->size());
+                size_t variant_size_hint = 0;
+                for (size_t i = 0; i != column->size(); ++i)
+                {
+                    if (null_map[i])
+                    {
+                        discriminators_data.push_back(ColumnVariant::NULL_DISCRIMINATOR);
+                        filter.push_back(0);
+                    }
+                    else
+                    {
+                        discriminators_data.push_back(variant_discr);
+                        filter.push_back(1);
+                        ++variant_size_hint;
+                    }
+                }
+
+                ColumnPtr variant_column;
+                /// If there were no NULLs, just use the column.
+                if (variant_size_hint == column->size())
+                    variant_column = column;
+                /// Otherwise we should use filtered column.
+                else
+                    variant_column = column->filter(filter, variant_size_hint);
+                return createVariantFromDescriptorsAndOneNonEmptyVariant(variant_types, std::move(discriminators), variant_column, variant_discr);
+            }
+            else if (isColumnLowCardinalityNullable(*arguments.front().column))
+            {
+                const auto & column = arguments.front().column;
+
+                /// Variant column cannot have LowCardinality(Nullable(...)) variant, as Variant column stores NULLs itself.
+                /// We should create a null-map, insert NULL_DISCRIMINATOR on NULL values and filter initial column.
+                const auto & col_lc = assert_cast<const ColumnLowCardinality &>(*column);
+                const auto & indexes = col_lc.getIndexes();
+                auto null_index = col_lc.getDictionary().getNullValueIndex();
+                IColumn::Filter filter;
+                filter.reserve(col_lc.size());
+                auto discriminators = ColumnVariant::ColumnDiscriminators::create();
+                auto & discriminators_data = discriminators->getData();
+                discriminators_data.reserve(col_lc.size());
+                size_t variant_size_hint = 0;
+                for (size_t i = 0; i != col_lc.size(); ++i)
+                {
+                    if (indexes.getUInt(i) == null_index)
+                    {
+                        discriminators_data.push_back(ColumnVariant::NULL_DISCRIMINATOR);
+                        filter.push_back(0);
+                    }
+                    else
+                    {
+                        discriminators_data.push_back(variant_discr);
+                        filter.push_back(1);
+                        ++variant_size_hint;
+                    }
+                }
+
+                MutableColumnPtr variant_column;
+                /// If there were no NULLs, we can just clone the column.
+                if (variant_size_hint == col_lc.size())
+                    variant_column = IColumn::mutate(column);
+                /// Otherwise we should filter column.
+                else
+                    variant_column = column->filter(filter, variant_size_hint)->assumeMutable();
+
+                assert_cast<ColumnLowCardinality &>(*variant_column).nestedRemoveNullable();
+                return createVariantFromDescriptorsAndOneNonEmptyVariant(variant_types, std::move(discriminators), std::move(variant_column), variant_discr);
+            }
+            else
+            {
+                const auto & column = arguments.front().column;
+                auto discriminators = ColumnVariant::ColumnDiscriminators::create();
+                discriminators->getData().resize_fill(column->size(), variant_discr);
+                return createVariantFromDescriptorsAndOneNonEmptyVariant(variant_types, std::move(discriminators), column, variant_discr);
+            }
+        };
+    }
+
+    /// Wrapper for conversion to/from Variant type
+    WrapperType createVariantWrapper(const DataTypePtr & from_type, const DataTypePtr & to_type) const
+    {
+        if (const auto * from_variant = checkAndGetDataType<DataTypeVariant>(from_type.get()))
+        {
+            if (const auto * to_variant = checkAndGetDataType<DataTypeVariant>(to_type.get()))
+                return createVariantToVariantWrapper(*from_variant, *to_variant);
+
+            return createVariantToColumnWrapper(*from_variant, to_type);
+        }
+
+        return createColumnToVariantWrapper(from_type, assert_cast<const DataTypeVariant &>(*to_type));
+    }
+
+    template <typename FieldType>
+    WrapperType createEnumWrapper(const DataTypePtr & from_type, const DataTypeEnum<FieldType> * to_type) const
+    {
+        using EnumType = DataTypeEnum<FieldType>;
+        using Function = typename FunctionTo<EnumType>::Type;
+
+        if (const auto * from_enum8 = checkAndGetDataType<DataTypeEnum8>(from_type.get()))
+            checkEnumToEnumConversion(from_enum8, to_type);
+        else if (const auto * from_enum16 = checkAndGetDataType<DataTypeEnum16>(from_type.get()))
+            checkEnumToEnumConversion(from_enum16, to_type);
+
+        if (checkAndGetDataType<DataTypeString>(from_type.get()))
+            return createStringToEnumWrapper<ColumnString, EnumType>();
+        else if (checkAndGetDataType<DataTypeFixedString>(from_type.get()))
+            return createStringToEnumWrapper<ColumnFixedString, EnumType>();
+        else if (isNativeNumber(from_type) || isEnum(from_type))
+        {
+            auto function = Function::create();
+            return createFunctionAdaptor(function, from_type);
+        }
+        else
+        {
+            if (cast_type == CastType::accurateOrNull)
+                return createToNullableColumnWrapper();
+            else
+                throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Conversion from {} to {} is not supported",
+                    from_type->getName(), to_type->getName());
+        }
+    }
+
+    template <typename EnumTypeFrom, typename EnumTypeTo>
+    void checkEnumToEnumConversion(const EnumTypeFrom * from_type, const EnumTypeTo * to_type) const
+    {
+        const auto & from_values = from_type->getValues();
+        const auto & to_values = to_type->getValues();
+
+        using ValueType = std::common_type_t<typename EnumTypeFrom::FieldType, typename EnumTypeTo::FieldType>;
+        using NameValuePair = std::pair<std::string, ValueType>;
+        using EnumValues = std::vector<NameValuePair>;
+
+        EnumValues name_intersection;
+        std::set_intersection(std::begin(from_values), std::end(from_values),
+            std::begin(to_values), std::end(to_values), std::back_inserter(name_intersection),
+            [] (auto && from, auto && to) { return from.first < to.first; });
+
+        for (const auto & name_value : name_intersection)
+        {
+            const auto & old_value = name_value.second;
+            const auto & new_value = to_type->getValue(name_value.first);
+            if (old_value != new_value)
+                throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Enum conversion changes value for element '{}' from {} to {}",
+                    name_value.first, toString(old_value), toString(new_value));
+        }
+    }
+
+    template <typename ColumnStringType, typename EnumType>
+    WrapperType createStringToEnumWrapper() const
+    {
+        const char * function_name = cast_name;
+        return [function_name] (
+            ColumnsWithTypeAndName & arguments, const DataTypePtr & res_type, const ColumnNullable * nullable_col, size_t /*input_rows_count*/)
+        {
+            const auto & first_col = arguments.front().column.get();
+            const auto & result_type = typeid_cast<const EnumType &>(*res_type);
+
+            const ColumnStringType * col = typeid_cast<const ColumnStringType *>(first_col);
+
+            if (col && nullable_col && nullable_col->size() != col->size())
+                throw Exception(ErrorCodes::LOGICAL_ERROR, "ColumnNullable is not compatible with original");
+
+            if (col)
+            {
+                const auto size = col->size();
+
+                auto res = result_type.createColumn();
+                auto & out_data = static_cast<typename EnumType::ColumnType &>(*res).getData();
+                out_data.resize(size);
+
+                auto default_enum_value = result_type.getValues().front().second;
+
+                if (nullable_col)
+                {
+                    for (size_t i = 0; i < size; ++i)
+                    {
+                        if (!nullable_col->isNullAt(i))
+                            out_data[i] = result_type.getValue(col->getDataAt(i));
+                        else
+                            out_data[i] = default_enum_value;
+                    }
+                }
+                else
+                {
+                    for (size_t i = 0; i < size; ++i)
+                        out_data[i] = result_type.getValue(col->getDataAt(i));
+                }
+
+                return res;
+            }
+            else
+                throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected column {} as first argument of function {}",
+                    first_col->getName(), function_name);
+        };
+    }
+
+    template <typename EnumType>
+    WrapperType createEnumToStringWrapper() const
+    {
+        const char * function_name = cast_name;
+        return [function_name] (
+            ColumnsWithTypeAndName & arguments, const DataTypePtr & res_type, const ColumnNullable * nullable_col, size_t /*input_rows_count*/)
+        {
+            using ColumnEnumType = EnumType::ColumnType;
+
+            const auto & first_col = arguments.front().column.get();
+            const auto & first_type = arguments.front().type.get();
+
+            const ColumnEnumType * enum_col = typeid_cast<const ColumnEnumType *>(first_col);
+            const EnumType * enum_type = typeid_cast<const EnumType *>(first_type);
+
+            if (enum_col && nullable_col && nullable_col->size() != enum_col->size())
+                throw Exception(ErrorCodes::LOGICAL_ERROR, "ColumnNullable is not compatible with original");
+
+            if (enum_col && enum_type)
+            {
+                const auto size = enum_col->size();
+                const auto & enum_data = enum_col->getData();
+
+                auto res = res_type->createColumn();
+
+                if (nullable_col)
+                {
+                    for (size_t i = 0; i < size; ++i)
+                    {
+                        if (!nullable_col->isNullAt(i))
+                        {
+                            const auto & value = enum_type->getNameForValue(enum_data[i]);
+                            res->insertData(value.data, value.size);
+                        }
+                        else
+                            res->insertDefault();
+                    }
+                }
+                else
+                {
+                    for (size_t i = 0; i < size; ++i)
+                    {
+                        const auto & value = enum_type->getNameForValue(enum_data[i]);
+                        res->insertData(value.data, value.size);
+                    }
+                }
+
+                return res;
+            }
+            else
+                throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected column {} as first argument of function {}",
+                    first_col->getName(), function_name);
+        };
+    }
+
+    static WrapperType createIdentityWrapper(const DataTypePtr &)
+    {
+        return [] (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable *, size_t /*input_rows_count*/)
+        {
+            return arguments.front().column;
+        };
+    }
+
+    static WrapperType createNothingWrapper(const IDataType * to_type)
+    {
+        ColumnPtr res = to_type->createColumnConstWithDefaultValue(1);
+        return [res] (ColumnsWithTypeAndName &, const DataTypePtr &, const ColumnNullable *, size_t input_rows_count)
+        {
+            /// Column of Nothing type is trivially convertible to any other column
+            return res->cloneResized(input_rows_count)->convertToFullColumnIfConst();
+        };
+    }
+
+    WrapperType prepareUnpackDictionaries(const DataTypePtr & from_type, const DataTypePtr & to_type) const
+    {
+        /// Conversion from/to Variant data type is processed in a special way.
+        /// We don't need to remove LowCardinality/Nullable.
+        if (isVariant(to_type) || isVariant(from_type))
+            return createVariantWrapper(from_type, to_type);
+
+        const auto * from_low_cardinality = typeid_cast<const DataTypeLowCardinality *>(from_type.get());
+        const auto * to_low_cardinality = typeid_cast<const DataTypeLowCardinality *>(to_type.get());
+        const auto & from_nested = from_low_cardinality ? from_low_cardinality->getDictionaryType() : from_type;
+        const auto & to_nested = to_low_cardinality ? to_low_cardinality->getDictionaryType() : to_type;
+
+        if (from_type->onlyNull())
+        {
+            if (!to_nested->isNullable() && !isVariant(to_type))
+            {
+                if (cast_type == CastType::accurateOrNull)
+                {
+                    return createToNullableColumnWrapper();
+                }
+                else
+                {
+                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Cannot convert NULL to a non-nullable type");
+                }
+            }
+
+            return [](ColumnsWithTypeAndName &, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count)
+            {
+                return result_type->createColumnConstWithDefaultValue(input_rows_count)->convertToFullColumnIfConst();
+            };
+        }
+
+        bool skip_not_null_check = false;
+
+        if (from_low_cardinality && from_nested->isNullable() && !to_nested->isNullable())
+            /// Disable check for dictionary. Will check that column doesn't contain NULL in wrapper below.
+            skip_not_null_check = true;
+
+        auto wrapper = prepareRemoveNullable(from_nested, to_nested, skip_not_null_check);
+        if (!from_low_cardinality && !to_low_cardinality)
+            return wrapper;
+
+        return [wrapper, from_low_cardinality, to_low_cardinality, skip_not_null_check]
+                (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable * nullable_source, size_t input_rows_count) -> ColumnPtr
+        {
+            ColumnsWithTypeAndName args = {arguments[0]};
+            auto & arg = args.front();
+            auto res_type = result_type;
+
+            ColumnPtr converted_column;
+
+            ColumnPtr res_indexes;
+            /// For some types default can't be casted (for example, String to Int). In that case convert column to full.
+            bool src_converted_to_full_column = false;
+
+            {
+                auto tmp_rows_count = input_rows_count;
+
+                if (to_low_cardinality)
+                    res_type = to_low_cardinality->getDictionaryType();
+
+                if (from_low_cardinality)
+                {
+                    const auto * col_low_cardinality = typeid_cast<const ColumnLowCardinality *>(arguments[0].column.get());
+
+                    if (skip_not_null_check && col_low_cardinality->containsNull())
+                        throw Exception(ErrorCodes::CANNOT_INSERT_NULL_IN_ORDINARY_COLUMN, "Cannot convert NULL value to non-Nullable type");
+
+                    arg.column = col_low_cardinality->getDictionary().getNestedColumn();
+                    arg.type = from_low_cardinality->getDictionaryType();
+
+                    /// TODO: Make map with defaults conversion.
+                    src_converted_to_full_column = !removeNullable(arg.type)->equals(*removeNullable(res_type));
+                    if (src_converted_to_full_column)
+                        arg.column = arg.column->index(col_low_cardinality->getIndexes(), 0);
+                    else
+                        res_indexes = col_low_cardinality->getIndexesPtr();
+
+                    tmp_rows_count = arg.column->size();
+                }
+
+                /// Perform the requested conversion.
+                converted_column = wrapper(args, res_type, nullable_source, tmp_rows_count);
+            }
+
+            if (to_low_cardinality)
+            {
+                auto res_column = to_low_cardinality->createColumn();
+                auto * col_low_cardinality = typeid_cast<ColumnLowCardinality *>(res_column.get());
+
+                if (from_low_cardinality && !src_converted_to_full_column)
+                {
+                    col_low_cardinality->insertRangeFromDictionaryEncodedColumn(*converted_column, *res_indexes);
+                }
+                else
+                    col_low_cardinality->insertRangeFromFullColumn(*converted_column, 0, converted_column->size());
+
+                return res_column;
+            }
+            else if (!src_converted_to_full_column)
+                return converted_column->index(*res_indexes, 0);
+            else
+                return converted_column;
+        };
+    }
+
+    WrapperType prepareRemoveNullable(const DataTypePtr & from_type, const DataTypePtr & to_type, bool skip_not_null_check) const
+    {
+        /// Determine whether pre-processing and/or post-processing must take place during conversion.
+
+        bool source_is_nullable = from_type->isNullable();
+        bool result_is_nullable = to_type->isNullable();
+
+        auto wrapper = prepareImpl(removeNullable(from_type), removeNullable(to_type), result_is_nullable);
+
+        if (result_is_nullable)
+        {
+            return [wrapper, source_is_nullable]
+                (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count) -> ColumnPtr
+            {
+                /// Create a temporary columns on which to perform the operation.
+                const auto & nullable_type = static_cast<const DataTypeNullable &>(*result_type);
+                const auto & nested_type = nullable_type.getNestedType();
+
+                ColumnsWithTypeAndName tmp_args;
+                if (source_is_nullable)
+                    tmp_args = createBlockWithNestedColumns(arguments);
+                else
+                    tmp_args = arguments;
+
+                const ColumnNullable * nullable_source = nullptr;
+
+                /// Add original ColumnNullable for createStringToEnumWrapper()
+                if (source_is_nullable)
+                {
+                    if (arguments.size() != 1)
+                        throw Exception(ErrorCodes::LOGICAL_ERROR, "Invalid number of arguments");
+                    nullable_source = typeid_cast<const ColumnNullable *>(arguments.front().column.get());
+                }
+
+                /// Perform the requested conversion.
+                auto tmp_res = wrapper(tmp_args, nested_type, nullable_source, input_rows_count);
+
+                /// May happen in fuzzy tests. For debug purpose.
+                if (!tmp_res)
+                    throw Exception(ErrorCodes::LOGICAL_ERROR, "Couldn't convert {} to {} in prepareRemoveNullable wrapper.",
+                                    arguments[0].type->getName(), nested_type->getName());
+
+                return wrapInNullable(tmp_res, arguments, nested_type, input_rows_count);
+            };
+        }
+        else if (source_is_nullable)
+        {
+            /// Conversion from Nullable to non-Nullable.
+
+            return [wrapper, skip_not_null_check]
+                (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count) -> ColumnPtr
+            {
+                auto tmp_args = createBlockWithNestedColumns(arguments);
+                auto nested_type = removeNullable(result_type);
+
+                /// Check that all values are not-NULL.
+                /// Check can be skipped in case if LowCardinality dictionary is transformed.
+                /// In that case, correctness will be checked beforehand.
+                if (!skip_not_null_check)
+                {
+                    const auto & col = arguments[0].column;
+                    const auto & nullable_col = assert_cast<const ColumnNullable &>(*col);
+                    const auto & null_map = nullable_col.getNullMapData();
+
+                    if (!memoryIsZero(null_map.data(), 0, null_map.size()))
+                        throw Exception(ErrorCodes::CANNOT_INSERT_NULL_IN_ORDINARY_COLUMN, "Cannot convert NULL value to non-Nullable type");
+                }
+                const ColumnNullable * nullable_source = typeid_cast<const ColumnNullable *>(arguments.front().column.get());
+                return wrapper(tmp_args, nested_type, nullable_source, input_rows_count);
+            };
+        }
+        else
+            return wrapper;
+    }
+
+    /// 'from_type' and 'to_type' are nested types in case of Nullable.
+    /// 'requested_result_is_nullable' is true if CAST to Nullable type is requested.
+    WrapperType prepareImpl(const DataTypePtr & from_type, const DataTypePtr & to_type, bool requested_result_is_nullable) const
+    {
+        if (isUInt8(from_type) && isBool(to_type))
+            return createUInt8ToBoolWrapper(from_type, to_type);
+
+        /// We can cast IPv6 into IPv6, IPv4 into IPv4, but we should not allow to cast FixedString(16) into IPv6 as part of identity cast
+        bool safe_convert_custom_types = true;
+
+        if (const auto * to_type_custom_name = to_type->getCustomName())
+            safe_convert_custom_types = from_type->getCustomName() && from_type->getCustomName()->getName() == to_type_custom_name->getName();
+        else if (const auto * from_type_custom_name = from_type->getCustomName())
+            safe_convert_custom_types = to_type->getCustomName() && from_type_custom_name->getName() == to_type->getCustomName()->getName();
+
+        if (from_type->equals(*to_type) && safe_convert_custom_types)
+        {
+            /// We can only use identity conversion for DataTypeAggregateFunction when they are strictly equivalent.
+            if (typeid_cast<const DataTypeAggregateFunction *>(from_type.get()))
+            {
+                if (DataTypeAggregateFunction::strictEquals(from_type, to_type))
+                    return createIdentityWrapper(from_type);
+            }
+            else
+                return createIdentityWrapper(from_type);
+        }
+        else if (WhichDataType(from_type).isNothing())
+            return createNothingWrapper(to_type.get());
+
+        WrapperType ret;
+
+        auto make_default_wrapper = [&](const auto & types) -> bool
+        {
+            using Types = std::decay_t<decltype(types)>;
+            using ToDataType = typename Types::LeftType;
+
+            if constexpr (
+                std::is_same_v<ToDataType, DataTypeUInt16> ||
+                std::is_same_v<ToDataType, DataTypeUInt32> ||
+                std::is_same_v<ToDataType, DataTypeUInt64> ||
+                std::is_same_v<ToDataType, DataTypeUInt128> ||
+                std::is_same_v<ToDataType, DataTypeUInt256> ||
+                std::is_same_v<ToDataType, DataTypeInt8> ||
+                std::is_same_v<ToDataType, DataTypeInt16> ||
+                std::is_same_v<ToDataType, DataTypeInt32> ||
+                std::is_same_v<ToDataType, DataTypeInt64> ||
+                std::is_same_v<ToDataType, DataTypeInt128> ||
+                std::is_same_v<ToDataType, DataTypeInt256> ||
+                std::is_same_v<ToDataType, DataTypeFloat32> ||
+                std::is_same_v<ToDataType, DataTypeFloat64> ||
+                std::is_same_v<ToDataType, DataTypeDate> ||
+                std::is_same_v<ToDataType, DataTypeDate32> ||
+                std::is_same_v<ToDataType, DataTypeDateTime> ||
+                std::is_same_v<ToDataType, DataTypeUUID> ||
+                std::is_same_v<ToDataType, DataTypeIPv4> ||
+                std::is_same_v<ToDataType, DataTypeIPv6>)
+            {
+                ret = createWrapper(from_type, checkAndGetDataType<ToDataType>(to_type.get()), requested_result_is_nullable);
+                return true;
+            }
+            if constexpr (std::is_same_v<ToDataType, DataTypeUInt8>)
+            {
+                if (isBool(to_type))
+                    ret = createBoolWrapper<ToDataType>(from_type, checkAndGetDataType<ToDataType>(to_type.get()), requested_result_is_nullable);
+                else
+                    ret = createWrapper(from_type, checkAndGetDataType<ToDataType>(to_type.get()), requested_result_is_nullable);
+                return true;
+            }
+            if constexpr (
+                std::is_same_v<ToDataType, DataTypeEnum8> ||
+                std::is_same_v<ToDataType, DataTypeEnum16>)
+            {
+                ret = createEnumWrapper(from_type, checkAndGetDataType<ToDataType>(to_type.get()));
+                return true;
+            }
+            if constexpr (
+                std::is_same_v<ToDataType, DataTypeDecimal<Decimal32>> ||
+                std::is_same_v<ToDataType, DataTypeDecimal<Decimal64>> ||
+                std::is_same_v<ToDataType, DataTypeDecimal<Decimal128>> ||
+                std::is_same_v<ToDataType, DataTypeDecimal<Decimal256>> ||
+                std::is_same_v<ToDataType, DataTypeDateTime64>)
+            {
+                ret = createDecimalWrapper(from_type, checkAndGetDataType<ToDataType>(to_type.get()), requested_result_is_nullable);
+                return true;
+            }
+
+            return false;
+        };
+
+        bool cast_ipv4_ipv6_default_on_conversion_error_value = context && context->getSettingsRef().cast_ipv4_ipv6_default_on_conversion_error;
+        bool input_format_ipv4_default_on_conversion_error_value = context && context->getSettingsRef().input_format_ipv4_default_on_conversion_error;
+        bool input_format_ipv6_default_on_conversion_error_value = context && context->getSettingsRef().input_format_ipv6_default_on_conversion_error;
+
+        auto make_custom_serialization_wrapper = [&, cast_ipv4_ipv6_default_on_conversion_error_value, input_format_ipv4_default_on_conversion_error_value, input_format_ipv6_default_on_conversion_error_value](const auto & types) -> bool
+        {
+            using Types = std::decay_t<decltype(types)>;
+            using ToDataType = typename Types::RightType;
+            using FromDataType = typename Types::LeftType;
+
+            if constexpr (WhichDataType(FromDataType::type_id).isStringOrFixedString())
+            {
+                if constexpr (std::is_same_v<ToDataType, DataTypeIPv4>)
+                {
+                    ret = [cast_ipv4_ipv6_default_on_conversion_error_value,
+                           input_format_ipv4_default_on_conversion_error_value,
+                           requested_result_is_nullable](
+                              ColumnsWithTypeAndName & arguments,
+                              const DataTypePtr & result_type,
+                              const ColumnNullable * column_nullable,
+                              size_t) -> ColumnPtr
+                    {
+                        if (!WhichDataType(result_type).isIPv4())
+                            throw Exception(ErrorCodes::TYPE_MISMATCH, "Wrong result type {}. Expected IPv4", result_type->getName());
+
+                        const auto * null_map = column_nullable ? &column_nullable->getNullMapData() : nullptr;
+                        if (requested_result_is_nullable)
+                            return convertToIPv4<IPStringToNumExceptionMode::Null>(arguments[0].column, null_map);
+                        else if (cast_ipv4_ipv6_default_on_conversion_error_value || input_format_ipv4_default_on_conversion_error_value)
+                            return convertToIPv4<IPStringToNumExceptionMode::Default>(arguments[0].column, null_map);
+                        else
+                            return convertToIPv4<IPStringToNumExceptionMode::Throw>(arguments[0].column, null_map);
+                    };
+
+                    return true;
+                }
+
+                if constexpr (std::is_same_v<ToDataType, DataTypeIPv6>)
+                {
+                    ret = [cast_ipv4_ipv6_default_on_conversion_error_value,
+                           input_format_ipv6_default_on_conversion_error_value,
+                           requested_result_is_nullable](
+                              ColumnsWithTypeAndName & arguments,
+                              const DataTypePtr & result_type,
+                              const ColumnNullable * column_nullable,
+                              size_t) -> ColumnPtr
+                    {
+                        if (!WhichDataType(result_type).isIPv6())
+                            throw Exception(
+                                ErrorCodes::TYPE_MISMATCH, "Wrong result type {}. Expected IPv6", result_type->getName());
+
+                        const auto * null_map = column_nullable ? &column_nullable->getNullMapData() : nullptr;
+                        if (requested_result_is_nullable)
+                            return convertToIPv6<IPStringToNumExceptionMode::Null>(arguments[0].column, null_map);
+                        else if (cast_ipv4_ipv6_default_on_conversion_error_value || input_format_ipv6_default_on_conversion_error_value)
+                            return convertToIPv6<IPStringToNumExceptionMode::Default>(arguments[0].column, null_map);
+                        else
+                            return convertToIPv6<IPStringToNumExceptionMode::Throw>(arguments[0].column, null_map);
+                    };
+
+                    return true;
+                }
+
+                if (to_type->getCustomSerialization() && to_type->getCustomName())
+                {
+                    ret = [requested_result_is_nullable](
+                              ColumnsWithTypeAndName & arguments,
+                              const DataTypePtr & result_type,
+                              const ColumnNullable * column_nullable,
+                              size_t input_rows_count) -> ColumnPtr
+                    {
+                        auto wrapped_result_type = result_type;
+                        if (requested_result_is_nullable)
+                            wrapped_result_type = makeNullable(result_type);
+                        return ConvertImplGenericFromString::execute(
+                            arguments, wrapped_result_type, column_nullable, input_rows_count);
+                    };
+                    return true;
+                }
+            }
+            else if constexpr (WhichDataType(FromDataType::type_id).isIPv6() && WhichDataType(ToDataType::type_id).isIPv4())
+            {
+                ret = [cast_ipv4_ipv6_default_on_conversion_error_value, requested_result_is_nullable](
+                                ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable * column_nullable, size_t)
+                        -> ColumnPtr
+                {
+                    if (!WhichDataType(result_type).isIPv4())
+                        throw Exception(
+                            ErrorCodes::TYPE_MISMATCH, "Wrong result type {}. Expected IPv4", result_type->getName());
+
+                    const auto * null_map = column_nullable ? &column_nullable->getNullMapData() : nullptr;
+                    if (requested_result_is_nullable)
+                        return convertIPv6ToIPv4<IPStringToNumExceptionMode::Null>(arguments[0].column, null_map);
+                    else if (cast_ipv4_ipv6_default_on_conversion_error_value)
+                        return convertIPv6ToIPv4<IPStringToNumExceptionMode::Default>(arguments[0].column, null_map);
+                    else
+                        return convertIPv6ToIPv4<IPStringToNumExceptionMode::Throw>(arguments[0].column, null_map);
+                };
+
+                return true;
+            }
+
+            if constexpr (WhichDataType(ToDataType::type_id).isStringOrFixedString())
+            {
+                if constexpr (WhichDataType(FromDataType::type_id).isEnum())
+                {
+                    ret = createEnumToStringWrapper<FromDataType>();
+                    return true;
+                }
+                else if (from_type->getCustomSerialization())
+                {
+                    ret = [](ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count) -> ColumnPtr
+                    {
+                        return ConvertImplGenericToString<typename ToDataType::ColumnType>::execute(arguments, result_type, input_rows_count);
+                    };
+                    return true;
+                }
+            }
+
+            return false;
+        };
+
+        if (callOnTwoTypeIndexes(from_type->getTypeId(), to_type->getTypeId(), make_custom_serialization_wrapper))
+            return ret;
+
+        if (callOnIndexAndDataType<void>(to_type->getTypeId(), make_default_wrapper))
+            return ret;
+
+        switch (to_type->getTypeId())
+        {
+            case TypeIndex::String:
+                return createStringWrapper(from_type);
+            case TypeIndex::FixedString:
+                return createFixedStringWrapper(from_type, checkAndGetDataType<DataTypeFixedString>(to_type.get())->getN());
+            case TypeIndex::Array:
+                return createArrayWrapper(from_type, static_cast<const DataTypeArray &>(*to_type));
+            case TypeIndex::Tuple:
+                return createTupleWrapper(from_type, checkAndGetDataType<DataTypeTuple>(to_type.get()));
+            case TypeIndex::Map:
+                return createMapWrapper(from_type, checkAndGetDataType<DataTypeMap>(to_type.get()));
+            case TypeIndex::Object:
+                return createObjectWrapper(from_type, checkAndGetDataType<DataTypeObject>(to_type.get()));
+            case TypeIndex::AggregateFunction:
+                return createAggregateFunctionWrapper(from_type, checkAndGetDataType<DataTypeAggregateFunction>(to_type.get()));
+            case TypeIndex::Interval:
+                return createIntervalWrapper(from_type, checkAndGetDataType<DataTypeInterval>(to_type.get())->getKind());
+            default:
+                break;
+        }
+
+        if (cast_type == CastType::accurateOrNull)
+            return createToNullableColumnWrapper();
+        else
+            throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Conversion from {} to {} is not supported",
+                from_type->getName(), to_type->getName());
+    }
+};
+
+class MonotonicityHelper
+{
+public:
+    using MonotonicityForRange = FunctionCastBase::MonotonicityForRange;
+
+    template <typename DataType>
+    static auto monotonicityForType(const DataType * const)
+    {
+        return FunctionTo<DataType>::Type::Monotonic::get;
+    }
+
+    static MonotonicityForRange getMonotonicityInformation(const DataTypePtr & from_type, const IDataType * to_type)
+    {
+        if (const auto * type = checkAndGetDataType<DataTypeUInt8>(to_type))
+            return monotonicityForType(type);
+        if (const auto * type = checkAndGetDataType<DataTypeUInt16>(to_type))
+            return monotonicityForType(type);
+        if (const auto * type = checkAndGetDataType<DataTypeUInt32>(to_type))
+            return monotonicityForType(type);
+        if (const auto * type = checkAndGetDataType<DataTypeUInt64>(to_type))
+            return monotonicityForType(type);
+        if (const auto * type = checkAndGetDataType<DataTypeUInt128>(to_type))
+            return monotonicityForType(type);
+        if (const auto * type = checkAndGetDataType<DataTypeUInt256>(to_type))
+            return monotonicityForType(type);
+        if (const auto * type = checkAndGetDataType<DataTypeInt8>(to_type))
+            return monotonicityForType(type);
+        if (const auto * type = checkAndGetDataType<DataTypeInt16>(to_type))
+            return monotonicityForType(type);
+        if (const auto * type = checkAndGetDataType<DataTypeInt32>(to_type))
+            return monotonicityForType(type);
+        if (const auto * type = checkAndGetDataType<DataTypeInt64>(to_type))
+            return monotonicityForType(type);
+        if (const auto * type = checkAndGetDataType<DataTypeInt128>(to_type))
+            return monotonicityForType(type);
+        if (const auto * type = checkAndGetDataType<DataTypeInt256>(to_type))
+            return monotonicityForType(type);
+        if (const auto * type = checkAndGetDataType<DataTypeFloat32>(to_type))
+            return monotonicityForType(type);
+        if (const auto * type = checkAndGetDataType<DataTypeFloat64>(to_type))
+            return monotonicityForType(type);
+        if (const auto * type = checkAndGetDataType<DataTypeDate>(to_type))
+            return monotonicityForType(type);
+        if (const auto * type = checkAndGetDataType<DataTypeDate32>(to_type))
+            return monotonicityForType(type);
+        if (const auto * type = checkAndGetDataType<DataTypeDateTime>(to_type))
+            return monotonicityForType(type);
+        if (const auto * type = checkAndGetDataType<DataTypeString>(to_type))
+            return monotonicityForType(type);
+        if (isEnum(from_type))
+        {
+            if (const auto * type = checkAndGetDataType<DataTypeEnum8>(to_type))
+                return monotonicityForType(type);
+            if (const auto * type = checkAndGetDataType<DataTypeEnum16>(to_type))
+                return monotonicityForType(type);
+        }
+        /// other types like Null, FixedString, Array and Tuple have no monotonicity defined
+        return {};
+    }
+};
+
+
 FunctionBasePtr createFunctionBaseCast(
     ContextPtr context
     , const ColumnsWithTypeAndName & arguments
@@ -146,54 +5102,4 @@ REGISTER_FUNCTION(Conversion)
     factory.registerFunction<FunctionConvert<DataTypeInterval, NameToIntervalYear, PositiveMonotonicity>>();
 }
 
-
-MonotonicityHelper::MonotonicityForRange MonotonicityHelper::getMonotonicityInformation(const DataTypePtr & from_type, const IDataType * to_type)
-{
-    if (const auto * type = checkAndGetDataType<DataTypeUInt8>(to_type))
-        return monotonicityForType(type);
-    if (const auto * type = checkAndGetDataType<DataTypeUInt16>(to_type))
-        return monotonicityForType(type);
-    if (const auto * type = checkAndGetDataType<DataTypeUInt32>(to_type))
-        return monotonicityForType(type);
-    if (const auto * type = checkAndGetDataType<DataTypeUInt64>(to_type))
-        return monotonicityForType(type);
-    if (const auto * type = checkAndGetDataType<DataTypeUInt128>(to_type))
-        return monotonicityForType(type);
-    if (const auto * type = checkAndGetDataType<DataTypeUInt256>(to_type))
-        return monotonicityForType(type);
-    if (const auto * type = checkAndGetDataType<DataTypeInt8>(to_type))
-        return monotonicityForType(type);
-    if (const auto * type = checkAndGetDataType<DataTypeInt16>(to_type))
-        return monotonicityForType(type);
-    if (const auto * type = checkAndGetDataType<DataTypeInt32>(to_type))
-        return monotonicityForType(type);
-    if (const auto * type = checkAndGetDataType<DataTypeInt64>(to_type))
-        return monotonicityForType(type);
-    if (const auto * type = checkAndGetDataType<DataTypeInt128>(to_type))
-        return monotonicityForType(type);
-    if (const auto * type = checkAndGetDataType<DataTypeInt256>(to_type))
-        return monotonicityForType(type);
-    if (const auto * type = checkAndGetDataType<DataTypeFloat32>(to_type))
-        return monotonicityForType(type);
-    if (const auto * type = checkAndGetDataType<DataTypeFloat64>(to_type))
-        return monotonicityForType(type);
-    if (const auto * type = checkAndGetDataType<DataTypeDate>(to_type))
-        return monotonicityForType(type);
-    if (const auto * type = checkAndGetDataType<DataTypeDate32>(to_type))
-        return monotonicityForType(type);
-    if (const auto * type = checkAndGetDataType<DataTypeDateTime>(to_type))
-        return monotonicityForType(type);
-    if (const auto * type = checkAndGetDataType<DataTypeString>(to_type))
-        return monotonicityForType(type);
-    if (isEnum(from_type))
-    {
-        if (const auto * type = checkAndGetDataType<DataTypeEnum8>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeEnum16>(to_type))
-            return monotonicityForType(type);
-    }
-    /// other types like Null, FixedString, Array and Tuple have no monotonicity defined
-    return {};
-}
-
 }
diff --git a/src/Functions/FunctionsConversion.h b/src/Functions/FunctionsConversion.h
deleted file mode 100644
index c21e85fb40e..00000000000
--- a/src/Functions/FunctionsConversion.h
+++ /dev/null
@@ -1,4924 +0,0 @@
-#pragma once
-
-#include <cstddef>
-#include <type_traits>
-
-#include <IO/WriteBufferFromVector.h>
-#include <IO/ReadBufferFromMemory.h>
-#include <IO/Operators.h>
-#include <IO/parseDateTimeBestEffort.h>
-#include <DataTypes/DataTypeFactory.h>
-#include <DataTypes/DataTypesNumber.h>
-#include <DataTypes/DataTypesDecimal.h>
-#include <DataTypes/DataTypeString.h>
-#include <DataTypes/DataTypeFixedString.h>
-#include <DataTypes/DataTypeDate.h>
-#include <DataTypes/DataTypeDate32.h>
-#include <DataTypes/DataTypeDateTime.h>
-#include <DataTypes/DataTypeDateTime64.h>
-#include <DataTypes/DataTypeEnum.h>
-#include <DataTypes/DataTypeArray.h>
-#include <DataTypes/DataTypeTuple.h>
-#include <DataTypes/DataTypeMap.h>
-#include <DataTypes/DataTypeNullable.h>
-#include <DataTypes/DataTypeNothing.h>
-#include <DataTypes/DataTypeUUID.h>
-#include <DataTypes/DataTypeInterval.h>
-#include <DataTypes/DataTypeAggregateFunction.h>
-#include <DataTypes/DataTypeObject.h>
-#include <DataTypes/ObjectUtils.h>
-#include <DataTypes/DataTypeNested.h>
-#include <DataTypes/DataTypeVariant.h>
-#include <DataTypes/Serializations/SerializationDecimal.h>
-#include <Formats/FormatSettings.h>
-#include <Columns/ColumnString.h>
-#include <Columns/ColumnFixedString.h>
-#include <Columns/ColumnConst.h>
-#include <Columns/ColumnAggregateFunction.h>
-#include <Columns/ColumnArray.h>
-#include <Columns/ColumnNullable.h>
-#include <Columns/ColumnTuple.h>
-#include <Columns/ColumnMap.h>
-#include <Columns/ColumnObject.h>
-#include <Columns/ColumnsCommon.h>
-#include <Columns/ColumnVariant.h>
-#include <Columns/ColumnStringHelpers.h>
-#include <Common/assert_cast.h>
-#include <Common/Concepts.h>
-#include <Common/quoteString.h>
-#include <Common/Exception.h>
-#include <Core/AccurateComparison.h>
-#include <Functions/IFunctionAdaptors.h>
-#include <Functions/FunctionHelpers.h>
-#include <Functions/DateTimeTransforms.h>
-#include <Functions/toFixedString.h>
-#include <Functions/TransformDateTime64.h>
-#include <Functions/FunctionsCodingIP.h>
-#include <Functions/CastOverloadResolver.h>
-#include <DataTypes/DataTypeLowCardinality.h>
-#include <Columns/ColumnLowCardinality.h>
-#include <Interpreters/Context.h>
-#include <Common/HashTable/HashMap.h>
-#include <DataTypes/DataTypeIPv4andIPv6.h>
-#include <Common/IPv6ToBinary.h>
-#include <Core/Types.h>
-
-
-namespace DB
-{
-
-namespace ErrorCodes
-{
-    extern const int ATTEMPT_TO_READ_AFTER_EOF;
-    extern const int CANNOT_PARSE_NUMBER;
-    extern const int CANNOT_READ_ARRAY_FROM_TEXT;
-    extern const int CANNOT_PARSE_INPUT_ASSERTION_FAILED;
-    extern const int CANNOT_PARSE_QUOTED_STRING;
-    extern const int CANNOT_PARSE_ESCAPE_SEQUENCE;
-    extern const int CANNOT_PARSE_DATE;
-    extern const int CANNOT_PARSE_DATETIME;
-    extern const int CANNOT_PARSE_TEXT;
-    extern const int CANNOT_PARSE_UUID;
-    extern const int CANNOT_PARSE_IPV4;
-    extern const int CANNOT_PARSE_IPV6;
-    extern const int TOO_FEW_ARGUMENTS_FOR_FUNCTION;
-    extern const int LOGICAL_ERROR;
-    extern const int TYPE_MISMATCH;
-    extern const int CANNOT_CONVERT_TYPE;
-    extern const int ILLEGAL_COLUMN;
-    extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH;
-    extern const int ILLEGAL_TYPE_OF_ARGUMENT;
-    extern const int NOT_IMPLEMENTED;
-    extern const int CANNOT_INSERT_NULL_IN_ORDINARY_COLUMN;
-    extern const int CANNOT_PARSE_BOOL;
-    extern const int VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE;
-}
-
-/** Type conversion functions.
-  * toType - conversion in "natural way";
-  */
-
-UInt32 extractToDecimalScale(const ColumnWithTypeAndName & named_column);
-
-/// Function toUnixTimestamp has exactly the same implementation as toDateTime of String type.
-struct NameToUnixTimestamp { static constexpr auto name = "toUnixTimestamp"; };
-
-struct AccurateConvertStrategyAdditions
-{
-    UInt32 scale { 0 };
-};
-
-struct AccurateOrNullConvertStrategyAdditions
-{
-    UInt32 scale { 0 };
-};
-
-
-struct ConvertDefaultBehaviorTag {};
-struct ConvertReturnNullOnErrorTag {};
-struct ConvertReturnZeroOnErrorTag {};
-
-/** Conversion of number types to each other, enums to numbers, dates and datetimes to numbers and back: done by straight assignment.
-  *  (Date is represented internally as number of days from some day; DateTime - as unix timestamp)
-  */
-template <typename FromDataType, typename ToDataType, typename Name,
-    typename SpecialTag = ConvertDefaultBehaviorTag,
-    FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior = default_date_time_overflow_behavior>
-struct ConvertImpl
-{
-    using FromFieldType = typename FromDataType::FieldType;
-    using ToFieldType = typename ToDataType::FieldType;
-
-    template <typename Additions = void *>
-    static ColumnPtr NO_SANITIZE_UNDEFINED execute(
-        const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type [[maybe_unused]], size_t input_rows_count,
-        Additions additions [[maybe_unused]] = Additions())
-    {
-        const ColumnWithTypeAndName & named_from = arguments[0];
-
-        using ColVecFrom = typename FromDataType::ColumnType;
-        using ColVecTo = typename ToDataType::ColumnType;
-
-        if constexpr ((IsDataTypeDecimal<FromDataType> || IsDataTypeDecimal<ToDataType>)
-            && !(std::is_same_v<DataTypeDateTime64, FromDataType> || std::is_same_v<DataTypeDateTime64, ToDataType>))
-        {
-            if constexpr (!IsDataTypeDecimalOrNumber<FromDataType> || !IsDataTypeDecimalOrNumber<ToDataType>)
-            {
-                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                    named_from.column->getName(), Name::name);
-            }
-        }
-
-        if (const ColVecFrom * col_from = checkAndGetColumn<ColVecFrom>(named_from.column.get()))
-        {
-            typename ColVecTo::MutablePtr col_to = nullptr;
-
-            if constexpr (IsDataTypeDecimal<ToDataType>)
-            {
-                UInt32 scale;
-
-                if constexpr (std::is_same_v<Additions, AccurateConvertStrategyAdditions>
-                    || std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-                {
-                    scale = additions.scale;
-                }
-                else
-                {
-                    scale = additions;
-                }
-
-                col_to = ColVecTo::create(0, scale);
-            }
-            else
-                col_to = ColVecTo::create();
-
-            const auto & vec_from = col_from->getData();
-            auto & vec_to = col_to->getData();
-            vec_to.resize(input_rows_count);
-
-            ColumnUInt8::MutablePtr col_null_map_to;
-            ColumnUInt8::Container * vec_null_map_to [[maybe_unused]] = nullptr;
-            if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-            {
-                col_null_map_to = ColumnUInt8::create(input_rows_count, false);
-                vec_null_map_to = &col_null_map_to->getData();
-            }
-
-            bool result_is_bool = isBool(result_type);
-            for (size_t i = 0; i < input_rows_count; ++i)
-            {
-                if constexpr (std::is_same_v<ToDataType, DataTypeUInt8>)
-                {
-                    if (result_is_bool)
-                    {
-                        vec_to[i] = vec_from[i] != FromFieldType(0);
-                        continue;
-                    }
-                }
-
-                if constexpr (std::is_same_v<FromDataType, DataTypeUUID> && std::is_same_v<ToDataType, DataTypeUInt128>)
-                {
-                    static_assert(
-                        std::is_same_v<DataTypeUInt128::FieldType, DataTypeUUID::FieldType::UnderlyingType>,
-                        "UInt128 and UUID types must be same");
-
-                    vec_to[i].items[1] = vec_from[i].toUnderType().items[0];
-                    vec_to[i].items[0] = vec_from[i].toUnderType().items[1];
-
-                    continue;
-                }
-
-                if constexpr (std::is_same_v<FromDataType, DataTypeIPv6> && std::is_same_v<ToDataType, DataTypeUInt128>)
-                {
-                    static_assert(
-                        std::is_same_v<DataTypeUInt128::FieldType, DataTypeUUID::FieldType::UnderlyingType>,
-                        "UInt128 and IPv6 types must be same");
-
-                    vec_to[i].items[1] = std::byteswap(vec_from[i].toUnderType().items[0]);
-                    vec_to[i].items[0] = std::byteswap(vec_from[i].toUnderType().items[1]);
-
-                    continue;
-                }
-
-                if constexpr (std::is_same_v<FromDataType, DataTypeUUID> != std::is_same_v<ToDataType, DataTypeUUID>)
-                {
-                    throw Exception(ErrorCodes::NOT_IMPLEMENTED,
-                                    "Conversion between numeric types and UUID is not supported. "
-                                    "Probably the passed UUID is unquoted");
-                }
-                else if constexpr (
-                    (std::is_same_v<FromDataType, DataTypeIPv4> != std::is_same_v<ToDataType, DataTypeIPv4>)
-                    && !(is_any_of<FromDataType, DataTypeUInt8, DataTypeUInt16, DataTypeUInt32, DataTypeUInt64, DataTypeIPv6> || is_any_of<ToDataType, DataTypeUInt32, DataTypeUInt64, DataTypeUInt128, DataTypeUInt256, DataTypeIPv6>)
-                )
-                {
-                    throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Conversion from {} to {} is not supported",
-                                    TypeName<typename FromDataType::FieldType>, TypeName<typename ToDataType::FieldType>);
-                }
-                else if constexpr (std::is_same_v<FromDataType, DataTypeIPv6> != std::is_same_v<ToDataType, DataTypeIPv6> && !(std::is_same_v<ToDataType, DataTypeIPv4> || std::is_same_v<FromDataType, DataTypeIPv4>))
-                {
-                    throw Exception(ErrorCodes::NOT_IMPLEMENTED,
-                                    "Conversion between numeric types and IPv6 is not supported. "
-                                    "Probably the passed IPv6 is unquoted");
-                }
-                else
-                {
-                    if constexpr (IsDataTypeDecimal<FromDataType> || IsDataTypeDecimal<ToDataType>)
-                    {
-                        if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-                        {
-                            ToFieldType result;
-                            bool convert_result = false;
-
-                            if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeDecimal<ToDataType>)
-                                convert_result = tryConvertDecimals<FromDataType, ToDataType>(vec_from[i], col_from->getScale(), col_to->getScale(), result);
-                            else if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeNumber<ToDataType>)
-                                convert_result = tryConvertFromDecimal<FromDataType, ToDataType>(vec_from[i], col_from->getScale(), result);
-                            else if constexpr (IsDataTypeNumber<FromDataType> && IsDataTypeDecimal<ToDataType>)
-                                convert_result = tryConvertToDecimal<FromDataType, ToDataType>(vec_from[i], col_to->getScale(), result);
-
-                            if (convert_result)
-                                vec_to[i] = result;
-                            else
-                            {
-                                vec_to[i] = static_cast<ToFieldType>(0);
-                                (*vec_null_map_to)[i] = true;
-                            }
-                        }
-                        else
-                        {
-                            if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeDecimal<ToDataType>)
-                                vec_to[i] = convertDecimals<FromDataType, ToDataType>(vec_from[i], col_from->getScale(), col_to->getScale());
-                            else if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeNumber<ToDataType>)
-                                vec_to[i] = convertFromDecimal<FromDataType, ToDataType>(vec_from[i], col_from->getScale());
-                            else if constexpr (IsDataTypeNumber<FromDataType> && IsDataTypeDecimal<ToDataType>)
-                                vec_to[i] = convertToDecimal<FromDataType, ToDataType>(vec_from[i], col_to->getScale());
-                            else
-                                throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Unsupported data type in conversion function");
-                        }
-                    }
-                    else
-                    {
-                        /// If From Data is Nan or Inf and we convert to integer type, throw exception
-                        if constexpr (std::is_floating_point_v<FromFieldType> && !std::is_floating_point_v<ToFieldType>)
-                        {
-                            if (!isFinite(vec_from[i]))
-                            {
-                                if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-                                {
-                                    vec_to[i] = 0;
-                                    (*vec_null_map_to)[i] = true;
-                                    continue;
-                                }
-                                else
-                                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Unexpected inf or nan to integer conversion");
-                            }
-                        }
-
-                        if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>
-                                || std::is_same_v<Additions, AccurateConvertStrategyAdditions>)
-                        {
-                            bool convert_result = accurate::convertNumeric(vec_from[i], vec_to[i]);
-
-                            if (!convert_result)
-                            {
-                                if (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-                                {
-                                    vec_to[i] = 0;
-                                    (*vec_null_map_to)[i] = true;
-                                }
-                                else
-                                {
-                                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Value in column {} cannot be safely converted into type {}",
-                                        named_from.column->getName(), result_type->getName());
-                                }
-                            }
-                        }
-                        else
-                        {
-                            if constexpr (std::is_same_v<ToDataType, DataTypeIPv4> && std::is_same_v<FromDataType, DataTypeIPv6>)
-                            {
-                                const uint8_t ip4_cidr[] {0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x00, 0x00, 0x00, 0x00};
-                                const uint8_t * src = reinterpret_cast<const uint8_t *>(&vec_from[i].toUnderType());
-                                if (!matchIPv6Subnet(src, ip4_cidr, 96))
-                                {
-                                    char addr[IPV6_MAX_TEXT_LENGTH + 1] {};
-                                    char * paddr = addr;
-                                    formatIPv6(src, paddr);
-
-                                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "IPv6 {} in column {} is not in IPv4 mapping block", addr, named_from.column->getName());
-                                }
-
-                                uint8_t * dst = reinterpret_cast<uint8_t *>(&vec_to[i].toUnderType());
-                                if constexpr (std::endian::native == std::endian::little)
-                                {
-                                    dst[0] = src[15];
-                                    dst[1] = src[14];
-                                    dst[2] = src[13];
-                                    dst[3] = src[12];
-                                }
-                                else
-                                {
-                                    dst[0] = src[12];
-                                    dst[1] = src[13];
-                                    dst[2] = src[14];
-                                    dst[3] = src[15];
-                                }
-                            }
-                            else if constexpr (std::is_same_v<ToDataType, DataTypeIPv6> && std::is_same_v<FromDataType, DataTypeIPv4>)
-                            {
-                                const uint8_t * src = reinterpret_cast<const uint8_t *>(&vec_from[i].toUnderType());
-                                uint8_t * dst = reinterpret_cast<uint8_t *>(&vec_to[i].toUnderType());
-                                std::memset(dst, '\0', IPV6_BINARY_LENGTH);
-                                dst[10] = dst[11] = 0xff;
-
-                                if constexpr (std::endian::native == std::endian::little)
-                                {
-                                    dst[12] = src[3];
-                                    dst[13] = src[2];
-                                    dst[14] = src[1];
-                                    dst[15] = src[0];
-                                }
-                                else
-                                {
-                                    dst[12] = src[0];
-                                    dst[13] = src[1];
-                                    dst[14] = src[2];
-                                    dst[15] = src[3];
-                                }
-                            }
-                            else if constexpr (std::is_same_v<ToDataType, DataTypeIPv4> && std::is_same_v<FromDataType, DataTypeUInt64>)
-                                vec_to[i] = static_cast<ToFieldType>(static_cast<IPv4::UnderlyingType>(vec_from[i]));
-                            else if constexpr (std::is_same_v<Name, NameToUnixTimestamp> && (std::is_same_v<FromDataType, DataTypeDate> || std::is_same_v<FromDataType, DataTypeDate32>))
-                                vec_to[i] = static_cast<ToFieldType>(vec_from[i] * DATE_SECONDS_PER_DAY);
-                            else
-                                vec_to[i] = static_cast<ToFieldType>(vec_from[i]);
-                        }
-                    }
-                }
-            }
-
-            if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-                return ColumnNullable::create(std::move(col_to), std::move(col_null_map_to));
-            else
-                return col_to;
-        }
-        else
-            throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                named_from.column->getName(), Name::name);
-    }
-};
-
-/** Conversion of DateTime to Date: throw off time component.
-  */
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDateTime, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDateTime, DataTypeDate, ToDateImpl<date_time_overflow_behavior>, false> {};
-
-/** Conversion of DateTime to Date32: throw off time component.
-  */
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDateTime, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDateTime, DataTypeDate32, ToDate32Impl, false> {};
-
-/** Conversion of Date to DateTime: adding 00:00:00 time component.
-  */
-template <FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior = default_date_time_overflow_behavior>
-struct ToDateTimeImpl
-{
-    static constexpr auto name = "toDateTime";
-
-    static UInt32 execute(UInt16 d, const DateLUTImpl & time_zone)
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (d > MAX_DATETIME_DAY_NUM) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Day number {} is out of bounds of type DateTime", d);
-        }
-        else if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Saturate)
-        {
-            if (d > MAX_DATETIME_DAY_NUM)
-                d = MAX_DATETIME_DAY_NUM;
-        }
-        return static_cast<UInt32>(time_zone.fromDayNum(DayNum(d)));
-    }
-
-    static UInt32 execute(Int32 d, const DateLUTImpl & time_zone)
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Saturate)
-        {
-            if (d < 0)
-                return 0;
-            else if (d > MAX_DATETIME_DAY_NUM)
-                d = MAX_DATETIME_DAY_NUM;
-        }
-        else if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (d < 0 || d > MAX_DATETIME_DAY_NUM) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type DateTime", d);
-        }
-        return static_cast<UInt32>(time_zone.fromDayNum(ExtendedDayNum(d)));
-    }
-
-    static UInt32 execute(UInt32 dt, const DateLUTImpl & /*time_zone*/)
-    {
-        return dt;
-    }
-
-    static UInt32 execute(Int64 dt64, const DateLUTImpl & /*time_zone*/)
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Ignore)
-            return static_cast<UInt32>(dt64);
-        else
-        {
-            if (dt64 < 0 || dt64 >= MAX_DATETIME_TIMESTAMP)
-            {
-                if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Saturate)
-                    return dt64 < 0 ? 0 : std::numeric_limits<UInt32>::max();
-                else
-                    throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type DateTime", dt64);
-            }
-            else
-                return static_cast<UInt32>(dt64);
-        }
-    }
-};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDate, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDate, DataTypeDateTime, ToDateTimeImpl<date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDate32, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDate32, DataTypeDateTime, ToDateTimeImpl<date_time_overflow_behavior>, false> {};
-
-/// Implementation of toDate function.
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTransform32Or64
-{
-    static constexpr auto name = "toDate";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from > MAX_DATETIME_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type Date", from);
-        }
-        /// if value is smaller (or equal) than maximum day value for Date, than treat it as day num,
-        /// otherwise treat it as unix timestamp. This is a bit weird, but we leave this behavior.
-        if (from <= DATE_LUT_MAX_DAY_NUM)
-            return from;
-        else
-            return time_zone.toDayNum(std::min(time_t(from), time_t(MAX_DATETIME_TIMESTAMP)));
-    }
-};
-
-/** Conversion of Date32 to Date.
-  */
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDate32, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDate32, DataTypeDate, ToDateImpl<date_time_overflow_behavior>, false> {};
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTransform32Or64Signed
-{
-    static constexpr auto name = "toDate";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
-    {
-        // TODO: decide narrow or extended range based on FromType
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from < 0 || from > MAX_DATE_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type Date", from);
-        }
-        else
-        {
-            if (from < 0)
-                return 0;
-        }
-        return (from <= DATE_LUT_MAX_DAY_NUM)
-            ? static_cast<ToType>(from)
-            : time_zone.toDayNum(std::min(time_t(from), time_t(MAX_DATE_TIMESTAMP)));
-    }
-};
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTransform8Or16Signed
-{
-    static constexpr auto name = "toDate";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
-    {
-        if (from < 0)
-        {
-            if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type Date", from);
-            else
-                return 0;
-        }
-        return from;
-    }
-};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDateTime64, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-        : DateTimeTransformImpl<DataTypeDateTime64, DataTypeDate32, TransformDateTime64<ToDate32Impl>, false> {};
-
-/// Implementation of toDate32 function.
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDate32Transform32Or64
-{
-    static constexpr auto name = "toDate32";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
-    {
-        if (from < DATE_LUT_MAX_EXTEND_DAY_NUM)
-            return static_cast<ToType>(from);
-        else
-        {
-            if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-            {
-                if (from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
-                    throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type Date32", from);
-            }
-            return time_zone.toDayNum(std::min(time_t(from), time_t(MAX_DATETIME64_TIMESTAMP)));
-        }
-    }
-};
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDate32Transform32Or64Signed
-{
-    static constexpr auto name = "toDate32";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
-    {
-        static const Int32 daynum_min_offset = -static_cast<Int32>(time_zone.getDayNumOffsetEpoch());
-
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from < daynum_min_offset || from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type Date32", from);
-        }
-
-        if (from < daynum_min_offset)
-            return daynum_min_offset;
-
-        return (from < DATE_LUT_MAX_EXTEND_DAY_NUM)
-            ? static_cast<ToType>(from)
-            : time_zone.toDayNum(std::min(time_t(Int64(from)), time_t(MAX_DATETIME64_TIMESTAMP)));
-    }
-};
-
-template <typename FromType, typename ToType>
-struct ToDate32Transform8Or16Signed
-{
-    static constexpr auto name = "toDate32";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
-    {
-        return from;
-    }
-};
-
-/** Special case of converting Int8, Int16, (U)Int32 or (U)Int64 (and also, for convenience,
-  * Float32, Float64) to Date. If the
-  * number is less than 65536, then it is treated as DayNum, and if it's greater or equals to 65536,
-  * then treated as unix timestamp. If the number exceeds UInt32, saturate to MAX_UINT32 then as DayNum.
-  * It's a bit illogical, as we actually have two functions in one.
-  * But allows to support frequent case,
-  *  when user write toDate(UInt32), expecting conversion of unix timestamp to Date.
-  *  (otherwise such usage would be frequent mistake).
-  */
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeUInt32, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeUInt32, DataTypeDate, ToDateTransform32Or64<UInt32, UInt16, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeUInt64, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeUInt64, DataTypeDate, ToDateTransform32Or64<UInt64, UInt16, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt8, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt8, DataTypeDate, ToDateTransform8Or16Signed<Int8, UInt16, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt16, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt16, DataTypeDate, ToDateTransform8Or16Signed<Int16, UInt16, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt32, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt32, DataTypeDate, ToDateTransform32Or64Signed<Int32, UInt16, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt64, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt64, DataTypeDate, ToDateTransform32Or64Signed<Int64, UInt16, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat32, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat32, DataTypeDate, ToDateTransform32Or64Signed<Float32, UInt16, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat64, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat64, DataTypeDate, ToDateTransform32Or64Signed<Float64, UInt16, default_date_time_overflow_behavior>, false> {};
-
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeUInt32, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeUInt32, DataTypeDate32, ToDate32Transform32Or64<UInt32, Int32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeUInt64, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeUInt64, DataTypeDate32, ToDate32Transform32Or64<UInt64, Int32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt8, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt8, DataTypeDate32, ToDate32Transform8Or16Signed<Int8, Int32>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt16, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt16, DataTypeDate32, ToDate32Transform8Or16Signed<Int16, Int32>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt32, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt32, DataTypeDate32, ToDate32Transform32Or64Signed<Int32, Int32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt64, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt64, DataTypeDate32, ToDate32Transform32Or64Signed<Int64, Int32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat32, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat32, DataTypeDate32, ToDate32Transform32Or64Signed<Float32, Int32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat64, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat64, DataTypeDate32, ToDate32Transform32Or64Signed<Float64, Int32, default_date_time_overflow_behavior>, false> {};
-
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTimeTransform64
-{
-    static constexpr auto name = "toDateTime";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from > MAX_DATETIME_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime", from);
-        }
-        return static_cast<ToType>(std::min(time_t(from), time_t(MAX_DATETIME_TIMESTAMP)));
-    }
-};
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTimeTransformSigned
-{
-    static constexpr auto name = "toDateTime";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
-    {
-        if (from < 0)
-        {
-            if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime", from);
-            else
-                return 0;
-        }
-        return from;
-    }
-};
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTimeTransform64Signed
-{
-    static constexpr auto name = "toDateTime";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from < 0 || from > MAX_DATETIME_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime", from);
-        }
-
-        if (from < 0)
-            return 0;
-        return static_cast<ToType>(std::min(time_t(from), time_t(MAX_DATETIME_TIMESTAMP)));
-    }
-};
-
-/// Special case of converting Int8, Int16, Int32 or (U)Int64 (and also, for convenience, Float32, Float64) to DateTime.
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt8, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt8, DataTypeDateTime, ToDateTimeTransformSigned<Int8, UInt32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt16, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt16, DataTypeDateTime, ToDateTimeTransformSigned<Int16, UInt32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt32, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt32, DataTypeDateTime, ToDateTimeTransformSigned<Int32, UInt32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt64, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt64, DataTypeDateTime, ToDateTimeTransform64Signed<Int64, UInt32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeUInt64, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeUInt64, DataTypeDateTime, ToDateTimeTransform64<UInt64, UInt32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat32, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat32, DataTypeDateTime, ToDateTimeTransform64Signed<Float32, UInt32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat64, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat64, DataTypeDateTime, ToDateTimeTransform64Signed<Float64, UInt32, default_date_time_overflow_behavior>, false> {};
-
-/** Conversion of numeric to DateTime64
-  */
-
-template <typename FromType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTime64TransformUnsigned
-{
-    static constexpr auto name = "toDateTime64";
-
-    const DateTime64::NativeType scale_multiplier = 1;
-
-    ToDateTime64TransformUnsigned(UInt32 scale = 0) /// NOLINT
-        : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
-    {}
-
-    NO_SANITIZE_UNDEFINED DateTime64::NativeType execute(FromType from, const DateLUTImpl &) const
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime64", from);
-            else
-                return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(from, 0, scale_multiplier);
-        }
-        else
-            return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(std::min<time_t>(from, MAX_DATETIME64_TIMESTAMP), 0, scale_multiplier);
-    }
-};
-template <typename FromType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTime64TransformSigned
-{
-    static constexpr auto name = "toDateTime64";
-
-    const DateTime64::NativeType scale_multiplier = 1;
-
-    ToDateTime64TransformSigned(UInt32 scale = 0) /// NOLINT
-        : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
-    {}
-
-    NO_SANITIZE_UNDEFINED DateTime64::NativeType execute(FromType from, const DateLUTImpl &) const
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from < MIN_DATETIME64_TIMESTAMP || from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime64", from);
-        }
-        from = static_cast<FromType>(std::max<time_t>(from, MIN_DATETIME64_TIMESTAMP));
-        from = static_cast<FromType>(std::min<time_t>(from, MAX_DATETIME64_TIMESTAMP));
-
-        return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(from, 0, scale_multiplier);
-    }
-};
-template <typename FromDataType, typename FromType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTime64TransformFloat
-{
-    static constexpr auto name = "toDateTime64";
-
-    const UInt32 scale = 1;
-
-    ToDateTime64TransformFloat(UInt32 scale_ = 0) /// NOLINT
-        : scale(scale_)
-    {}
-
-    NO_SANITIZE_UNDEFINED DateTime64::NativeType execute(FromType from, const DateLUTImpl &) const
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from < MIN_DATETIME64_TIMESTAMP || from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime64", from);
-        }
-
-        from = std::max(from, static_cast<FromType>(MIN_DATETIME64_TIMESTAMP));
-        from = std::min(from, static_cast<FromType>(MAX_DATETIME64_TIMESTAMP));
-        return convertToDecimal<FromDataType, DataTypeDateTime64>(from, scale);
-    }
-};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt8, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt8, DataTypeDateTime64, ToDateTime64TransformSigned<Int8, date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt16, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt16, DataTypeDateTime64, ToDateTime64TransformSigned<Int16, date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt32, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt32, DataTypeDateTime64, ToDateTime64TransformSigned<Int32, date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt64, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt64, DataTypeDateTime64, ToDateTime64TransformSigned<Int64, date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeUInt64, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeUInt64, DataTypeDateTime64, ToDateTime64TransformUnsigned<UInt64, date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat32, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat32, DataTypeDateTime64, ToDateTime64TransformFloat<DataTypeFloat32, Float32, date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat64, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat64, DataTypeDateTime64, ToDateTime64TransformFloat<DataTypeFloat64, Float64, date_time_overflow_behavior>, false> {};
-
-
-/** Conversion of DateTime64 to Date or DateTime: discards fractional part.
- */
-template <typename Transform>
-struct FromDateTime64Transform
-{
-    static constexpr auto name = Transform::name;
-
-    const DateTime64::NativeType scale_multiplier = 1;
-
-    FromDateTime64Transform(UInt32 scale) /// NOLINT
-        : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
-    {}
-
-    auto execute(DateTime64::NativeType dt, const DateLUTImpl & time_zone) const
-    {
-        const auto c = DecimalUtils::splitWithScaleMultiplier(DateTime64(dt), scale_multiplier);
-        return Transform::execute(static_cast<UInt32>(c.whole), time_zone);
-    }
-};
-
-/** Conversion of DateTime64 to Date or DateTime: discards fractional part.
- */
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDateTime64, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDateTime64, DataTypeDate, TransformDateTime64<ToDateImpl<date_time_overflow_behavior>>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDateTime64, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDateTime64, DataTypeDateTime, TransformDateTime64<ToDateTimeImpl<date_time_overflow_behavior>>, false> {};
-
-struct ToDateTime64Transform
-{
-    static constexpr auto name = "toDateTime64";
-
-    const DateTime64::NativeType scale_multiplier = 1;
-
-    ToDateTime64Transform(UInt32 scale = 0) /// NOLINT
-        : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
-    {}
-
-    DateTime64::NativeType execute(UInt16 d, const DateLUTImpl & time_zone) const
-    {
-        const auto dt = ToDateTimeImpl<>::execute(d, time_zone);
-        return execute(dt, time_zone);
-    }
-
-    DateTime64::NativeType execute(Int32 d, const DateLUTImpl & time_zone) const
-    {
-        Int64 dt = static_cast<Int64>(time_zone.fromDayNum(ExtendedDayNum(d)));
-        return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(dt, 0, scale_multiplier);
-    }
-
-    DateTime64::NativeType execute(UInt32 dt, const DateLUTImpl & /*time_zone*/) const
-    {
-        return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(dt, 0, scale_multiplier);
-    }
-};
-
-/** Conversion of Date or DateTime to DateTime64: add zero sub-second part.
-  */
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDate, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDate, DataTypeDateTime64, ToDateTime64Transform> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDate32, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDate32, DataTypeDateTime64, ToDateTime64Transform> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDateTime, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDateTime, DataTypeDateTime64, ToDateTime64Transform> {};
-
-
-/** Transformation of numbers, dates, datetimes to strings: through formatting.
-  */
-template <typename DataType>
-struct FormatImpl
-{
-    template <typename ReturnType = void>
-    static ReturnType execute(const typename DataType::FieldType x, WriteBuffer & wb, const DataType *, const DateLUTImpl *)
-    {
-        writeText(x, wb);
-        return ReturnType(true);
-    }
-};
-
-template <>
-struct FormatImpl<DataTypeDate>
-{
-    template <typename ReturnType = void>
-    static ReturnType execute(const DataTypeDate::FieldType x, WriteBuffer & wb, const DataTypeDate *, const DateLUTImpl * time_zone)
-    {
-        writeDateText(DayNum(x), wb, *time_zone);
-        return ReturnType(true);
-    }
-};
-
-template <>
-struct FormatImpl<DataTypeDate32>
-{
-    template <typename ReturnType = void>
-    static ReturnType execute(const DataTypeDate32::FieldType x, WriteBuffer & wb, const DataTypeDate32 *, const DateLUTImpl * time_zone)
-    {
-        writeDateText(ExtendedDayNum(x), wb, *time_zone);
-        return ReturnType(true);
-    }
-};
-
-template <>
-struct FormatImpl<DataTypeDateTime>
-{
-    template <typename ReturnType = void>
-    static ReturnType execute(const DataTypeDateTime::FieldType x, WriteBuffer & wb, const DataTypeDateTime *, const DateLUTImpl * time_zone)
-    {
-        writeDateTimeText(x, wb, *time_zone);
-        return ReturnType(true);
-    }
-};
-
-template <>
-struct FormatImpl<DataTypeDateTime64>
-{
-    template <typename ReturnType = void>
-    static ReturnType execute(const DataTypeDateTime64::FieldType x, WriteBuffer & wb, const DataTypeDateTime64 * type, const DateLUTImpl * time_zone)
-    {
-        writeDateTimeText(DateTime64(x), type->getScale(), wb, *time_zone);
-        return ReturnType(true);
-    }
-};
-
-
-template <typename FieldType>
-struct FormatImpl<DataTypeEnum<FieldType>>
-{
-    template <typename ReturnType = void>
-    static ReturnType execute(const FieldType x, WriteBuffer & wb, const DataTypeEnum<FieldType> * type, const DateLUTImpl *)
-    {
-        static constexpr bool throw_exception = std::is_same_v<ReturnType, void>;
-
-        if constexpr (throw_exception)
-        {
-            writeString(type->getNameForValue(x), wb);
-        }
-        else
-        {
-            StringRef res;
-            bool is_ok = type->getNameForValue(x, res);
-            if (is_ok)
-                writeString(res, wb);
-            return ReturnType(is_ok);
-        }
-    }
-};
-
-template <typename FieldType>
-struct FormatImpl<DataTypeDecimal<FieldType>>
-{
-    template <typename ReturnType = void>
-    static ReturnType execute(const FieldType x, WriteBuffer & wb, const DataTypeDecimal<FieldType> * type, const DateLUTImpl *)
-    {
-        writeText(x, type->getScale(), wb, false);
-        return ReturnType(true);
-    }
-};
-
-
-/// DataTypeEnum<T> to DataType<T> free conversion
-template <typename FieldType, typename Name>
-struct ConvertImpl<DataTypeEnum<FieldType>, DataTypeNumber<FieldType>, Name, ConvertDefaultBehaviorTag>
-{
-    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/)
-    {
-        return arguments[0].column;
-    }
-};
-
-static inline ColumnUInt8::MutablePtr copyNullMap(ColumnPtr col)
-{
-    ColumnUInt8::MutablePtr null_map = nullptr;
-    if (const auto * col_null = checkAndGetColumn<ColumnNullable>(col.get()))
-    {
-        null_map = ColumnUInt8::create();
-        null_map->insertRangeFrom(col_null->getNullMapColumn(), 0, col_null->size());
-    }
-    return null_map;
-}
-
-template <typename FromDataType, typename Name>
-requires (!std::is_same_v<FromDataType, DataTypeString>)
-struct ConvertImpl<FromDataType, DataTypeString, Name, ConvertDefaultBehaviorTag>
-{
-    using FromFieldType = typename FromDataType::FieldType;
-    using ColVecType = ColumnVectorOrDecimal<FromFieldType>;
-
-    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/)
-    {
-        if constexpr (IsDataTypeDateOrDateTime<FromDataType>)
-        {
-            auto datetime_arg = arguments[0];
-
-            const DateLUTImpl * time_zone = nullptr;
-            const ColumnConst * time_zone_column = nullptr;
-
-            if (arguments.size() == 1)
-            {
-                auto non_null_args = createBlockWithNestedColumns(arguments);
-                time_zone = &extractTimeZoneFromFunctionArguments(non_null_args, 1, 0);
-            }
-            else /// When we have a column for timezone
-            {
-                datetime_arg.column = datetime_arg.column->convertToFullColumnIfConst();
-
-                if constexpr (std::is_same_v<FromDataType, DataTypeDate> || std::is_same_v<FromDataType, DataTypeDate32>)
-                    time_zone = &DateLUT::instance();
-                /// For argument of Date or DateTime type, second argument with time zone could be specified.
-                if constexpr (std::is_same_v<FromDataType, DataTypeDateTime> || std::is_same_v<FromDataType, DataTypeDateTime64>)
-                {
-                    if ((time_zone_column = checkAndGetColumnConst<ColumnString>(arguments[1].column.get())))
-                    {
-                        auto non_null_args = createBlockWithNestedColumns(arguments);
-                        time_zone = &extractTimeZoneFromFunctionArguments(non_null_args, 1, 0);
-                    }
-                }
-            }
-            const auto & col_with_type_and_name = columnGetNested(datetime_arg);
-
-            if (const auto col_from = checkAndGetColumn<ColVecType>(col_with_type_and_name.column.get()))
-            {
-                auto col_to = ColumnString::create();
-
-                const typename ColVecType::Container & vec_from = col_from->getData();
-                ColumnString::Chars & data_to = col_to->getChars();
-                ColumnString::Offsets & offsets_to = col_to->getOffsets();
-                size_t size = vec_from.size();
-
-                if constexpr (std::is_same_v<FromDataType, DataTypeDate>)
-                    data_to.resize(size * (strlen("YYYY-MM-DD") + 1));
-                else if constexpr (std::is_same_v<FromDataType, DataTypeDate32>)
-                    data_to.resize(size * (strlen("YYYY-MM-DD") + 1));
-                else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime>)
-                    data_to.resize(size * (strlen("YYYY-MM-DD hh:mm:ss") + 1));
-                else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime64>)
-                    data_to.resize(size * (strlen("YYYY-MM-DD hh:mm:ss.") + col_from->getScale() + 1));
-                else
-                    data_to.resize(size * 3);   /// Arbitrary
-
-                offsets_to.resize(size);
-
-                WriteBufferFromVector<ColumnString::Chars> write_buffer(data_to);
-                const auto & type = static_cast<const FromDataType &>(*col_with_type_and_name.type);
-
-                ColumnUInt8::MutablePtr null_map = copyNullMap(datetime_arg.column);
-
-                if (!null_map && arguments.size() > 1)
-                    null_map = copyNullMap(arguments[1].column->convertToFullColumnIfConst());
-
-                if (null_map)
-                {
-                    for (size_t i = 0; i < size; ++i)
-                    {
-                        if (!time_zone_column && arguments.size() > 1)
-                        {
-                            if (!arguments[1].column.get()->getDataAt(i).toString().empty())
-                                time_zone = &DateLUT::instance(arguments[1].column.get()->getDataAt(i).toString());
-                            else
-                                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Provided time zone must be non-empty");
-                        }
-                        bool is_ok = FormatImpl<FromDataType>::template execute<bool>(vec_from[i], write_buffer, &type, time_zone);
-                        null_map->getData()[i] |= !is_ok;
-                        writeChar(0, write_buffer);
-                        offsets_to[i] = write_buffer.count();
-                    }
-                }
-                else
-                {
-                    for (size_t i = 0; i < size; ++i)
-                    {
-                        if (!time_zone_column && arguments.size() > 1)
-                        {
-                            if (!arguments[1].column.get()->getDataAt(i).toString().empty())
-                                time_zone = &DateLUT::instance(arguments[1].column.get()->getDataAt(i).toString());
-                            else
-                                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Provided time zone must be non-empty");
-                        }
-                        FormatImpl<FromDataType>::template execute<void>(vec_from[i], write_buffer, &type, time_zone);
-                        writeChar(0, write_buffer);
-                        offsets_to[i] = write_buffer.count();
-                    }
-                }
-
-                write_buffer.finalize();
-
-                if (null_map)
-                    return ColumnNullable::create(std::move(col_to), std::move(null_map));
-                return col_to;
-            }
-            else
-                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                        arguments[0].column->getName(), Name::name);
-        }
-        else
-        {
-            ColumnUInt8::MutablePtr null_map = copyNullMap(arguments[0].column);
-
-            const auto & col_with_type_and_name = columnGetNested(arguments[0]);
-            const auto & type = static_cast<const FromDataType &>(*col_with_type_and_name.type);
-
-            if (const auto col_from = checkAndGetColumn<ColVecType>(col_with_type_and_name.column.get()))
-            {
-                auto col_to = ColumnString::create();
-
-                const typename ColVecType::Container & vec_from = col_from->getData();
-                ColumnString::Chars & data_to = col_to->getChars();
-                ColumnString::Offsets & offsets_to = col_to->getOffsets();
-                size_t size = vec_from.size();
-
-                data_to.resize(size * 3);
-                offsets_to.resize(size);
-
-                WriteBufferFromVector<ColumnString::Chars> write_buffer(data_to);
-
-                if (null_map)
-                {
-                    for (size_t i = 0; i < size; ++i)
-                    {
-                        bool is_ok = FormatImpl<FromDataType>::template execute<bool>(vec_from[i], write_buffer, &type, nullptr);
-                        /// We don't use timezones in this branch
-                        null_map->getData()[i] |= !is_ok;
-                        writeChar(0, write_buffer);
-                        offsets_to[i] = write_buffer.count();
-                    }
-                }
-                else
-                {
-                    for (size_t i = 0; i < size; ++i)
-                    {
-                        FormatImpl<FromDataType>::template execute<void>(vec_from[i], write_buffer, &type, nullptr);
-                        writeChar(0, write_buffer);
-                        offsets_to[i] = write_buffer.count();
-                    }
-                }
-
-                write_buffer.finalize();
-
-                if (null_map)
-                    return ColumnNullable::create(std::move(col_to), std::move(null_map));
-                return col_to;
-            }
-            else
-                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                        arguments[0].column->getName(), Name::name);
-        }
-    }
-};
-
-
-/// Generic conversion of any type to String or FixedString via serialization to text.
-template <typename StringColumnType>
-struct ConvertImplGenericToString
-{
-    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t /*input_rows_count*/)
-    {
-        static_assert(std::is_same_v<StringColumnType, ColumnString> || std::is_same_v<StringColumnType, ColumnFixedString>,
-                "Can be used only to serialize to ColumnString or ColumnFixedString");
-
-        ColumnUInt8::MutablePtr null_map = copyNullMap(arguments[0].column);
-
-        const auto & col_with_type_and_name = columnGetNested(arguments[0]);
-        const IDataType & type = *col_with_type_and_name.type;
-        const IColumn & col_from = *col_with_type_and_name.column;
-
-        size_t size = col_from.size();
-        auto col_to = removeNullable(result_type)->createColumn();
-
-        {
-            ColumnStringHelpers::WriteHelper write_helper(
-                    assert_cast<StringColumnType &>(*col_to),
-                    size);
-
-            auto & write_buffer = write_helper.getWriteBuffer();
-
-            FormatSettings format_settings;
-            auto serialization = type.getDefaultSerialization();
-            for (size_t row = 0; row < size; ++row)
-            {
-                serialization->serializeText(col_from, row, write_buffer, format_settings);
-                write_helper.rowWritten();
-            }
-
-            write_helper.finalize();
-        }
-
-        if (result_type->isNullable() && null_map)
-            return ColumnNullable::create(std::move(col_to), std::move(null_map));
-        return col_to;
-    }
-};
-
-/** Conversion of time_t to UInt16, Int32, UInt32
-  */
-template <typename DataType>
-void convertFromTime(typename DataType::FieldType & x, time_t & time)
-{
-    x = time;
-}
-
-template <>
-inline void convertFromTime<DataTypeDate>(DataTypeDate::FieldType & x, time_t & time)
-{
-    if (unlikely(time < 0))
-        x = 0;
-    else if (unlikely(time > 0xFFFF))
-        x = 0xFFFF;
-    else
-        x = time;
-}
-
-template <>
-inline void convertFromTime<DataTypeDate32>(DataTypeDate32::FieldType & x, time_t & time)
-{
-    x = static_cast<UInt32>(time);
-}
-
-template <>
-inline void convertFromTime<DataTypeDateTime>(DataTypeDateTime::FieldType & x, time_t & time)
-{
-    if (unlikely(time < 0))
-        x = 0;
-    else if (unlikely(time > MAX_DATETIME_TIMESTAMP))
-        x = MAX_DATETIME_TIMESTAMP;
-    else
-        x = static_cast<UInt32>(time);
-}
-
-/** Conversion of strings to numbers, dates, datetimes: through parsing.
-  */
-template <typename DataType>
-void parseImpl(typename DataType::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool precise_float_parsing)
-{
-    if constexpr (std::is_floating_point_v<typename DataType::FieldType>)
-    {
-        if (precise_float_parsing)
-            readFloatTextPrecise(x, rb);
-        else
-            readFloatTextFast(x, rb);
-    }
-    else
-        readText(x, rb);
-}
-
-template <>
-inline void parseImpl<DataTypeDate>(DataTypeDate::FieldType & x, ReadBuffer & rb, const DateLUTImpl * time_zone, bool)
-{
-    DayNum tmp(0);
-    readDateText(tmp, rb, *time_zone);
-    x = tmp;
-}
-
-template <>
-inline void parseImpl<DataTypeDate32>(DataTypeDate32::FieldType & x, ReadBuffer & rb, const DateLUTImpl * time_zone, bool)
-{
-    ExtendedDayNum tmp(0);
-    readDateText(tmp, rb, *time_zone);
-    x = tmp;
-}
-
-
-// NOTE: no need of extra overload of DateTime64, since readDateTimeText64 has different signature and that case is explicitly handled in the calling code.
-template <>
-inline void parseImpl<DataTypeDateTime>(DataTypeDateTime::FieldType & x, ReadBuffer & rb, const DateLUTImpl * time_zone, bool)
-{
-    time_t time = 0;
-    readDateTimeText(time, rb, *time_zone);
-    convertFromTime<DataTypeDateTime>(x, time);
-}
-
-template <>
-inline void parseImpl<DataTypeUUID>(DataTypeUUID::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool)
-{
-    UUID tmp;
-    readUUIDText(tmp, rb);
-    x = tmp.toUnderType();
-}
-
-template <>
-inline void parseImpl<DataTypeIPv4>(DataTypeIPv4::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool)
-{
-    IPv4 tmp;
-    readIPv4Text(tmp, rb);
-    x = tmp.toUnderType();
-}
-
-template <>
-inline void parseImpl<DataTypeIPv6>(DataTypeIPv6::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool)
-{
-    IPv6 tmp;
-    readIPv6Text(tmp, rb);
-    x = tmp;
-}
-
-template <typename DataType>
-bool tryParseImpl(typename DataType::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool precise_float_parsing)
-{
-    if constexpr (std::is_floating_point_v<typename DataType::FieldType>)
-    {
-        if (precise_float_parsing)
-            return tryReadFloatTextPrecise(x, rb);
-        else
-            return tryReadFloatTextFast(x, rb);
-    }
-    else /*if constexpr (is_integer_v<typename DataType::FieldType>)*/
-        return tryReadIntText(x, rb);
-}
-
-template <>
-inline bool tryParseImpl<DataTypeDate>(DataTypeDate::FieldType & x, ReadBuffer & rb, const DateLUTImpl * time_zone, bool)
-{
-    DayNum tmp(0);
-    if (!tryReadDateText(tmp, rb, *time_zone))
-        return false;
-    x = tmp;
-    return true;
-}
-
-template <>
-inline bool tryParseImpl<DataTypeDate32>(DataTypeDate32::FieldType & x, ReadBuffer & rb, const DateLUTImpl * time_zone, bool)
-{
-    ExtendedDayNum tmp(0);
-    if (!tryReadDateText(tmp, rb, *time_zone))
-        return false;
-    x = tmp;
-    return true;
-}
-
-template <>
-inline bool tryParseImpl<DataTypeDateTime>(DataTypeDateTime::FieldType & x, ReadBuffer & rb, const DateLUTImpl * time_zone, bool)
-{
-    time_t time = 0;
-    if (!tryReadDateTimeText(time, rb, *time_zone))
-        return false;
-    convertFromTime<DataTypeDateTime>(x, time);
-    return true;
-}
-
-template <>
-inline bool tryParseImpl<DataTypeUUID>(DataTypeUUID::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool)
-{
-    UUID tmp;
-    if (!tryReadUUIDText(tmp, rb))
-        return false;
-
-    x = tmp.toUnderType();
-    return true;
-}
-
-template <>
-inline bool tryParseImpl<DataTypeIPv4>(DataTypeIPv4::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool)
-{
-    IPv4 tmp;
-    if (!tryReadIPv4Text(tmp, rb))
-        return false;
-
-    x = tmp.toUnderType();
-    return true;
-}
-
-template <>
-inline bool tryParseImpl<DataTypeIPv6>(DataTypeIPv6::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool)
-{
-    IPv6 tmp;
-    if (!tryReadIPv6Text(tmp, rb))
-        return false;
-
-    x = tmp;
-    return true;
-}
-
-
-/** Throw exception with verbose message when string value is not parsed completely.
-  */
-[[noreturn]] inline void throwExceptionForIncompletelyParsedValue(ReadBuffer & read_buffer, const IDataType & result_type)
-{
-    WriteBufferFromOwnString message_buf;
-    message_buf << "Cannot parse string " << quote << String(read_buffer.buffer().begin(), read_buffer.buffer().size())
-                << " as " << result_type.getName()
-                << ": syntax error";
-
-    if (read_buffer.offset())
-        message_buf << " at position " << read_buffer.offset()
-                    << " (parsed just " << quote << String(read_buffer.buffer().begin(), read_buffer.offset()) << ")";
-    else
-        message_buf << " at begin of string";
-
-    // Currently there are no functions toIPv{4,6}Or{Null,Zero}
-    if (isNativeNumber(result_type) && !(result_type.getName() == "IPv4" || result_type.getName() == "IPv6"))
-        message_buf << ". Note: there are to" << result_type.getName() << "OrZero and to" << result_type.getName() << "OrNull functions, which returns zero/NULL instead of throwing exception.";
-
-    throw Exception(PreformattedMessage{message_buf.str(), "Cannot parse string {} as {}: syntax error {}"}, ErrorCodes::CANNOT_PARSE_TEXT);
-}
-
-
-enum class ConvertFromStringExceptionMode
-{
-    Throw,  /// Throw exception if value cannot be parsed.
-    Zero,   /// Fill with zero or default if value cannot be parsed.
-    Null    /// Return ColumnNullable with NULLs when value cannot be parsed.
-};
-
-enum class ConvertFromStringParsingMode
-{
-    Normal,
-    BestEffort,  /// Only applicable for DateTime. Will use sophisticated method, that is slower.
-    BestEffortUS
-};
-
-template <typename FromDataType, typename ToDataType, typename Name,
-    ConvertFromStringExceptionMode exception_mode, ConvertFromStringParsingMode parsing_mode>
-struct ConvertThroughParsing
-{
-    static_assert(std::is_same_v<FromDataType, DataTypeString> || std::is_same_v<FromDataType, DataTypeFixedString>,
-        "ConvertThroughParsing is only applicable for String or FixedString data types");
-
-    static constexpr bool to_datetime64 = std::is_same_v<ToDataType, DataTypeDateTime64>;
-
-    static bool isAllRead(ReadBuffer & in)
-    {
-        /// In case of FixedString, skip zero bytes at end.
-        if constexpr (std::is_same_v<FromDataType, DataTypeFixedString>)
-            while (!in.eof() && *in.position() == 0)
-                ++in.position();
-
-        if (in.eof())
-            return true;
-
-        /// Special case, that allows to parse string with DateTime or DateTime64 as Date or Date32.
-        if constexpr (std::is_same_v<ToDataType, DataTypeDate> || std::is_same_v<ToDataType, DataTypeDate32>)
-        {
-            if (!in.eof() && (*in.position() == ' ' || *in.position() == 'T'))
-            {
-                if (in.buffer().size() == strlen("YYYY-MM-DD hh:mm:ss"))
-                    return true;
-
-                if (in.buffer().size() >= strlen("YYYY-MM-DD hh:mm:ss.x")
-                    && in.buffer().begin()[19] == '.')
-                {
-                    in.position() = in.buffer().begin() + 20;
-
-                    while (!in.eof() && isNumericASCII(*in.position()))
-                        ++in.position();
-
-                    if (in.eof())
-                        return true;
-                }
-            }
-        }
-
-        return false;
-    }
-
-    template <typename Additions = void *>
-    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr & res_type, size_t input_rows_count,
-                        Additions additions [[maybe_unused]] = Additions())
-    {
-        using ColVecTo = typename ToDataType::ColumnType;
-
-        const DateLUTImpl * local_time_zone [[maybe_unused]] = nullptr;
-        const DateLUTImpl * utc_time_zone [[maybe_unused]] = nullptr;
-
-        /// For conversion to Date or DateTime type, second argument with time zone could be specified.
-        if constexpr (std::is_same_v<ToDataType, DataTypeDateTime> || to_datetime64)
-        {
-            const auto result_type = removeNullable(res_type);
-            // Time zone is already figured out during result type resolution, no need to do it here.
-            if (const auto dt_col = checkAndGetDataType<ToDataType>(result_type.get()))
-                local_time_zone = &dt_col->getTimeZone();
-            else
-                local_time_zone = &extractTimeZoneFromFunctionArguments(arguments, 1, 0);
-
-            if constexpr (parsing_mode == ConvertFromStringParsingMode::BestEffort || parsing_mode == ConvertFromStringParsingMode::BestEffortUS)
-                utc_time_zone = &DateLUT::instance("UTC");
-        }
-        else if constexpr (std::is_same_v<ToDataType, DataTypeDate> || std::is_same_v<ToDataType, DataTypeDate32>)
-        {
-            // Timezone is more or less dummy when parsing Date/Date32 from string.
-            local_time_zone = &DateLUT::instance();
-            utc_time_zone = &DateLUT::instance("UTC");
-        }
-
-        const IColumn * col_from = arguments[0].column.get();
-        const ColumnString * col_from_string = checkAndGetColumn<ColumnString>(col_from);
-        const ColumnFixedString * col_from_fixed_string = checkAndGetColumn<ColumnFixedString>(col_from);
-
-        if (std::is_same_v<FromDataType, DataTypeString> && !col_from_string)
-            throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                col_from->getName(), Name::name);
-
-        if (std::is_same_v<FromDataType, DataTypeFixedString> && !col_from_fixed_string)
-            throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                col_from->getName(), Name::name);
-
-        size_t size = input_rows_count;
-        typename ColVecTo::MutablePtr col_to = nullptr;
-
-        if constexpr (IsDataTypeDecimal<ToDataType>)
-        {
-            UInt32 scale = additions;
-            if constexpr (to_datetime64)
-            {
-                ToDataType check_bounds_in_ctor(scale, local_time_zone ? local_time_zone->getTimeZone() : String{});
-            }
-            else
-            {
-                ToDataType check_bounds_in_ctor(ToDataType::maxPrecision(), scale);
-            }
-            col_to = ColVecTo::create(size, scale);
-        }
-        else
-            col_to = ColVecTo::create(size);
-
-        typename ColVecTo::Container & vec_to = col_to->getData();
-
-        ColumnUInt8::MutablePtr col_null_map_to;
-        ColumnUInt8::Container * vec_null_map_to [[maybe_unused]] = nullptr;
-        if constexpr (exception_mode == ConvertFromStringExceptionMode::Null)
-        {
-            col_null_map_to = ColumnUInt8::create(size);
-            vec_null_map_to = &col_null_map_to->getData();
-        }
-
-        const ColumnString::Chars * chars = nullptr;
-        const IColumn::Offsets * offsets = nullptr;
-        size_t fixed_string_size = 0;
-
-        if constexpr (std::is_same_v<FromDataType, DataTypeString>)
-        {
-            chars = &col_from_string->getChars();
-            offsets = &col_from_string->getOffsets();
-        }
-        else
-        {
-            chars = &col_from_fixed_string->getChars();
-            fixed_string_size = col_from_fixed_string->getN();
-        }
-
-        size_t current_offset = 0;
-
-        bool precise_float_parsing = false;
-
-        if (DB::CurrentThread::isInitialized())
-        {
-            const DB::ContextPtr query_context = DB::CurrentThread::get().getQueryContext();
-
-            if (query_context)
-                precise_float_parsing = query_context->getSettingsRef().precise_float_parsing;
-        }
-
-        for (size_t i = 0; i < size; ++i)
-        {
-            size_t next_offset = std::is_same_v<FromDataType, DataTypeString> ? (*offsets)[i] : (current_offset + fixed_string_size);
-            size_t string_size = std::is_same_v<FromDataType, DataTypeString> ? next_offset - current_offset - 1 : fixed_string_size;
-
-            ReadBufferFromMemory read_buffer(&(*chars)[current_offset], string_size);
-
-            if constexpr (exception_mode == ConvertFromStringExceptionMode::Throw)
-            {
-                if constexpr (parsing_mode == ConvertFromStringParsingMode::BestEffort)
-                {
-                    if constexpr (to_datetime64)
-                    {
-                        DateTime64 res = 0;
-                        parseDateTime64BestEffort(res, col_to->getScale(), read_buffer, *local_time_zone, *utc_time_zone);
-                        vec_to[i] = res;
-                    }
-                    else
-                    {
-                        time_t res;
-                        parseDateTimeBestEffort(res, read_buffer, *local_time_zone, *utc_time_zone);
-                        convertFromTime<ToDataType>(vec_to[i], res);
-                    }
-                }
-                else if constexpr (parsing_mode == ConvertFromStringParsingMode::BestEffortUS)
-                {
-                    if constexpr (to_datetime64)
-                    {
-                        DateTime64 res = 0;
-                        parseDateTime64BestEffortUS(res, col_to->getScale(), read_buffer, *local_time_zone, *utc_time_zone);
-                        vec_to[i] = res;
-                    }
-                    else
-                    {
-                        time_t res;
-                        parseDateTimeBestEffortUS(res, read_buffer, *local_time_zone, *utc_time_zone);
-                        convertFromTime<ToDataType>(vec_to[i], res);
-                    }
-                }
-                else
-                {
-                    if constexpr (to_datetime64)
-                    {
-                        DateTime64 value = 0;
-                        readDateTime64Text(value, col_to->getScale(), read_buffer, *local_time_zone);
-                        vec_to[i] = value;
-                    }
-                    else if constexpr (IsDataTypeDecimal<ToDataType>)
-                    {
-                        SerializationDecimal<typename ToDataType::FieldType>::readText(
-                            vec_to[i], read_buffer, ToDataType::maxPrecision(), col_to->getScale());
-                    }
-                    else
-                    {
-                        /// we want to utilize constexpr condition here, which is not mixable with value comparison
-                        do
-                        {
-                            if constexpr (std::is_same_v<FromDataType, DataTypeFixedString> && std::is_same_v<ToDataType, DataTypeIPv6>)
-                            {
-                                if (fixed_string_size == IPV6_BINARY_LENGTH)
-                                {
-                                    readBinary(vec_to[i], read_buffer);
-                                    break;
-                                }
-                            }
-                            parseImpl<ToDataType>(vec_to[i], read_buffer, local_time_zone, precise_float_parsing);
-                        } while (false);
-                    }
-                }
-
-                if (!isAllRead(read_buffer))
-                    throwExceptionForIncompletelyParsedValue(read_buffer, *res_type);
-            }
-            else
-            {
-                bool parsed;
-
-                if constexpr (parsing_mode == ConvertFromStringParsingMode::BestEffort)
-                {
-                    if constexpr (to_datetime64)
-                    {
-                        DateTime64 res = 0;
-                        parsed = tryParseDateTime64BestEffort(res, col_to->getScale(), read_buffer, *local_time_zone, *utc_time_zone);
-                        vec_to[i] = res;
-                    }
-                    else
-                    {
-                        time_t res;
-                        parsed = tryParseDateTimeBestEffort(res, read_buffer, *local_time_zone, *utc_time_zone);
-                        convertFromTime<ToDataType>(vec_to[i],res);
-                    }
-                }
-                else if constexpr (parsing_mode == ConvertFromStringParsingMode::BestEffortUS)
-                {
-                    if constexpr (to_datetime64)
-                    {
-                        DateTime64 res = 0;
-                        parsed = tryParseDateTime64BestEffortUS(res, col_to->getScale(), read_buffer, *local_time_zone, *utc_time_zone);
-                        vec_to[i] = res;
-                    }
-                    else
-                    {
-                        time_t res;
-                        parsed = tryParseDateTimeBestEffortUS(res, read_buffer, *local_time_zone, *utc_time_zone);
-                        convertFromTime<ToDataType>(vec_to[i],res);
-                    }
-                }
-                else
-                {
-                    if constexpr (to_datetime64)
-                    {
-                        DateTime64 value = 0;
-                        parsed = tryReadDateTime64Text(value, col_to->getScale(), read_buffer, *local_time_zone);
-                        vec_to[i] = value;
-                    }
-                    else if constexpr (IsDataTypeDecimal<ToDataType>)
-                    {
-                        parsed = SerializationDecimal<typename ToDataType::FieldType>::tryReadText(
-                            vec_to[i], read_buffer, ToDataType::maxPrecision(), col_to->getScale());
-                    }
-                    else
-                    {
-                        /// we want to utilize constexpr condition here, which is not mixable with value comparison
-                        do
-                        {
-                            if constexpr (std::is_same_v<FromDataType, DataTypeFixedString> && std::is_same_v<ToDataType, DataTypeIPv6>)
-                            {
-                                if (fixed_string_size == IPV6_BINARY_LENGTH)
-                                {
-                                    readBinary(vec_to[i], read_buffer);
-                                    parsed = true;
-                                    break;
-                                }
-                            }
-
-                            parsed = tryParseImpl<ToDataType>(vec_to[i], read_buffer, local_time_zone, precise_float_parsing);
-                        } while (false);
-                    }
-                }
-
-                if (!isAllRead(read_buffer))
-                    parsed = false;
-
-                if (!parsed)
-                {
-                    if constexpr (std::is_same_v<ToDataType, DataTypeDate32>)
-                    {
-                        vec_to[i] = -static_cast<Int32>(DateLUT::instance().getDayNumOffsetEpoch());
-                    }
-                    else
-                    {
-                        vec_to[i] = static_cast<typename ToDataType::FieldType>(0);
-                    }
-                }
-
-                if constexpr (exception_mode == ConvertFromStringExceptionMode::Null)
-                    (*vec_null_map_to)[i] = !parsed;
-            }
-
-            current_offset = next_offset;
-        }
-
-        if constexpr (exception_mode == ConvertFromStringExceptionMode::Null)
-            return ColumnNullable::create(std::move(col_to), std::move(col_null_map_to));
-        else
-            return col_to;
-    }
-};
-
-
-template <typename ToDataType, typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-requires (!std::is_same_v<ToDataType, DataTypeString>)
-struct ConvertImpl<DataTypeString, ToDataType, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : ConvertThroughParsing<DataTypeString, ToDataType, Name, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::Normal> {};
-
-template <typename ToDataType, typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-requires (!std::is_same_v<ToDataType, DataTypeFixedString>)
-struct ConvertImpl<DataTypeFixedString, ToDataType, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : ConvertThroughParsing<DataTypeFixedString, ToDataType, Name, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::Normal> {};
-
-template <typename ToDataType, typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-requires (!std::is_same_v<ToDataType, DataTypeString>)
-struct ConvertImpl<DataTypeString, ToDataType, Name, ConvertReturnNullOnErrorTag, date_time_overflow_behavior>
-    : ConvertThroughParsing<DataTypeString, ToDataType, Name, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::Normal> {};
-
-template <typename ToDataType, typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-requires (!std::is_same_v<ToDataType, DataTypeFixedString>)
-struct ConvertImpl<DataTypeFixedString, ToDataType, Name, ConvertReturnNullOnErrorTag, date_time_overflow_behavior>
-    : ConvertThroughParsing<DataTypeFixedString, ToDataType, Name, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::Normal> {};
-
-template <typename FromDataType, typename ToDataType, typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-requires (is_any_of<FromDataType, DataTypeString, DataTypeFixedString> && is_any_of<ToDataType, DataTypeIPv4, DataTypeIPv6>)
-struct ConvertImpl<FromDataType, ToDataType, Name, ConvertReturnZeroOnErrorTag, date_time_overflow_behavior>
-    : ConvertThroughParsing<FromDataType, ToDataType, Name, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::Normal> {};
-
-/// Generic conversion of any type from String. Used for complex types: Array and Tuple or types with custom serialization.
-struct ConvertImplGenericFromString
-{
-    static ColumnPtr execute(ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable * column_nullable, size_t input_rows_count)
-    {
-        const IColumn & column_from = *arguments[0].column;
-        const IDataType & data_type_to = *result_type;
-        auto res = data_type_to.createColumn();
-        auto serialization = data_type_to.getDefaultSerialization();
-        const auto * null_map = column_nullable ? &column_nullable->getNullMapData() : nullptr;
-
-        executeImpl(column_from, *res, *serialization, input_rows_count, null_map, result_type.get());
-        return res;
-    }
-
-    static void executeImpl(
-        const IColumn & column_from,
-        IColumn & column_to,
-        const ISerialization & serialization_from,
-        size_t input_rows_count,
-        const PaddedPODArray<UInt8> * null_map,
-        const IDataType * result_type)
-    {
-        column_to.reserve(input_rows_count);
-
-        FormatSettings format_settings;
-        for (size_t i = 0; i < input_rows_count; ++i)
-        {
-            if (null_map && (*null_map)[i])
-            {
-                column_to.insertDefault();
-                continue;
-            }
-
-            const auto & val = column_from.getDataAt(i);
-            ReadBufferFromMemory read_buffer(val.data, val.size);
-            try
-            {
-                serialization_from.deserializeWholeText(column_to, read_buffer, format_settings);
-            }
-            catch (const Exception & e)
-            {
-                auto * nullable_column = typeid_cast<ColumnNullable *>(&column_to);
-                if (e.code() == ErrorCodes::CANNOT_PARSE_BOOL && nullable_column)
-                {
-                    auto & col_nullmap = nullable_column->getNullMapData();
-                    if (col_nullmap.size() != nullable_column->size())
-                        col_nullmap.resize_fill(nullable_column->size());
-                    if (nullable_column->size() == (i + 1))
-                        nullable_column->popBack(1);
-                    nullable_column->insertDefault();
-                    continue;
-                }
-                throw;
-            }
-
-            if (!read_buffer.eof())
-            {
-                if (result_type)
-                    throwExceptionForIncompletelyParsedValue(read_buffer, *result_type);
-                else
-                    throw Exception(ErrorCodes::CANNOT_PARSE_TEXT,
-                        "Cannot parse string to column {}. Expected eof", column_to.getName());
-            }
-        }
-    }
-};
-
-
-template <>
-struct ConvertImpl<DataTypeString, DataTypeUInt32, NameToUnixTimestamp, ConvertDefaultBehaviorTag>
-    : ConvertImpl<DataTypeString, DataTypeDateTime, NameToUnixTimestamp, ConvertDefaultBehaviorTag> {};
-
-template <>
-struct ConvertImpl<DataTypeString, DataTypeUInt32, NameToUnixTimestamp, ConvertReturnNullOnErrorTag>
-    : ConvertImpl<DataTypeString, DataTypeDateTime, NameToUnixTimestamp, ConvertReturnNullOnErrorTag> {};
-
-/** If types are identical, just take reference to column.
-  */
-template <typename T, typename Name>
-requires (!T::is_parametric)
-struct ConvertImpl<T, T, Name, ConvertDefaultBehaviorTag>
-{
-    template <typename Additions = void *>
-    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/,
-        Additions additions [[maybe_unused]] = Additions())
-    {
-        return arguments[0].column;
-    }
-};
-
-template <typename Name>
-struct ConvertImpl<DataTypeUInt8, DataTypeUInt8, Name, ConvertDefaultBehaviorTag>
-{
-    template <typename Additions = void *>
-    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/,
-                             Additions additions [[maybe_unused]] = Additions())
-    {
-
-        return arguments[0].column;
-    }
-};
-
-
-/** Conversion from FixedString to String.
-  * Cutting sequences of zero bytes from end of strings.
-  */
-template <typename Name>
-struct ConvertImpl<DataTypeFixedString, DataTypeString, Name, ConvertDefaultBehaviorTag>
-{
-    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr & return_type, size_t /*input_rows_count*/)
-    {
-        ColumnUInt8::MutablePtr null_map = copyNullMap(arguments[0].column);
-        const auto & nested =  columnGetNested(arguments[0]);
-        if (const ColumnFixedString * col_from = checkAndGetColumn<ColumnFixedString>(nested.column.get()))
-        {
-            auto col_to = ColumnString::create();
-
-            const ColumnFixedString::Chars & data_from = col_from->getChars();
-            ColumnString::Chars & data_to = col_to->getChars();
-            ColumnString::Offsets & offsets_to = col_to->getOffsets();
-            size_t size = col_from->size();
-            size_t n = col_from->getN();
-            data_to.resize(size * (n + 1)); /// + 1 - zero terminator
-            offsets_to.resize(size);
-
-            size_t offset_from = 0;
-            size_t offset_to = 0;
-            for (size_t i = 0; i < size; ++i)
-            {
-                if (!null_map || !null_map->getData()[i])
-                {
-                    size_t bytes_to_copy = n;
-                    while (bytes_to_copy > 0 && data_from[offset_from + bytes_to_copy - 1] == 0)
-                        --bytes_to_copy;
-
-                    memcpy(&data_to[offset_to], &data_from[offset_from], bytes_to_copy);
-                    offset_to += bytes_to_copy;
-                }
-                data_to[offset_to] = 0;
-                ++offset_to;
-                offsets_to[i] = offset_to;
-                offset_from += n;
-            }
-
-            data_to.resize(offset_to);
-            if (return_type->isNullable() && null_map)
-                return ColumnNullable::create(std::move(col_to), std::move(null_map));
-            return col_to;
-        }
-        else
-            throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                    arguments[0].column->getName(), Name::name);
-    }
-};
-
-
-/// Declared early because used below.
-struct NameToDate { static constexpr auto name = "toDate"; };
-struct NameToDate32 { static constexpr auto name = "toDate32"; };
-struct NameToDateTime { static constexpr auto name = "toDateTime"; };
-struct NameToDateTime32 { static constexpr auto name = "toDateTime32"; };
-struct NameToDateTime64 { static constexpr auto name = "toDateTime64"; };
-struct NameToString { static constexpr auto name = "toString"; };
-struct NameToDecimal32 { static constexpr auto name = "toDecimal32"; };
-struct NameToDecimal64 { static constexpr auto name = "toDecimal64"; };
-struct NameToDecimal128 { static constexpr auto name = "toDecimal128"; };
-struct NameToDecimal256 { static constexpr auto name = "toDecimal256"; };
-
-
-#define DEFINE_NAME_TO_INTERVAL(INTERVAL_KIND) \
-    struct NameToInterval ## INTERVAL_KIND \
-    { \
-        static constexpr auto name = "toInterval" #INTERVAL_KIND; \
-        static constexpr auto kind = IntervalKind::Kind::INTERVAL_KIND; \
-    };
-
-DEFINE_NAME_TO_INTERVAL(Nanosecond)
-DEFINE_NAME_TO_INTERVAL(Microsecond)
-DEFINE_NAME_TO_INTERVAL(Millisecond)
-DEFINE_NAME_TO_INTERVAL(Second)
-DEFINE_NAME_TO_INTERVAL(Minute)
-DEFINE_NAME_TO_INTERVAL(Hour)
-DEFINE_NAME_TO_INTERVAL(Day)
-DEFINE_NAME_TO_INTERVAL(Week)
-DEFINE_NAME_TO_INTERVAL(Month)
-DEFINE_NAME_TO_INTERVAL(Quarter)
-DEFINE_NAME_TO_INTERVAL(Year)
-
-#undef DEFINE_NAME_TO_INTERVAL
-
-struct NameParseDateTimeBestEffort;
-struct NameParseDateTimeBestEffortOrZero;
-struct NameParseDateTimeBestEffortOrNull;
-
-template<typename Name, typename ToDataType>
-static inline bool isDateTime64(const ColumnsWithTypeAndName & arguments)
-{
-    if constexpr (std::is_same_v<ToDataType, DataTypeDateTime64>)
-        return true;
-    else if constexpr (std::is_same_v<Name, NameToDateTime> || std::is_same_v<Name, NameParseDateTimeBestEffort>
-        || std::is_same_v<Name, NameParseDateTimeBestEffortOrZero> || std::is_same_v<Name, NameParseDateTimeBestEffortOrNull>)
-    {
-        return (arguments.size() == 2 && isUInt(arguments[1].type)) || arguments.size() == 3;
-    }
-
-    return false;
-}
-
-template <typename ToDataType, typename Name, typename MonotonicityImpl>
-class FunctionConvert : public IFunction
-{
-public:
-    using Monotonic = MonotonicityImpl;
-
-    static constexpr auto name = Name::name;
-    static constexpr bool to_decimal =
-        std::is_same_v<Name, NameToDecimal32> || std::is_same_v<Name, NameToDecimal64>
-         || std::is_same_v<Name, NameToDecimal128> || std::is_same_v<Name, NameToDecimal256>;
-
-    static constexpr bool to_datetime64 = std::is_same_v<ToDataType, DataTypeDateTime64>;
-
-    static constexpr bool to_string_or_fixed_string = std::is_same_v<ToDataType, DataTypeFixedString> ||
-                                                      std::is_same_v<ToDataType, DataTypeString>;
-
-    static constexpr bool to_date_or_datetime = std::is_same_v<ToDataType, DataTypeDate> ||
-                                                std::is_same_v<ToDataType, DataTypeDate32> ||
-                                                std::is_same_v<ToDataType, DataTypeDateTime>;
-
-    static FunctionPtr create(ContextPtr context) { return std::make_shared<FunctionConvert>(context); }
-    static FunctionPtr create() { return std::make_shared<FunctionConvert>(); }
-
-    FunctionConvert() = default;
-    explicit FunctionConvert(ContextPtr context_) : context(context_) {}
-
-    String getName() const override
-    {
-        return name;
-    }
-
-    bool isVariadic() const override { return true; }
-    size_t getNumberOfArguments() const override { return 0; }
-    bool isInjective(const ColumnsWithTypeAndName &) const override { return std::is_same_v<Name, NameToString>; }
-    bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & arguments) const override
-    {
-        /// TODO: We can make more optimizations here.
-        return !(to_date_or_datetime && isNumber(*arguments[0].type));
-    }
-
-    using DefaultReturnTypeGetter = std::function<DataTypePtr(const ColumnsWithTypeAndName &)>;
-    static DataTypePtr getReturnTypeDefaultImplementationForNulls(const ColumnsWithTypeAndName & arguments, const DefaultReturnTypeGetter & getter)
-    {
-        NullPresence null_presence = getNullPresense(arguments);
-
-        if (null_presence.has_null_constant)
-        {
-            return makeNullable(std::make_shared<DataTypeNothing>());
-        }
-        if (null_presence.has_nullable)
-        {
-            auto nested_columns = Block(createBlockWithNestedColumns(arguments));
-            auto return_type = getter(ColumnsWithTypeAndName(nested_columns.begin(), nested_columns.end()));
-            return makeNullable(return_type);
-        }
-
-        return getter(arguments);
-    }
-
-    DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override
-    {
-        auto getter = [&] (const auto & args) { return getReturnTypeImplRemovedNullable(args); };
-        auto res = getReturnTypeDefaultImplementationForNulls(arguments, getter);
-        to_nullable = res->isNullable();
-        checked_return_type = true;
-        return res;
-    }
-
-    DataTypePtr getReturnTypeImplRemovedNullable(const ColumnsWithTypeAndName & arguments) const
-    {
-        FunctionArgumentDescriptors mandatory_args = {{"Value", nullptr, nullptr, nullptr}};
-        FunctionArgumentDescriptors optional_args;
-
-        if constexpr (to_decimal)
-        {
-            mandatory_args.push_back({"scale", static_cast<FunctionArgumentDescriptor::TypeValidator>(&isNativeInteger), &isColumnConst, "const Integer"});
-        }
-
-        if (!to_decimal && isDateTime64<Name, ToDataType>(arguments))
-        {
-            mandatory_args.push_back({"scale", static_cast<FunctionArgumentDescriptor::TypeValidator>(&isNativeInteger), &isColumnConst, "const Integer"});
-        }
-
-        // toString(DateTime or DateTime64, [timezone: String])
-        if ((std::is_same_v<Name, NameToString> && !arguments.empty() && (isDateTime64(arguments[0].type) || isDateTime(arguments[0].type)))
-            // toUnixTimestamp(value[, timezone : String])
-            || std::is_same_v<Name, NameToUnixTimestamp>
-            // toDate(value[, timezone : String])
-            || std::is_same_v<ToDataType, DataTypeDate> // TODO: shall we allow timestamp argument for toDate? DateTime knows nothing about timezones and this argument is ignored below.
-            // toDate32(value[, timezone : String])
-            || std::is_same_v<ToDataType, DataTypeDate32>
-            // toDateTime(value[, timezone: String])
-            || std::is_same_v<ToDataType, DataTypeDateTime>
-            // toDateTime64(value, scale : Integer[, timezone: String])
-            || std::is_same_v<ToDataType, DataTypeDateTime64>)
-        {
-            optional_args.push_back({"timezone", static_cast<FunctionArgumentDescriptor::TypeValidator>(&isString), nullptr, "String"});
-        }
-
-        validateFunctionArgumentTypes(*this, arguments, mandatory_args, optional_args);
-
-        if constexpr (std::is_same_v<ToDataType, DataTypeInterval>)
-        {
-            return std::make_shared<DataTypeInterval>(Name::kind);
-        }
-        else if constexpr (to_decimal)
-        {
-            UInt64 scale = extractToDecimalScale(arguments[1]);
-
-            if constexpr (std::is_same_v<Name, NameToDecimal32>)
-                return createDecimalMaxPrecision<Decimal32>(scale);
-            else if constexpr (std::is_same_v<Name, NameToDecimal64>)
-                return createDecimalMaxPrecision<Decimal64>(scale);
-            else if constexpr (std::is_same_v<Name, NameToDecimal128>)
-                return createDecimalMaxPrecision<Decimal128>(scale);
-            else if constexpr (std::is_same_v<Name, NameToDecimal256>)
-                return createDecimalMaxPrecision<Decimal256>(scale);
-
-            throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected branch in code of conversion function: it is a bug.");
-        }
-        else
-        {
-            // Optional second argument with time zone for DateTime.
-            UInt8 timezone_arg_position = 1;
-            UInt32 scale [[maybe_unused]] = DataTypeDateTime64::default_scale;
-
-            // DateTime64 requires more arguments: scale and timezone. Since timezone is optional, scale should be first.
-            if (isDateTime64<Name, ToDataType>(arguments))
-            {
-                timezone_arg_position += 1;
-                scale = static_cast<UInt32>(arguments[1].column->get64(0));
-
-                if (to_datetime64 || scale != 0) /// toDateTime('xxxx-xx-xx xx:xx:xx', 0) return DateTime
-                    return std::make_shared<DataTypeDateTime64>(scale,
-                        extractTimeZoneNameFromFunctionArguments(arguments, timezone_arg_position, 0, false));
-
-                return std::make_shared<DataTypeDateTime>(extractTimeZoneNameFromFunctionArguments(arguments, timezone_arg_position, 0, false));
-            }
-
-            if constexpr (std::is_same_v<ToDataType, DataTypeDateTime>)
-                return std::make_shared<DataTypeDateTime>(extractTimeZoneNameFromFunctionArguments(arguments, timezone_arg_position, 0, false));
-            else if constexpr (std::is_same_v<ToDataType, DataTypeDateTime64>)
-                throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected branch in code of conversion function: it is a bug.");
-            else
-                return std::make_shared<ToDataType>();
-        }
-    }
-
-    /// Function actually uses default implementation for nulls,
-    /// but we need to know if return type is Nullable or not,
-    /// so we use checked_return_type only to intercept the first call to getReturnTypeImpl(...).
-    bool useDefaultImplementationForNulls() const override
-    {
-        bool to_nullable_string = to_nullable && std::is_same_v<ToDataType, DataTypeString>;
-        return checked_return_type && !to_nullable_string;
-    }
-
-    bool useDefaultImplementationForConstants() const override { return true; }
-    ColumnNumbers getArgumentsThatAreAlwaysConstant() const override
-    {
-        if constexpr (std::is_same_v<ToDataType, DataTypeString>)
-            return {};
-        else if constexpr (std::is_same_v<ToDataType, DataTypeDateTime64>)
-            return {2};
-        return {1};
-    }
-    bool canBeExecutedOnDefaultArguments() const override { return false; }
-
-    ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const override
-    {
-        try
-        {
-            return executeInternal(arguments, result_type, input_rows_count);
-        }
-        catch (Exception & e)
-        {
-            /// More convenient error message.
-            if (e.code() == ErrorCodes::ATTEMPT_TO_READ_AFTER_EOF)
-            {
-                e.addMessage("Cannot parse "
-                    + result_type->getName() + " from "
-                    + arguments[0].type->getName()
-                    + ", because value is too short");
-            }
-            else if (e.code() == ErrorCodes::CANNOT_PARSE_NUMBER
-                || e.code() == ErrorCodes::CANNOT_READ_ARRAY_FROM_TEXT
-                || e.code() == ErrorCodes::CANNOT_PARSE_INPUT_ASSERTION_FAILED
-                || e.code() == ErrorCodes::CANNOT_PARSE_QUOTED_STRING
-                || e.code() == ErrorCodes::CANNOT_PARSE_ESCAPE_SEQUENCE
-                || e.code() == ErrorCodes::CANNOT_PARSE_DATE
-                || e.code() == ErrorCodes::CANNOT_PARSE_DATETIME
-                || e.code() == ErrorCodes::CANNOT_PARSE_UUID
-                || e.code() == ErrorCodes::CANNOT_PARSE_IPV4
-                || e.code() == ErrorCodes::CANNOT_PARSE_IPV6)
-            {
-                e.addMessage("Cannot parse "
-                    + result_type->getName() + " from "
-                    + arguments[0].type->getName());
-            }
-
-            throw;
-        }
-    }
-
-    bool hasInformationAboutMonotonicity() const override
-    {
-        return Monotonic::has();
-    }
-
-    Monotonicity getMonotonicityForRange(const IDataType & type, const Field & left, const Field & right) const override
-    {
-        return Monotonic::get(type, left, right);
-    }
-
-private:
-    ContextPtr context;
-    mutable bool checked_return_type = false;
-    mutable bool to_nullable = false;
-
-    ColumnPtr executeInternal(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const
-    {
-        if (arguments.empty())
-            throw Exception(ErrorCodes::TOO_FEW_ARGUMENTS_FOR_FUNCTION, "Function {} expects at least 1 argument", getName());
-
-        if (result_type->onlyNull())
-            return result_type->createColumnConstWithDefaultValue(input_rows_count);
-
-        const DataTypePtr from_type = removeNullable(arguments[0].type);
-        ColumnPtr result_column;
-
-        [[maybe_unused]] FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior = default_date_time_overflow_behavior;
-
-        if (context)
-            date_time_overflow_behavior = context->getSettingsRef().date_time_overflow_behavior.value;
-
-        auto call = [&](const auto & types, const auto & tag) -> bool
-        {
-            using Types = std::decay_t<decltype(types)>;
-            using LeftDataType = typename Types::LeftType;
-            using RightDataType = typename Types::RightType;
-            using SpecialTag = std::decay_t<decltype(tag)>;
-
-            if constexpr (IsDataTypeDecimal<RightDataType>)
-            {
-                if constexpr (std::is_same_v<RightDataType, DataTypeDateTime64>)
-                {
-                    /// Account for optional timezone argument.
-                    if (arguments.size() != 2 && arguments.size() != 3)
-                        throw Exception(ErrorCodes::TOO_FEW_ARGUMENTS_FOR_FUNCTION, "Function {} expects 2 or 3 arguments for DataTypeDateTime64.", getName());
-                }
-                else if (arguments.size() != 2)
-                {
-                    throw Exception(ErrorCodes::TOO_FEW_ARGUMENTS_FOR_FUNCTION, "Function {} expects 2 arguments for Decimal.", getName());
-                }
-
-                const ColumnWithTypeAndName & scale_column = arguments[1];
-                UInt32 scale = extractToDecimalScale(scale_column);
-
-                switch (date_time_overflow_behavior)
-                {
-                    case FormatSettings::DateTimeOverflowBehavior::Throw:
-                        result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag, FormatSettings::DateTimeOverflowBehavior::Throw>::execute(arguments, result_type, input_rows_count, scale);
-                        break;
-                    case FormatSettings::DateTimeOverflowBehavior::Ignore:
-                        result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag, FormatSettings::DateTimeOverflowBehavior::Ignore>::execute(arguments, result_type, input_rows_count, scale);
-                        break;
-                    case FormatSettings::DateTimeOverflowBehavior::Saturate:
-                        result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag, FormatSettings::DateTimeOverflowBehavior::Saturate>::execute(arguments, result_type, input_rows_count, scale);
-                        break;
-                }
-
-            }
-            else if constexpr (IsDataTypeDateOrDateTime<RightDataType> && std::is_same_v<LeftDataType, DataTypeDateTime64>)
-            {
-                const auto * dt64 = assert_cast<const DataTypeDateTime64 *>(arguments[0].type.get());
-                switch (date_time_overflow_behavior)
-                {
-                    case FormatSettings::DateTimeOverflowBehavior::Throw:
-                        result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag, FormatSettings::DateTimeOverflowBehavior::Throw>::execute(arguments, result_type, input_rows_count, dt64->getScale());
-                        break;
-                    case FormatSettings::DateTimeOverflowBehavior::Ignore:
-                        result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag, FormatSettings::DateTimeOverflowBehavior::Ignore>::execute(arguments, result_type, input_rows_count, dt64->getScale());
-                        break;
-                    case FormatSettings::DateTimeOverflowBehavior::Saturate:
-                        result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag, FormatSettings::DateTimeOverflowBehavior::Saturate>::execute(arguments, result_type, input_rows_count, dt64->getScale());
-                        break;
-                }
-            }
-#define GENERATE_OVERFLOW_MODE_CASE(OVERFLOW_MODE) \
-            case FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE: \
-                result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag, FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE>::execute( \
-                arguments, result_type, input_rows_count); \
-                break;
-
-            else if constexpr (IsDataTypeDecimalOrNumber<LeftDataType> && IsDataTypeDecimalOrNumber<RightDataType>)
-            {
-                using LeftT = typename LeftDataType::FieldType;
-                using RightT = typename RightDataType::FieldType;
-
-                static constexpr bool bad_left =
-                    is_decimal<LeftT> || std::is_floating_point_v<LeftT> || is_big_int_v<LeftT> || is_signed_v<LeftT>;
-                static constexpr bool bad_right =
-                    is_decimal<RightT> || std::is_floating_point_v<RightT> || is_big_int_v<RightT> || is_signed_v<RightT>;
-
-                /// Disallow int vs UUID conversion (but support int vs UInt128 conversion)
-                if constexpr ((bad_left && std::is_same_v<RightDataType, DataTypeUUID>) ||
-                              (bad_right && std::is_same_v<LeftDataType, DataTypeUUID>))
-                {
-                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Wrong UUID conversion");
-                }
-                else
-                {
-                    switch (date_time_overflow_behavior)
-                    {
-                        GENERATE_OVERFLOW_MODE_CASE(Throw)
-                        GENERATE_OVERFLOW_MODE_CASE(Ignore)
-                        GENERATE_OVERFLOW_MODE_CASE(Saturate)
-                    }
-                }
-            }
-            else if constexpr ((IsDataTypeNumber<LeftDataType> || IsDataTypeDateOrDateTime<LeftDataType>)
-                               && IsDataTypeDateOrDateTime<RightDataType>)
-            {
-                switch (date_time_overflow_behavior)
-                {
-                    GENERATE_OVERFLOW_MODE_CASE(Throw)
-                    GENERATE_OVERFLOW_MODE_CASE(Ignore)
-                    GENERATE_OVERFLOW_MODE_CASE(Saturate)
-                }
-            }
-#undef GENERATE_OVERFLOW_MODE_CASE
-            else
-                  result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag>::execute(arguments, result_type, input_rows_count);
-
-            return true;
-        };
-
-        if (isDateTime64<Name, ToDataType>(arguments))
-        {
-            /// For toDateTime('xxxx-xx-xx xx:xx:xx.00', 2[, 'timezone']) we need to it convert to DateTime64
-            const ColumnWithTypeAndName & scale_column = arguments[1];
-            UInt32 scale = extractToDecimalScale(scale_column);
-
-            if (to_datetime64 || scale != 0) /// When scale = 0, the data type is DateTime otherwise the data type is DateTime64
-            {
-                if (!callOnIndexAndDataType<DataTypeDateTime64>(from_type->getTypeId(), call, ConvertDefaultBehaviorTag{}))
-                    throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of argument of function {}",
-                                    arguments[0].type->getName(), getName());
-
-                return result_column;
-            }
-        }
-
-        if constexpr (std::is_same_v<ToDataType, DataTypeString>)
-        {
-            if (from_type->getCustomSerialization())
-                return ConvertImplGenericToString<ColumnString>::execute(arguments, result_type, input_rows_count);
-        }
-
-        bool done = false;
-        if constexpr (to_string_or_fixed_string)
-        {
-            done = callOnIndexAndDataType<ToDataType>(from_type->getTypeId(), call, ConvertDefaultBehaviorTag{});
-        }
-        else
-        {
-            bool cast_ipv4_ipv6_default_on_conversion_error = false;
-            if constexpr (is_any_of<ToDataType, DataTypeIPv4, DataTypeIPv6>)
-                if (context && (cast_ipv4_ipv6_default_on_conversion_error = context->getSettingsRef().cast_ipv4_ipv6_default_on_conversion_error))
-                    done = callOnIndexAndDataType<ToDataType>(from_type->getTypeId(), call, ConvertReturnZeroOnErrorTag{});
-
-            if (!cast_ipv4_ipv6_default_on_conversion_error)
-            {
-                /// We should use ConvertFromStringExceptionMode::Null mode when converting from String (or FixedString)
-                /// to Nullable type, to avoid 'value is too short' error on attempt to parse empty string from NULL values.
-                if (to_nullable && WhichDataType(from_type).isStringOrFixedString())
-                    done = callOnIndexAndDataType<ToDataType>(from_type->getTypeId(), call, ConvertReturnNullOnErrorTag{});
-                else
-                    done = callOnIndexAndDataType<ToDataType>(from_type->getTypeId(), call, ConvertDefaultBehaviorTag{});
-            }
-        }
-
-        if (!done)
-        {
-            /// Generic conversion of any type to String.
-            if (std::is_same_v<ToDataType, DataTypeString>)
-            {
-                return ConvertImplGenericToString<ColumnString>::execute(arguments, result_type, input_rows_count);
-            }
-            else
-                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of argument of function {}",
-                    arguments[0].type->getName(), getName());
-        }
-
-        return result_column;
-    }
-};
-
-
-/** Function toTOrZero (where T is number of date or datetime type):
-  *  try to convert from String to type T through parsing,
-  *  if cannot parse, return default value instead of throwing exception.
-  * Function toTOrNull will return Nullable type with NULL when cannot parse.
-  * NOTE Also need to implement tryToUnixTimestamp with timezone.
-  */
-template <typename ToDataType, typename Name,
-    ConvertFromStringExceptionMode exception_mode,
-    ConvertFromStringParsingMode parsing_mode = ConvertFromStringParsingMode::Normal>
-class FunctionConvertFromString : public IFunction
-{
-public:
-    static constexpr auto name = Name::name;
-    static constexpr bool to_decimal =
-        std::is_same_v<ToDataType, DataTypeDecimal<Decimal32>> ||
-        std::is_same_v<ToDataType, DataTypeDecimal<Decimal64>> ||
-        std::is_same_v<ToDataType, DataTypeDecimal<Decimal128>> ||
-        std::is_same_v<ToDataType, DataTypeDecimal<Decimal256>>;
-
-    static constexpr bool to_datetime64 = std::is_same_v<ToDataType, DataTypeDateTime64>;
-
-    static FunctionPtr create(ContextPtr) { return std::make_shared<FunctionConvertFromString>(); }
-    static FunctionPtr create() { return std::make_shared<FunctionConvertFromString>(); }
-
-    String getName() const override
-    {
-        return name;
-    }
-
-    bool isVariadic() const override { return true; }
-    bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return true; }
-    size_t getNumberOfArguments() const override { return 0; }
-
-    bool useDefaultImplementationForConstants() const override { return true; }
-    bool canBeExecutedOnDefaultArguments() const override { return false; }
-
-    ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {1}; }
-
-    DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override
-    {
-        DataTypePtr res;
-
-        if (isDateTime64<Name, ToDataType>(arguments))
-        {
-            validateFunctionArgumentTypes(*this, arguments,
-                FunctionArgumentDescriptors{{"string", static_cast<FunctionArgumentDescriptor::TypeValidator>(&isStringOrFixedString), nullptr, "String or FixedString"}},
-                // optional
-                FunctionArgumentDescriptors{
-                    {"precision", static_cast<FunctionArgumentDescriptor::TypeValidator>(&isUInt8), isColumnConst, "const UInt8"},
-                    {"timezone", static_cast<FunctionArgumentDescriptor::TypeValidator>(&isStringOrFixedString), isColumnConst, "const String or FixedString"},
-                });
-
-            UInt64 scale = to_datetime64 ? DataTypeDateTime64::default_scale : 0;
-            if (arguments.size() > 1)
-                scale = extractToDecimalScale(arguments[1]);
-            const auto timezone = extractTimeZoneNameFromFunctionArguments(arguments, 2, 0, false);
-
-            res = scale == 0 ? res = std::make_shared<DataTypeDateTime>(timezone) : std::make_shared<DataTypeDateTime64>(scale, timezone);
-        }
-        else
-        {
-            if ((arguments.size() != 1 && arguments.size() != 2) || (to_decimal && arguments.size() != 2))
-                throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH,
-                    "Number of arguments for function {} doesn't match: passed {}, should be 1 or 2. "
-                    "Second argument only make sense for DateTime (time zone, optional) and Decimal (scale).",
-                    getName(), arguments.size());
-
-            if (!isStringOrFixedString(arguments[0].type))
-            {
-                if (this->getName().find("OrZero") != std::string::npos ||
-                    this->getName().find("OrNull") != std::string::npos)
-                    throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of first argument of function {}. "
-                            "Conversion functions with postfix 'OrZero' or 'OrNull' should take String argument",
-                            arguments[0].type->getName(), getName());
-                else
-                    throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of first argument of function {}",
-                            arguments[0].type->getName(), getName());
-            }
-
-            if (arguments.size() == 2)
-            {
-                if constexpr (std::is_same_v<ToDataType, DataTypeDateTime>)
-                {
-                    if (!isString(arguments[1].type))
-                        throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of 2nd argument of function {}",
-                            arguments[1].type->getName(), getName());
-                }
-                else if constexpr (to_decimal)
-                {
-                    if (!isInteger(arguments[1].type))
-                        throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of 2nd argument of function {}",
-                            arguments[1].type->getName(), getName());
-                    if (!arguments[1].column)
-                        throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Second argument for function {} must be constant", getName());
-                }
-                else
-                {
-                    throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH,
-                        "Number of arguments for function {} doesn't match: passed {}, should be 1. "
-                        "Second argument makes sense only for DateTime and Decimal.",
-                        getName(), arguments.size());
-                }
-            }
-
-            if constexpr (std::is_same_v<ToDataType, DataTypeDateTime>)
-                res = std::make_shared<DataTypeDateTime>(extractTimeZoneNameFromFunctionArguments(arguments, 1, 0, false));
-            else if constexpr (std::is_same_v<ToDataType, DataTypeDateTime64>)
-                throw Exception(ErrorCodes::LOGICAL_ERROR, "MaterializedMySQL is a bug.");
-            else if constexpr (to_decimal)
-            {
-                UInt64 scale = extractToDecimalScale(arguments[1]);
-                res = createDecimalMaxPrecision<typename ToDataType::FieldType>(scale);
-                if (!res)
-                    throw Exception(ErrorCodes::LOGICAL_ERROR, "Something wrong with toDecimalNNOrZero() or toDecimalNNOrNull()");
-            }
-            else
-                res = std::make_shared<ToDataType>();
-        }
-
-        if constexpr (exception_mode == ConvertFromStringExceptionMode::Null)
-            res = std::make_shared<DataTypeNullable>(res);
-
-        return res;
-    }
-
-    template <typename ConvertToDataType>
-    ColumnPtr executeInternal(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count, UInt32 scale = 0) const
-    {
-        const IDataType * from_type = arguments[0].type.get();
-
-        if (checkAndGetDataType<DataTypeString>(from_type))
-        {
-            return ConvertThroughParsing<DataTypeString, ConvertToDataType, Name, exception_mode, parsing_mode>::execute(
-                arguments, result_type, input_rows_count, scale);
-        }
-        else if (checkAndGetDataType<DataTypeFixedString>(from_type))
-        {
-            return ConvertThroughParsing<DataTypeFixedString, ConvertToDataType, Name, exception_mode, parsing_mode>::execute(
-                arguments, result_type, input_rows_count, scale);
-        }
-
-        return nullptr;
-    }
-
-    ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const override
-    {
-        ColumnPtr result_column;
-
-        if constexpr (to_decimal)
-            result_column = executeInternal<ToDataType>(arguments, result_type, input_rows_count,
-                assert_cast<const ToDataType &>(*removeNullable(result_type)).getScale());
-        else
-        {
-            if (isDateTime64<Name, ToDataType>(arguments))
-            {
-                UInt64 scale = to_datetime64 ? DataTypeDateTime64::default_scale : 0;
-                if (arguments.size() > 1)
-                    scale = extractToDecimalScale(arguments[1]);
-
-                if (scale == 0)
-                    result_column = executeInternal<DataTypeDateTime>(arguments, result_type, input_rows_count);
-                else
-                {
-                    result_column = executeInternal<DataTypeDateTime64>(arguments, result_type, input_rows_count, static_cast<UInt32>(scale));
-                }
-            }
-            else
-            {
-                result_column = executeInternal<ToDataType>(arguments, result_type, input_rows_count);
-            }
-        }
-
-        if (!result_column)
-            throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of argument of function {}. "
-                "Only String or FixedString argument is accepted for try-conversion function. For other arguments, "
-                "use function without 'orZero' or 'orNull'.", arguments[0].type->getName(), getName());
-
-        return result_column;
-    }
-};
-
-
-/// Monotonicity.
-
-struct PositiveMonotonicity
-{
-    static bool has() { return true; }
-    static IFunction::Monotonicity get(const IDataType &, const Field &, const Field &)
-    {
-        return { .is_monotonic = true };
-    }
-};
-
-struct UnknownMonotonicity
-{
-    static bool has() { return false; }
-    static IFunction::Monotonicity get(const IDataType &, const Field &, const Field &)
-    {
-        return { };
-    }
-};
-
-template <typename T>
-struct ToNumberMonotonicity
-{
-    static bool has() { return true; }
-
-    static UInt64 divideByRangeOfType(UInt64 x)
-    {
-        if constexpr (sizeof(T) < sizeof(UInt64))
-            return x >> (sizeof(T) * 8);
-        else
-            return 0;
-    }
-
-    static IFunction::Monotonicity get(const IDataType & type, const Field & left, const Field & right)
-    {
-        if (!type.isValueRepresentedByNumber())
-            return {};
-
-        /// If type is same, the conversion is always monotonic.
-        /// (Enum has separate case, because it is different data type)
-        if (checkAndGetDataType<DataTypeNumber<T>>(&type) ||
-            checkAndGetDataType<DataTypeEnum<T>>(&type))
-            return { .is_monotonic = true, .is_always_monotonic = true };
-
-        /// Float cases.
-
-        /// When converting to Float, the conversion is always monotonic.
-        if constexpr (std::is_floating_point_v<T>)
-            return { .is_monotonic = true, .is_always_monotonic = true };
-
-        const auto * low_cardinality = typeid_cast<const DataTypeLowCardinality *>(&type);
-        const IDataType * low_cardinality_dictionary_type = nullptr;
-        if (low_cardinality)
-            low_cardinality_dictionary_type = low_cardinality->getDictionaryType().get();
-
-        WhichDataType which_type(type);
-        WhichDataType which_inner_type = low_cardinality
-            ? WhichDataType(low_cardinality_dictionary_type)
-            : WhichDataType(type);
-
-        /// If converting from Float, for monotonicity, arguments must fit in range of result type.
-        if (which_inner_type.isFloat())
-        {
-            if (left.isNull() || right.isNull())
-                return {};
-
-            Float64 left_float = left.get<Float64>();
-            Float64 right_float = right.get<Float64>();
-
-            if (left_float >= static_cast<Float64>(std::numeric_limits<T>::min())
-                && left_float <= static_cast<Float64>(std::numeric_limits<T>::max())
-                && right_float >= static_cast<Float64>(std::numeric_limits<T>::min())
-                && right_float <= static_cast<Float64>(std::numeric_limits<T>::max()))
-                return { .is_monotonic = true };
-
-            return {};
-        }
-
-        /// Integer cases.
-
-        /// Only support types represented by native integers.
-        /// It can be extended to big integers, decimals and DateTime64 later.
-        /// By the way, NULLs are representing unbounded ranges.
-        if (!((left.isNull() || left.getType() == Field::Types::UInt64 || left.getType() == Field::Types::Int64)
-            && (right.isNull() || right.getType() == Field::Types::UInt64 || right.getType() == Field::Types::Int64)))
-            return {};
-
-        const bool from_is_unsigned = type.isValueRepresentedByUnsignedInteger();
-        const bool to_is_unsigned = is_unsigned_v<T>;
-
-        const size_t size_of_from = type.getSizeOfValueInMemory();
-        const size_t size_of_to = sizeof(T);
-
-        const bool left_in_first_half = left.isNull()
-            ? from_is_unsigned
-            : (left.get<Int64>() >= 0);
-
-        const bool right_in_first_half = right.isNull()
-            ? !from_is_unsigned
-            : (right.get<Int64>() >= 0);
-
-        /// Size of type is the same.
-        if (size_of_from == size_of_to)
-        {
-            if (from_is_unsigned == to_is_unsigned)
-                return { .is_monotonic = true, .is_always_monotonic = true };
-
-            if (left_in_first_half == right_in_first_half)
-                return { .is_monotonic = true };
-
-            return {};
-        }
-
-        /// Size of type is expanded.
-        if (size_of_from < size_of_to)
-        {
-            if (from_is_unsigned == to_is_unsigned)
-                return { .is_monotonic = true, .is_always_monotonic = true };
-
-            if (!to_is_unsigned)
-                return { .is_monotonic = true, .is_always_monotonic = true };
-
-            /// signed -> unsigned. If arguments from the same half, then function is monotonic.
-            if (left_in_first_half == right_in_first_half)
-                return { .is_monotonic = true };
-
-            return {};
-        }
-
-        /// Size of type is shrunk.
-        if (size_of_from > size_of_to)
-        {
-            /// Function cannot be monotonic on unbounded ranges.
-            if (left.isNull() || right.isNull())
-                return {};
-
-            /// Function cannot be monotonic when left and right are not on the same ranges.
-            if (divideByRangeOfType(left.get<UInt64>()) != divideByRangeOfType(right.get<UInt64>()))
-                return {};
-
-            if (to_is_unsigned)
-                return { .is_monotonic = true };
-            else
-            {
-                // If To is signed, it's possible that the signedness is different after conversion. So we check it explicitly.
-                const bool is_monotonic = (T(left.get<UInt64>()) >= 0) == (T(right.get<UInt64>()) >= 0);
-
-                return { .is_monotonic = is_monotonic };
-            }
-        }
-
-        UNREACHABLE();
-    }
-};
-
-struct ToDateMonotonicity
-{
-    static bool has() { return true; }
-
-    static IFunction::Monotonicity get(const IDataType & type, const Field & left, const Field & right)
-    {
-        auto which = WhichDataType(type);
-        if (which.isDateOrDate32() || which.isDateTime() || which.isDateTime64() || which.isInt8() || which.isInt16() || which.isUInt8()
-            || which.isUInt16())
-        {
-            return {.is_monotonic = true, .is_always_monotonic = true};
-        }
-        else if (
-            ((left.getType() == Field::Types::UInt64 || left.isNull()) && (right.getType() == Field::Types::UInt64 || right.isNull())
-             && ((left.isNull() || left.get<UInt64>() < 0xFFFF) && (right.isNull() || right.get<UInt64>() >= 0xFFFF)))
-            || ((left.getType() == Field::Types::Int64 || left.isNull()) && (right.getType() == Field::Types::Int64 || right.isNull())
-                && ((left.isNull() || left.get<Int64>() < 0xFFFF) && (right.isNull() || right.get<Int64>() >= 0xFFFF)))
-            || ((
-                (left.getType() == Field::Types::Float64 || left.isNull())
-                && (right.getType() == Field::Types::Float64 || right.isNull())
-                && ((left.isNull() || left.get<Float64>() < 0xFFFF) && (right.isNull() || right.get<Float64>() >= 0xFFFF))))
-            || !isNativeNumber(type))
-        {
-            return {};
-        }
-        else
-        {
-            return {.is_monotonic = true, .is_always_monotonic = true};
-        }
-    }
-};
-
-struct ToDateTimeMonotonicity
-{
-    static bool has() { return true; }
-
-    static IFunction::Monotonicity get(const IDataType & type, const Field &, const Field &)
-    {
-        if (type.isValueRepresentedByNumber())
-            return {.is_monotonic = true, .is_always_monotonic = true};
-        else
-            return {};
-    }
-};
-
-/** The monotonicity for the `toString` function is mainly determined for test purposes.
-  * It is doubtful that anyone is looking to optimize queries with conditions `toString(CounterID) = 34`.
-  */
-struct ToStringMonotonicity
-{
-    static bool has() { return true; }
-
-    static IFunction::Monotonicity get(const IDataType & type, const Field & left, const Field & right)
-    {
-        IFunction::Monotonicity positive{ .is_monotonic = true };
-        IFunction::Monotonicity not_monotonic;
-
-        const auto * type_ptr = &type;
-        if (const auto * low_cardinality_type = checkAndGetDataType<DataTypeLowCardinality>(type_ptr))
-            type_ptr = low_cardinality_type->getDictionaryType().get();
-
-        /// Order on enum values (which is the order on integers) is completely arbitrary in respect to the order on strings.
-        if (WhichDataType(type).isEnum())
-            return not_monotonic;
-
-        /// `toString` function is monotonous if the argument is Date or Date32 or DateTime or String, or non-negative numbers with the same number of symbols.
-        if (checkDataTypes<DataTypeDate, DataTypeDate32, DataTypeDateTime, DataTypeString>(type_ptr))
-            return positive;
-
-        if (left.isNull() || right.isNull())
-            return {};
-
-        if (left.getType() == Field::Types::UInt64
-            && right.getType() == Field::Types::UInt64)
-        {
-            return (left.get<Int64>() == 0 && right.get<Int64>() == 0)
-                || (floor(log10(left.get<UInt64>())) == floor(log10(right.get<UInt64>())))
-                ? positive : not_monotonic;
-        }
-
-        if (left.getType() == Field::Types::Int64
-            && right.getType() == Field::Types::Int64)
-        {
-            return (left.get<Int64>() == 0 && right.get<Int64>() == 0)
-                || (left.get<Int64>() > 0 && right.get<Int64>() > 0 && floor(log10(left.get<Int64>())) == floor(log10(right.get<Int64>())))
-                ? positive : not_monotonic;
-        }
-
-        return not_monotonic;
-    }
-};
-
-
-struct NameToUInt8 { static constexpr auto name = "toUInt8"; };
-struct NameToUInt16 { static constexpr auto name = "toUInt16"; };
-struct NameToUInt32 { static constexpr auto name = "toUInt32"; };
-struct NameToUInt64 { static constexpr auto name = "toUInt64"; };
-struct NameToUInt128 { static constexpr auto name = "toUInt128"; };
-struct NameToUInt256 { static constexpr auto name = "toUInt256"; };
-struct NameToInt8 { static constexpr auto name = "toInt8"; };
-struct NameToInt16 { static constexpr auto name = "toInt16"; };
-struct NameToInt32 { static constexpr auto name = "toInt32"; };
-struct NameToInt64 { static constexpr auto name = "toInt64"; };
-struct NameToInt128 { static constexpr auto name = "toInt128"; };
-struct NameToInt256 { static constexpr auto name = "toInt256"; };
-struct NameToFloat32 { static constexpr auto name = "toFloat32"; };
-struct NameToFloat64 { static constexpr auto name = "toFloat64"; };
-struct NameToUUID { static constexpr auto name = "toUUID"; };
-struct NameToIPv4 { static constexpr auto name = "toIPv4"; };
-struct NameToIPv6 { static constexpr auto name = "toIPv6"; };
-
-using FunctionToUInt8 = FunctionConvert<DataTypeUInt8, NameToUInt8, ToNumberMonotonicity<UInt8>>;
-using FunctionToUInt16 = FunctionConvert<DataTypeUInt16, NameToUInt16, ToNumberMonotonicity<UInt16>>;
-using FunctionToUInt32 = FunctionConvert<DataTypeUInt32, NameToUInt32, ToNumberMonotonicity<UInt32>>;
-using FunctionToUInt64 = FunctionConvert<DataTypeUInt64, NameToUInt64, ToNumberMonotonicity<UInt64>>;
-using FunctionToUInt128 = FunctionConvert<DataTypeUInt128, NameToUInt128, ToNumberMonotonicity<UInt128>>;
-using FunctionToUInt256 = FunctionConvert<DataTypeUInt256, NameToUInt256, ToNumberMonotonicity<UInt256>>;
-using FunctionToInt8 = FunctionConvert<DataTypeInt8, NameToInt8, ToNumberMonotonicity<Int8>>;
-using FunctionToInt16 = FunctionConvert<DataTypeInt16, NameToInt16, ToNumberMonotonicity<Int16>>;
-using FunctionToInt32 = FunctionConvert<DataTypeInt32, NameToInt32, ToNumberMonotonicity<Int32>>;
-using FunctionToInt64 = FunctionConvert<DataTypeInt64, NameToInt64, ToNumberMonotonicity<Int64>>;
-using FunctionToInt128 = FunctionConvert<DataTypeInt128, NameToInt128, ToNumberMonotonicity<Int128>>;
-using FunctionToInt256 = FunctionConvert<DataTypeInt256, NameToInt256, ToNumberMonotonicity<Int256>>;
-using FunctionToFloat32 = FunctionConvert<DataTypeFloat32, NameToFloat32, ToNumberMonotonicity<Float32>>;
-using FunctionToFloat64 = FunctionConvert<DataTypeFloat64, NameToFloat64, ToNumberMonotonicity<Float64>>;
-
-using FunctionToDate = FunctionConvert<DataTypeDate, NameToDate, ToDateMonotonicity>;
-
-using FunctionToDate32 = FunctionConvert<DataTypeDate32, NameToDate32, ToDateMonotonicity>;
-
-using FunctionToDateTime = FunctionConvert<DataTypeDateTime, NameToDateTime, ToDateTimeMonotonicity>;
-
-using FunctionToDateTime32 = FunctionConvert<DataTypeDateTime, NameToDateTime32, ToDateTimeMonotonicity>;
-
-using FunctionToDateTime64 = FunctionConvert<DataTypeDateTime64, NameToDateTime64, ToDateTimeMonotonicity>;
-
-using FunctionToUUID = FunctionConvert<DataTypeUUID, NameToUUID, ToNumberMonotonicity<UInt128>>;
-using FunctionToIPv4 = FunctionConvert<DataTypeIPv4, NameToIPv4, ToNumberMonotonicity<UInt32>>;
-using FunctionToIPv6 = FunctionConvert<DataTypeIPv6, NameToIPv6, ToNumberMonotonicity<UInt128>>;
-using FunctionToString = FunctionConvert<DataTypeString, NameToString, ToStringMonotonicity>;
-using FunctionToUnixTimestamp = FunctionConvert<DataTypeUInt32, NameToUnixTimestamp, ToNumberMonotonicity<UInt32>>;
-using FunctionToDecimal32 = FunctionConvert<DataTypeDecimal<Decimal32>, NameToDecimal32, UnknownMonotonicity>;
-using FunctionToDecimal64 = FunctionConvert<DataTypeDecimal<Decimal64>, NameToDecimal64, UnknownMonotonicity>;
-using FunctionToDecimal128 = FunctionConvert<DataTypeDecimal<Decimal128>, NameToDecimal128, UnknownMonotonicity>;
-using FunctionToDecimal256 = FunctionConvert<DataTypeDecimal<Decimal256>, NameToDecimal256, UnknownMonotonicity>;
-
-template <typename DataType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior = default_date_time_overflow_behavior> struct FunctionTo;
-
-template <> struct FunctionTo<DataTypeUInt8> { using Type = FunctionToUInt8; };
-template <> struct FunctionTo<DataTypeUInt16> { using Type = FunctionToUInt16; };
-template <> struct FunctionTo<DataTypeUInt32> { using Type = FunctionToUInt32; };
-template <> struct FunctionTo<DataTypeUInt64> { using Type = FunctionToUInt64; };
-template <> struct FunctionTo<DataTypeUInt128> { using Type = FunctionToUInt128; };
-template <> struct FunctionTo<DataTypeUInt256> { using Type = FunctionToUInt256; };
-template <> struct FunctionTo<DataTypeInt8> { using Type = FunctionToInt8; };
-template <> struct FunctionTo<DataTypeInt16> { using Type = FunctionToInt16; };
-template <> struct FunctionTo<DataTypeInt32> { using Type = FunctionToInt32; };
-template <> struct FunctionTo<DataTypeInt64> { using Type = FunctionToInt64; };
-template <> struct FunctionTo<DataTypeInt128> { using Type = FunctionToInt128; };
-template <> struct FunctionTo<DataTypeInt256> { using Type = FunctionToInt256; };
-template <> struct FunctionTo<DataTypeFloat32> { using Type = FunctionToFloat32; };
-template <> struct FunctionTo<DataTypeFloat64> { using Type = FunctionToFloat64; };
-
-template <FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct FunctionTo<DataTypeDate, date_time_overflow_behavior> { using Type = FunctionToDate; };
-
-template <FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct FunctionTo<DataTypeDate32, date_time_overflow_behavior> { using Type = FunctionToDate32; };
-
-template <FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct FunctionTo<DataTypeDateTime, date_time_overflow_behavior> { using Type = FunctionToDateTime; };
-
-template <FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct FunctionTo<DataTypeDateTime64, date_time_overflow_behavior> { using Type = FunctionToDateTime64; };
-
-template <> struct FunctionTo<DataTypeUUID> { using Type = FunctionToUUID; };
-template <> struct FunctionTo<DataTypeIPv4> { using Type = FunctionToIPv4; };
-template <> struct FunctionTo<DataTypeIPv6> { using Type = FunctionToIPv6; };
-template <> struct FunctionTo<DataTypeString> { using Type = FunctionToString; };
-template <> struct FunctionTo<DataTypeFixedString> { using Type = FunctionToFixedString; };
-template <> struct FunctionTo<DataTypeDecimal<Decimal32>> { using Type = FunctionToDecimal32; };
-template <> struct FunctionTo<DataTypeDecimal<Decimal64>> { using Type = FunctionToDecimal64; };
-template <> struct FunctionTo<DataTypeDecimal<Decimal128>> { using Type = FunctionToDecimal128; };
-template <> struct FunctionTo<DataTypeDecimal<Decimal256>> { using Type = FunctionToDecimal256; };
-
-template <typename FieldType> struct FunctionTo<DataTypeEnum<FieldType>>
-    : FunctionTo<DataTypeNumber<FieldType>>
-{
-};
-
-struct NameToUInt8OrZero { static constexpr auto name = "toUInt8OrZero"; };
-struct NameToUInt16OrZero { static constexpr auto name = "toUInt16OrZero"; };
-struct NameToUInt32OrZero { static constexpr auto name = "toUInt32OrZero"; };
-struct NameToUInt64OrZero { static constexpr auto name = "toUInt64OrZero"; };
-struct NameToUInt128OrZero { static constexpr auto name = "toUInt128OrZero"; };
-struct NameToUInt256OrZero { static constexpr auto name = "toUInt256OrZero"; };
-struct NameToInt8OrZero { static constexpr auto name = "toInt8OrZero"; };
-struct NameToInt16OrZero { static constexpr auto name = "toInt16OrZero"; };
-struct NameToInt32OrZero { static constexpr auto name = "toInt32OrZero"; };
-struct NameToInt64OrZero { static constexpr auto name = "toInt64OrZero"; };
-struct NameToInt128OrZero { static constexpr auto name = "toInt128OrZero"; };
-struct NameToInt256OrZero { static constexpr auto name = "toInt256OrZero"; };
-struct NameToFloat32OrZero { static constexpr auto name = "toFloat32OrZero"; };
-struct NameToFloat64OrZero { static constexpr auto name = "toFloat64OrZero"; };
-struct NameToDateOrZero { static constexpr auto name = "toDateOrZero"; };
-struct NameToDate32OrZero { static constexpr auto name = "toDate32OrZero"; };
-struct NameToDateTimeOrZero { static constexpr auto name = "toDateTimeOrZero"; };
-struct NameToDateTime64OrZero { static constexpr auto name = "toDateTime64OrZero"; };
-struct NameToDecimal32OrZero { static constexpr auto name = "toDecimal32OrZero"; };
-struct NameToDecimal64OrZero { static constexpr auto name = "toDecimal64OrZero"; };
-struct NameToDecimal128OrZero { static constexpr auto name = "toDecimal128OrZero"; };
-struct NameToDecimal256OrZero { static constexpr auto name = "toDecimal256OrZero"; };
-struct NameToUUIDOrZero { static constexpr auto name = "toUUIDOrZero"; };
-struct NameToIPv4OrZero { static constexpr auto name = "toIPv4OrZero"; };
-struct NameToIPv6OrZero { static constexpr auto name = "toIPv6OrZero"; };
-
-using FunctionToUInt8OrZero = FunctionConvertFromString<DataTypeUInt8, NameToUInt8OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToUInt16OrZero = FunctionConvertFromString<DataTypeUInt16, NameToUInt16OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToUInt32OrZero = FunctionConvertFromString<DataTypeUInt32, NameToUInt32OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToUInt64OrZero = FunctionConvertFromString<DataTypeUInt64, NameToUInt64OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToUInt128OrZero = FunctionConvertFromString<DataTypeUInt128, NameToUInt128OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToUInt256OrZero = FunctionConvertFromString<DataTypeUInt256, NameToUInt256OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToInt8OrZero = FunctionConvertFromString<DataTypeInt8, NameToInt8OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToInt16OrZero = FunctionConvertFromString<DataTypeInt16, NameToInt16OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToInt32OrZero = FunctionConvertFromString<DataTypeInt32, NameToInt32OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToInt64OrZero = FunctionConvertFromString<DataTypeInt64, NameToInt64OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToInt128OrZero = FunctionConvertFromString<DataTypeInt128, NameToInt128OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToInt256OrZero = FunctionConvertFromString<DataTypeInt256, NameToInt256OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToFloat32OrZero = FunctionConvertFromString<DataTypeFloat32, NameToFloat32OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToFloat64OrZero = FunctionConvertFromString<DataTypeFloat64, NameToFloat64OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToDateOrZero = FunctionConvertFromString<DataTypeDate, NameToDateOrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToDate32OrZero = FunctionConvertFromString<DataTypeDate32, NameToDate32OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToDateTimeOrZero = FunctionConvertFromString<DataTypeDateTime, NameToDateTimeOrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToDateTime64OrZero = FunctionConvertFromString<DataTypeDateTime64, NameToDateTime64OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToDecimal32OrZero = FunctionConvertFromString<DataTypeDecimal<Decimal32>, NameToDecimal32OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToDecimal64OrZero = FunctionConvertFromString<DataTypeDecimal<Decimal64>, NameToDecimal64OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToDecimal128OrZero = FunctionConvertFromString<DataTypeDecimal<Decimal128>, NameToDecimal128OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToDecimal256OrZero = FunctionConvertFromString<DataTypeDecimal<Decimal256>, NameToDecimal256OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToUUIDOrZero = FunctionConvertFromString<DataTypeUUID, NameToUUIDOrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToIPv4OrZero = FunctionConvertFromString<DataTypeIPv4, NameToIPv4OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToIPv6OrZero = FunctionConvertFromString<DataTypeIPv6, NameToIPv6OrZero, ConvertFromStringExceptionMode::Zero>;
-
-struct NameToUInt8OrNull { static constexpr auto name = "toUInt8OrNull"; };
-struct NameToUInt16OrNull { static constexpr auto name = "toUInt16OrNull"; };
-struct NameToUInt32OrNull { static constexpr auto name = "toUInt32OrNull"; };
-struct NameToUInt64OrNull { static constexpr auto name = "toUInt64OrNull"; };
-struct NameToUInt128OrNull { static constexpr auto name = "toUInt128OrNull"; };
-struct NameToUInt256OrNull { static constexpr auto name = "toUInt256OrNull"; };
-struct NameToInt8OrNull { static constexpr auto name = "toInt8OrNull"; };
-struct NameToInt16OrNull { static constexpr auto name = "toInt16OrNull"; };
-struct NameToInt32OrNull { static constexpr auto name = "toInt32OrNull"; };
-struct NameToInt64OrNull { static constexpr auto name = "toInt64OrNull"; };
-struct NameToInt128OrNull { static constexpr auto name = "toInt128OrNull"; };
-struct NameToInt256OrNull { static constexpr auto name = "toInt256OrNull"; };
-struct NameToFloat32OrNull { static constexpr auto name = "toFloat32OrNull"; };
-struct NameToFloat64OrNull { static constexpr auto name = "toFloat64OrNull"; };
-struct NameToDateOrNull { static constexpr auto name = "toDateOrNull"; };
-struct NameToDate32OrNull { static constexpr auto name = "toDate32OrNull"; };
-struct NameToDateTimeOrNull { static constexpr auto name = "toDateTimeOrNull"; };
-struct NameToDateTime64OrNull { static constexpr auto name = "toDateTime64OrNull"; };
-struct NameToDecimal32OrNull { static constexpr auto name = "toDecimal32OrNull"; };
-struct NameToDecimal64OrNull { static constexpr auto name = "toDecimal64OrNull"; };
-struct NameToDecimal128OrNull { static constexpr auto name = "toDecimal128OrNull"; };
-struct NameToDecimal256OrNull { static constexpr auto name = "toDecimal256OrNull"; };
-struct NameToUUIDOrNull { static constexpr auto name = "toUUIDOrNull"; };
-struct NameToIPv4OrNull { static constexpr auto name = "toIPv4OrNull"; };
-struct NameToIPv6OrNull { static constexpr auto name = "toIPv6OrNull"; };
-
-using FunctionToUInt8OrNull = FunctionConvertFromString<DataTypeUInt8, NameToUInt8OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToUInt16OrNull = FunctionConvertFromString<DataTypeUInt16, NameToUInt16OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToUInt32OrNull = FunctionConvertFromString<DataTypeUInt32, NameToUInt32OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToUInt64OrNull = FunctionConvertFromString<DataTypeUInt64, NameToUInt64OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToUInt128OrNull = FunctionConvertFromString<DataTypeUInt128, NameToUInt128OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToUInt256OrNull = FunctionConvertFromString<DataTypeUInt256, NameToUInt256OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToInt8OrNull = FunctionConvertFromString<DataTypeInt8, NameToInt8OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToInt16OrNull = FunctionConvertFromString<DataTypeInt16, NameToInt16OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToInt32OrNull = FunctionConvertFromString<DataTypeInt32, NameToInt32OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToInt64OrNull = FunctionConvertFromString<DataTypeInt64, NameToInt64OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToInt128OrNull = FunctionConvertFromString<DataTypeInt128, NameToInt128OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToInt256OrNull = FunctionConvertFromString<DataTypeInt256, NameToInt256OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToFloat32OrNull = FunctionConvertFromString<DataTypeFloat32, NameToFloat32OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToFloat64OrNull = FunctionConvertFromString<DataTypeFloat64, NameToFloat64OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToDateOrNull = FunctionConvertFromString<DataTypeDate, NameToDateOrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToDate32OrNull = FunctionConvertFromString<DataTypeDate32, NameToDate32OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToDateTimeOrNull = FunctionConvertFromString<DataTypeDateTime, NameToDateTimeOrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToDateTime64OrNull = FunctionConvertFromString<DataTypeDateTime64, NameToDateTime64OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToDecimal32OrNull = FunctionConvertFromString<DataTypeDecimal<Decimal32>, NameToDecimal32OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToDecimal64OrNull = FunctionConvertFromString<DataTypeDecimal<Decimal64>, NameToDecimal64OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToDecimal128OrNull = FunctionConvertFromString<DataTypeDecimal<Decimal128>, NameToDecimal128OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToDecimal256OrNull = FunctionConvertFromString<DataTypeDecimal<Decimal256>, NameToDecimal256OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToUUIDOrNull = FunctionConvertFromString<DataTypeUUID, NameToUUIDOrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToIPv4OrNull = FunctionConvertFromString<DataTypeIPv4, NameToIPv4OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToIPv6OrNull = FunctionConvertFromString<DataTypeIPv6, NameToIPv6OrNull, ConvertFromStringExceptionMode::Null>;
-
-struct NameParseDateTimeBestEffort { static constexpr auto name = "parseDateTimeBestEffort"; };
-struct NameParseDateTimeBestEffortOrZero { static constexpr auto name = "parseDateTimeBestEffortOrZero"; };
-struct NameParseDateTimeBestEffortOrNull { static constexpr auto name = "parseDateTimeBestEffortOrNull"; };
-struct NameParseDateTimeBestEffortUS { static constexpr auto name = "parseDateTimeBestEffortUS"; };
-struct NameParseDateTimeBestEffortUSOrZero { static constexpr auto name = "parseDateTimeBestEffortUSOrZero"; };
-struct NameParseDateTimeBestEffortUSOrNull { static constexpr auto name = "parseDateTimeBestEffortUSOrNull"; };
-struct NameParseDateTime32BestEffort { static constexpr auto name = "parseDateTime32BestEffort"; };
-struct NameParseDateTime32BestEffortOrZero { static constexpr auto name = "parseDateTime32BestEffortOrZero"; };
-struct NameParseDateTime32BestEffortOrNull { static constexpr auto name = "parseDateTime32BestEffortOrNull"; };
-struct NameParseDateTime64BestEffort { static constexpr auto name = "parseDateTime64BestEffort"; };
-struct NameParseDateTime64BestEffortOrZero { static constexpr auto name = "parseDateTime64BestEffortOrZero"; };
-struct NameParseDateTime64BestEffortOrNull { static constexpr auto name = "parseDateTime64BestEffortOrNull"; };
-struct NameParseDateTime64BestEffortUS { static constexpr auto name = "parseDateTime64BestEffortUS"; };
-struct NameParseDateTime64BestEffortUSOrZero { static constexpr auto name = "parseDateTime64BestEffortUSOrZero"; };
-struct NameParseDateTime64BestEffortUSOrNull { static constexpr auto name = "parseDateTime64BestEffortUSOrNull"; };
-
-
-using FunctionParseDateTimeBestEffort = FunctionConvertFromString<
-    DataTypeDateTime, NameParseDateTimeBestEffort, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::BestEffort>;
-using FunctionParseDateTimeBestEffortOrZero = FunctionConvertFromString<
-    DataTypeDateTime, NameParseDateTimeBestEffortOrZero, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::BestEffort>;
-using FunctionParseDateTimeBestEffortOrNull = FunctionConvertFromString<
-    DataTypeDateTime, NameParseDateTimeBestEffortOrNull, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::BestEffort>;
-
-using FunctionParseDateTimeBestEffortUS = FunctionConvertFromString<
-    DataTypeDateTime, NameParseDateTimeBestEffortUS, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::BestEffortUS>;
-using FunctionParseDateTimeBestEffortUSOrZero = FunctionConvertFromString<
-    DataTypeDateTime, NameParseDateTimeBestEffortUSOrZero, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::BestEffortUS>;
-using FunctionParseDateTimeBestEffortUSOrNull = FunctionConvertFromString<
-    DataTypeDateTime, NameParseDateTimeBestEffortUSOrNull, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::BestEffortUS>;
-
-using FunctionParseDateTime32BestEffort = FunctionConvertFromString<
-    DataTypeDateTime, NameParseDateTime32BestEffort, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::BestEffort>;
-using FunctionParseDateTime32BestEffortOrZero = FunctionConvertFromString<
-    DataTypeDateTime, NameParseDateTime32BestEffortOrZero, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::BestEffort>;
-using FunctionParseDateTime32BestEffortOrNull = FunctionConvertFromString<
-    DataTypeDateTime, NameParseDateTime32BestEffortOrNull, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::BestEffort>;
-
-using FunctionParseDateTime64BestEffort = FunctionConvertFromString<
-    DataTypeDateTime64, NameParseDateTime64BestEffort, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::BestEffort>;
-using FunctionParseDateTime64BestEffortOrZero = FunctionConvertFromString<
-    DataTypeDateTime64, NameParseDateTime64BestEffortOrZero, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::BestEffort>;
-using FunctionParseDateTime64BestEffortOrNull = FunctionConvertFromString<
-    DataTypeDateTime64, NameParseDateTime64BestEffortOrNull, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::BestEffort>;
-
-using FunctionParseDateTime64BestEffortUS = FunctionConvertFromString<
-    DataTypeDateTime64, NameParseDateTime64BestEffortUS, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::BestEffortUS>;
-using FunctionParseDateTime64BestEffortUSOrZero = FunctionConvertFromString<
-    DataTypeDateTime64, NameParseDateTime64BestEffortUSOrZero, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::BestEffortUS>;
-using FunctionParseDateTime64BestEffortUSOrNull = FunctionConvertFromString<
-    DataTypeDateTime64, NameParseDateTime64BestEffortUSOrNull, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::BestEffortUS>;
-
-
-class ExecutableFunctionCast : public IExecutableFunction
-{
-public:
-    using WrapperType = std::function<ColumnPtr(ColumnsWithTypeAndName &, const DataTypePtr &, const ColumnNullable *, size_t)>;
-
-    explicit ExecutableFunctionCast(
-            WrapperType && wrapper_function_, const char * name_, std::optional<CastDiagnostic> diagnostic_)
-            : wrapper_function(std::move(wrapper_function_)), name(name_), diagnostic(std::move(diagnostic_)) {}
-
-    String getName() const override { return name; }
-
-protected:
-    ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const override
-    {
-        /// drop second argument, pass others
-        ColumnsWithTypeAndName new_arguments{arguments.front()};
-        if (arguments.size() > 2)
-            new_arguments.insert(std::end(new_arguments), std::next(std::begin(arguments), 2), std::end(arguments));
-
-        try
-        {
-            return wrapper_function(new_arguments, result_type, nullptr, input_rows_count);
-        }
-        catch (Exception & e)
-        {
-            if (diagnostic)
-                e.addMessage("while converting source column " + backQuoteIfNeed(diagnostic->column_from) +
-                             " to destination column " + backQuoteIfNeed(diagnostic->column_to));
-            throw;
-        }
-    }
-
-    bool useDefaultImplementationForNulls() const override { return false; }
-    /// CAST(Nothing, T) -> T
-    bool useDefaultImplementationForNothing() const override { return false; }
-    bool useDefaultImplementationForConstants() const override { return true; }
-    bool useDefaultImplementationForLowCardinalityColumns() const override { return false; }
-    ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {1}; }
-
-private:
-    WrapperType wrapper_function;
-    const char * name;
-    std::optional<CastDiagnostic> diagnostic;
-};
-
-
-struct FunctionCastName
-{
-    static constexpr auto name = "CAST";
-};
-
-class FunctionCastBase : public IFunctionBase
-{
-public:
-    using MonotonicityForRange = std::function<Monotonicity(const IDataType &, const Field &, const Field &)>;
-};
-
-class FunctionCast final : public FunctionCastBase
-{
-public:
-    using WrapperType = std::function<ColumnPtr(ColumnsWithTypeAndName &, const DataTypePtr &, const ColumnNullable *, size_t)>;
-
-    FunctionCast(ContextPtr context_
-            , const char * cast_name_
-            , MonotonicityForRange && monotonicity_for_range_
-            , const DataTypes & argument_types_
-            , const DataTypePtr & return_type_
-            , std::optional<CastDiagnostic> diagnostic_
-            , CastType cast_type_)
-        : cast_name(cast_name_), monotonicity_for_range(std::move(monotonicity_for_range_))
-        , argument_types(argument_types_), return_type(return_type_), diagnostic(std::move(diagnostic_))
-        , cast_type(cast_type_)
-        , context(context_)
-    {
-    }
-
-    const DataTypes & getArgumentTypes() const override { return argument_types; }
-    const DataTypePtr & getResultType() const override { return return_type; }
-
-    ExecutableFunctionPtr prepare(const ColumnsWithTypeAndName & /*sample_columns*/) const override
-    {
-        try
-        {
-            return std::make_unique<ExecutableFunctionCast>(
-                prepareUnpackDictionaries(getArgumentTypes()[0], getResultType()), cast_name, diagnostic);
-        }
-        catch (Exception & e)
-        {
-            if (diagnostic)
-                e.addMessage("while converting source column " + backQuoteIfNeed(diagnostic->column_from) +
-                             " to destination column " + backQuoteIfNeed(diagnostic->column_to));
-            throw;
-        }
-    }
-
-    String getName() const override { return cast_name; }
-
-    bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return true; }
-
-    bool hasInformationAboutMonotonicity() const override
-    {
-        return static_cast<bool>(monotonicity_for_range);
-    }
-
-    Monotonicity getMonotonicityForRange(const IDataType & type, const Field & left, const Field & right) const override
-    {
-        return monotonicity_for_range(type, left, right);
-    }
-
-private:
-
-    const char * cast_name;
-    MonotonicityForRange monotonicity_for_range;
-
-    DataTypes argument_types;
-    DataTypePtr return_type;
-
-    std::optional<CastDiagnostic> diagnostic;
-    CastType cast_type;
-    ContextPtr context;
-
-    static WrapperType createFunctionAdaptor(FunctionPtr function, const DataTypePtr & from_type)
-    {
-        auto function_adaptor = std::make_unique<FunctionToOverloadResolverAdaptor>(function)->build({ColumnWithTypeAndName{nullptr, from_type, ""}});
-
-        return [function_adaptor]
-            (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count)
-        {
-            return function_adaptor->execute(arguments, result_type, input_rows_count);
-        };
-    }
-
-    static WrapperType createToNullableColumnWrapper()
-    {
-        return [] (ColumnsWithTypeAndName &, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count)
-        {
-            ColumnPtr res = result_type->createColumn();
-            ColumnUInt8::Ptr col_null_map_to = ColumnUInt8::create(input_rows_count, true);
-            return ColumnNullable::create(res->cloneResized(input_rows_count), std::move(col_null_map_to));
-        };
-    }
-
-    template <typename ToDataType>
-    WrapperType createWrapper(const DataTypePtr & from_type, const ToDataType * const to_type, bool requested_result_is_nullable) const
-    {
-        TypeIndex from_type_index = from_type->getTypeId();
-        WhichDataType which(from_type_index);
-        bool can_apply_accurate_cast = (cast_type == CastType::accurate || cast_type == CastType::accurateOrNull)
-            && (which.isInt() || which.isUInt() || which.isFloat());
-
-        FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior = default_date_time_overflow_behavior;
-        if (context)
-            date_time_overflow_behavior = context->getSettingsRef().date_time_overflow_behavior;
-
-        if (requested_result_is_nullable && checkAndGetDataType<DataTypeString>(from_type.get()))
-        {
-            /// In case when converting to Nullable type, we apply different parsing rule,
-            /// that will not throw an exception but return NULL in case of malformed input.
-            FunctionPtr function = FunctionConvertFromString<ToDataType, FunctionCastName, ConvertFromStringExceptionMode::Null>::create();
-            return createFunctionAdaptor(function, from_type);
-        }
-        else if (!can_apply_accurate_cast)
-        {
-            FunctionPtr function = FunctionTo<ToDataType>::Type::create(context);
-            return createFunctionAdaptor(function, from_type);
-        }
-
-        auto wrapper_cast_type = cast_type;
-
-        return [wrapper_cast_type, from_type_index, to_type, date_time_overflow_behavior]
-            (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *column_nullable, size_t input_rows_count)
-        {
-            ColumnPtr result_column;
-            auto res = callOnIndexAndDataType<ToDataType>(from_type_index, [&](const auto & types) -> bool {
-                using Types = std::decay_t<decltype(types)>;
-                using LeftDataType = typename Types::LeftType;
-                using RightDataType = typename Types::RightType;
-
-                if constexpr (IsDataTypeNumber<LeftDataType>)
-                {
-                    if constexpr (IsDataTypeNumber<RightDataType>)
-                    {
-#define GENERATE_OVERFLOW_MODE_CASE(OVERFLOW_MODE, ADDITIONS) \
-            case FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE: \
-                result_column = ConvertImpl<LeftDataType, RightDataType, FunctionCastName, ConvertDefaultBehaviorTag, FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE>::execute( \
-                arguments, result_type, input_rows_count, ADDITIONS()); \
-                break;
-                        if (wrapper_cast_type == CastType::accurate)
-                        {
-                            switch (date_time_overflow_behavior)
-                            {
-                                GENERATE_OVERFLOW_MODE_CASE(Throw, AccurateConvertStrategyAdditions)
-                                GENERATE_OVERFLOW_MODE_CASE(Ignore, AccurateConvertStrategyAdditions)
-                                GENERATE_OVERFLOW_MODE_CASE(Saturate, AccurateConvertStrategyAdditions)
-                            }
-                        }
-                        else
-                        {
-                            switch (date_time_overflow_behavior)
-                            {
-                                GENERATE_OVERFLOW_MODE_CASE(Throw, AccurateOrNullConvertStrategyAdditions)
-                                GENERATE_OVERFLOW_MODE_CASE(Ignore, AccurateOrNullConvertStrategyAdditions)
-                                GENERATE_OVERFLOW_MODE_CASE(Saturate, AccurateOrNullConvertStrategyAdditions)
-                            }
-                        }
-#undef GENERATE_OVERFLOW_MODE_CASE
-
-                        return true;
-                    }
-
-                    if constexpr (std::is_same_v<RightDataType, DataTypeDate> || std::is_same_v<RightDataType, DataTypeDateTime>)
-                    {
-#define GENERATE_OVERFLOW_MODE_CASE(OVERFLOW_MODE, ADDITIONS) \
-            case FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE: \
-            result_column = ConvertImpl<LeftDataType, RightDataType, FunctionCastName, ConvertDefaultBehaviorTag, FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE>::template execute<ADDITIONS>( \
-arguments, result_type, input_rows_count); \
-                break;
-                        if (wrapper_cast_type == CastType::accurate)
-                        {
-                            switch (date_time_overflow_behavior)
-                            {
-                                GENERATE_OVERFLOW_MODE_CASE(Throw, DateTimeAccurateConvertStrategyAdditions)
-                                GENERATE_OVERFLOW_MODE_CASE(Ignore, DateTimeAccurateConvertStrategyAdditions)
-                                GENERATE_OVERFLOW_MODE_CASE(Saturate, DateTimeAccurateConvertStrategyAdditions)
-                            }
-                        }
-                        else
-                        {
-                            switch (date_time_overflow_behavior)
-                            {
-                                GENERATE_OVERFLOW_MODE_CASE(Throw, DateTimeAccurateOrNullConvertStrategyAdditions)
-                                GENERATE_OVERFLOW_MODE_CASE(Ignore, DateTimeAccurateOrNullConvertStrategyAdditions)
-                                GENERATE_OVERFLOW_MODE_CASE(Saturate, DateTimeAccurateOrNullConvertStrategyAdditions)
-                            }
-                        }
-#undef GENERATE_OVERFLOW_MODE_CASE
-                        return true;
-                    }
-                }
-
-                return false;
-            });
-
-            /// Additionally check if callOnIndexAndDataType wasn't called at all.
-            if (!res)
-            {
-                if (wrapper_cast_type == CastType::accurateOrNull)
-                {
-                    auto nullable_column_wrapper = FunctionCast::createToNullableColumnWrapper();
-                    return nullable_column_wrapper(arguments, result_type, column_nullable, input_rows_count);
-                }
-                else
-                {
-                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE,
-                        "Conversion from {} to {} is not supported",
-                        from_type_index, to_type->getName());
-                }
-            }
-
-            return result_column;
-        };
-    }
-
-    template <typename ToDataType>
-    WrapperType createBoolWrapper(const DataTypePtr & from_type, const ToDataType * const to_type, bool requested_result_is_nullable) const
-    {
-        if (checkAndGetDataType<DataTypeString>(from_type.get()))
-        {
-            return &ConvertImplGenericFromString::execute;
-        }
-
-        return createWrapper<ToDataType>(from_type, to_type, requested_result_is_nullable);
-    }
-
-    WrapperType createUInt8ToBoolWrapper(const DataTypePtr from_type, const DataTypePtr to_type) const
-    {
-        return [from_type, to_type] (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable *, size_t /*input_rows_count*/) -> ColumnPtr
-        {
-            /// Special case when we convert UInt8 column to Bool column.
-            /// both columns have type UInt8, but we shouldn't use identity wrapper,
-            /// because Bool column can contain only 0 and 1.
-            auto res_column = to_type->createColumn();
-            const auto & data_from = checkAndGetColumn<ColumnUInt8>(arguments[0].column.get())->getData();
-            auto & data_to = assert_cast<ColumnUInt8 *>(res_column.get())->getData();
-            data_to.resize(data_from.size());
-            for (size_t i = 0; i != data_from.size(); ++i)
-                data_to[i] = static_cast<bool>(data_from[i]);
-            return res_column;
-        };
-    }
-
-    static WrapperType createStringWrapper(const DataTypePtr & from_type)
-    {
-        FunctionPtr function = FunctionToString::create();
-        return createFunctionAdaptor(function, from_type);
-    }
-
-    WrapperType createFixedStringWrapper(const DataTypePtr & from_type, const size_t N) const
-    {
-        if (!isStringOrFixedString(from_type))
-            throw Exception(ErrorCodes::NOT_IMPLEMENTED, "CAST AS FixedString is only implemented for types String and FixedString");
-
-        bool exception_mode_null = cast_type == CastType::accurateOrNull;
-        return [exception_mode_null, N] (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable *, size_t /*input_rows_count*/)
-        {
-            if (exception_mode_null)
-                return FunctionToFixedString::executeForN<ConvertToFixedStringExceptionMode::Null>(arguments, N);
-            else
-                return FunctionToFixedString::executeForN<ConvertToFixedStringExceptionMode::Throw>(arguments, N);
-        };
-    }
-
-#define GENERATE_INTERVAL_CASE(INTERVAL_KIND) \
-            case IntervalKind::Kind::INTERVAL_KIND: \
-                return createFunctionAdaptor(FunctionConvert<DataTypeInterval, NameToInterval##INTERVAL_KIND, PositiveMonotonicity>::create(), from_type);
-
-    static WrapperType createIntervalWrapper(const DataTypePtr & from_type, IntervalKind kind)
-    {
-        switch (kind)
-        {
-            GENERATE_INTERVAL_CASE(Nanosecond)
-            GENERATE_INTERVAL_CASE(Microsecond)
-            GENERATE_INTERVAL_CASE(Millisecond)
-            GENERATE_INTERVAL_CASE(Second)
-            GENERATE_INTERVAL_CASE(Minute)
-            GENERATE_INTERVAL_CASE(Hour)
-            GENERATE_INTERVAL_CASE(Day)
-            GENERATE_INTERVAL_CASE(Week)
-            GENERATE_INTERVAL_CASE(Month)
-            GENERATE_INTERVAL_CASE(Quarter)
-            GENERATE_INTERVAL_CASE(Year)
-        }
-        throw Exception{ErrorCodes::CANNOT_CONVERT_TYPE, "Conversion to unexpected IntervalKind: {}", kind.toString()};
-    }
-
-#undef GENERATE_INTERVAL_CASE
-
-    template <typename ToDataType>
-    requires IsDataTypeDecimal<ToDataType>
-    WrapperType createDecimalWrapper(const DataTypePtr & from_type, const ToDataType * to_type, bool requested_result_is_nullable) const
-    {
-        TypeIndex type_index = from_type->getTypeId();
-        UInt32 scale = to_type->getScale();
-
-        WhichDataType which(type_index);
-        bool ok = which.isNativeInt() || which.isNativeUInt() || which.isDecimal() || which.isFloat() || which.isDateOrDate32() || which.isDateTime() || which.isDateTime64()
-            || which.isStringOrFixedString();
-        if (!ok)
-        {
-            if (cast_type == CastType::accurateOrNull)
-                return createToNullableColumnWrapper();
-            else
-                throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Conversion from {} to {} is not supported",
-                    from_type->getName(), to_type->getName());
-        }
-
-        auto wrapper_cast_type = cast_type;
-
-        return [wrapper_cast_type, type_index, scale, to_type, requested_result_is_nullable]
-            (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *column_nullable, size_t input_rows_count)
-        {
-            ColumnPtr result_column;
-            auto res = callOnIndexAndDataType<ToDataType>(type_index, [&](const auto & types) -> bool
-            {
-                using Types = std::decay_t<decltype(types)>;
-                using LeftDataType = typename Types::LeftType;
-                using RightDataType = typename Types::RightType;
-
-                if constexpr (IsDataTypeDecimalOrNumber<LeftDataType> && IsDataTypeDecimalOrNumber<RightDataType> && !std::is_same_v<DataTypeDateTime64, RightDataType>)
-                {
-                    if (wrapper_cast_type == CastType::accurate)
-                    {
-                        AccurateConvertStrategyAdditions additions;
-                        additions.scale = scale;
-                        result_column = ConvertImpl<LeftDataType, RightDataType, FunctionCastName>::execute(
-                            arguments, result_type, input_rows_count, additions);
-
-                        return true;
-                    }
-                    else if (wrapper_cast_type == CastType::accurateOrNull)
-                    {
-                        AccurateOrNullConvertStrategyAdditions additions;
-                        additions.scale = scale;
-                        result_column = ConvertImpl<LeftDataType, RightDataType, FunctionCastName>::execute(
-                            arguments, result_type, input_rows_count, additions);
-
-                        return true;
-                    }
-                }
-                else if constexpr (std::is_same_v<LeftDataType, DataTypeString>)
-                {
-                    if (requested_result_is_nullable)
-                    {
-                        /// Consistent with CAST(Nullable(String) AS Nullable(Numbers))
-                        /// In case when converting to Nullable type, we apply different parsing rule,
-                        /// that will not throw an exception but return NULL in case of malformed input.
-                        result_column = ConvertImpl<LeftDataType, RightDataType, FunctionCastName, ConvertReturnNullOnErrorTag>::execute(
-                            arguments, result_type, input_rows_count, scale);
-
-                        return true;
-                    }
-                }
-
-                result_column = ConvertImpl<LeftDataType, RightDataType, FunctionCastName>::execute(arguments, result_type, input_rows_count, scale);
-
-                return true;
-            });
-
-            /// Additionally check if callOnIndexAndDataType wasn't called at all.
-            if (!res)
-            {
-                if (wrapper_cast_type == CastType::accurateOrNull)
-                {
-                    auto nullable_column_wrapper = FunctionCast::createToNullableColumnWrapper();
-                    return nullable_column_wrapper(arguments, result_type, column_nullable, input_rows_count);
-                }
-                else
-                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE,
-                        "Conversion from {} to {} is not supported",
-                        type_index, to_type->getName());
-            }
-
-            return result_column;
-        };
-    }
-
-    WrapperType createAggregateFunctionWrapper(const DataTypePtr & from_type_untyped, const DataTypeAggregateFunction * to_type) const
-    {
-        /// Conversion from String through parsing.
-        if (checkAndGetDataType<DataTypeString>(from_type_untyped.get()))
-        {
-            return &ConvertImplGenericFromString::execute;
-        }
-        else if (const auto * agg_type = checkAndGetDataType<DataTypeAggregateFunction>(from_type_untyped.get()))
-        {
-            if (agg_type->getFunction()->haveSameStateRepresentation(*to_type->getFunction()))
-            {
-                return [function = to_type->getFunction()](
-                           ColumnsWithTypeAndName & arguments,
-                           const DataTypePtr & /* result_type */,
-                           const ColumnNullable * /* nullable_source */,
-                           size_t /*input_rows_count*/) -> ColumnPtr
-                {
-                    const auto & argument_column = arguments.front();
-                    const auto * col_agg = checkAndGetColumn<ColumnAggregateFunction>(argument_column.column.get());
-                    if (col_agg)
-                    {
-                        auto new_col_agg = ColumnAggregateFunction::create(*col_agg);
-                        new_col_agg->set(function);
-                        return new_col_agg;
-                    }
-                    else
-                    {
-                        throw Exception(
-                            ErrorCodes::LOGICAL_ERROR,
-                            "Illegal column {} for function CAST AS AggregateFunction",
-                            argument_column.column->getName());
-                    }
-                };
-            }
-        }
-
-        if (cast_type == CastType::accurateOrNull)
-            return createToNullableColumnWrapper();
-        else
-            throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Conversion from {} to {} is not supported",
-                from_type_untyped->getName(), to_type->getName());
-    }
-
-    WrapperType createArrayWrapper(const DataTypePtr & from_type_untyped, const DataTypeArray & to_type) const
-    {
-        /// Conversion from String through parsing.
-        if (checkAndGetDataType<DataTypeString>(from_type_untyped.get()))
-        {
-            return &ConvertImplGenericFromString::execute;
-        }
-
-        DataTypePtr from_type_holder;
-        const auto * from_type = checkAndGetDataType<DataTypeArray>(from_type_untyped.get());
-        const auto * from_type_map = checkAndGetDataType<DataTypeMap>(from_type_untyped.get());
-
-        /// Convert from Map
-        if (from_type_map)
-        {
-            /// Recreate array of unnamed tuples because otherwise it may work
-            /// unexpectedly while converting to array of named tuples.
-            from_type_holder = from_type_map->getNestedTypeWithUnnamedTuple();
-            from_type = assert_cast<const DataTypeArray *>(from_type_holder.get());
-        }
-
-        if (!from_type)
-        {
-            throw Exception(ErrorCodes::TYPE_MISMATCH,
-                "CAST AS Array can only be performed between same-dimensional Array, Map or String types");
-        }
-
-        DataTypePtr from_nested_type = from_type->getNestedType();
-
-        /// In query SELECT CAST([] AS Array(Array(String))) from type is Array(Nothing)
-        bool from_empty_array = isNothing(from_nested_type);
-
-        if (from_type->getNumberOfDimensions() != to_type.getNumberOfDimensions() && !from_empty_array)
-            throw Exception(ErrorCodes::TYPE_MISMATCH,
-                "CAST AS Array can only be performed between same-dimensional array types");
-
-        const DataTypePtr & to_nested_type = to_type.getNestedType();
-
-        /// Prepare nested type conversion
-        const auto nested_function = prepareUnpackDictionaries(from_nested_type, to_nested_type);
-
-        return [nested_function, from_nested_type, to_nested_type](
-                ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * nullable_source, size_t /*input_rows_count*/) -> ColumnPtr
-        {
-            const auto & argument_column = arguments.front();
-
-            const ColumnArray * col_array = nullptr;
-
-            if (const ColumnMap * col_map = checkAndGetColumn<ColumnMap>(argument_column.column.get()))
-                col_array = &col_map->getNestedColumn();
-            else
-                col_array = checkAndGetColumn<ColumnArray>(argument_column.column.get());
-
-            if (col_array)
-            {
-                /// create columns for converting nested column containing original and result columns
-                ColumnsWithTypeAndName nested_columns{{ col_array->getDataPtr(), from_nested_type, "" }};
-
-                /// convert nested column
-                auto result_column = nested_function(nested_columns, to_nested_type, nullable_source, nested_columns.front().column->size());
-
-                /// set converted nested column to result
-                return ColumnArray::create(result_column, col_array->getOffsetsPtr());
-            }
-            else
-            {
-                throw Exception(ErrorCodes::LOGICAL_ERROR,
-                    "Illegal column {} for function CAST AS Array",
-                    argument_column.column->getName());
-            }
-        };
-    }
-
-    using ElementWrappers = std::vector<WrapperType>;
-
-    ElementWrappers getElementWrappers(const DataTypes & from_element_types, const DataTypes & to_element_types) const
-    {
-        ElementWrappers element_wrappers;
-        element_wrappers.reserve(from_element_types.size());
-
-        /// Create conversion wrapper for each element in tuple
-        for (size_t i = 0; i < from_element_types.size(); ++i)
-        {
-            const DataTypePtr & from_element_type = from_element_types[i];
-            const DataTypePtr & to_element_type = to_element_types[i];
-            element_wrappers.push_back(prepareUnpackDictionaries(from_element_type, to_element_type));
-        }
-
-        return element_wrappers;
-    }
-
-    WrapperType createTupleWrapper(const DataTypePtr & from_type_untyped, const DataTypeTuple * to_type) const
-    {
-        /// Conversion from String through parsing.
-        if (checkAndGetDataType<DataTypeString>(from_type_untyped.get()))
-        {
-            return &ConvertImplGenericFromString::execute;
-        }
-
-        const auto * from_type = checkAndGetDataType<DataTypeTuple>(from_type_untyped.get());
-        if (!from_type)
-            throw Exception(ErrorCodes::TYPE_MISMATCH, "CAST AS Tuple can only be performed between tuple types or from String.\n"
-                            "Left type: {}, right type: {}", from_type_untyped->getName(), to_type->getName());
-
-        const auto & from_element_types = from_type->getElements();
-        const auto & to_element_types = to_type->getElements();
-
-        std::vector<WrapperType> element_wrappers;
-        std::vector<std::optional<size_t>> to_reverse_index;
-
-        /// For named tuples allow conversions for tuples with
-        /// different sets of elements. If element exists in @to_type
-        /// and doesn't exist in @to_type it will be filled by default values.
-        if (from_type->haveExplicitNames() && to_type->haveExplicitNames())
-        {
-            const auto & from_names = from_type->getElementNames();
-            std::unordered_map<String, size_t> from_positions;
-            from_positions.reserve(from_names.size());
-            for (size_t i = 0; i < from_names.size(); ++i)
-                from_positions[from_names[i]] = i;
-
-            const auto & to_names = to_type->getElementNames();
-            element_wrappers.reserve(to_names.size());
-            to_reverse_index.reserve(from_names.size());
-
-            for (size_t i = 0; i < to_names.size(); ++i)
-            {
-                auto it = from_positions.find(to_names[i]);
-                if (it != from_positions.end())
-                {
-                    element_wrappers.emplace_back(prepareUnpackDictionaries(from_element_types[it->second], to_element_types[i]));
-                    to_reverse_index.emplace_back(it->second);
-                }
-                else
-                {
-                    element_wrappers.emplace_back();
-                    to_reverse_index.emplace_back();
-                }
-            }
-        }
-        else
-        {
-            if (from_element_types.size() != to_element_types.size())
-                throw Exception(ErrorCodes::TYPE_MISMATCH, "CAST AS Tuple can only be performed between tuple types "
-                                "with the same number of elements or from String.\nLeft type: {}, right type: {}",
-                                from_type->getName(), to_type->getName());
-
-            element_wrappers = getElementWrappers(from_element_types, to_element_types);
-            to_reverse_index.reserve(to_element_types.size());
-            for (size_t i = 0; i < to_element_types.size(); ++i)
-                to_reverse_index.emplace_back(i);
-        }
-
-        return [element_wrappers, from_element_types, to_element_types, to_reverse_index]
-            (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * nullable_source, size_t input_rows_count) -> ColumnPtr
-        {
-            const auto * col = arguments.front().column.get();
-
-            size_t tuple_size = to_element_types.size();
-            const ColumnTuple & column_tuple = typeid_cast<const ColumnTuple &>(*col);
-
-            Columns converted_columns(tuple_size);
-
-            /// invoke conversion for each element
-            for (size_t i = 0; i < tuple_size; ++i)
-            {
-                if (to_reverse_index[i])
-                {
-                    size_t from_idx = *to_reverse_index[i];
-                    ColumnsWithTypeAndName element = {{column_tuple.getColumns()[from_idx], from_element_types[from_idx], "" }};
-                    converted_columns[i] = element_wrappers[i](element, to_element_types[i], nullable_source, input_rows_count);
-                }
-                else
-                {
-                    converted_columns[i] = to_element_types[i]->createColumn()->cloneResized(input_rows_count);
-                }
-            }
-
-            return ColumnTuple::create(converted_columns);
-        };
-    }
-
-    /// The case of: tuple([key1, key2, ..., key_n], [value1, value2, ..., value_n])
-    WrapperType createTupleToMapWrapper(const DataTypes & from_kv_types, const DataTypes & to_kv_types) const
-    {
-        return [element_wrappers = getElementWrappers(from_kv_types, to_kv_types), from_kv_types, to_kv_types]
-            (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * nullable_source, size_t /*input_rows_count*/) -> ColumnPtr
-        {
-            const auto * col = arguments.front().column.get();
-            const auto & column_tuple = assert_cast<const ColumnTuple &>(*col);
-
-            Columns offsets(2);
-            Columns converted_columns(2);
-            for (size_t i = 0; i < 2; ++i)
-            {
-                const auto & column_array = assert_cast<const ColumnArray &>(column_tuple.getColumn(i));
-                ColumnsWithTypeAndName element = {{column_array.getDataPtr(), from_kv_types[i], ""}};
-                converted_columns[i] = element_wrappers[i](element, to_kv_types[i], nullable_source, (element[0].column)->size());
-                offsets[i] = column_array.getOffsetsPtr();
-            }
-
-            const auto & keys_offsets = assert_cast<const ColumnArray::ColumnOffsets &>(*offsets[0]).getData();
-            const auto & values_offsets = assert_cast<const ColumnArray::ColumnOffsets &>(*offsets[1]).getData();
-            if (keys_offsets != values_offsets)
-                throw Exception(ErrorCodes::TYPE_MISMATCH,
-                    "CAST AS Map can only be performed from tuple of arrays with equal sizes.");
-
-            return ColumnMap::create(converted_columns[0], converted_columns[1], offsets[0]);
-        };
-    }
-
-    WrapperType createMapToMapWrapper(const DataTypes & from_kv_types, const DataTypes & to_kv_types) const
-    {
-        return [element_wrappers = getElementWrappers(from_kv_types, to_kv_types), from_kv_types, to_kv_types]
-            (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * nullable_source, size_t /*input_rows_count*/) -> ColumnPtr
-        {
-            const auto * col = arguments.front().column.get();
-            const auto & column_map = typeid_cast<const ColumnMap &>(*col);
-            const auto & nested_data = column_map.getNestedData();
-
-            Columns converted_columns(2);
-            for (size_t i = 0; i < 2; ++i)
-            {
-                ColumnsWithTypeAndName element = {{nested_data.getColumnPtr(i), from_kv_types[i], ""}};
-                converted_columns[i] = element_wrappers[i](element, to_kv_types[i], nullable_source, (element[0].column)->size());
-            }
-
-            return ColumnMap::create(converted_columns[0], converted_columns[1], column_map.getNestedColumn().getOffsetsPtr());
-        };
-    }
-
-    /// The case of: [(key1, value1), (key2, value2), ...]
-    WrapperType createArrayToMapWrapper(const DataTypes & from_kv_types, const DataTypes & to_kv_types) const
-    {
-        return [element_wrappers = getElementWrappers(from_kv_types, to_kv_types), from_kv_types, to_kv_types]
-            (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * nullable_source, size_t /*input_rows_count*/) -> ColumnPtr
-        {
-            const auto * col = arguments.front().column.get();
-            const auto & column_array = typeid_cast<const ColumnArray &>(*col);
-            const auto & nested_data = typeid_cast<const ColumnTuple &>(column_array.getData());
-
-            Columns converted_columns(2);
-            for (size_t i = 0; i < 2; ++i)
-            {
-                ColumnsWithTypeAndName element = {{nested_data.getColumnPtr(i), from_kv_types[i], ""}};
-                converted_columns[i] = element_wrappers[i](element, to_kv_types[i], nullable_source, (element[0].column)->size());
-            }
-
-            return ColumnMap::create(converted_columns[0], converted_columns[1], column_array.getOffsetsPtr());
-        };
-    }
-
-
-    WrapperType createMapWrapper(const DataTypePtr & from_type_untyped, const DataTypeMap * to_type) const
-    {
-        if (const auto * from_tuple = checkAndGetDataType<DataTypeTuple>(from_type_untyped.get()))
-        {
-            if (from_tuple->getElements().size() != 2)
-                throw Exception(
-                    ErrorCodes::TYPE_MISMATCH,
-                    "CAST AS Map from tuple requires 2 elements. "
-                    "Left type: {}, right type: {}",
-                    from_tuple->getName(),
-                    to_type->getName());
-
-            DataTypes from_kv_types;
-            const auto & to_kv_types = to_type->getKeyValueTypes();
-
-            for (const auto & elem : from_tuple->getElements())
-            {
-                const auto * type_array = checkAndGetDataType<DataTypeArray>(elem.get());
-                if (!type_array)
-                    throw Exception(ErrorCodes::TYPE_MISMATCH,
-                        "CAST AS Map can only be performed from tuples of array. Got: {}", from_tuple->getName());
-
-                from_kv_types.push_back(type_array->getNestedType());
-            }
-
-            return createTupleToMapWrapper(from_kv_types, to_kv_types);
-        }
-        else if (const auto * from_array = typeid_cast<const DataTypeArray *>(from_type_untyped.get()))
-        {
-            const auto * nested_tuple = typeid_cast<const DataTypeTuple *>(from_array->getNestedType().get());
-            if (!nested_tuple || nested_tuple->getElements().size() != 2)
-                throw Exception(
-                    ErrorCodes::TYPE_MISMATCH,
-                    "CAST AS Map from array requires nested tuple of 2 elements. "
-                    "Left type: {}, right type: {}",
-                    from_array->getName(),
-                    to_type->getName());
-
-            return createArrayToMapWrapper(nested_tuple->getElements(), to_type->getKeyValueTypes());
-        }
-        else if (const auto * from_type = checkAndGetDataType<DataTypeMap>(from_type_untyped.get()))
-        {
-            return createMapToMapWrapper(from_type->getKeyValueTypes(), to_type->getKeyValueTypes());
-        }
-        else
-        {
-            throw Exception(ErrorCodes::TYPE_MISMATCH, "Unsupported types to CAST AS Map. "
-                "Left type: {}, right type: {}", from_type_untyped->getName(), to_type->getName());
-        }
-    }
-
-    WrapperType createTupleToObjectWrapper(const DataTypeTuple & from_tuple, bool has_nullable_subcolumns) const
-    {
-        if (!from_tuple.haveExplicitNames())
-            throw Exception(ErrorCodes::TYPE_MISMATCH,
-            "Cast to Object can be performed only from flatten Named Tuple. Got: {}", from_tuple.getName());
-
-        PathsInData paths;
-        DataTypes from_types;
-
-        std::tie(paths, from_types) = flattenTuple(from_tuple.getPtr());
-        auto to_types = from_types;
-
-        for (auto & type : to_types)
-        {
-            if (isTuple(type) || isNested(type))
-                throw Exception(ErrorCodes::TYPE_MISMATCH,
-                    "Cast to Object can be performed only from flatten Named Tuple. Got: {}",
-                    from_tuple.getName());
-
-            type = recursiveRemoveLowCardinality(type);
-        }
-
-        return [element_wrappers = getElementWrappers(from_types, to_types),
-            has_nullable_subcolumns, from_types, to_types, paths]
-            (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * nullable_source, size_t input_rows_count)
-        {
-            size_t tuple_size = to_types.size();
-            auto flattened_column = flattenTuple(arguments.front().column);
-            const auto & column_tuple = assert_cast<const ColumnTuple &>(*flattened_column);
-
-            if (tuple_size != column_tuple.getColumns().size())
-                throw Exception(ErrorCodes::TYPE_MISMATCH,
-                    "Expected tuple with {} subcolumn, but got {} subcolumns",
-                    tuple_size, column_tuple.getColumns().size());
-
-            auto res = ColumnObject::create(has_nullable_subcolumns);
-            for (size_t i = 0; i < tuple_size; ++i)
-            {
-                ColumnsWithTypeAndName element = {{column_tuple.getColumns()[i], from_types[i], "" }};
-                auto converted_column = element_wrappers[i](element, to_types[i], nullable_source, input_rows_count);
-                res->addSubcolumn(paths[i], converted_column->assumeMutable());
-            }
-
-            return res;
-        };
-    }
-
-    WrapperType createMapToObjectWrapper(const DataTypeMap & from_map, bool has_nullable_subcolumns) const
-    {
-        auto key_value_types = from_map.getKeyValueTypes();
-
-        if (!isStringOrFixedString(key_value_types[0]))
-            throw Exception(ErrorCodes::TYPE_MISMATCH,
-                "Cast to Object from Map can be performed only from Map "
-                "with String or FixedString key. Got: {}", from_map.getName());
-
-        const auto & value_type = key_value_types[1];
-        auto to_value_type = value_type;
-
-        if (!has_nullable_subcolumns && value_type->isNullable())
-            to_value_type = removeNullable(value_type);
-
-        if (has_nullable_subcolumns && !value_type->isNullable())
-            to_value_type = makeNullable(value_type);
-
-        DataTypes to_key_value_types{std::make_shared<DataTypeString>(), std::move(to_value_type)};
-        auto element_wrappers = getElementWrappers(key_value_types, to_key_value_types);
-
-        return [has_nullable_subcolumns, element_wrappers, key_value_types, to_key_value_types]
-            (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * nullable_source, size_t) -> ColumnPtr
-        {
-            const auto & column_map = assert_cast<const ColumnMap &>(*arguments.front().column);
-            const auto & offsets = column_map.getNestedColumn().getOffsets();
-            auto key_value_columns = column_map.getNestedData().getColumnsCopy();
-
-            for (size_t i = 0; i < 2; ++i)
-            {
-                ColumnsWithTypeAndName element{{key_value_columns[i], key_value_types[i], ""}};
-                key_value_columns[i] = element_wrappers[i](element, to_key_value_types[i], nullable_source, key_value_columns[i]->size());
-            }
-
-            const auto & key_column_str = assert_cast<const ColumnString &>(*key_value_columns[0]);
-            const auto & value_column = *key_value_columns[1];
-
-            using SubcolumnsMap = HashMap<StringRef, MutableColumnPtr, StringRefHash>;
-            SubcolumnsMap subcolumns;
-
-            for (size_t row = 0; row < offsets.size(); ++row)
-            {
-                for (size_t i = offsets[static_cast<ssize_t>(row) - 1]; i < offsets[row]; ++i)
-                {
-                    auto ref = key_column_str.getDataAt(i);
-
-                    bool inserted;
-                    SubcolumnsMap::LookupResult it;
-                    subcolumns.emplace(ref, it, inserted);
-                    auto & subcolumn = it->getMapped();
-
-                    if (inserted)
-                        subcolumn = value_column.cloneEmpty()->cloneResized(row);
-
-                    /// Map can have duplicated keys. We insert only first one.
-                    if (subcolumn->size() == row)
-                        subcolumn->insertFrom(value_column, i);
-                }
-
-                /// Insert default values for keys missed in current row.
-                for (const auto & [_, subcolumn] : subcolumns)
-                    if (subcolumn->size() == row)
-                        subcolumn->insertDefault();
-            }
-
-            auto column_object = ColumnObject::create(has_nullable_subcolumns);
-            for (auto && [key, subcolumn] : subcolumns)
-            {
-                PathInData path(key.toView());
-                column_object->addSubcolumn(path, std::move(subcolumn));
-            }
-
-            return column_object;
-        };
-    }
-
-    WrapperType createObjectWrapper(const DataTypePtr & from_type, const DataTypeObject * to_type) const
-    {
-        if (const auto * from_tuple = checkAndGetDataType<DataTypeTuple>(from_type.get()))
-        {
-            return createTupleToObjectWrapper(*from_tuple, to_type->hasNullableSubcolumns());
-        }
-        else if (const auto * from_map = checkAndGetDataType<DataTypeMap>(from_type.get()))
-        {
-            return createMapToObjectWrapper(*from_map, to_type->hasNullableSubcolumns());
-        }
-        else if (checkAndGetDataType<DataTypeString>(from_type.get()))
-        {
-            return [] (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable * nullable_source, size_t input_rows_count)
-            {
-                auto res = ConvertImplGenericFromString::execute(arguments, result_type, nullable_source, input_rows_count)->assumeMutable();
-                res->finalize();
-                return res;
-            };
-        }
-        else if (checkAndGetDataType<DataTypeObject>(from_type.get()))
-        {
-            return [is_nullable = to_type->hasNullableSubcolumns()] (ColumnsWithTypeAndName & arguments, const DataTypePtr & , const ColumnNullable * , size_t) -> ColumnPtr
-            {
-                auto & column_object = assert_cast<const ColumnObject &>(*arguments.front().column);
-                auto res = ColumnObject::create(is_nullable);
-                for (size_t i = 0; i < column_object.size(); i++)
-                    res->insert(column_object[i]);
-
-                res->finalize();
-                return res;
-            };
-        }
-
-        throw Exception(ErrorCodes::TYPE_MISMATCH,
-            "Cast to Object can be performed only from flatten named Tuple, Map or String. Got: {}", from_type->getName());
-    }
-
-    WrapperType createVariantToVariantWrapper(const DataTypeVariant & from_variant, const DataTypeVariant & to_variant) const
-    {
-        /// We support only extension of variant type, so, only new types can be added.
-        /// For example: Variant(T1, T2) -> Variant(T1, T2, T3) is supported, but Variant(T1, T2) -> Variant(T1, T3) is not supported.
-        /// We want to extend Variant type for free without rewriting the data, but we sort data types inside Variant during type creation
-        /// (we do it because we want Variant(T1, T2) to be the same as Variant(T2, T1)), but after extension the order of variant types
-        /// (and so their discriminators) can be different. For example: Variant(T1, T3) -> Variant(T1, T2, T3).
-        /// To avoid full rewrite of discriminators column, ColumnVariant supports it's local order of variant columns (and so local
-        /// discriminators) and stores mapping global order -> local order.
-        /// So, to extend Variant with new types for free, we should keep old local order for old variants, append new variants and change
-        /// mapping global order -> local order according to the new global order.
-
-        /// Create map (new variant type) -> (it's global discriminator in new order).
-        const auto & new_variants = to_variant.getVariants();
-        std::unordered_map<String, ColumnVariant::Discriminator> new_variant_types_to_new_global_discriminator;
-        new_variant_types_to_new_global_discriminator.reserve(new_variants.size());
-        for (size_t i = 0; i != new_variants.size(); ++i)
-            new_variant_types_to_new_global_discriminator[new_variants[i]->getName()] = i;
-
-        /// Create set of old variant types.
-        const auto & old_variants = from_variant.getVariants();
-        std::unordered_map<String, ColumnVariant::Discriminator> old_variant_types_to_old_global_discriminator;
-        old_variant_types_to_old_global_discriminator.reserve(old_variants.size());
-        for (size_t i = 0; i != old_variants.size(); ++i)
-            old_variant_types_to_old_global_discriminator[old_variants[i]->getName()] = i;
-
-        /// Check that the set of old variants types is a subset of new variant types and collect new global discriminator for each old global discriminator.
-        std::unordered_map<ColumnVariant::Discriminator, ColumnVariant::Discriminator> old_global_discriminator_to_new;
-        old_global_discriminator_to_new.reserve(old_variants.size());
-        for (const auto & [old_variant_type, old_discriminator] : old_variant_types_to_old_global_discriminator)
-        {
-            auto it = new_variant_types_to_new_global_discriminator.find(old_variant_type);
-            if (it == new_variant_types_to_new_global_discriminator.end())
-                throw Exception(
-                    ErrorCodes::CANNOT_CONVERT_TYPE,
-                    "Cannot convert type {} to {}. Conversion between Variant types is allowed only when new Variant type is an extension "
-                    "of an initial one", from_variant.getName(), to_variant.getName());
-            old_global_discriminator_to_new[old_discriminator] = it->second;
-        }
-
-        /// Collect variant types and their global discriminators that should be added to the old Variant to get the new Variant.
-        std::vector<std::pair<DataTypePtr, ColumnVariant::Discriminator>> variant_types_and_discriminators_to_add;
-        variant_types_and_discriminators_to_add.reserve(new_variants.size() - old_variants.size());
-        for (size_t i = 0; i != new_variants.size(); ++i)
-        {
-            if (!old_variant_types_to_old_global_discriminator.contains(new_variants[i]->getName()))
-                variant_types_and_discriminators_to_add.emplace_back(new_variants[i], i);
-        }
-
-        return [old_global_discriminator_to_new, variant_types_and_discriminators_to_add]
-               (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable *, size_t) -> ColumnPtr
-        {
-            const auto & column_variant = assert_cast<const ColumnVariant &>(*arguments.front().column.get());
-            size_t num_old_variants = column_variant.getNumVariants();
-            Columns new_variant_columns;
-            new_variant_columns.reserve(num_old_variants + variant_types_and_discriminators_to_add.size());
-            std::vector<ColumnVariant::Discriminator> new_local_to_global_discriminators;
-            new_local_to_global_discriminators.reserve(num_old_variants + variant_types_and_discriminators_to_add.size());
-            for (size_t i = 0; i != num_old_variants; ++i)
-            {
-                new_variant_columns.push_back(column_variant.getVariantPtrByLocalDiscriminator(i));
-                new_local_to_global_discriminators.push_back(old_global_discriminator_to_new.at(column_variant.globalDiscriminatorByLocal(i)));
-            }
-
-            for (const auto & [new_variant_type, new_global_discriminator] : variant_types_and_discriminators_to_add)
-            {
-                new_variant_columns.push_back(new_variant_type->createColumn());
-                new_local_to_global_discriminators.push_back(new_global_discriminator);
-            }
-
-            return ColumnVariant::create(column_variant.getLocalDiscriminatorsPtr(), column_variant.getOffsetsPtr(), new_variant_columns, new_local_to_global_discriminators);
-        };
-    }
-
-    WrapperType createVariantToColumnWrapper(const DataTypeVariant & from_variant, const DataTypePtr & to_type) const
-    {
-        const auto & variant_types = from_variant.getVariants();
-        std::vector<WrapperType> variant_wrappers;
-        variant_wrappers.reserve(variant_types.size());
-
-        /// Create conversion wrapper for each variant.
-        for (const auto & variant_type : variant_types)
-            variant_wrappers.push_back(prepareUnpackDictionaries(variant_type, to_type));
-
-        return [variant_wrappers, variant_types, to_type]
-               (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count) -> ColumnPtr
-        {
-            const auto & column_variant = assert_cast<const ColumnVariant &>(*arguments.front().column.get());
-
-            /// First, cast each variant to the result type.
-            std::vector<ColumnPtr> casted_variant_columns;
-            casted_variant_columns.reserve(variant_types.size());
-            for (size_t i = 0; i != variant_types.size(); ++i)
-            {
-                auto variant_col = column_variant.getVariantPtrByLocalDiscriminator(i);
-                ColumnsWithTypeAndName variant = {{variant_col, variant_types[i], "" }};
-                const auto & variant_wrapper = variant_wrappers[column_variant.globalDiscriminatorByLocal(i)];
-                casted_variant_columns.push_back(variant_wrapper(variant, result_type, nullptr, variant_col->size()));
-            }
-
-            /// Second, construct resulting column from casted variant columns according to discriminators.
-            const auto & local_discriminators = column_variant.getLocalDiscriminators();
-            auto res = result_type->createColumn();
-            res->reserve(input_rows_count);
-            for (size_t i = 0; i != input_rows_count; ++i)
-            {
-                auto local_discr = local_discriminators[i];
-                if (local_discr == ColumnVariant::NULL_DISCRIMINATOR)
-                    res->insertDefault();
-                else
-                    res->insertFrom(*casted_variant_columns[local_discr], column_variant.offsetAt(i));
-            }
-
-            return res;
-        };
-    }
-
-    static ColumnPtr createVariantFromDescriptorsAndOneNonEmptyVariant(const DataTypes & variant_types, const ColumnPtr & discriminators, const ColumnPtr & variant, ColumnVariant::Discriminator variant_discr)
-    {
-        Columns variants;
-        variants.reserve(variant_types.size());
-        for (size_t i = 0; i != variant_types.size(); ++i)
-        {
-            if (i == variant_discr)
-                variants.emplace_back(variant);
-            else
-                variants.push_back(variant_types[i]->createColumn());
-        }
-
-        return ColumnVariant::create(discriminators, variants);
-    }
-
-    WrapperType createColumnToVariantWrapper(const DataTypePtr & from_type, const DataTypeVariant & to_variant) const
-    {
-        /// We allow converting NULL to Variant(...) as Variant can store NULLs.
-        if (from_type->onlyNull())
-        {
-            return [](ColumnsWithTypeAndName &, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count) -> ColumnPtr
-            {
-                auto result_column = result_type->createColumn();
-                result_column->insertManyDefaults(input_rows_count);
-                return result_column;
-            };
-        }
-
-        auto variant_discr_opt = to_variant.tryGetVariantDiscriminator(*removeNullableOrLowCardinalityNullable(from_type));
-        if (!variant_discr_opt)
-            throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Cannot convert type {} to {}. Conversion to Variant allowed only for types from this Variant", from_type->getName(), to_variant.getName());
-
-        return [variant_discr = *variant_discr_opt]
-               (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *, size_t) -> ColumnPtr
-        {
-            const auto & result_variant_type = assert_cast<const DataTypeVariant &>(*result_type);
-            const auto & variant_types = result_variant_type.getVariants();
-            if (const ColumnNullable * col_nullable = typeid_cast<const ColumnNullable *>(arguments.front().column.get()))
-            {
-                const auto & column = col_nullable->getNestedColumnPtr();
-                const auto & null_map = col_nullable->getNullMapData();
-                IColumn::Filter filter;
-                filter.reserve(column->size());
-                auto discriminators = ColumnVariant::ColumnDiscriminators::create();
-                auto & discriminators_data = discriminators->getData();
-                discriminators_data.reserve(column->size());
-                size_t variant_size_hint = 0;
-                for (size_t i = 0; i != column->size(); ++i)
-                {
-                    if (null_map[i])
-                    {
-                        discriminators_data.push_back(ColumnVariant::NULL_DISCRIMINATOR);
-                        filter.push_back(0);
-                    }
-                    else
-                    {
-                        discriminators_data.push_back(variant_discr);
-                        filter.push_back(1);
-                        ++variant_size_hint;
-                    }
-                }
-
-                ColumnPtr variant_column;
-                /// If there were no NULLs, just use the column.
-                if (variant_size_hint == column->size())
-                    variant_column = column;
-                /// Otherwise we should use filtered column.
-                else
-                    variant_column = column->filter(filter, variant_size_hint);
-                return createVariantFromDescriptorsAndOneNonEmptyVariant(variant_types, std::move(discriminators), variant_column, variant_discr);
-            }
-            else if (isColumnLowCardinalityNullable(*arguments.front().column))
-            {
-                const auto & column = arguments.front().column;
-
-                /// Variant column cannot have LowCardinality(Nullable(...)) variant, as Variant column stores NULLs itself.
-                /// We should create a null-map, insert NULL_DISCRIMINATOR on NULL values and filter initial column.
-                const auto & col_lc = assert_cast<const ColumnLowCardinality &>(*column);
-                const auto & indexes = col_lc.getIndexes();
-                auto null_index = col_lc.getDictionary().getNullValueIndex();
-                IColumn::Filter filter;
-                filter.reserve(col_lc.size());
-                auto discriminators = ColumnVariant::ColumnDiscriminators::create();
-                auto & discriminators_data = discriminators->getData();
-                discriminators_data.reserve(col_lc.size());
-                size_t variant_size_hint = 0;
-                for (size_t i = 0; i != col_lc.size(); ++i)
-                {
-                    if (indexes.getUInt(i) == null_index)
-                    {
-                        discriminators_data.push_back(ColumnVariant::NULL_DISCRIMINATOR);
-                        filter.push_back(0);
-                    }
-                    else
-                    {
-                        discriminators_data.push_back(variant_discr);
-                        filter.push_back(1);
-                        ++variant_size_hint;
-                    }
-                }
-
-                MutableColumnPtr variant_column;
-                /// If there were no NULLs, we can just clone the column.
-                if (variant_size_hint == col_lc.size())
-                    variant_column = IColumn::mutate(column);
-                /// Otherwise we should filter column.
-                else
-                    variant_column = column->filter(filter, variant_size_hint)->assumeMutable();
-
-                assert_cast<ColumnLowCardinality &>(*variant_column).nestedRemoveNullable();
-                return createVariantFromDescriptorsAndOneNonEmptyVariant(variant_types, std::move(discriminators), std::move(variant_column), variant_discr);
-            }
-            else
-            {
-                const auto & column = arguments.front().column;
-                auto discriminators = ColumnVariant::ColumnDiscriminators::create();
-                discriminators->getData().resize_fill(column->size(), variant_discr);
-                return createVariantFromDescriptorsAndOneNonEmptyVariant(variant_types, std::move(discriminators), column, variant_discr);
-            }
-        };
-    }
-
-    /// Wrapper for conversion to/from Variant type
-    WrapperType createVariantWrapper(const DataTypePtr & from_type, const DataTypePtr & to_type) const
-    {
-        if (const auto * from_variant = checkAndGetDataType<DataTypeVariant>(from_type.get()))
-        {
-            if (const auto * to_variant = checkAndGetDataType<DataTypeVariant>(to_type.get()))
-                return createVariantToVariantWrapper(*from_variant, *to_variant);
-
-            return createVariantToColumnWrapper(*from_variant, to_type);
-        }
-
-        return createColumnToVariantWrapper(from_type, assert_cast<const DataTypeVariant &>(*to_type));
-    }
-
-    template <typename FieldType>
-    WrapperType createEnumWrapper(const DataTypePtr & from_type, const DataTypeEnum<FieldType> * to_type) const
-    {
-        using EnumType = DataTypeEnum<FieldType>;
-        using Function = typename FunctionTo<EnumType>::Type;
-
-        if (const auto * from_enum8 = checkAndGetDataType<DataTypeEnum8>(from_type.get()))
-            checkEnumToEnumConversion(from_enum8, to_type);
-        else if (const auto * from_enum16 = checkAndGetDataType<DataTypeEnum16>(from_type.get()))
-            checkEnumToEnumConversion(from_enum16, to_type);
-
-        if (checkAndGetDataType<DataTypeString>(from_type.get()))
-            return createStringToEnumWrapper<ColumnString, EnumType>();
-        else if (checkAndGetDataType<DataTypeFixedString>(from_type.get()))
-            return createStringToEnumWrapper<ColumnFixedString, EnumType>();
-        else if (isNativeNumber(from_type) || isEnum(from_type))
-        {
-            auto function = Function::create();
-            return createFunctionAdaptor(function, from_type);
-        }
-        else
-        {
-            if (cast_type == CastType::accurateOrNull)
-                return createToNullableColumnWrapper();
-            else
-                throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Conversion from {} to {} is not supported",
-                    from_type->getName(), to_type->getName());
-        }
-    }
-
-    template <typename EnumTypeFrom, typename EnumTypeTo>
-    void checkEnumToEnumConversion(const EnumTypeFrom * from_type, const EnumTypeTo * to_type) const
-    {
-        const auto & from_values = from_type->getValues();
-        const auto & to_values = to_type->getValues();
-
-        using ValueType = std::common_type_t<typename EnumTypeFrom::FieldType, typename EnumTypeTo::FieldType>;
-        using NameValuePair = std::pair<std::string, ValueType>;
-        using EnumValues = std::vector<NameValuePair>;
-
-        EnumValues name_intersection;
-        std::set_intersection(std::begin(from_values), std::end(from_values),
-            std::begin(to_values), std::end(to_values), std::back_inserter(name_intersection),
-            [] (auto && from, auto && to) { return from.first < to.first; });
-
-        for (const auto & name_value : name_intersection)
-        {
-            const auto & old_value = name_value.second;
-            const auto & new_value = to_type->getValue(name_value.first);
-            if (old_value != new_value)
-                throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Enum conversion changes value for element '{}' from {} to {}",
-                    name_value.first, toString(old_value), toString(new_value));
-        }
-    }
-
-    template <typename ColumnStringType, typename EnumType>
-    WrapperType createStringToEnumWrapper() const
-    {
-        const char * function_name = cast_name;
-        return [function_name] (
-            ColumnsWithTypeAndName & arguments, const DataTypePtr & res_type, const ColumnNullable * nullable_col, size_t /*input_rows_count*/)
-        {
-            const auto & first_col = arguments.front().column.get();
-            const auto & result_type = typeid_cast<const EnumType &>(*res_type);
-
-            const ColumnStringType * col = typeid_cast<const ColumnStringType *>(first_col);
-
-            if (col && nullable_col && nullable_col->size() != col->size())
-                throw Exception(ErrorCodes::LOGICAL_ERROR, "ColumnNullable is not compatible with original");
-
-            if (col)
-            {
-                const auto size = col->size();
-
-                auto res = result_type.createColumn();
-                auto & out_data = static_cast<typename EnumType::ColumnType &>(*res).getData();
-                out_data.resize(size);
-
-                auto default_enum_value = result_type.getValues().front().second;
-
-                if (nullable_col)
-                {
-                    for (size_t i = 0; i < size; ++i)
-                    {
-                        if (!nullable_col->isNullAt(i))
-                            out_data[i] = result_type.getValue(col->getDataAt(i));
-                        else
-                            out_data[i] = default_enum_value;
-                    }
-                }
-                else
-                {
-                    for (size_t i = 0; i < size; ++i)
-                        out_data[i] = result_type.getValue(col->getDataAt(i));
-                }
-
-                return res;
-            }
-            else
-                throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected column {} as first argument of function {}",
-                    first_col->getName(), function_name);
-        };
-    }
-
-    template <typename EnumType>
-    WrapperType createEnumToStringWrapper() const
-    {
-        const char * function_name = cast_name;
-        return [function_name] (
-            ColumnsWithTypeAndName & arguments, const DataTypePtr & res_type, const ColumnNullable * nullable_col, size_t /*input_rows_count*/)
-        {
-            using ColumnEnumType = EnumType::ColumnType;
-
-            const auto & first_col = arguments.front().column.get();
-            const auto & first_type = arguments.front().type.get();
-
-            const ColumnEnumType * enum_col = typeid_cast<const ColumnEnumType *>(first_col);
-            const EnumType * enum_type = typeid_cast<const EnumType *>(first_type);
-
-            if (enum_col && nullable_col && nullable_col->size() != enum_col->size())
-                throw Exception(ErrorCodes::LOGICAL_ERROR, "ColumnNullable is not compatible with original");
-
-            if (enum_col && enum_type)
-            {
-                const auto size = enum_col->size();
-                const auto & enum_data = enum_col->getData();
-
-                auto res = res_type->createColumn();
-
-                if (nullable_col)
-                {
-                    for (size_t i = 0; i < size; ++i)
-                    {
-                        if (!nullable_col->isNullAt(i))
-                        {
-                            const auto & value = enum_type->getNameForValue(enum_data[i]);
-                            res->insertData(value.data, value.size);
-                        }
-                        else
-                            res->insertDefault();
-                    }
-                }
-                else
-                {
-                    for (size_t i = 0; i < size; ++i)
-                    {
-                        const auto & value = enum_type->getNameForValue(enum_data[i]);
-                        res->insertData(value.data, value.size);
-                    }
-                }
-
-                return res;
-            }
-            else
-                throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected column {} as first argument of function {}",
-                    first_col->getName(), function_name);
-        };
-    }
-
-    static WrapperType createIdentityWrapper(const DataTypePtr &)
-    {
-        return [] (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable *, size_t /*input_rows_count*/)
-        {
-            return arguments.front().column;
-        };
-    }
-
-    static WrapperType createNothingWrapper(const IDataType * to_type)
-    {
-        ColumnPtr res = to_type->createColumnConstWithDefaultValue(1);
-        return [res] (ColumnsWithTypeAndName &, const DataTypePtr &, const ColumnNullable *, size_t input_rows_count)
-        {
-            /// Column of Nothing type is trivially convertible to any other column
-            return res->cloneResized(input_rows_count)->convertToFullColumnIfConst();
-        };
-    }
-
-    WrapperType prepareUnpackDictionaries(const DataTypePtr & from_type, const DataTypePtr & to_type) const
-    {
-        /// Conversion from/to Variant data type is processed in a special way.
-        /// We don't need to remove LowCardinality/Nullable.
-        if (isVariant(to_type) || isVariant(from_type))
-            return createVariantWrapper(from_type, to_type);
-
-        const auto * from_low_cardinality = typeid_cast<const DataTypeLowCardinality *>(from_type.get());
-        const auto * to_low_cardinality = typeid_cast<const DataTypeLowCardinality *>(to_type.get());
-        const auto & from_nested = from_low_cardinality ? from_low_cardinality->getDictionaryType() : from_type;
-        const auto & to_nested = to_low_cardinality ? to_low_cardinality->getDictionaryType() : to_type;
-
-        if (from_type->onlyNull())
-        {
-            if (!to_nested->isNullable() && !isVariant(to_type))
-            {
-                if (cast_type == CastType::accurateOrNull)
-                {
-                    return createToNullableColumnWrapper();
-                }
-                else
-                {
-                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Cannot convert NULL to a non-nullable type");
-                }
-            }
-
-            return [](ColumnsWithTypeAndName &, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count)
-            {
-                return result_type->createColumnConstWithDefaultValue(input_rows_count)->convertToFullColumnIfConst();
-            };
-        }
-
-        bool skip_not_null_check = false;
-
-        if (from_low_cardinality && from_nested->isNullable() && !to_nested->isNullable())
-            /// Disable check for dictionary. Will check that column doesn't contain NULL in wrapper below.
-            skip_not_null_check = true;
-
-        auto wrapper = prepareRemoveNullable(from_nested, to_nested, skip_not_null_check);
-        if (!from_low_cardinality && !to_low_cardinality)
-            return wrapper;
-
-        return [wrapper, from_low_cardinality, to_low_cardinality, skip_not_null_check]
-                (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable * nullable_source, size_t input_rows_count) -> ColumnPtr
-        {
-            ColumnsWithTypeAndName args = {arguments[0]};
-            auto & arg = args.front();
-            auto res_type = result_type;
-
-            ColumnPtr converted_column;
-
-            ColumnPtr res_indexes;
-            /// For some types default can't be casted (for example, String to Int). In that case convert column to full.
-            bool src_converted_to_full_column = false;
-
-            {
-                auto tmp_rows_count = input_rows_count;
-
-                if (to_low_cardinality)
-                    res_type = to_low_cardinality->getDictionaryType();
-
-                if (from_low_cardinality)
-                {
-                    const auto * col_low_cardinality = typeid_cast<const ColumnLowCardinality *>(arguments[0].column.get());
-
-                    if (skip_not_null_check && col_low_cardinality->containsNull())
-                        throw Exception(ErrorCodes::CANNOT_INSERT_NULL_IN_ORDINARY_COLUMN, "Cannot convert NULL value to non-Nullable type");
-
-                    arg.column = col_low_cardinality->getDictionary().getNestedColumn();
-                    arg.type = from_low_cardinality->getDictionaryType();
-
-                    /// TODO: Make map with defaults conversion.
-                    src_converted_to_full_column = !removeNullable(arg.type)->equals(*removeNullable(res_type));
-                    if (src_converted_to_full_column)
-                        arg.column = arg.column->index(col_low_cardinality->getIndexes(), 0);
-                    else
-                        res_indexes = col_low_cardinality->getIndexesPtr();
-
-                    tmp_rows_count = arg.column->size();
-                }
-
-                /// Perform the requested conversion.
-                converted_column = wrapper(args, res_type, nullable_source, tmp_rows_count);
-            }
-
-            if (to_low_cardinality)
-            {
-                auto res_column = to_low_cardinality->createColumn();
-                auto * col_low_cardinality = typeid_cast<ColumnLowCardinality *>(res_column.get());
-
-                if (from_low_cardinality && !src_converted_to_full_column)
-                {
-                    col_low_cardinality->insertRangeFromDictionaryEncodedColumn(*converted_column, *res_indexes);
-                }
-                else
-                    col_low_cardinality->insertRangeFromFullColumn(*converted_column, 0, converted_column->size());
-
-                return res_column;
-            }
-            else if (!src_converted_to_full_column)
-                return converted_column->index(*res_indexes, 0);
-            else
-                return converted_column;
-        };
-    }
-
-    WrapperType prepareRemoveNullable(const DataTypePtr & from_type, const DataTypePtr & to_type, bool skip_not_null_check) const
-    {
-        /// Determine whether pre-processing and/or post-processing must take place during conversion.
-
-        bool source_is_nullable = from_type->isNullable();
-        bool result_is_nullable = to_type->isNullable();
-
-        auto wrapper = prepareImpl(removeNullable(from_type), removeNullable(to_type), result_is_nullable);
-
-        if (result_is_nullable)
-        {
-            return [wrapper, source_is_nullable]
-                (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count) -> ColumnPtr
-            {
-                /// Create a temporary columns on which to perform the operation.
-                const auto & nullable_type = static_cast<const DataTypeNullable &>(*result_type);
-                const auto & nested_type = nullable_type.getNestedType();
-
-                ColumnsWithTypeAndName tmp_args;
-                if (source_is_nullable)
-                    tmp_args = createBlockWithNestedColumns(arguments);
-                else
-                    tmp_args = arguments;
-
-                const ColumnNullable * nullable_source = nullptr;
-
-                /// Add original ColumnNullable for createStringToEnumWrapper()
-                if (source_is_nullable)
-                {
-                    if (arguments.size() != 1)
-                        throw Exception(ErrorCodes::LOGICAL_ERROR, "Invalid number of arguments");
-                    nullable_source = typeid_cast<const ColumnNullable *>(arguments.front().column.get());
-                }
-
-                /// Perform the requested conversion.
-                auto tmp_res = wrapper(tmp_args, nested_type, nullable_source, input_rows_count);
-
-                /// May happen in fuzzy tests. For debug purpose.
-                if (!tmp_res)
-                    throw Exception(ErrorCodes::LOGICAL_ERROR, "Couldn't convert {} to {} in prepareRemoveNullable wrapper.",
-                                    arguments[0].type->getName(), nested_type->getName());
-
-                return wrapInNullable(tmp_res, arguments, nested_type, input_rows_count);
-            };
-        }
-        else if (source_is_nullable)
-        {
-            /// Conversion from Nullable to non-Nullable.
-
-            return [wrapper, skip_not_null_check]
-                (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count) -> ColumnPtr
-            {
-                auto tmp_args = createBlockWithNestedColumns(arguments);
-                auto nested_type = removeNullable(result_type);
-
-                /// Check that all values are not-NULL.
-                /// Check can be skipped in case if LowCardinality dictionary is transformed.
-                /// In that case, correctness will be checked beforehand.
-                if (!skip_not_null_check)
-                {
-                    const auto & col = arguments[0].column;
-                    const auto & nullable_col = assert_cast<const ColumnNullable &>(*col);
-                    const auto & null_map = nullable_col.getNullMapData();
-
-                    if (!memoryIsZero(null_map.data(), 0, null_map.size()))
-                        throw Exception(ErrorCodes::CANNOT_INSERT_NULL_IN_ORDINARY_COLUMN, "Cannot convert NULL value to non-Nullable type");
-                }
-                const ColumnNullable * nullable_source = typeid_cast<const ColumnNullable *>(arguments.front().column.get());
-                return wrapper(tmp_args, nested_type, nullable_source, input_rows_count);
-            };
-        }
-        else
-            return wrapper;
-    }
-
-    /// 'from_type' and 'to_type' are nested types in case of Nullable.
-    /// 'requested_result_is_nullable' is true if CAST to Nullable type is requested.
-    WrapperType prepareImpl(const DataTypePtr & from_type, const DataTypePtr & to_type, bool requested_result_is_nullable) const
-    {
-        if (isUInt8(from_type) && isBool(to_type))
-            return createUInt8ToBoolWrapper(from_type, to_type);
-
-        /// We can cast IPv6 into IPv6, IPv4 into IPv4, but we should not allow to cast FixedString(16) into IPv6 as part of identity cast
-        bool safe_convert_custom_types = true;
-
-        if (const auto * to_type_custom_name = to_type->getCustomName())
-            safe_convert_custom_types = from_type->getCustomName() && from_type->getCustomName()->getName() == to_type_custom_name->getName();
-        else if (const auto * from_type_custom_name = from_type->getCustomName())
-            safe_convert_custom_types = to_type->getCustomName() && from_type_custom_name->getName() == to_type->getCustomName()->getName();
-
-        if (from_type->equals(*to_type) && safe_convert_custom_types)
-        {
-            /// We can only use identity conversion for DataTypeAggregateFunction when they are strictly equivalent.
-            if (typeid_cast<const DataTypeAggregateFunction *>(from_type.get()))
-            {
-                if (DataTypeAggregateFunction::strictEquals(from_type, to_type))
-                    return createIdentityWrapper(from_type);
-            }
-            else
-                return createIdentityWrapper(from_type);
-        }
-        else if (WhichDataType(from_type).isNothing())
-            return createNothingWrapper(to_type.get());
-
-        WrapperType ret;
-
-        auto make_default_wrapper = [&](const auto & types) -> bool
-        {
-            using Types = std::decay_t<decltype(types)>;
-            using ToDataType = typename Types::LeftType;
-
-            if constexpr (
-                std::is_same_v<ToDataType, DataTypeUInt16> ||
-                std::is_same_v<ToDataType, DataTypeUInt32> ||
-                std::is_same_v<ToDataType, DataTypeUInt64> ||
-                std::is_same_v<ToDataType, DataTypeUInt128> ||
-                std::is_same_v<ToDataType, DataTypeUInt256> ||
-                std::is_same_v<ToDataType, DataTypeInt8> ||
-                std::is_same_v<ToDataType, DataTypeInt16> ||
-                std::is_same_v<ToDataType, DataTypeInt32> ||
-                std::is_same_v<ToDataType, DataTypeInt64> ||
-                std::is_same_v<ToDataType, DataTypeInt128> ||
-                std::is_same_v<ToDataType, DataTypeInt256> ||
-                std::is_same_v<ToDataType, DataTypeFloat32> ||
-                std::is_same_v<ToDataType, DataTypeFloat64> ||
-                std::is_same_v<ToDataType, DataTypeDate> ||
-                std::is_same_v<ToDataType, DataTypeDate32> ||
-                std::is_same_v<ToDataType, DataTypeDateTime> ||
-                std::is_same_v<ToDataType, DataTypeUUID> ||
-                std::is_same_v<ToDataType, DataTypeIPv4> ||
-                std::is_same_v<ToDataType, DataTypeIPv6>)
-            {
-                ret = createWrapper(from_type, checkAndGetDataType<ToDataType>(to_type.get()), requested_result_is_nullable);
-                return true;
-            }
-            if constexpr (std::is_same_v<ToDataType, DataTypeUInt8>)
-            {
-                if (isBool(to_type))
-                    ret = createBoolWrapper<ToDataType>(from_type, checkAndGetDataType<ToDataType>(to_type.get()), requested_result_is_nullable);
-                else
-                    ret = createWrapper(from_type, checkAndGetDataType<ToDataType>(to_type.get()), requested_result_is_nullable);
-                return true;
-            }
-            if constexpr (
-                std::is_same_v<ToDataType, DataTypeEnum8> ||
-                std::is_same_v<ToDataType, DataTypeEnum16>)
-            {
-                ret = createEnumWrapper(from_type, checkAndGetDataType<ToDataType>(to_type.get()));
-                return true;
-            }
-            if constexpr (
-                std::is_same_v<ToDataType, DataTypeDecimal<Decimal32>> ||
-                std::is_same_v<ToDataType, DataTypeDecimal<Decimal64>> ||
-                std::is_same_v<ToDataType, DataTypeDecimal<Decimal128>> ||
-                std::is_same_v<ToDataType, DataTypeDecimal<Decimal256>> ||
-                std::is_same_v<ToDataType, DataTypeDateTime64>)
-            {
-                ret = createDecimalWrapper(from_type, checkAndGetDataType<ToDataType>(to_type.get()), requested_result_is_nullable);
-                return true;
-            }
-
-            return false;
-        };
-
-        bool cast_ipv4_ipv6_default_on_conversion_error_value = context && context->getSettingsRef().cast_ipv4_ipv6_default_on_conversion_error;
-        bool input_format_ipv4_default_on_conversion_error_value = context && context->getSettingsRef().input_format_ipv4_default_on_conversion_error;
-        bool input_format_ipv6_default_on_conversion_error_value = context && context->getSettingsRef().input_format_ipv6_default_on_conversion_error;
-
-        auto make_custom_serialization_wrapper = [&, cast_ipv4_ipv6_default_on_conversion_error_value, input_format_ipv4_default_on_conversion_error_value, input_format_ipv6_default_on_conversion_error_value](const auto & types) -> bool
-        {
-            using Types = std::decay_t<decltype(types)>;
-            using ToDataType = typename Types::RightType;
-            using FromDataType = typename Types::LeftType;
-
-            if constexpr (WhichDataType(FromDataType::type_id).isStringOrFixedString())
-            {
-                if constexpr (std::is_same_v<ToDataType, DataTypeIPv4>)
-                {
-                    ret = [cast_ipv4_ipv6_default_on_conversion_error_value,
-                           input_format_ipv4_default_on_conversion_error_value,
-                           requested_result_is_nullable](
-                              ColumnsWithTypeAndName & arguments,
-                              const DataTypePtr & result_type,
-                              const ColumnNullable * column_nullable,
-                              size_t) -> ColumnPtr
-                    {
-                        if (!WhichDataType(result_type).isIPv4())
-                            throw Exception(ErrorCodes::TYPE_MISMATCH, "Wrong result type {}. Expected IPv4", result_type->getName());
-
-                        const auto * null_map = column_nullable ? &column_nullable->getNullMapData() : nullptr;
-                        if (requested_result_is_nullable)
-                            return convertToIPv4<IPStringToNumExceptionMode::Null>(arguments[0].column, null_map);
-                        else if (cast_ipv4_ipv6_default_on_conversion_error_value || input_format_ipv4_default_on_conversion_error_value)
-                            return convertToIPv4<IPStringToNumExceptionMode::Default>(arguments[0].column, null_map);
-                        else
-                            return convertToIPv4<IPStringToNumExceptionMode::Throw>(arguments[0].column, null_map);
-                    };
-
-                    return true;
-                }
-
-                if constexpr (std::is_same_v<ToDataType, DataTypeIPv6>)
-                {
-                    ret = [cast_ipv4_ipv6_default_on_conversion_error_value,
-                           input_format_ipv6_default_on_conversion_error_value,
-                           requested_result_is_nullable](
-                              ColumnsWithTypeAndName & arguments,
-                              const DataTypePtr & result_type,
-                              const ColumnNullable * column_nullable,
-                              size_t) -> ColumnPtr
-                    {
-                        if (!WhichDataType(result_type).isIPv6())
-                            throw Exception(
-                                ErrorCodes::TYPE_MISMATCH, "Wrong result type {}. Expected IPv6", result_type->getName());
-
-                        const auto * null_map = column_nullable ? &column_nullable->getNullMapData() : nullptr;
-                        if (requested_result_is_nullable)
-                            return convertToIPv6<IPStringToNumExceptionMode::Null>(arguments[0].column, null_map);
-                        else if (cast_ipv4_ipv6_default_on_conversion_error_value || input_format_ipv6_default_on_conversion_error_value)
-                            return convertToIPv6<IPStringToNumExceptionMode::Default>(arguments[0].column, null_map);
-                        else
-                            return convertToIPv6<IPStringToNumExceptionMode::Throw>(arguments[0].column, null_map);
-                    };
-
-                    return true;
-                }
-
-                if (to_type->getCustomSerialization() && to_type->getCustomName())
-                {
-                    ret = [requested_result_is_nullable](
-                              ColumnsWithTypeAndName & arguments,
-                              const DataTypePtr & result_type,
-                              const ColumnNullable * column_nullable,
-                              size_t input_rows_count) -> ColumnPtr
-                    {
-                        auto wrapped_result_type = result_type;
-                        if (requested_result_is_nullable)
-                            wrapped_result_type = makeNullable(result_type);
-                        return ConvertImplGenericFromString::execute(
-                            arguments, wrapped_result_type, column_nullable, input_rows_count);
-                    };
-                    return true;
-                }
-            }
-            else if constexpr (WhichDataType(FromDataType::type_id).isIPv6() && WhichDataType(ToDataType::type_id).isIPv4())
-            {
-                ret = [cast_ipv4_ipv6_default_on_conversion_error_value, requested_result_is_nullable](
-                                ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable * column_nullable, size_t)
-                        -> ColumnPtr
-                {
-                    if (!WhichDataType(result_type).isIPv4())
-                        throw Exception(
-                            ErrorCodes::TYPE_MISMATCH, "Wrong result type {}. Expected IPv4", result_type->getName());
-
-                    const auto * null_map = column_nullable ? &column_nullable->getNullMapData() : nullptr;
-                    if (requested_result_is_nullable)
-                        return convertIPv6ToIPv4<IPStringToNumExceptionMode::Null>(arguments[0].column, null_map);
-                    else if (cast_ipv4_ipv6_default_on_conversion_error_value)
-                        return convertIPv6ToIPv4<IPStringToNumExceptionMode::Default>(arguments[0].column, null_map);
-                    else
-                        return convertIPv6ToIPv4<IPStringToNumExceptionMode::Throw>(arguments[0].column, null_map);
-                };
-
-                return true;
-            }
-
-            if constexpr (WhichDataType(ToDataType::type_id).isStringOrFixedString())
-            {
-                if constexpr (WhichDataType(FromDataType::type_id).isEnum())
-                {
-                    ret = createEnumToStringWrapper<FromDataType>();
-                    return true;
-                }
-                else if (from_type->getCustomSerialization())
-                {
-                    ret = [](ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count) -> ColumnPtr
-                    {
-                        return ConvertImplGenericToString<typename ToDataType::ColumnType>::execute(arguments, result_type, input_rows_count);
-                    };
-                    return true;
-                }
-            }
-
-            return false;
-        };
-
-        if (callOnTwoTypeIndexes(from_type->getTypeId(), to_type->getTypeId(), make_custom_serialization_wrapper))
-            return ret;
-
-        if (callOnIndexAndDataType<void>(to_type->getTypeId(), make_default_wrapper))
-            return ret;
-
-        switch (to_type->getTypeId())
-        {
-            case TypeIndex::String:
-                return createStringWrapper(from_type);
-            case TypeIndex::FixedString:
-                return createFixedStringWrapper(from_type, checkAndGetDataType<DataTypeFixedString>(to_type.get())->getN());
-            case TypeIndex::Array:
-                return createArrayWrapper(from_type, static_cast<const DataTypeArray &>(*to_type));
-            case TypeIndex::Tuple:
-                return createTupleWrapper(from_type, checkAndGetDataType<DataTypeTuple>(to_type.get()));
-            case TypeIndex::Map:
-                return createMapWrapper(from_type, checkAndGetDataType<DataTypeMap>(to_type.get()));
-            case TypeIndex::Object:
-                return createObjectWrapper(from_type, checkAndGetDataType<DataTypeObject>(to_type.get()));
-            case TypeIndex::AggregateFunction:
-                return createAggregateFunctionWrapper(from_type, checkAndGetDataType<DataTypeAggregateFunction>(to_type.get()));
-            case TypeIndex::Interval:
-                return createIntervalWrapper(from_type, checkAndGetDataType<DataTypeInterval>(to_type.get())->getKind());
-            default:
-                break;
-        }
-
-        if (cast_type == CastType::accurateOrNull)
-            return createToNullableColumnWrapper();
-        else
-            throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Conversion from {} to {} is not supported",
-                from_type->getName(), to_type->getName());
-    }
-};
-
-class MonotonicityHelper
-{
-public:
-    using MonotonicityForRange = FunctionCastBase::MonotonicityForRange;
-
-    template <typename DataType>
-    static auto monotonicityForType(const DataType * const)
-    {
-        return FunctionTo<DataType>::Type::Monotonic::get;
-    }
-
-    static MonotonicityForRange getMonotonicityInformation(const DataTypePtr & from_type, const IDataType * to_type);
-};
-
-FunctionBasePtr createFunctionBaseCast(
-    ContextPtr context
-    , const ColumnsWithTypeAndName & arguments
-    , const DataTypePtr & return_type
-    , std::optional<CastDiagnostic> diagnostic
-    , CastType cast_type);
-
-}

From 70796e497f1970d186d017226e93a67f1c6d170f Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 9 Mar 2024 08:32:13 +0100
Subject: [PATCH 106/443] Miscellaneous

---
 programs/library-bridge/createFunctionBaseCast.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/programs/library-bridge/createFunctionBaseCast.cpp b/programs/library-bridge/createFunctionBaseCast.cpp
index 473aa1ca81d..194fc4bfcf7 100644
--- a/programs/library-bridge/createFunctionBaseCast.cpp
+++ b/programs/library-bridge/createFunctionBaseCast.cpp
@@ -13,7 +13,7 @@ namespace ErrorCodes
 FunctionBasePtr createFunctionBaseCast(
     ContextPtr, const ColumnsWithTypeAndName &, const DataTypePtr &, std::optional<CastDiagnostic>, CastType)
 {
-    throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Type conversions are not implemented for ODBC Bridge");
+    throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Type conversions are not implemented for Library Bridge");
 }
 
 }

From 3cd70f48b0d2ef1584a9351291de4d230e71c021 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 9 Mar 2024 08:49:55 +0100
Subject: [PATCH 107/443] Anonymous

---
 src/Functions/CastOverloadResolver.cpp | 15 -----------
 src/Functions/FunctionsConversion.cpp  | 37 +++++++++++++-------------
 2 files changed, 19 insertions(+), 33 deletions(-)

diff --git a/src/Functions/CastOverloadResolver.cpp b/src/Functions/CastOverloadResolver.cpp
index 4a081d684f6..ad4b28d11c1 100644
--- a/src/Functions/CastOverloadResolver.cpp
+++ b/src/Functions/CastOverloadResolver.cpp
@@ -117,21 +117,6 @@ private:
 };
 
 
-UInt32 extractToDecimalScale(const ColumnWithTypeAndName & named_column)
-{
-    const auto * arg_type = named_column.type.get();
-    bool ok = checkAndGetDataType<DataTypeUInt64>(arg_type)
-        || checkAndGetDataType<DataTypeUInt32>(arg_type)
-        || checkAndGetDataType<DataTypeUInt16>(arg_type)
-        || checkAndGetDataType<DataTypeUInt8>(arg_type);
-    if (!ok)
-        throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type of toDecimal() scale {}", named_column.type->getName());
-
-    Field field;
-    named_column.column->get(0, field);
-    return static_cast<UInt32>(field.get<UInt32>());
-}
-
 FunctionOverloadResolverPtr createInternalCastOverloadResolver(CastType type, std::optional<CastDiagnostic> diagnostic)
 {
     return CastOverloadResolverImpl::create(ContextPtr{}, type, true, diagnostic);
diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 865f7db8e12..3da6b9abe2d 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -93,11 +93,27 @@ namespace ErrorCodes
     extern const int VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE;
 }
 
+namespace
+{
+
 /** Type conversion functions.
   * toType - conversion in "natural way";
   */
 
-UInt32 extractToDecimalScale(const ColumnWithTypeAndName & named_column);
+UInt32 extractToDecimalScale(const ColumnWithTypeAndName & named_column)
+{
+    const auto * arg_type = named_column.type.get();
+    bool ok = checkAndGetDataType<DataTypeUInt64>(arg_type)
+        || checkAndGetDataType<DataTypeUInt32>(arg_type)
+        || checkAndGetDataType<DataTypeUInt16>(arg_type)
+        || checkAndGetDataType<DataTypeUInt8>(arg_type);
+    if (!ok)
+        throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type of toDecimal() scale {}", named_column.type->getName());
+
+    Field field;
+    named_column.column->get(0, field);
+    return static_cast<UInt32>(field.get<UInt32>());
+}
 
 /// Function toUnixTimestamp has exactly the same implementation as toDateTime of String type.
 struct NameToUnixTimestamp { static constexpr auto name = "toUnixTimestamp"; };
@@ -1271,23 +1287,6 @@ void convertFromTime(typename DataType::FieldType & x, time_t & time)
     x = time;
 }
 
-template <>
-inline void convertFromTime<DataTypeDate>(DataTypeDate::FieldType & x, time_t & time)
-{
-    if (unlikely(time < 0))
-        x = 0;
-    else if (unlikely(time > 0xFFFF))
-        x = 0xFFFF;
-    else
-        x = time;
-}
-
-template <>
-inline void convertFromTime<DataTypeDate32>(DataTypeDate32::FieldType & x, time_t & time)
-{
-    x = static_cast<UInt32>(time);
-}
-
 template <>
 inline void convertFromTime<DataTypeDateTime>(DataTypeDateTime::FieldType & x, time_t & time)
 {
@@ -4960,6 +4959,8 @@ public:
     }
 };
 
+}
+
 
 FunctionBasePtr createFunctionBaseCast(
     ContextPtr context

From d2e19a5da342236eb377e40a79a1fcd467b13b7a Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 9 Mar 2024 08:55:53 +0100
Subject: [PATCH 108/443] Remove unused header

---
 src/Functions/CastOverloadResolver.cpp | 1 -
 1 file changed, 1 deletion(-)

diff --git a/src/Functions/CastOverloadResolver.cpp b/src/Functions/CastOverloadResolver.cpp
index ad4b28d11c1..1e1fbd06d93 100644
--- a/src/Functions/CastOverloadResolver.cpp
+++ b/src/Functions/CastOverloadResolver.cpp
@@ -2,7 +2,6 @@
 #include <Functions/CastOverloadResolver.h>
 #include <Functions/FunctionHelpers.h>
 #include <DataTypes/DataTypeFactory.h>
-#include <DataTypes/DataTypesNumber.h>
 #include <DataTypes/DataTypeNullable.h>
 #include <Columns/ColumnString.h>
 #include <Interpreters/parseColumnsListForTableFunction.h>

From 8a46eace13458a735ccce5f423f6b750b0db869a Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 9 Mar 2024 09:15:56 +0100
Subject: [PATCH 109/443] Less exceptions

---
 utils/check-style/check-large-objects.sh | 1 -
 1 file changed, 1 deletion(-)

diff --git a/utils/check-style/check-large-objects.sh b/utils/check-style/check-large-objects.sh
index 5c1276e5732..4eb9190512f 100755
--- a/utils/check-style/check-large-objects.sh
+++ b/utils/check-style/check-large-objects.sh
@@ -3,7 +3,6 @@
 # Check that there are no new translation units compiled to an object file larger than a certain size.
 
 TU_EXCLUDES=(
-    CastOverloadResolver
     AggregateFunctionUniq
     FunctionsConversion
 

From 194de2066c1e60a15d48f21333e97e5ec2d2f82b Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 9 Mar 2024 09:22:14 +0100
Subject: [PATCH 110/443] Loosen

---
 CMakeLists.txt | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/CMakeLists.txt b/CMakeLists.txt
index b55e9810361..9ffb4789dc9 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -61,8 +61,8 @@ if (ENABLE_CHECK_HEAVY_BUILDS)
     # set CPU time limit to 1000 seconds
     set (RLIMIT_CPU 1000)
 
-    # -fsanitize=memory is too heavy
-    if (SANITIZE STREQUAL "memory")
+    # -fsanitize=memory and address are too heavy
+    if (SANITIZE)
        set (RLIMIT_DATA 10000000000) # 10G
     endif()
 

From e9ab3ed2dd8ec6e81a5125c27ae4ad30882ecddf Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 9 Mar 2024 09:27:11 +0100
Subject: [PATCH 111/443] Even better

---
 programs/library-bridge/LibraryBridge.h            | 2 +-
 programs/library-bridge/createFunctionBaseCast.cpp | 6 +++++-
 programs/odbc-bridge/createFunctionBaseCast.cpp    | 8 ++++++--
 3 files changed, 12 insertions(+), 4 deletions(-)

diff --git a/programs/library-bridge/LibraryBridge.h b/programs/library-bridge/LibraryBridge.h
index 04860a042a3..a8d15a87e07 100644
--- a/programs/library-bridge/LibraryBridge.h
+++ b/programs/library-bridge/LibraryBridge.h
@@ -1,6 +1,6 @@
 #pragma once
 
-#include <Interpreters/Context.h>
+#include <Interpreters/Context_fwd.h>
 #include <Bridge/IBridge.h>
 #include "LibraryBridgeHandlerFactory.h"
 
diff --git a/programs/library-bridge/createFunctionBaseCast.cpp b/programs/library-bridge/createFunctionBaseCast.cpp
index 194fc4bfcf7..54319ca707a 100644
--- a/programs/library-bridge/createFunctionBaseCast.cpp
+++ b/programs/library-bridge/createFunctionBaseCast.cpp
@@ -1,5 +1,6 @@
-#include <Functions/IFunction.h>
+#include <memory>
 #include <Functions/CastOverloadResolver.h>
+#include <Core/ColumnsWithTypeAndName.h>
 
 
 namespace DB
@@ -10,6 +11,9 @@ namespace ErrorCodes
     extern const int NOT_IMPLEMENTED;
 }
 
+class IFunctionBase;
+using FunctionBasePtr = std::shared_ptr<const IFunctionBase>;
+
 FunctionBasePtr createFunctionBaseCast(
     ContextPtr, const ColumnsWithTypeAndName &, const DataTypePtr &, std::optional<CastDiagnostic>, CastType)
 {
diff --git a/programs/odbc-bridge/createFunctionBaseCast.cpp b/programs/odbc-bridge/createFunctionBaseCast.cpp
index 473aa1ca81d..54319ca707a 100644
--- a/programs/odbc-bridge/createFunctionBaseCast.cpp
+++ b/programs/odbc-bridge/createFunctionBaseCast.cpp
@@ -1,5 +1,6 @@
-#include <Functions/IFunction.h>
+#include <memory>
 #include <Functions/CastOverloadResolver.h>
+#include <Core/ColumnsWithTypeAndName.h>
 
 
 namespace DB
@@ -10,10 +11,13 @@ namespace ErrorCodes
     extern const int NOT_IMPLEMENTED;
 }
 
+class IFunctionBase;
+using FunctionBasePtr = std::shared_ptr<const IFunctionBase>;
+
 FunctionBasePtr createFunctionBaseCast(
     ContextPtr, const ColumnsWithTypeAndName &, const DataTypePtr &, std::optional<CastDiagnostic>, CastType)
 {
-    throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Type conversions are not implemented for ODBC Bridge");
+    throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Type conversions are not implemented for Library Bridge");
 }
 
 }

From 3ac4f56cfa65443119bd0fa9d23d528adf73d7bb Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 9 Mar 2024 18:53:31 +0100
Subject: [PATCH 112/443] Fix tests

---
 .../library-bridge/createFunctionBaseCast.cpp |  2 +-
 .../odbc-bridge/createFunctionBaseCast.cpp    |  2 +-
 src/Functions/CastOverloadResolver.cpp        | 20 ++++++++++++-------
 src/Functions/FunctionsConversion.cpp         | 13 ++++++------
 4 files changed, 22 insertions(+), 15 deletions(-)

diff --git a/programs/library-bridge/createFunctionBaseCast.cpp b/programs/library-bridge/createFunctionBaseCast.cpp
index 54319ca707a..dcdd47d79ce 100644
--- a/programs/library-bridge/createFunctionBaseCast.cpp
+++ b/programs/library-bridge/createFunctionBaseCast.cpp
@@ -15,7 +15,7 @@ class IFunctionBase;
 using FunctionBasePtr = std::shared_ptr<const IFunctionBase>;
 
 FunctionBasePtr createFunctionBaseCast(
-    ContextPtr, const ColumnsWithTypeAndName &, const DataTypePtr &, std::optional<CastDiagnostic>, CastType)
+    ContextPtr, const char *, const ColumnsWithTypeAndName &, const DataTypePtr &, std::optional<CastDiagnostic>, CastType)
 {
     throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Type conversions are not implemented for Library Bridge");
 }
diff --git a/programs/odbc-bridge/createFunctionBaseCast.cpp b/programs/odbc-bridge/createFunctionBaseCast.cpp
index 54319ca707a..dcdd47d79ce 100644
--- a/programs/odbc-bridge/createFunctionBaseCast.cpp
+++ b/programs/odbc-bridge/createFunctionBaseCast.cpp
@@ -15,7 +15,7 @@ class IFunctionBase;
 using FunctionBasePtr = std::shared_ptr<const IFunctionBase>;
 
 FunctionBasePtr createFunctionBaseCast(
-    ContextPtr, const ColumnsWithTypeAndName &, const DataTypePtr &, std::optional<CastDiagnostic>, CastType)
+    ContextPtr, const char *, const ColumnsWithTypeAndName &, const DataTypePtr &, std::optional<CastDiagnostic>, CastType)
 {
     throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Type conversions are not implemented for Library Bridge");
 }
diff --git a/src/Functions/CastOverloadResolver.cpp b/src/Functions/CastOverloadResolver.cpp
index 1e1fbd06d93..4e8e99b4d95 100644
--- a/src/Functions/CastOverloadResolver.cpp
+++ b/src/Functions/CastOverloadResolver.cpp
@@ -17,11 +17,12 @@ namespace ErrorCodes
 }
 
 FunctionBasePtr createFunctionBaseCast(
-    ContextPtr context
-    , const ColumnsWithTypeAndName & arguments
-    , const DataTypePtr & return_type
-    , std::optional<CastDiagnostic> diagnostic
-    , CastType cast_type);
+    ContextPtr context,
+    const char * name,
+    const ColumnsWithTypeAndName & arguments,
+    const DataTypePtr & return_type,
+    std::optional<CastDiagnostic> diagnostic,
+    CastType cast_type);
 
 
 /** CastInternal does not preserve nullability of the data type,
@@ -33,7 +34,7 @@ FunctionBasePtr createFunctionBaseCast(
 class CastOverloadResolverImpl : public IFunctionOverloadResolver
 {
 public:
-    String getName() const override
+    const char * getNameImpl() const
     {
         if (cast_type == CastType::accurate)
             return "accurateCast";
@@ -45,6 +46,11 @@ public:
             return "CAST";
     }
 
+    String getName() const override
+    {
+        return getNameImpl();
+    }
+
     size_t getNumberOfArguments() const override { return 2; }
 
     ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {1}; }
@@ -72,7 +78,7 @@ public:
 protected:
     FunctionBasePtr buildImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & return_type) const override
     {
-        return createFunctionBaseCast(context, arguments, return_type, diagnostic, cast_type);
+        return createFunctionBaseCast(context, getNameImpl(), arguments, return_type, diagnostic, cast_type);
     }
 
     DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override
diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 3da6b9abe2d..d1d00780205 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -4963,11 +4963,12 @@ public:
 
 
 FunctionBasePtr createFunctionBaseCast(
-    ContextPtr context
-    , const ColumnsWithTypeAndName & arguments
-    , const DataTypePtr & return_type
-    , std::optional<CastDiagnostic> diagnostic
-    , CastType cast_type)
+    ContextPtr context,
+    const char * name,
+    const ColumnsWithTypeAndName & arguments,
+    const DataTypePtr & return_type,
+    std::optional<CastDiagnostic> diagnostic,
+    CastType cast_type)
 {
     DataTypes data_types(arguments.size());
 
@@ -4975,7 +4976,7 @@ FunctionBasePtr createFunctionBaseCast(
         data_types[i] = arguments[i].type;
 
     auto monotonicity = MonotonicityHelper::getMonotonicityInformation(arguments.front().type, return_type.get());
-    return std::make_unique<FunctionCast>(context, "CAST", std::move(monotonicity), data_types, return_type, diagnostic, cast_type);
+    return std::make_unique<FunctionCast>(context, name, std::move(monotonicity), data_types, return_type, diagnostic, cast_type);
 }
 
 REGISTER_FUNCTION(Conversion)

From e30a1c9cbcc465370aa3be5c96437598c1a3ac80 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 10 Mar 2024 02:26:18 +0100
Subject: [PATCH 113/443] Remove garbage

---
 src/Functions/FunctionsConversion.cpp | 9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index d1d00780205..5347e1f739d 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -3158,15 +3158,10 @@ struct FunctionCastName
     static constexpr auto name = "CAST";
 };
 
-class FunctionCastBase : public IFunctionBase
+class FunctionCast final : public IFunctionBase
 {
 public:
     using MonotonicityForRange = std::function<Monotonicity(const IDataType &, const Field &, const Field &)>;
-};
-
-class FunctionCast final : public FunctionCastBase
-{
-public:
     using WrapperType = std::function<ColumnPtr(ColumnsWithTypeAndName &, const DataTypePtr &, const ColumnNullable *, size_t)>;
 
     FunctionCast(ContextPtr context_
@@ -4901,7 +4896,7 @@ arguments, result_type, input_rows_count); \
 class MonotonicityHelper
 {
 public:
-    using MonotonicityForRange = FunctionCastBase::MonotonicityForRange;
+    using MonotonicityForRange = FunctionCast::MonotonicityForRange;
 
     template <typename DataType>
     static auto monotonicityForType(const DataType * const)

From bbeecf4ed5361154c2eed61f476ed9ca2c50b773 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 10 Mar 2024 02:51:32 +0100
Subject: [PATCH 114/443] Fix error

---
 src/Functions/CastOverloadResolver.cpp | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/src/Functions/CastOverloadResolver.cpp b/src/Functions/CastOverloadResolver.cpp
index 4e8e99b4d95..5ca4b0bc579 100644
--- a/src/Functions/CastOverloadResolver.cpp
+++ b/src/Functions/CastOverloadResolver.cpp
@@ -67,12 +67,15 @@ public:
 
     static FunctionOverloadResolverPtr create(ContextPtr context, CastType cast_type, bool internal, std::optional<CastDiagnostic> diagnostic)
     {
-        const auto & settings_ref = context->getSettingsRef();
-
         if (internal)
+        {
             return std::make_unique<CastOverloadResolverImpl>(context, cast_type, internal, diagnostic, false /*keep_nullable*/, DataTypeValidationSettings{});
+        }
         else
+        {
+            const auto & settings_ref = context->getSettingsRef();
             return std::make_unique<CastOverloadResolverImpl>(context, cast_type, internal, diagnostic, settings_ref.cast_keep_nullable, DataTypeValidationSettings(settings_ref));
+        }
     }
 
 protected:

From 6c0f8773426021d93229b2c713641e189c4ffe3e Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 10 Mar 2024 03:02:38 +0100
Subject: [PATCH 115/443] I did not understand this code and removed it

---
 src/Functions/FunctionsConversion.cpp | 12 ------------
 1 file changed, 12 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 5347e1f739d..38c18ddf850 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -1906,18 +1906,6 @@ struct ConvertImpl<T, T, Name, ConvertDefaultBehaviorTag>
     }
 };
 
-template <typename Name>
-struct ConvertImpl<DataTypeUInt8, DataTypeUInt8, Name, ConvertDefaultBehaviorTag>
-{
-    template <typename Additions = void *>
-    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/,
-                             Additions additions [[maybe_unused]] = Additions())
-    {
-
-        return arguments[0].column;
-    }
-};
-
 
 /** Conversion from FixedString to String.
   * Cutting sequences of zero bytes from end of strings.

From 7c8a97a91a0842556e1ed67234c85c7aead67700 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 10 Mar 2024 03:13:15 +0100
Subject: [PATCH 116/443] Maybe better

---
 src/Functions/FunctionsConversion.cpp | 425 +++++++++++++-------------
 1 file changed, 206 insertions(+), 219 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 38c18ddf850..1d849446254 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -151,255 +151,256 @@ struct ConvertImpl
     {
         const ColumnWithTypeAndName & named_from = arguments[0];
 
+        /// If types are the same, reuse the columns.
+        if (result_type->equals(*named_from.type))
+            return named_from.column;
+
         using ColVecFrom = typename FromDataType::ColumnType;
         using ColVecTo = typename ToDataType::ColumnType;
 
         if constexpr ((IsDataTypeDecimal<FromDataType> || IsDataTypeDecimal<ToDataType>)
-            && !(std::is_same_v<DataTypeDateTime64, FromDataType> || std::is_same_v<DataTypeDateTime64, ToDataType>))
+            && !(std::is_same_v<DataTypeDateTime64, FromDataType> || std::is_same_v<DataTypeDateTime64, ToDataType>)
+            && (!IsDataTypeDecimalOrNumber<FromDataType> || !IsDataTypeDecimalOrNumber<ToDataType>))
         {
-            if constexpr (!IsDataTypeDecimalOrNumber<FromDataType> || !IsDataTypeDecimalOrNumber<ToDataType>)
-            {
-                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                    named_from.column->getName(), Name::name);
-            }
+            throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
+                named_from.column->getName(), Name::name);
         }
 
-        if (const ColVecFrom * col_from = checkAndGetColumn<ColVecFrom>(named_from.column.get()))
+        const ColVecFrom * col_from = checkAndGetColumn<ColVecFrom>(named_from.column.get());
+        if (!col_from)
+            throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
+                named_from.column->getName(), Name::name);
+
+        typename ColVecTo::MutablePtr col_to = nullptr;
+
+        if constexpr (IsDataTypeDecimal<ToDataType>)
         {
-            typename ColVecTo::MutablePtr col_to = nullptr;
+            UInt32 scale;
 
-            if constexpr (IsDataTypeDecimal<ToDataType>)
+            if constexpr (std::is_same_v<Additions, AccurateConvertStrategyAdditions>
+                || std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
             {
-                UInt32 scale;
-
-                if constexpr (std::is_same_v<Additions, AccurateConvertStrategyAdditions>
-                    || std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-                {
-                    scale = additions.scale;
-                }
-                else
-                {
-                    scale = additions;
-                }
-
-                col_to = ColVecTo::create(0, scale);
+                scale = additions.scale;
             }
             else
-                col_to = ColVecTo::create();
-
-            const auto & vec_from = col_from->getData();
-            auto & vec_to = col_to->getData();
-            vec_to.resize(input_rows_count);
-
-            ColumnUInt8::MutablePtr col_null_map_to;
-            ColumnUInt8::Container * vec_null_map_to [[maybe_unused]] = nullptr;
-            if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
             {
-                col_null_map_to = ColumnUInt8::create(input_rows_count, false);
-                vec_null_map_to = &col_null_map_to->getData();
+                scale = additions;
             }
 
-            bool result_is_bool = isBool(result_type);
-            for (size_t i = 0; i < input_rows_count; ++i)
+            col_to = ColVecTo::create(0, scale);
+        }
+        else
+            col_to = ColVecTo::create();
+
+        const auto & vec_from = col_from->getData();
+        auto & vec_to = col_to->getData();
+        vec_to.resize(input_rows_count);
+
+        ColumnUInt8::MutablePtr col_null_map_to;
+        ColumnUInt8::Container * vec_null_map_to [[maybe_unused]] = nullptr;
+        if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
+        {
+            col_null_map_to = ColumnUInt8::create(input_rows_count, false);
+            vec_null_map_to = &col_null_map_to->getData();
+        }
+
+        bool result_is_bool = isBool(result_type);
+        for (size_t i = 0; i < input_rows_count; ++i)
+        {
+            if constexpr (std::is_same_v<ToDataType, DataTypeUInt8>)
             {
-                if constexpr (std::is_same_v<ToDataType, DataTypeUInt8>)
+                if (result_is_bool)
                 {
-                    if (result_is_bool)
-                    {
-                        vec_to[i] = vec_from[i] != FromFieldType(0);
-                        continue;
-                    }
-                }
-
-                if constexpr (std::is_same_v<FromDataType, DataTypeUUID> && std::is_same_v<ToDataType, DataTypeUInt128>)
-                {
-                    static_assert(
-                        std::is_same_v<DataTypeUInt128::FieldType, DataTypeUUID::FieldType::UnderlyingType>,
-                        "UInt128 and UUID types must be same");
-
-                    vec_to[i].items[1] = vec_from[i].toUnderType().items[0];
-                    vec_to[i].items[0] = vec_from[i].toUnderType().items[1];
-
+                    vec_to[i] = vec_from[i] != FromFieldType(0);
                     continue;
                 }
+            }
 
-                if constexpr (std::is_same_v<FromDataType, DataTypeIPv6> && std::is_same_v<ToDataType, DataTypeUInt128>)
-                {
-                    static_assert(
-                        std::is_same_v<DataTypeUInt128::FieldType, DataTypeUUID::FieldType::UnderlyingType>,
-                        "UInt128 and IPv6 types must be same");
+            if constexpr (std::is_same_v<FromDataType, DataTypeUUID> && std::is_same_v<ToDataType, DataTypeUInt128>)
+            {
+                static_assert(
+                    std::is_same_v<DataTypeUInt128::FieldType, DataTypeUUID::FieldType::UnderlyingType>,
+                    "UInt128 and UUID types must be same");
 
-                    vec_to[i].items[1] = std::byteswap(vec_from[i].toUnderType().items[0]);
-                    vec_to[i].items[0] = std::byteswap(vec_from[i].toUnderType().items[1]);
+                vec_to[i].items[1] = vec_from[i].toUnderType().items[0];
+                vec_to[i].items[0] = vec_from[i].toUnderType().items[1];
 
-                    continue;
-                }
+                continue;
+            }
 
-                if constexpr (std::is_same_v<FromDataType, DataTypeUUID> != std::is_same_v<ToDataType, DataTypeUUID>)
+            if constexpr (std::is_same_v<FromDataType, DataTypeIPv6> && std::is_same_v<ToDataType, DataTypeUInt128>)
+            {
+                static_assert(
+                    std::is_same_v<DataTypeUInt128::FieldType, DataTypeUUID::FieldType::UnderlyingType>,
+                    "UInt128 and IPv6 types must be same");
+
+                vec_to[i].items[1] = std::byteswap(vec_from[i].toUnderType().items[0]);
+                vec_to[i].items[0] = std::byteswap(vec_from[i].toUnderType().items[1]);
+
+                continue;
+            }
+
+            if constexpr (std::is_same_v<FromDataType, DataTypeUUID> != std::is_same_v<ToDataType, DataTypeUUID>)
+            {
+                throw Exception(ErrorCodes::NOT_IMPLEMENTED,
+                                "Conversion between numeric types and UUID is not supported. "
+                                "Probably the passed UUID is unquoted");
+            }
+            else if constexpr (
+                (std::is_same_v<FromDataType, DataTypeIPv4> != std::is_same_v<ToDataType, DataTypeIPv4>)
+                && !(is_any_of<FromDataType, DataTypeUInt8, DataTypeUInt16, DataTypeUInt32, DataTypeUInt64, DataTypeIPv6> || is_any_of<ToDataType, DataTypeUInt32, DataTypeUInt64, DataTypeUInt128, DataTypeUInt256, DataTypeIPv6>)
+            )
+            {
+                throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Conversion from {} to {} is not supported",
+                                TypeName<typename FromDataType::FieldType>, TypeName<typename ToDataType::FieldType>);
+            }
+            else if constexpr (std::is_same_v<FromDataType, DataTypeIPv6> != std::is_same_v<ToDataType, DataTypeIPv6> && !(std::is_same_v<ToDataType, DataTypeIPv4> || std::is_same_v<FromDataType, DataTypeIPv4>))
+            {
+                throw Exception(ErrorCodes::NOT_IMPLEMENTED,
+                                "Conversion between numeric types and IPv6 is not supported. "
+                                "Probably the passed IPv6 is unquoted");
+            }
+            else
+            {
+                if constexpr (IsDataTypeDecimal<FromDataType> || IsDataTypeDecimal<ToDataType>)
                 {
-                    throw Exception(ErrorCodes::NOT_IMPLEMENTED,
-                                    "Conversion between numeric types and UUID is not supported. "
-                                    "Probably the passed UUID is unquoted");
-                }
-                else if constexpr (
-                    (std::is_same_v<FromDataType, DataTypeIPv4> != std::is_same_v<ToDataType, DataTypeIPv4>)
-                    && !(is_any_of<FromDataType, DataTypeUInt8, DataTypeUInt16, DataTypeUInt32, DataTypeUInt64, DataTypeIPv6> || is_any_of<ToDataType, DataTypeUInt32, DataTypeUInt64, DataTypeUInt128, DataTypeUInt256, DataTypeIPv6>)
-                )
-                {
-                    throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Conversion from {} to {} is not supported",
-                                    TypeName<typename FromDataType::FieldType>, TypeName<typename ToDataType::FieldType>);
-                }
-                else if constexpr (std::is_same_v<FromDataType, DataTypeIPv6> != std::is_same_v<ToDataType, DataTypeIPv6> && !(std::is_same_v<ToDataType, DataTypeIPv4> || std::is_same_v<FromDataType, DataTypeIPv4>))
-                {
-                    throw Exception(ErrorCodes::NOT_IMPLEMENTED,
-                                    "Conversion between numeric types and IPv6 is not supported. "
-                                    "Probably the passed IPv6 is unquoted");
-                }
-                else
-                {
-                    if constexpr (IsDataTypeDecimal<FromDataType> || IsDataTypeDecimal<ToDataType>)
+                    if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
                     {
-                        if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-                        {
-                            ToFieldType result;
-                            bool convert_result = false;
+                        ToFieldType result;
+                        bool convert_result = false;
 
-                            if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeDecimal<ToDataType>)
-                                convert_result = tryConvertDecimals<FromDataType, ToDataType>(vec_from[i], col_from->getScale(), col_to->getScale(), result);
-                            else if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeNumber<ToDataType>)
-                                convert_result = tryConvertFromDecimal<FromDataType, ToDataType>(vec_from[i], col_from->getScale(), result);
-                            else if constexpr (IsDataTypeNumber<FromDataType> && IsDataTypeDecimal<ToDataType>)
-                                convert_result = tryConvertToDecimal<FromDataType, ToDataType>(vec_from[i], col_to->getScale(), result);
+                        if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeDecimal<ToDataType>)
+                            convert_result = tryConvertDecimals<FromDataType, ToDataType>(vec_from[i], col_from->getScale(), col_to->getScale(), result);
+                        else if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeNumber<ToDataType>)
+                            convert_result = tryConvertFromDecimal<FromDataType, ToDataType>(vec_from[i], col_from->getScale(), result);
+                        else if constexpr (IsDataTypeNumber<FromDataType> && IsDataTypeDecimal<ToDataType>)
+                            convert_result = tryConvertToDecimal<FromDataType, ToDataType>(vec_from[i], col_to->getScale(), result);
 
-                            if (convert_result)
-                                vec_to[i] = result;
-                            else
-                            {
-                                vec_to[i] = static_cast<ToFieldType>(0);
-                                (*vec_null_map_to)[i] = true;
-                            }
-                        }
+                        if (convert_result)
+                            vec_to[i] = result;
                         else
                         {
-                            if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeDecimal<ToDataType>)
-                                vec_to[i] = convertDecimals<FromDataType, ToDataType>(vec_from[i], col_from->getScale(), col_to->getScale());
-                            else if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeNumber<ToDataType>)
-                                vec_to[i] = convertFromDecimal<FromDataType, ToDataType>(vec_from[i], col_from->getScale());
-                            else if constexpr (IsDataTypeNumber<FromDataType> && IsDataTypeDecimal<ToDataType>)
-                                vec_to[i] = convertToDecimal<FromDataType, ToDataType>(vec_from[i], col_to->getScale());
-                            else
-                                throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Unsupported data type in conversion function");
+                            vec_to[i] = static_cast<ToFieldType>(0);
+                            (*vec_null_map_to)[i] = true;
                         }
                     }
                     else
                     {
-                        /// If From Data is Nan or Inf and we convert to integer type, throw exception
-                        if constexpr (std::is_floating_point_v<FromFieldType> && !std::is_floating_point_v<ToFieldType>)
-                        {
-                            if (!isFinite(vec_from[i]))
-                            {
-                                if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-                                {
-                                    vec_to[i] = 0;
-                                    (*vec_null_map_to)[i] = true;
-                                    continue;
-                                }
-                                else
-                                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Unexpected inf or nan to integer conversion");
-                            }
-                        }
-
-                        if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>
-                                || std::is_same_v<Additions, AccurateConvertStrategyAdditions>)
-                        {
-                            bool convert_result = accurate::convertNumeric(vec_from[i], vec_to[i]);
-
-                            if (!convert_result)
-                            {
-                                if (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-                                {
-                                    vec_to[i] = 0;
-                                    (*vec_null_map_to)[i] = true;
-                                }
-                                else
-                                {
-                                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Value in column {} cannot be safely converted into type {}",
-                                        named_from.column->getName(), result_type->getName());
-                                }
-                            }
-                        }
+                        if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeDecimal<ToDataType>)
+                            vec_to[i] = convertDecimals<FromDataType, ToDataType>(vec_from[i], col_from->getScale(), col_to->getScale());
+                        else if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeNumber<ToDataType>)
+                            vec_to[i] = convertFromDecimal<FromDataType, ToDataType>(vec_from[i], col_from->getScale());
+                        else if constexpr (IsDataTypeNumber<FromDataType> && IsDataTypeDecimal<ToDataType>)
+                            vec_to[i] = convertToDecimal<FromDataType, ToDataType>(vec_from[i], col_to->getScale());
                         else
+                            throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Unsupported data type in conversion function");
+                    }
+                }
+                else
+                {
+                    /// If From Data is Nan or Inf and we convert to integer type, throw exception
+                    if constexpr (std::is_floating_point_v<FromFieldType> && !std::is_floating_point_v<ToFieldType>)
+                    {
+                        if (!isFinite(vec_from[i]))
                         {
-                            if constexpr (std::is_same_v<ToDataType, DataTypeIPv4> && std::is_same_v<FromDataType, DataTypeIPv6>)
+                            if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
                             {
-                                const uint8_t ip4_cidr[] {0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x00, 0x00, 0x00, 0x00};
-                                const uint8_t * src = reinterpret_cast<const uint8_t *>(&vec_from[i].toUnderType());
-                                if (!matchIPv6Subnet(src, ip4_cidr, 96))
-                                {
-                                    char addr[IPV6_MAX_TEXT_LENGTH + 1] {};
-                                    char * paddr = addr;
-                                    formatIPv6(src, paddr);
-
-                                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "IPv6 {} in column {} is not in IPv4 mapping block", addr, named_from.column->getName());
-                                }
-
-                                uint8_t * dst = reinterpret_cast<uint8_t *>(&vec_to[i].toUnderType());
-                                if constexpr (std::endian::native == std::endian::little)
-                                {
-                                    dst[0] = src[15];
-                                    dst[1] = src[14];
-                                    dst[2] = src[13];
-                                    dst[3] = src[12];
-                                }
-                                else
-                                {
-                                    dst[0] = src[12];
-                                    dst[1] = src[13];
-                                    dst[2] = src[14];
-                                    dst[3] = src[15];
-                                }
+                                vec_to[i] = 0;
+                                (*vec_null_map_to)[i] = true;
+                                continue;
                             }
-                            else if constexpr (std::is_same_v<ToDataType, DataTypeIPv6> && std::is_same_v<FromDataType, DataTypeIPv4>)
-                            {
-                                const uint8_t * src = reinterpret_cast<const uint8_t *>(&vec_from[i].toUnderType());
-                                uint8_t * dst = reinterpret_cast<uint8_t *>(&vec_to[i].toUnderType());
-                                std::memset(dst, '\0', IPV6_BINARY_LENGTH);
-                                dst[10] = dst[11] = 0xff;
-
-                                if constexpr (std::endian::native == std::endian::little)
-                                {
-                                    dst[12] = src[3];
-                                    dst[13] = src[2];
-                                    dst[14] = src[1];
-                                    dst[15] = src[0];
-                                }
-                                else
-                                {
-                                    dst[12] = src[0];
-                                    dst[13] = src[1];
-                                    dst[14] = src[2];
-                                    dst[15] = src[3];
-                                }
-                            }
-                            else if constexpr (std::is_same_v<ToDataType, DataTypeIPv4> && std::is_same_v<FromDataType, DataTypeUInt64>)
-                                vec_to[i] = static_cast<ToFieldType>(static_cast<IPv4::UnderlyingType>(vec_from[i]));
-                            else if constexpr (std::is_same_v<Name, NameToUnixTimestamp> && (std::is_same_v<FromDataType, DataTypeDate> || std::is_same_v<FromDataType, DataTypeDate32>))
-                                vec_to[i] = static_cast<ToFieldType>(vec_from[i] * DATE_SECONDS_PER_DAY);
                             else
-                                vec_to[i] = static_cast<ToFieldType>(vec_from[i]);
+                                throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Unexpected inf or nan to integer conversion");
                         }
                     }
+
+                    if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>
+                            || std::is_same_v<Additions, AccurateConvertStrategyAdditions>)
+                    {
+                        bool convert_result = accurate::convertNumeric(vec_from[i], vec_to[i]);
+
+                        if (!convert_result)
+                        {
+                            if (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
+                            {
+                                vec_to[i] = 0;
+                                (*vec_null_map_to)[i] = true;
+                            }
+                            else
+                            {
+                                throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Value in column {} cannot be safely converted into type {}",
+                                    named_from.column->getName(), result_type->getName());
+                            }
+                        }
+                    }
+                    else
+                    {
+                        if constexpr (std::is_same_v<ToDataType, DataTypeIPv4> && std::is_same_v<FromDataType, DataTypeIPv6>)
+                        {
+                            const uint8_t ip4_cidr[] {0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x00, 0x00, 0x00, 0x00};
+                            const uint8_t * src = reinterpret_cast<const uint8_t *>(&vec_from[i].toUnderType());
+                            if (!matchIPv6Subnet(src, ip4_cidr, 96))
+                            {
+                                char addr[IPV6_MAX_TEXT_LENGTH + 1] {};
+                                char * paddr = addr;
+                                formatIPv6(src, paddr);
+
+                                throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "IPv6 {} in column {} is not in IPv4 mapping block", addr, named_from.column->getName());
+                            }
+
+                            uint8_t * dst = reinterpret_cast<uint8_t *>(&vec_to[i].toUnderType());
+                            if constexpr (std::endian::native == std::endian::little)
+                            {
+                                dst[0] = src[15];
+                                dst[1] = src[14];
+                                dst[2] = src[13];
+                                dst[3] = src[12];
+                            }
+                            else
+                            {
+                                dst[0] = src[12];
+                                dst[1] = src[13];
+                                dst[2] = src[14];
+                                dst[3] = src[15];
+                            }
+                        }
+                        else if constexpr (std::is_same_v<ToDataType, DataTypeIPv6> && std::is_same_v<FromDataType, DataTypeIPv4>)
+                        {
+                            const uint8_t * src = reinterpret_cast<const uint8_t *>(&vec_from[i].toUnderType());
+                            uint8_t * dst = reinterpret_cast<uint8_t *>(&vec_to[i].toUnderType());
+                            std::memset(dst, '\0', IPV6_BINARY_LENGTH);
+                            dst[10] = dst[11] = 0xff;
+
+                            if constexpr (std::endian::native == std::endian::little)
+                            {
+                                dst[12] = src[3];
+                                dst[13] = src[2];
+                                dst[14] = src[1];
+                                dst[15] = src[0];
+                            }
+                            else
+                            {
+                                dst[12] = src[0];
+                                dst[13] = src[1];
+                                dst[14] = src[2];
+                                dst[15] = src[3];
+                            }
+                        }
+                        else if constexpr (std::is_same_v<ToDataType, DataTypeIPv4> && std::is_same_v<FromDataType, DataTypeUInt64>)
+                            vec_to[i] = static_cast<ToFieldType>(static_cast<IPv4::UnderlyingType>(vec_from[i]));
+                        else if constexpr (std::is_same_v<Name, NameToUnixTimestamp> && (std::is_same_v<FromDataType, DataTypeDate> || std::is_same_v<FromDataType, DataTypeDate32>))
+                            vec_to[i] = static_cast<ToFieldType>(vec_from[i] * DATE_SECONDS_PER_DAY);
+                        else
+                            vec_to[i] = static_cast<ToFieldType>(vec_from[i]);
+                    }
                 }
             }
-
-            if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-                return ColumnNullable::create(std::move(col_to), std::move(col_null_map_to));
-            else
-                return col_to;
         }
+
+        if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
+            return ColumnNullable::create(std::move(col_to), std::move(col_null_map_to));
         else
-            throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                named_from.column->getName(), Name::name);
+            return col_to;
     }
 };
 
@@ -1892,20 +1893,6 @@ template <>
 struct ConvertImpl<DataTypeString, DataTypeUInt32, NameToUnixTimestamp, ConvertReturnNullOnErrorTag>
     : ConvertImpl<DataTypeString, DataTypeDateTime, NameToUnixTimestamp, ConvertReturnNullOnErrorTag> {};
 
-/** If types are identical, just take reference to column.
-  */
-template <typename T, typename Name>
-requires (!T::is_parametric)
-struct ConvertImpl<T, T, Name, ConvertDefaultBehaviorTag>
-{
-    template <typename Additions = void *>
-    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/,
-        Additions additions [[maybe_unused]] = Additions())
-    {
-        return arguments[0].column;
-    }
-};
-
 
 /** Conversion from FixedString to String.
   * Cutting sequences of zero bytes from end of strings.

From 3c9e6cfc96c8ba7cbc805260bfdec9ca18776ad0 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 10 Mar 2024 03:52:14 +0100
Subject: [PATCH 117/443] Maybe better

---
 src/Functions/FunctionsConversion.cpp | 322 +++++++++++++-------------
 1 file changed, 161 insertions(+), 161 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 1d849446254..66ba2d18f08 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -152,117 +152,113 @@ struct ConvertImpl
         const ColumnWithTypeAndName & named_from = arguments[0];
 
         /// If types are the same, reuse the columns.
-        if (result_type->equals(*named_from.type))
+        if constexpr (std::is_same_v<FromDataType, ToDataType> && !FromDataType::is_parametric)
+        {
             return named_from.column;
-
-        using ColVecFrom = typename FromDataType::ColumnType;
-        using ColVecTo = typename ToDataType::ColumnType;
-
-        if constexpr ((IsDataTypeDecimal<FromDataType> || IsDataTypeDecimal<ToDataType>)
-            && !(std::is_same_v<DataTypeDateTime64, FromDataType> || std::is_same_v<DataTypeDateTime64, ToDataType>)
-            && (!IsDataTypeDecimalOrNumber<FromDataType> || !IsDataTypeDecimalOrNumber<ToDataType>))
-        {
-            throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                named_from.column->getName(), Name::name);
-        }
-
-        const ColVecFrom * col_from = checkAndGetColumn<ColVecFrom>(named_from.column.get());
-        if (!col_from)
-            throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                named_from.column->getName(), Name::name);
-
-        typename ColVecTo::MutablePtr col_to = nullptr;
-
-        if constexpr (IsDataTypeDecimal<ToDataType>)
-        {
-            UInt32 scale;
-
-            if constexpr (std::is_same_v<Additions, AccurateConvertStrategyAdditions>
-                || std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-            {
-                scale = additions.scale;
-            }
-            else
-            {
-                scale = additions;
-            }
-
-            col_to = ColVecTo::create(0, scale);
         }
         else
-            col_to = ColVecTo::create();
-
-        const auto & vec_from = col_from->getData();
-        auto & vec_to = col_to->getData();
-        vec_to.resize(input_rows_count);
-
-        ColumnUInt8::MutablePtr col_null_map_to;
-        ColumnUInt8::Container * vec_null_map_to [[maybe_unused]] = nullptr;
-        if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
         {
-            col_null_map_to = ColumnUInt8::create(input_rows_count, false);
-            vec_null_map_to = &col_null_map_to->getData();
-        }
+            using ColVecFrom = typename FromDataType::ColumnType;
+            using ColVecTo = typename ToDataType::ColumnType;
 
-        bool result_is_bool = isBool(result_type);
-        for (size_t i = 0; i < input_rows_count; ++i)
-        {
-            if constexpr (std::is_same_v<ToDataType, DataTypeUInt8>)
+            if constexpr ((IsDataTypeDecimal<FromDataType> || IsDataTypeDecimal<ToDataType>)
+                && !(std::is_same_v<DataTypeDateTime64, FromDataType> || std::is_same_v<DataTypeDateTime64, ToDataType>)
+                && (!IsDataTypeDecimalOrNumber<FromDataType> || !IsDataTypeDecimalOrNumber<ToDataType>))
             {
-                if (result_is_bool)
+                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
+                    named_from.column->getName(), Name::name);
+            }
+
+            const ColVecFrom * col_from = checkAndGetColumn<ColVecFrom>(named_from.column.get());
+            if (!col_from)
+                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
+                    named_from.column->getName(), Name::name);
+
+            typename ColVecTo::MutablePtr col_to = nullptr;
+
+            if constexpr (IsDataTypeDecimal<ToDataType>)
+            {
+                UInt32 scale;
+
+                if constexpr (std::is_same_v<Additions, AccurateConvertStrategyAdditions>
+                    || std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
                 {
-                    vec_to[i] = vec_from[i] != FromFieldType(0);
-                    continue;
+                    scale = additions.scale;
+                }
+                else
+                {
+                    scale = additions;
                 }
-            }
 
-            if constexpr (std::is_same_v<FromDataType, DataTypeUUID> && std::is_same_v<ToDataType, DataTypeUInt128>)
-            {
-                static_assert(
-                    std::is_same_v<DataTypeUInt128::FieldType, DataTypeUUID::FieldType::UnderlyingType>,
-                    "UInt128 and UUID types must be same");
-
-                vec_to[i].items[1] = vec_from[i].toUnderType().items[0];
-                vec_to[i].items[0] = vec_from[i].toUnderType().items[1];
-
-                continue;
-            }
-
-            if constexpr (std::is_same_v<FromDataType, DataTypeIPv6> && std::is_same_v<ToDataType, DataTypeUInt128>)
-            {
-                static_assert(
-                    std::is_same_v<DataTypeUInt128::FieldType, DataTypeUUID::FieldType::UnderlyingType>,
-                    "UInt128 and IPv6 types must be same");
-
-                vec_to[i].items[1] = std::byteswap(vec_from[i].toUnderType().items[0]);
-                vec_to[i].items[0] = std::byteswap(vec_from[i].toUnderType().items[1]);
-
-                continue;
-            }
-
-            if constexpr (std::is_same_v<FromDataType, DataTypeUUID> != std::is_same_v<ToDataType, DataTypeUUID>)
-            {
-                throw Exception(ErrorCodes::NOT_IMPLEMENTED,
-                                "Conversion between numeric types and UUID is not supported. "
-                                "Probably the passed UUID is unquoted");
-            }
-            else if constexpr (
-                (std::is_same_v<FromDataType, DataTypeIPv4> != std::is_same_v<ToDataType, DataTypeIPv4>)
-                && !(is_any_of<FromDataType, DataTypeUInt8, DataTypeUInt16, DataTypeUInt32, DataTypeUInt64, DataTypeIPv6> || is_any_of<ToDataType, DataTypeUInt32, DataTypeUInt64, DataTypeUInt128, DataTypeUInt256, DataTypeIPv6>)
-            )
-            {
-                throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Conversion from {} to {} is not supported",
-                                TypeName<typename FromDataType::FieldType>, TypeName<typename ToDataType::FieldType>);
-            }
-            else if constexpr (std::is_same_v<FromDataType, DataTypeIPv6> != std::is_same_v<ToDataType, DataTypeIPv6> && !(std::is_same_v<ToDataType, DataTypeIPv4> || std::is_same_v<FromDataType, DataTypeIPv4>))
-            {
-                throw Exception(ErrorCodes::NOT_IMPLEMENTED,
-                                "Conversion between numeric types and IPv6 is not supported. "
-                                "Probably the passed IPv6 is unquoted");
+                col_to = ColVecTo::create(0, scale);
             }
             else
+                col_to = ColVecTo::create();
+
+            const auto & vec_from = col_from->getData();
+            auto & vec_to = col_to->getData();
+            vec_to.resize(input_rows_count);
+
+            ColumnUInt8::MutablePtr col_null_map_to;
+            ColumnUInt8::Container * vec_null_map_to [[maybe_unused]] = nullptr;
+            if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
             {
-                if constexpr (IsDataTypeDecimal<FromDataType> || IsDataTypeDecimal<ToDataType>)
+                col_null_map_to = ColumnUInt8::create(input_rows_count, false);
+                vec_null_map_to = &col_null_map_to->getData();
+            }
+
+            bool result_is_bool = isBool(result_type);
+            for (size_t i = 0; i < input_rows_count; ++i)
+            {
+                if constexpr (std::is_same_v<ToDataType, DataTypeUInt8>)
+                {
+                    if (result_is_bool)
+                    {
+                        vec_to[i] = vec_from[i] != FromFieldType(0);
+                        continue;
+                    }
+                }
+
+                if constexpr (std::is_same_v<FromDataType, DataTypeUUID> && std::is_same_v<ToDataType, DataTypeUInt128>)
+                {
+                    static_assert(
+                        std::is_same_v<DataTypeUInt128::FieldType, DataTypeUUID::FieldType::UnderlyingType>,
+                        "UInt128 and UUID types must be same");
+
+                    vec_to[i].items[1] = vec_from[i].toUnderType().items[0];
+                    vec_to[i].items[0] = vec_from[i].toUnderType().items[1];
+                }
+                else if constexpr (std::is_same_v<FromDataType, DataTypeIPv6> && std::is_same_v<ToDataType, DataTypeUInt128>)
+                {
+                    static_assert(
+                        std::is_same_v<DataTypeUInt128::FieldType, DataTypeUUID::FieldType::UnderlyingType>,
+                        "UInt128 and IPv6 types must be same");
+
+                    vec_to[i].items[1] = std::byteswap(vec_from[i].toUnderType().items[0]);
+                    vec_to[i].items[0] = std::byteswap(vec_from[i].toUnderType().items[1]);
+                }
+                else if constexpr (std::is_same_v<FromDataType, DataTypeUUID> != std::is_same_v<ToDataType, DataTypeUUID>)
+                {
+                    throw Exception(ErrorCodes::NOT_IMPLEMENTED,
+                                    "Conversion between numeric types and UUID is not supported. "
+                                    "Probably the passed UUID is unquoted");
+                }
+                else if constexpr (
+                    (std::is_same_v<FromDataType, DataTypeIPv4> != std::is_same_v<ToDataType, DataTypeIPv4>)
+                    && !(is_any_of<FromDataType, DataTypeUInt8, DataTypeUInt16, DataTypeUInt32, DataTypeUInt64, DataTypeIPv6>
+                        || is_any_of<ToDataType, DataTypeUInt32, DataTypeUInt64, DataTypeUInt128, DataTypeUInt256, DataTypeIPv6>))
+                {
+                    throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Conversion from {} to {} is not supported",
+                                    TypeName<typename FromDataType::FieldType>, TypeName<typename ToDataType::FieldType>);
+                }
+                else if constexpr (std::is_same_v<FromDataType, DataTypeIPv6> != std::is_same_v<ToDataType, DataTypeIPv6>
+                    && !(std::is_same_v<ToDataType, DataTypeIPv4> || std::is_same_v<FromDataType, DataTypeIPv4>))
+                {
+                    throw Exception(ErrorCodes::NOT_IMPLEMENTED,
+                                    "Conversion between numeric types and IPv6 is not supported. "
+                                    "Probably the passed IPv6 is unquoted");
+                }
+                else if constexpr (IsDataTypeDecimal<FromDataType> || IsDataTypeDecimal<ToDataType>)
                 {
                     if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
                     {
@@ -296,6 +292,66 @@ struct ConvertImpl
                             throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Unsupported data type in conversion function");
                     }
                 }
+                else if constexpr (std::is_same_v<ToDataType, DataTypeIPv4> && std::is_same_v<FromDataType, DataTypeIPv6>)
+                {
+                    const uint8_t ip4_cidr[] {0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x00, 0x00, 0x00, 0x00};
+                    const uint8_t * src = reinterpret_cast<const uint8_t *>(&vec_from[i].toUnderType());
+                    if (!matchIPv6Subnet(src, ip4_cidr, 96))
+                    {
+                        char addr[IPV6_MAX_TEXT_LENGTH + 1] {};
+                        char * paddr = addr;
+                        formatIPv6(src, paddr);
+
+                        throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "IPv6 {} in column {} is not in IPv4 mapping block", addr, named_from.column->getName());
+                    }
+
+                    uint8_t * dst = reinterpret_cast<uint8_t *>(&vec_to[i].toUnderType());
+                    if constexpr (std::endian::native == std::endian::little)
+                    {
+                        dst[0] = src[15];
+                        dst[1] = src[14];
+                        dst[2] = src[13];
+                        dst[3] = src[12];
+                    }
+                    else
+                    {
+                        dst[0] = src[12];
+                        dst[1] = src[13];
+                        dst[2] = src[14];
+                        dst[3] = src[15];
+                    }
+                }
+                else if constexpr (std::is_same_v<ToDataType, DataTypeIPv6> && std::is_same_v<FromDataType, DataTypeIPv4>)
+                {
+                    const uint8_t * src = reinterpret_cast<const uint8_t *>(&vec_from[i].toUnderType());
+                    uint8_t * dst = reinterpret_cast<uint8_t *>(&vec_to[i].toUnderType());
+                    std::memset(dst, '\0', IPV6_BINARY_LENGTH);
+                    dst[10] = dst[11] = 0xff;
+
+                    if constexpr (std::endian::native == std::endian::little)
+                    {
+                        dst[12] = src[3];
+                        dst[13] = src[2];
+                        dst[14] = src[1];
+                        dst[15] = src[0];
+                    }
+                    else
+                    {
+                        dst[12] = src[0];
+                        dst[13] = src[1];
+                        dst[14] = src[2];
+                        dst[15] = src[3];
+                    }
+                }
+                else if constexpr (std::is_same_v<ToDataType, DataTypeIPv4> && std::is_same_v<FromDataType, DataTypeUInt64>)
+                {
+                    vec_to[i] = static_cast<ToFieldType>(static_cast<IPv4::UnderlyingType>(vec_from[i]));
+                }
+                else if constexpr (std::is_same_v<Name, NameToUnixTimestamp>
+                    && (std::is_same_v<FromDataType, DataTypeDate> || std::is_same_v<FromDataType, DataTypeDate32>))
+                {
+                    vec_to[i] = static_cast<ToFieldType>(vec_from[i] * DATE_SECONDS_PER_DAY);
+                }
                 else
                 {
                     /// If From Data is Nan or Inf and we convert to integer type, throw exception
@@ -335,72 +391,16 @@ struct ConvertImpl
                     }
                     else
                     {
-                        if constexpr (std::is_same_v<ToDataType, DataTypeIPv4> && std::is_same_v<FromDataType, DataTypeIPv6>)
-                        {
-                            const uint8_t ip4_cidr[] {0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x00, 0x00, 0x00, 0x00};
-                            const uint8_t * src = reinterpret_cast<const uint8_t *>(&vec_from[i].toUnderType());
-                            if (!matchIPv6Subnet(src, ip4_cidr, 96))
-                            {
-                                char addr[IPV6_MAX_TEXT_LENGTH + 1] {};
-                                char * paddr = addr;
-                                formatIPv6(src, paddr);
-
-                                throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "IPv6 {} in column {} is not in IPv4 mapping block", addr, named_from.column->getName());
-                            }
-
-                            uint8_t * dst = reinterpret_cast<uint8_t *>(&vec_to[i].toUnderType());
-                            if constexpr (std::endian::native == std::endian::little)
-                            {
-                                dst[0] = src[15];
-                                dst[1] = src[14];
-                                dst[2] = src[13];
-                                dst[3] = src[12];
-                            }
-                            else
-                            {
-                                dst[0] = src[12];
-                                dst[1] = src[13];
-                                dst[2] = src[14];
-                                dst[3] = src[15];
-                            }
-                        }
-                        else if constexpr (std::is_same_v<ToDataType, DataTypeIPv6> && std::is_same_v<FromDataType, DataTypeIPv4>)
-                        {
-                            const uint8_t * src = reinterpret_cast<const uint8_t *>(&vec_from[i].toUnderType());
-                            uint8_t * dst = reinterpret_cast<uint8_t *>(&vec_to[i].toUnderType());
-                            std::memset(dst, '\0', IPV6_BINARY_LENGTH);
-                            dst[10] = dst[11] = 0xff;
-
-                            if constexpr (std::endian::native == std::endian::little)
-                            {
-                                dst[12] = src[3];
-                                dst[13] = src[2];
-                                dst[14] = src[1];
-                                dst[15] = src[0];
-                            }
-                            else
-                            {
-                                dst[12] = src[0];
-                                dst[13] = src[1];
-                                dst[14] = src[2];
-                                dst[15] = src[3];
-                            }
-                        }
-                        else if constexpr (std::is_same_v<ToDataType, DataTypeIPv4> && std::is_same_v<FromDataType, DataTypeUInt64>)
-                            vec_to[i] = static_cast<ToFieldType>(static_cast<IPv4::UnderlyingType>(vec_from[i]));
-                        else if constexpr (std::is_same_v<Name, NameToUnixTimestamp> && (std::is_same_v<FromDataType, DataTypeDate> || std::is_same_v<FromDataType, DataTypeDate32>))
-                            vec_to[i] = static_cast<ToFieldType>(vec_from[i] * DATE_SECONDS_PER_DAY);
-                        else
-                            vec_to[i] = static_cast<ToFieldType>(vec_from[i]);
+                        vec_to[i] = static_cast<ToFieldType>(vec_from[i]);
                     }
                 }
             }
-        }
 
-        if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-            return ColumnNullable::create(std::move(col_to), std::move(col_null_map_to));
-        else
-            return col_to;
+            if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
+                return ColumnNullable::create(std::move(col_to), std::move(col_null_map_to));
+            else
+                return col_to;
+        }
     }
 };
 

From d2e29525c85458c07f385208582e820de10297e4 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 10 Mar 2024 04:08:34 +0100
Subject: [PATCH 118/443] Maybe better

---
 src/Functions/FunctionsConversion.cpp | 176 +++++++++++++-------------
 1 file changed, 87 insertions(+), 89 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 66ba2d18f08..d2d72558500 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -1,4 +1,3 @@
-#include <cstddef>
 #include <type_traits>
 
 #include <IO/WriteBufferFromVector.h>
@@ -115,6 +114,71 @@ UInt32 extractToDecimalScale(const ColumnWithTypeAndName & named_column)
     return static_cast<UInt32>(field.get<UInt32>());
 }
 
+
+/** Conversion of Date to DateTime: adding 00:00:00 time component.
+  */
+template <FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior = default_date_time_overflow_behavior>
+struct ToDateTimeImpl
+{
+    static constexpr auto name = "toDateTime";
+
+    static UInt32 execute(UInt16 d, const DateLUTImpl & time_zone)
+    {
+        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+        {
+            if (d > MAX_DATETIME_DAY_NUM) [[unlikely]]
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Day number {} is out of bounds of type DateTime", d);
+        }
+        else if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Saturate)
+        {
+            if (d > MAX_DATETIME_DAY_NUM)
+                d = MAX_DATETIME_DAY_NUM;
+        }
+        return static_cast<UInt32>(time_zone.fromDayNum(DayNum(d)));
+    }
+
+    static UInt32 execute(Int32 d, const DateLUTImpl & time_zone)
+    {
+        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Saturate)
+        {
+            if (d < 0)
+                return 0;
+            else if (d > MAX_DATETIME_DAY_NUM)
+                d = MAX_DATETIME_DAY_NUM;
+        }
+        else if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+        {
+            if (d < 0 || d > MAX_DATETIME_DAY_NUM) [[unlikely]]
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type DateTime", d);
+        }
+        return static_cast<UInt32>(time_zone.fromDayNum(ExtendedDayNum(d)));
+    }
+
+    static UInt32 execute(UInt32 dt, const DateLUTImpl & /*time_zone*/)
+    {
+        return dt;
+    }
+
+    static UInt32 execute(Int64 dt64, const DateLUTImpl & /*time_zone*/)
+    {
+        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Ignore)
+            return static_cast<UInt32>(dt64);
+        else
+        {
+            if (dt64 < 0 || dt64 >= MAX_DATETIME_TIMESTAMP)
+            {
+                if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Saturate)
+                    return dt64 < 0 ? 0 : std::numeric_limits<UInt32>::max();
+                else
+                    throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type DateTime", dt64);
+            }
+            else
+                return static_cast<UInt32>(dt64);
+        }
+    }
+};
+
+
 /// Function toUnixTimestamp has exactly the same implementation as toDateTime of String type.
 struct NameToUnixTimestamp { static constexpr auto name = "toUnixTimestamp"; };
 
@@ -151,11 +215,32 @@ struct ConvertImpl
     {
         const ColumnWithTypeAndName & named_from = arguments[0];
 
-        /// If types are the same, reuse the columns.
         if constexpr (std::is_same_v<FromDataType, ToDataType> && !FromDataType::is_parametric)
         {
+            /// If types are the same, reuse the columns.
             return named_from.column;
         }
+        else if constexpr ((std::is_same_v<FromDataType, DataTypeDateTime> || std::is_same_v<FromDataType, DataTypeDate32>)
+            && std::is_same_v<ToDataType, DataTypeDate>)
+        {
+            /// Conversion of DateTime to Date: throw off time component.
+            /// Conversion of Date32 to Date.
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateImpl<date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime> && std::is_same_v<ToDataType, DataTypeDate32>)
+        {
+            /// Conversion of DateTime to Date: throw off time component.
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Impl, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr ((std::is_same_v<FromDataType, DataTypeDate> || std::is_same_v<FromDataType, DataTypeDate32>)
+            && std::is_same_v<ToDataType, DataTypeDateTime>)
+        {
+            /// Conversion from Date/Date32 to DateTime.
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeImpl<date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
         else
         {
             using ColVecFrom = typename FromDataType::ColumnType;
@@ -404,88 +489,6 @@ struct ConvertImpl
     }
 };
 
-/** Conversion of DateTime to Date: throw off time component.
-  */
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDateTime, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDateTime, DataTypeDate, ToDateImpl<date_time_overflow_behavior>, false> {};
-
-/** Conversion of DateTime to Date32: throw off time component.
-  */
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDateTime, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDateTime, DataTypeDate32, ToDate32Impl, false> {};
-
-/** Conversion of Date to DateTime: adding 00:00:00 time component.
-  */
-template <FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior = default_date_time_overflow_behavior>
-struct ToDateTimeImpl
-{
-    static constexpr auto name = "toDateTime";
-
-    static UInt32 execute(UInt16 d, const DateLUTImpl & time_zone)
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (d > MAX_DATETIME_DAY_NUM) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Day number {} is out of bounds of type DateTime", d);
-        }
-        else if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Saturate)
-        {
-            if (d > MAX_DATETIME_DAY_NUM)
-                d = MAX_DATETIME_DAY_NUM;
-        }
-        return static_cast<UInt32>(time_zone.fromDayNum(DayNum(d)));
-    }
-
-    static UInt32 execute(Int32 d, const DateLUTImpl & time_zone)
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Saturate)
-        {
-            if (d < 0)
-                return 0;
-            else if (d > MAX_DATETIME_DAY_NUM)
-                d = MAX_DATETIME_DAY_NUM;
-        }
-        else if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (d < 0 || d > MAX_DATETIME_DAY_NUM) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type DateTime", d);
-        }
-        return static_cast<UInt32>(time_zone.fromDayNum(ExtendedDayNum(d)));
-    }
-
-    static UInt32 execute(UInt32 dt, const DateLUTImpl & /*time_zone*/)
-    {
-        return dt;
-    }
-
-    static UInt32 execute(Int64 dt64, const DateLUTImpl & /*time_zone*/)
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Ignore)
-            return static_cast<UInt32>(dt64);
-        else
-        {
-            if (dt64 < 0 || dt64 >= MAX_DATETIME_TIMESTAMP)
-            {
-                if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Saturate)
-                    return dt64 < 0 ? 0 : std::numeric_limits<UInt32>::max();
-                else
-                    throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type DateTime", dt64);
-            }
-            else
-                return static_cast<UInt32>(dt64);
-        }
-    }
-};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDate, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDate, DataTypeDateTime, ToDateTimeImpl<date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDate32, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDate32, DataTypeDateTime, ToDateTimeImpl<date_time_overflow_behavior>, false> {};
 
 /// Implementation of toDate function.
 
@@ -510,11 +513,6 @@ struct ToDateTransform32Or64
     }
 };
 
-/** Conversion of Date32 to Date.
-  */
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDate32, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDate32, DataTypeDate, ToDateImpl<date_time_overflow_behavior>, false> {};
 
 template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
 struct ToDateTransform32Or64Signed

From d337379385de7270a894687481e7c0251a326349 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 10 Mar 2024 06:44:26 +0100
Subject: [PATCH 119/443] Remove in-memory data parts, step 2

---
 docs/en/operations/system-tables/metrics.md   |   4 -
 src/Common/CurrentMetrics.cpp                 |   1 -
 src/Storages/MergeTree/ColumnSizeEstimator.h  |   3 +-
 src/Storages/MergeTree/DataPartsExchange.cpp  | 119 +------
 src/Storages/MergeTree/DataPartsExchange.h    |  16 -
 src/Storages/MergeTree/IMergeTreeDataPart.cpp |  12 -
 src/Storages/MergeTree/IMergeTreeDataPart.h   |   1 -
 .../IMergeTreeDataPartInfoForReader.h         |   2 -
 .../LoadedMergeTreeDataPartInfoForReader.h    |   2 -
 src/Storages/MergeTree/MergeTask.cpp          |   2 +-
 src/Storages/MergeTree/MergeTreeData.cpp      |  53 +--
 src/Storages/MergeTree/MergeTreeData.h        |   6 +-
 .../MergeTree/MergeTreeDataPartBuilder.cpp    |   3 -
 .../MergeTree/MergeTreeDataPartInMemory.cpp   | 115 ------
 .../MergeTree/MergeTreeDataPartInMemory.h     |  70 ----
 .../MergeTree/MergeTreeDataPartType.h         |   3 -
 .../MergeTree/MergeTreeDataPartWriterWide.h   |   3 +
 .../MergeTree/MergeTreeDataWriter.cpp         |  17 +-
 .../MergeTreeIndexGranularityInfo.cpp         |   4 -
 .../MergeTree/MergeTreeReaderInMemory.cpp     | 119 -------
 .../MergeTree/MergeTreeReaderInMemory.h       |  40 ---
 src/Storages/MergeTree/MergeTreeSink.cpp      |   1 -
 .../MergeTree/MergeTreeWriteAheadLog.cpp      | 336 ------------------
 .../MergeTree/MergeTreeWriteAheadLog.h        | 105 ------
 src/Storages/MergeTree/MutateTask.cpp         |  10 +-
 .../MergeTree/ReplicatedMergeTreeQueue.cpp    |   5 +-
 src/Storages/MergeTree/checkDataPart.cpp      |  12 -
 .../MergeTree/examples/CMakeLists.txt         |   2 -
 .../examples/wal_action_metadata.cpp          |  61 ----
 src/Storages/StorageJoin.cpp                  |   2 +
 src/Storages/StorageMemory.cpp                |   2 +
 src/Storages/StorageMergeTree.cpp             |   1 -
 .../aspell-ignore/en/aspell-dict.txt          |   1 -
 33 files changed, 25 insertions(+), 1108 deletions(-)
 delete mode 100644 src/Storages/MergeTree/MergeTreeDataPartInMemory.cpp
 delete mode 100644 src/Storages/MergeTree/MergeTreeDataPartInMemory.h
 delete mode 100644 src/Storages/MergeTree/MergeTreeReaderInMemory.cpp
 delete mode 100644 src/Storages/MergeTree/MergeTreeReaderInMemory.h
 delete mode 100644 src/Storages/MergeTree/MergeTreeWriteAheadLog.cpp
 delete mode 100644 src/Storages/MergeTree/MergeTreeWriteAheadLog.h
 delete mode 100644 src/Storages/MergeTree/examples/CMakeLists.txt
 delete mode 100644 src/Storages/MergeTree/examples/wal_action_metadata.cpp

diff --git a/docs/en/operations/system-tables/metrics.md b/docs/en/operations/system-tables/metrics.md
index 898e6ae2e2c..83ce817b7db 100644
--- a/docs/en/operations/system-tables/metrics.md
+++ b/docs/en/operations/system-tables/metrics.md
@@ -513,10 +513,6 @@ Part was moved to another disk and should be deleted in own destructor.
 
 Not active data part with identity refcounter, it is deleting right now by a cleaner.
 
-### PartsInMemory
-
-In-memory parts.
-
 ### PartsOutdated
 
 Not active data part, but could be used by only current SELECTs, could be deleted after SELECTs finishes.
diff --git a/src/Common/CurrentMetrics.cpp b/src/Common/CurrentMetrics.cpp
index 82da4c4bbad..b58cea306c6 100644
--- a/src/Common/CurrentMetrics.cpp
+++ b/src/Common/CurrentMetrics.cpp
@@ -226,7 +226,6 @@
     M(PartsDeleteOnDestroy, "Part was moved to another disk and should be deleted in own destructor.") \
     M(PartsWide, "Wide parts.") \
     M(PartsCompact, "Compact parts.") \
-    M(PartsInMemory, "In-memory parts.") \
     M(MMappedFiles, "Total number of mmapped files.") \
     M(MMappedFileBytes, "Sum size of mmapped file regions.") \
     M(AsynchronousReadWait, "Number of threads waiting for asynchronous read.") \
diff --git a/src/Storages/MergeTree/ColumnSizeEstimator.h b/src/Storages/MergeTree/ColumnSizeEstimator.h
index 597dc80e525..a9c4f75ed6b 100644
--- a/src/Storages/MergeTree/ColumnSizeEstimator.h
+++ b/src/Storages/MergeTree/ColumnSizeEstimator.h
@@ -1,7 +1,6 @@
 #pragma once
 
 #include <Storages/MergeTree/IMergeTreeDataPart.h>
-#include <Storages/MergeTree/MergeTreeDataPartInMemory.h>
 
 
 namespace DB
@@ -10,7 +9,7 @@ namespace DB
 /* Allow to compute more accurate progress statistics */
 class ColumnSizeEstimator
 {
-    using ColumnToSize = MergeTreeDataPartInMemory::ColumnToSize;
+    using ColumnToSize = std::map<String, size_t>;
     ColumnToSize map;
 public:
 
diff --git a/src/Storages/MergeTree/DataPartsExchange.cpp b/src/Storages/MergeTree/DataPartsExchange.cpp
index 168c5f729ce..cc2967003a4 100644
--- a/src/Storages/MergeTree/DataPartsExchange.cpp
+++ b/src/Storages/MergeTree/DataPartsExchange.cpp
@@ -10,7 +10,6 @@
 #include <IO/S3Common.h>
 #include <Server/HTTP/HTMLForm.h>
 #include <Server/HTTP/HTTPServerResponse.h>
-#include <Storages/MergeTree/MergeTreeDataPartInMemory.h>
 #include <Storages/MergeTree/MergedBlockOutputStream.h>
 #include <Storages/MergeTree/ReplicatedFetchList.h>
 #include <Storages/StorageReplicatedMergeTree.h>
@@ -191,8 +190,6 @@ void Service::processQuery(const HTMLForm & params, ReadBuffer & /*body*/, Write
         }
 
         if (data_settings->allow_remote_fs_zero_copy_replication &&
-            /// In memory data part does not have metadata yet.
-            !isInMemoryPart(part) &&
             client_protocol_version >= REPLICATION_PROTOCOL_VERSION_WITH_PARTS_ZERO_COPY)
         {
             auto disk_type = part->getDataPartStorage().getDiskType();
@@ -205,11 +202,7 @@ void Service::processQuery(const HTMLForm & params, ReadBuffer & /*body*/, Write
             }
         }
 
-        if (isInMemoryPart(part))
-            sendPartFromMemory(part, out, send_projections);
-        else
-            sendPartFromDisk(part, out, client_protocol_version, false, send_projections);
-
+        sendPartFromDisk(part, out, client_protocol_version, false, send_projections);
         data.addLastSentPart(part->info);
     }
     catch (const NetException &)
@@ -231,36 +224,6 @@ void Service::processQuery(const HTMLForm & params, ReadBuffer & /*body*/, Write
     }
 }
 
-void Service::sendPartFromMemory(
-    const MergeTreeData::DataPartPtr & part, WriteBuffer & out, bool send_projections)
-{
-    auto metadata_snapshot = data.getInMemoryMetadataPtr();
-    if (send_projections)
-    {
-        for (const auto & [name, projection] : part->getProjectionParts())
-        {
-            auto projection_sample_block = metadata_snapshot->projections.get(name).sample_block;
-            auto part_in_memory = asInMemoryPart(projection);
-            if (!part_in_memory)
-                throw Exception(ErrorCodes::LOGICAL_ERROR, "Projection {} of part {} is not stored in memory", name, part->name);
-
-            writeStringBinary(name, out);
-            projection->checksums.write(out);
-            NativeWriter block_out(out, 0, projection_sample_block);
-            block_out.write(part_in_memory->block);
-        }
-    }
-
-    auto part_in_memory = asInMemoryPart(part);
-    if (!part_in_memory)
-        throw Exception(ErrorCodes::LOGICAL_ERROR, "Part {} is not stored in memory", part->name);
-
-    NativeWriter block_out(out, 0, metadata_snapshot->getSampleBlock());
-    part->checksums.write(out);
-    block_out.write(part_in_memory->block);
-
-    data.getSendsThrottler()->add(part_in_memory->block.bytes());
-}
 
 MergeTreeData::DataPart::Checksums Service::sendPartFromDisk(
     const MergeTreeData::DataPartPtr & part,
@@ -641,8 +604,6 @@ std::pair<MergeTreeData::MutableDataPartPtr, scope_guard> Fetcher::fetchSelected
                             remote_fs_metadata, fmt::join(capability, ", "));
         if (server_protocol_version < REPLICATION_PROTOCOL_VERSION_WITH_PARTS_ZERO_COPY)
             throw Exception(ErrorCodes::LOGICAL_ERROR, "Got 'remote_fs_metadata' cookie with old protocol version {}", server_protocol_version);
-        if (part_type == PartType::InMemory)
-            throw Exception(ErrorCodes::INCORRECT_PART_TYPE, "Got 'remote_fs_metadata' cookie for in-memory part");
 
         try
         {
@@ -701,7 +662,7 @@ std::pair<MergeTreeData::MutableDataPartPtr, scope_guard> Fetcher::fetchSelected
     }
 
     auto storage_id = data.getStorageID();
-    String new_part_path = part_type == PartType::InMemory ? "memory" : fs::path(data.getFullPathOnDisk(disk)) / part_name / "";
+    String new_part_path = fs::path(data.getFullPathOnDisk(disk)) / part_name / "";
     auto entry = data.getContext()->getReplicatedFetchList().insert(
         storage_id.getDatabaseName(), storage_id.getTableName(),
         part_info.partition_id, part_name, new_part_path,
@@ -709,22 +670,6 @@ std::pair<MergeTreeData::MutableDataPartPtr, scope_guard> Fetcher::fetchSelected
 
     in->setNextCallback(ReplicatedFetchReadCallback(*entry));
 
-    if (part_type == PartType::InMemory)
-    {
-        auto volume = std::make_shared<SingleDiskVolume>("volume_" + part_name, disk, 0);
-
-        auto data_part_storage = std::make_shared<DataPartStorageOnDiskFull>(
-            volume,
-            data.getRelativeDataPath(),
-            part_name);
-
-        return std::make_pair(downloadPartToMemory(
-            data_part_storage, part_name,
-            MergeTreePartInfo::fromPartName(part_name, data.format_version),
-            part_uuid, metadata_snapshot, context, *in,
-            projections, false, throttler), std::move(temporary_directory_lock));
-    }
-
     auto output_buffer_getter = [](IDataPartStorage & part_storage, const String & file_name, size_t file_size)
     {
         return part_storage.writeFile(file_name, std::min<UInt64>(file_size, DBMS_DEFAULT_BUFFER_SIZE), {});
@@ -736,66 +681,6 @@ std::pair<MergeTreeData::MutableDataPartPtr, scope_guard> Fetcher::fetchSelected
         projections, throttler, sync),std::move(temporary_directory_lock));
 }
 
-MergeTreeData::MutableDataPartPtr Fetcher::downloadPartToMemory(
-    MutableDataPartStoragePtr data_part_storage,
-    const String & part_name,
-    const MergeTreePartInfo & part_info,
-    const UUID & part_uuid,
-    const StorageMetadataPtr & metadata_snapshot,
-    ContextPtr context,
-    PooledReadWriteBufferFromHTTP & in,
-    size_t projections,
-    bool is_projection,
-    ThrottlerPtr throttler)
-{
-    auto new_data_part = std::make_shared<MergeTreeDataPartInMemory>(data, part_name, part_info, data_part_storage);
-
-    for (size_t i = 0; i < projections; ++i)
-    {
-        String projection_name;
-        readStringBinary(projection_name, in);
-
-        MergeTreePartInfo new_part_info("all", 0, 0, 0);
-        auto projection_part_storage = data_part_storage->getProjection(projection_name + ".proj");
-
-        auto new_projection_part = downloadPartToMemory(
-            projection_part_storage, projection_name,
-            new_part_info, part_uuid, metadata_snapshot,
-            context, in, 0, true, throttler);
-
-        new_data_part->addProjectionPart(projection_name, std::move(new_projection_part));
-    }
-
-    MergeTreeData::DataPart::Checksums checksums;
-    if (!checksums.read(in))
-        throw Exception(ErrorCodes::CORRUPTED_DATA, "Cannot deserialize checksums");
-
-    NativeReader block_in(in, 0);
-    auto block = block_in.read();
-    throttler->add(block.bytes());
-
-    new_data_part->setColumns(block.getNamesAndTypesList(), {}, metadata_snapshot->getMetadataVersion());
-
-    if (!is_projection)
-    {
-        new_data_part->version.setCreationTID(Tx::PrehistoricTID, nullptr);
-        new_data_part->uuid = part_uuid;
-        new_data_part->is_temp = true;
-        new_data_part->minmax_idx->update(block, data.getMinMaxColumnsNames(metadata_snapshot->getPartitionKey()));
-        new_data_part->partition.create(metadata_snapshot, block, 0, context);
-    }
-
-    MergedBlockOutputStream part_out(
-        new_data_part, metadata_snapshot, block.getNamesAndTypesList(), {}, {},
-        CompressionCodecFactory::instance().get("NONE", {}), NO_TRANSACTION_PTR);
-
-    part_out.write(block);
-    part_out.finalizePart(new_data_part, false);
-    new_data_part->checksums.checkEqual(checksums, /* have_uncompressed = */ true);
-
-    return new_data_part;
-}
-
 void Fetcher::downloadBaseOrProjectionPartToDisk(
     const String & replica_path,
     const MutableDataPartStoragePtr & data_part_storage,
diff --git a/src/Storages/MergeTree/DataPartsExchange.h b/src/Storages/MergeTree/DataPartsExchange.h
index 8c15dc3cfdb..21f47826344 100644
--- a/src/Storages/MergeTree/DataPartsExchange.h
+++ b/src/Storages/MergeTree/DataPartsExchange.h
@@ -40,10 +40,6 @@ public:
 
 private:
     MergeTreeData::DataPartPtr findPart(const String & name);
-    void sendPartFromMemory(
-        const MergeTreeData::DataPartPtr & part,
-        WriteBuffer & out,
-        bool send_projections);
 
     MergeTreeData::DataPart::Checksums sendPartFromDisk(
         const MergeTreeData::DataPartPtr & part,
@@ -113,18 +109,6 @@ private:
         ThrottlerPtr throttler,
         bool sync);
 
-    MergeTreeData::MutableDataPartPtr downloadPartToMemory(
-       MutableDataPartStoragePtr data_part_storage,
-       const String & part_name,
-       const MergeTreePartInfo & part_info,
-       const UUID & part_uuid,
-       const StorageMetadataPtr & metadata_snapshot,
-       ContextPtr context,
-       PooledReadWriteBufferFromHTTP & in,
-       size_t projections,
-       bool is_projection,
-       ThrottlerPtr throttler);
-
     MergeTreeData::MutableDataPartPtr downloadPartToDiskRemoteMeta(
        const String & part_name,
        const String & replica_path,
diff --git a/src/Storages/MergeTree/IMergeTreeDataPart.cpp b/src/Storages/MergeTree/IMergeTreeDataPart.cpp
index 526ea542b0e..3fea6d04944 100644
--- a/src/Storages/MergeTree/IMergeTreeDataPart.cpp
+++ b/src/Storages/MergeTree/IMergeTreeDataPart.cpp
@@ -51,7 +51,6 @@ namespace CurrentMetrics
 
     extern const Metric PartsWide;
     extern const Metric PartsCompact;
-    extern const Metric PartsInMemory;
 }
 
 namespace DB
@@ -278,9 +277,6 @@ static void incrementTypeMetric(MergeTreeDataPartType type)
         case MergeTreeDataPartType::Compact:
             CurrentMetrics::add(CurrentMetrics::PartsCompact);
             return;
-        case MergeTreeDataPartType::InMemory:
-            CurrentMetrics::add(CurrentMetrics::PartsInMemory);
-            return;
         case MergeTreeDataPartType::Unknown:
             return;
     }
@@ -296,9 +292,6 @@ static void decrementTypeMetric(MergeTreeDataPartType type)
         case MergeTreeDataPartType::Compact:
             CurrentMetrics::sub(CurrentMetrics::PartsCompact);
             return;
-        case MergeTreeDataPartType::InMemory:
-            CurrentMetrics::sub(CurrentMetrics::PartsInMemory);
-            return;
         case MergeTreeDataPartType::Unknown:
             return;
     }
@@ -2207,11 +2200,6 @@ bool isWidePart(const MergeTreeDataPartPtr & data_part)
     return (data_part && data_part->getType() == MergeTreeDataPartType::Wide);
 }
 
-bool isInMemoryPart(const MergeTreeDataPartPtr & data_part)
-{
-    return (data_part && data_part->getType() == MergeTreeDataPartType::InMemory);
-}
-
 std::optional<std::string> getIndexExtensionFromFilesystem(const IDataPartStorage & data_part_storage)
 {
     if (data_part_storage.exists())
diff --git a/src/Storages/MergeTree/IMergeTreeDataPart.h b/src/Storages/MergeTree/IMergeTreeDataPart.h
index fba1e6ddbb1..aaae64a5970 100644
--- a/src/Storages/MergeTree/IMergeTreeDataPart.h
+++ b/src/Storages/MergeTree/IMergeTreeDataPart.h
@@ -710,7 +710,6 @@ using MergeTreeMutableDataPartPtr = std::shared_ptr<IMergeTreeDataPart>;
 
 bool isCompactPart(const MergeTreeDataPartPtr & data_part);
 bool isWidePart(const MergeTreeDataPartPtr & data_part);
-bool isInMemoryPart(const MergeTreeDataPartPtr & data_part);
 
 inline String getIndexExtension(bool is_compressed_primary_key) { return is_compressed_primary_key ? ".cidx" : ".idx"; }
 std::optional<String> getIndexExtensionFromFilesystem(const IDataPartStorage & data_part_storage);
diff --git a/src/Storages/MergeTree/IMergeTreeDataPartInfoForReader.h b/src/Storages/MergeTree/IMergeTreeDataPartInfoForReader.h
index 2cc73556f04..7807f3d8c25 100644
--- a/src/Storages/MergeTree/IMergeTreeDataPartInfoForReader.h
+++ b/src/Storages/MergeTree/IMergeTreeDataPartInfoForReader.h
@@ -34,8 +34,6 @@ public:
 
     virtual bool isWidePart() const = 0;
 
-    virtual bool isInMemoryPart() const = 0;
-
     virtual bool isProjectionPart() const = 0;
 
     virtual DataPartStoragePtr getDataPartStorage() const = 0;
diff --git a/src/Storages/MergeTree/LoadedMergeTreeDataPartInfoForReader.h b/src/Storages/MergeTree/LoadedMergeTreeDataPartInfoForReader.h
index f1cb8b34ecf..f5111ccaacc 100644
--- a/src/Storages/MergeTree/LoadedMergeTreeDataPartInfoForReader.h
+++ b/src/Storages/MergeTree/LoadedMergeTreeDataPartInfoForReader.h
@@ -22,8 +22,6 @@ public:
 
     bool isWidePart() const override { return DB::isWidePart(data_part); }
 
-    bool isInMemoryPart() const override { return DB::isInMemoryPart(data_part); }
-
     bool isProjectionPart() const override { return data_part->isProjectionPart(); }
 
     DataPartStoragePtr getDataPartStorage() const override { return data_part->getDataPartStoragePtr(); }
diff --git a/src/Storages/MergeTree/MergeTask.cpp b/src/Storages/MergeTree/MergeTask.cpp
index aa38198334e..77bc680eae2 100644
--- a/src/Storages/MergeTree/MergeTask.cpp
+++ b/src/Storages/MergeTree/MergeTask.cpp
@@ -309,7 +309,7 @@ bool MergeTask::ExecuteAndFinalizeHorizontalPart::prepare()
             ctx->rows_sources_uncompressed_write_buf = ctx->tmp_disk->createRawStream();
             ctx->rows_sources_write_buf = std::make_unique<CompressedWriteBuffer>(*ctx->rows_sources_uncompressed_write_buf);
 
-            MergeTreeDataPartInMemory::ColumnToSize local_merged_column_to_size;
+            std::map<String, size_t> local_merged_column_to_size;
             for (const MergeTreeData::DataPartPtr & part : global_ctx->future_part->parts)
                 part->accumulateColumnSizes(local_merged_column_to_size);
 
diff --git a/src/Storages/MergeTree/MergeTreeData.cpp b/src/Storages/MergeTree/MergeTreeData.cpp
index 9bc360cbcac..d56cf761cf4 100644
--- a/src/Storages/MergeTree/MergeTreeData.cpp
+++ b/src/Storages/MergeTree/MergeTreeData.cpp
@@ -72,7 +72,6 @@
 #include <Storages/MergeTree/DataPartStorageOnDiskFull.h>
 #include <Storages/MergeTree/MergeTreeDataPartBuilder.h>
 #include <Storages/MergeTree/MergeTreeDataPartCompact.h>
-#include <Storages/MergeTree/MergeTreeDataPartInMemory.h>
 #include <Storages/Statistics/Estimator.h>
 #include <Storages/MergeTree/MergeTreeSelectProcessor.h>
 #include <Storages/MergeTree/checkDataPart.h>
@@ -1707,8 +1706,7 @@ void MergeTreeData::loadDataParts(bool skip_sanity_checks, std::optional<std::un
             {
                 /// Skip temporary directories, file 'format_version.txt' and directory 'detached'.
                 if (startsWith(it->name(), "tmp") || it->name() == MergeTreeData::FORMAT_VERSION_FILE_NAME
-                    || it->name() == MergeTreeData::DETACHED_DIR_NAME
-                    || startsWith(it->name(), MergeTreeWriteAheadLog::WAL_FILE_NAME))
+                    || it->name() == MergeTreeData::DETACHED_DIR_NAME)
                     continue;
 
                 if (auto part_info = MergeTreePartInfo::tryParsePartName(it->name(), format_version))
@@ -2261,7 +2259,6 @@ MergeTreeData::DataPartsVector MergeTreeData::grabOldParts(bool force)
             bool reached_removal_time = part_remove_time <= time_now && time_now - part_remove_time >= getSettings()->old_parts_lifetime.totalSeconds();
             if ((reached_removal_time && !has_skipped_mutation_parent(part))
                 || force
-                || isInMemoryPart(part)     /// Remove in-memory parts immediately to not store excessive data in RAM
                 || (part->version.creation_csn == Tx::RolledBackCSN && getSettings()->remove_rolled_back_parts_immediately))
             {
                 part->removal_state.store(DataPartRemovalState::REMOVED, std::memory_order_relaxed);
@@ -5225,14 +5222,14 @@ Pipe MergeTreeData::alterPartition(
             case PartitionCommand::FREEZE_PARTITION:
             {
                 auto lock = lockForShare(query_context->getCurrentQueryId(), query_context->getSettingsRef().lock_acquire_timeout);
-                current_command_results = freezePartition(command.partition, metadata_snapshot, command.with_name, query_context, lock);
+                current_command_results = freezePartition(command.partition, command.with_name, query_context, lock);
             }
             break;
 
             case PartitionCommand::FREEZE_ALL_PARTITIONS:
             {
                 auto lock = lockForShare(query_context->getCurrentQueryId(), query_context->getSettingsRef().lock_acquire_timeout);
-                current_command_results = freezeAll(command.with_name, metadata_snapshot, query_context, lock);
+                current_command_results = freezeAll(command.with_name, query_context, lock);
             }
             break;
 
@@ -7103,27 +7100,6 @@ std::pair<MergeTreeData::MutableDataPartPtr, scope_guard> MergeTreeData::cloneAn
     scope_guard src_flushed_tmp_dir_lock;
     MergeTreeData::MutableDataPartPtr src_flushed_tmp_part;
 
-    /// If source part is in memory, flush it to disk and clone it already in on-disk format
-    /// Protect tmp dir from removing by cleanup thread with src_flushed_tmp_dir_lock
-    /// Construct src_flushed_tmp_part in order to delete part with its directory at destructor
-    if (auto src_part_in_memory = asInMemoryPart(src_part))
-    {
-        auto flushed_part_path = *src_part_in_memory->getRelativePathForPrefix(tmp_part_prefix);
-
-        auto tmp_src_part_file_name = fs::path(tmp_dst_part_name).filename();
-        src_flushed_tmp_dir_lock = src_part->storage.getTemporaryPartDirectoryHolder(tmp_src_part_file_name);
-
-        auto flushed_part_storage = src_part_in_memory->flushToDisk(flushed_part_path, metadata_snapshot);
-
-        src_flushed_tmp_part = MergeTreeDataPartBuilder(*this, src_part->name, flushed_part_storage)
-            .withPartInfo(src_part->info)
-            .withPartFormatFromDisk()
-            .build();
-
-        src_flushed_tmp_part->is_temp = true;
-        src_part_storage = flushed_part_storage;
-    }
-
     String with_copy;
     if (params.copy_instead_of_hardlink)
         with_copy = " (copying data)";
@@ -7305,26 +7281,23 @@ MergeTreeData::MatcherFn MergeTreeData::getPartitionMatcher(const ASTPtr & parti
 
 PartitionCommandsResultInfo MergeTreeData::freezePartition(
     const ASTPtr & partition_ast,
-    const StorageMetadataPtr & metadata_snapshot,
     const String & with_name,
     ContextPtr local_context,
     TableLockHolder &)
 {
-    return freezePartitionsByMatcher(getPartitionMatcher(partition_ast, local_context), metadata_snapshot, with_name, local_context);
+    return freezePartitionsByMatcher(getPartitionMatcher(partition_ast, local_context), with_name, local_context);
 }
 
 PartitionCommandsResultInfo MergeTreeData::freezeAll(
     const String & with_name,
-    const StorageMetadataPtr & metadata_snapshot,
     ContextPtr local_context,
     TableLockHolder &)
 {
-    return freezePartitionsByMatcher([] (const String &) { return true; }, metadata_snapshot, with_name, local_context);
+    return freezePartitionsByMatcher([] (const String &) { return true; }, with_name, local_context);
 }
 
 PartitionCommandsResultInfo MergeTreeData::freezePartitionsByMatcher(
     MatcherFn matcher,
-    const StorageMetadataPtr & metadata_snapshot,
     const String & with_name,
     ContextPtr local_context)
 {
@@ -7376,22 +7349,6 @@ PartitionCommandsResultInfo MergeTreeData::freezePartitionsByMatcher(
         scope_guard src_flushed_tmp_dir_lock;
         MergeTreeData::MutableDataPartPtr src_flushed_tmp_part;
 
-        if (auto part_in_memory = asInMemoryPart(part))
-        {
-            auto flushed_part_path = *part_in_memory->getRelativePathForPrefix("tmp_freeze");
-            src_flushed_tmp_dir_lock = part->storage.getTemporaryPartDirectoryHolder("tmp_freeze" + part->name);
-
-            auto flushed_part_storage = part_in_memory->flushToDisk(flushed_part_path, metadata_snapshot);
-
-            src_flushed_tmp_part = MergeTreeDataPartBuilder(*this, part->name, flushed_part_storage)
-                .withPartInfo(part->info)
-                .withPartFormatFromDisk()
-                .build();
-
-            src_flushed_tmp_part->is_temp = true;
-            data_part_storage = flushed_part_storage;
-        }
-
         auto callback = [this, &part, &backup_part_path](const DiskPtr & disk)
         {
             // Store metadata for replicated table.
diff --git a/src/Storages/MergeTree/MergeTreeData.h b/src/Storages/MergeTree/MergeTreeData.h
index dc84505f38f..c864fce4659 100644
--- a/src/Storages/MergeTree/MergeTreeData.h
+++ b/src/Storages/MergeTree/MergeTreeData.h
@@ -23,9 +23,7 @@
 #include <Storages/MergeTree/MergeList.h>
 #include <Storages/MergeTree/IMergeTreeDataPart.h>
 #include <Storages/MergeTree/MergeTreeDataPartBuilder.h>
-#include <Storages/MergeTree/MergeTreeDataPartInMemory.h>
 #include <Storages/MergeTree/MergeTreePartsMover.h>
-#include <Storages/MergeTree/MergeTreeWriteAheadLog.h>
 #include <Storages/MergeTree/PinnedPartUUIDs.h>
 #include <Storages/MergeTree/ZeroCopyLock.h>
 #include <Storages/MergeTree/TemporaryParts.h>
@@ -752,7 +750,6 @@ public:
       */
     PartitionCommandsResultInfo freezePartition(
         const ASTPtr & partition,
-        const StorageMetadataPtr & metadata_snapshot,
         const String & with_name,
         ContextPtr context,
         TableLockHolder & table_lock_holder);
@@ -760,7 +757,6 @@ public:
     /// Freezes all parts.
     PartitionCommandsResultInfo freezeAll(
         const String & with_name,
-        const StorageMetadataPtr & metadata_snapshot,
         ContextPtr context,
         TableLockHolder & table_lock_holder);
 
@@ -1307,7 +1303,7 @@ protected:
     bool isPrimaryOrMinMaxKeyColumnPossiblyWrappedInFunctions(const ASTPtr & node, const StorageMetadataPtr & metadata_snapshot) const;
 
     /// Common part for |freezePartition()| and |freezeAll()|.
-    PartitionCommandsResultInfo freezePartitionsByMatcher(MatcherFn matcher, const StorageMetadataPtr & metadata_snapshot, const String & with_name, ContextPtr context);
+    PartitionCommandsResultInfo freezePartitionsByMatcher(MatcherFn matcher, const String & with_name, ContextPtr context);
     PartitionCommandsResultInfo unfreezePartitionsByMatcher(MatcherFn matcher, const String & backup_name, ContextPtr context);
 
     // Partition helpers
diff --git a/src/Storages/MergeTree/MergeTreeDataPartBuilder.cpp b/src/Storages/MergeTree/MergeTreeDataPartBuilder.cpp
index d55248df0af..146c15bbb69 100644
--- a/src/Storages/MergeTree/MergeTreeDataPartBuilder.cpp
+++ b/src/Storages/MergeTree/MergeTreeDataPartBuilder.cpp
@@ -1,5 +1,4 @@
 #include <Storages/MergeTree/MergeTreeDataPartBuilder.h>
-#include <Storages/MergeTree/MergeTreeDataPartInMemory.h>
 #include <Storages/MergeTree/MergeTreeDataPartCompact.h>
 #include <Storages/MergeTree/MergeTreeDataPartWide.h>
 #include <Storages/MergeTree/DataPartStorageOnDiskFull.h>
@@ -64,8 +63,6 @@ std::shared_ptr<IMergeTreeDataPart> MergeTreeDataPartBuilder::build()
             return std::make_shared<MergeTreeDataPartWide>(data, name, *part_info, part_storage, parent_part);
         case PartType::Compact:
             return std::make_shared<MergeTreeDataPartCompact>(data, name, *part_info, part_storage, parent_part);
-        case PartType::InMemory:
-            return std::make_shared<MergeTreeDataPartInMemory>(data, name, *part_info, part_storage, parent_part);
         default:
             throw Exception(ErrorCodes::UNKNOWN_PART_TYPE,
                 "Unknown type of part {}", part_storage->getRelativePath());
diff --git a/src/Storages/MergeTree/MergeTreeDataPartInMemory.cpp b/src/Storages/MergeTree/MergeTreeDataPartInMemory.cpp
deleted file mode 100644
index e023ae9be0e..00000000000
--- a/src/Storages/MergeTree/MergeTreeDataPartInMemory.cpp
+++ /dev/null
@@ -1,115 +0,0 @@
-#include <Storages/MergeTree/MergeTreeDataPartInMemory.h>
-#include <Storages/MergeTree/MergeTreeReaderInMemory.h>
-#include <Storages/MergeTree/MergedBlockOutputStream.h>
-#include <Storages/MergeTree/IMergeTreeReader.h>
-#include <Storages/MergeTree/LoadedMergeTreeDataPartInfoForReader.h>
-#include <Storages/MergeTree/DataPartStorageOnDiskFull.h>
-#include <DataTypes/NestedUtils.h>
-#include <Disks/createVolume.h>
-#include <Interpreters/Context.h>
-#include <Poco/Logger.h>
-
-
-namespace DB
-{
-
-namespace ErrorCodes
-{
-    extern const int NOT_IMPLEMENTED;
-}
-
-MergeTreeDataPartInMemory::MergeTreeDataPartInMemory(
-        const MergeTreeData & storage_,
-        const String & name_,
-        const MergeTreePartInfo & info_,
-        const MutableDataPartStoragePtr & data_part_storage_,
-        const IMergeTreeDataPart * parent_part_)
-    : IMergeTreeDataPart(storage_, name_, info_, data_part_storage_, Type::InMemory, parent_part_)
-{
-    default_codec = CompressionCodecFactory::instance().get("NONE", {});
-}
-
-IMergeTreeDataPart::MergeTreeReaderPtr MergeTreeDataPartInMemory::getReader(
-    const NamesAndTypesList & columns_to_read,
-    const StorageSnapshotPtr & storage_snapshot,
-    const MarkRanges & mark_ranges,
-    const VirtualFields & virtual_fields,
-    UncompressedCache * /* uncompressed_cache */,
-    MarkCache * /* mark_cache */,
-    const AlterConversionsPtr & alter_conversions,
-    const MergeTreeReaderSettings & reader_settings,
-    const ValueSizeMap & /* avg_value_size_hints */,
-    const ReadBufferFromFileBase::ProfileCallback & /* profile_callback */) const
-{
-    auto read_info = std::make_shared<LoadedMergeTreeDataPartInfoForReader>(shared_from_this(), alter_conversions);
-    auto ptr = std::static_pointer_cast<const MergeTreeDataPartInMemory>(shared_from_this());
-
-    return std::make_unique<MergeTreeReaderInMemory>(
-        read_info,
-        ptr,
-        columns_to_read,
-        virtual_fields,
-        storage_snapshot,
-        mark_ranges,
-        reader_settings);
-}
-
-IMergeTreeDataPart::MergeTreeWriterPtr MergeTreeDataPartInMemory::getWriter(
-    const NamesAndTypesList &,
-    const StorageMetadataPtr &,
-    const std::vector<MergeTreeIndexPtr> &,
-    const Statistics &,
-    const CompressionCodecPtr &,
-    const MergeTreeWriterSettings &,
-    const MergeTreeIndexGranularity &)
-{
-    throw Exception(ErrorCodes::NOT_IMPLEMENTED, "In-memory data parts are obsolete and no longer supported for writing");
-}
-
-MutableDataPartStoragePtr MergeTreeDataPartInMemory::flushToDisk(const String &, const StorageMetadataPtr &) const
-{
-    throw Exception(ErrorCodes::NOT_IMPLEMENTED, "In-memory data parts are obsolete and no longer supported for writing");
-}
-
-DataPartStoragePtr MergeTreeDataPartInMemory::makeCloneInDetached(const String & prefix,
-                                                                  const StorageMetadataPtr & metadata_snapshot,
-                                                                  const DiskTransactionPtr & disk_transaction) const
-{
-    if (disk_transaction)
-        throw Exception(ErrorCodes::NOT_IMPLEMENTED, "InMemory parts are not compatible with disk transactions");
-    String detached_path = *getRelativePathForDetachedPart(prefix, /* broken */ false);
-    return flushToDisk(detached_path, metadata_snapshot);
-}
-
-void MergeTreeDataPartInMemory::renameTo(const String & new_relative_path, bool /* remove_new_dir_if_exists */)
-{
-    getDataPartStorage().setRelativePath(new_relative_path);
-}
-
-void MergeTreeDataPartInMemory::calculateEachColumnSizes(ColumnSizeByName & each_columns_size, ColumnSize & total_size) const
-{
-    auto it = checksums.files.find("data.bin");
-    if (it != checksums.files.end())
-        total_size.data_uncompressed += it->second.uncompressed_size;
-
-    for (const auto & column : columns)
-        each_columns_size[column.name].data_uncompressed += block.getByName(column.name).column->byteSize();
-}
-
-IMergeTreeDataPart::Checksum MergeTreeDataPartInMemory::calculateBlockChecksum() const
-{
-    SipHash hash;
-    IMergeTreeDataPart::Checksum checksum;
-    for (const auto & column : block)
-        column.column->updateHashFast(hash);
-
-    checksum.uncompressed_size = block.bytes();
-    checksum.uncompressed_hash = getSipHash128AsPair(hash);
-    return checksum;
-}
-
-DataPartInMemoryPtr asInMemoryPart(const MergeTreeDataPartPtr & part)
-{
-    return std::dynamic_pointer_cast<const MergeTreeDataPartInMemory>(part);
-}
-}
diff --git a/src/Storages/MergeTree/MergeTreeDataPartInMemory.h b/src/Storages/MergeTree/MergeTreeDataPartInMemory.h
deleted file mode 100644
index 90b4b0e3471..00000000000
--- a/src/Storages/MergeTree/MergeTreeDataPartInMemory.h
+++ /dev/null
@@ -1,70 +0,0 @@
-#pragma once
-
-#include <Storages/MergeTree/IMergeTreeDataPart.h>
-
-namespace DB
-{
-
-class UncompressedCache;
-
-class MergeTreeDataPartInMemory : public IMergeTreeDataPart
-{
-public:
-    MergeTreeDataPartInMemory(
-        const MergeTreeData & storage_,
-        const String & name_,
-        const MergeTreePartInfo & info_,
-        const MutableDataPartStoragePtr & data_part_storage_,
-        const IMergeTreeDataPart * parent_part_ = nullptr);
-
-    MergeTreeReaderPtr getReader(
-        const NamesAndTypesList & columns,
-        const StorageSnapshotPtr & storage_snapshot,
-        const MarkRanges & mark_ranges,
-        const VirtualFields & virtual_fields,
-        UncompressedCache * uncompressed_cache,
-        MarkCache * mark_cache,
-        const AlterConversionsPtr & alter_conversions,
-        const MergeTreeReaderSettings & reader_settings_,
-        const ValueSizeMap & avg_value_size_hints,
-        const ReadBufferFromFileBase::ProfileCallback & profile_callback) const override;
-
-    MergeTreeWriterPtr getWriter(
-        const NamesAndTypesList & columns_list,
-        const StorageMetadataPtr & metadata_snapshot,
-        const std::vector<MergeTreeIndexPtr> & indices_to_recalc,
-        const Statistics & stats_to_recalc_,
-        const CompressionCodecPtr & default_codec_,
-        const MergeTreeWriterSettings & writer_settings,
-        const MergeTreeIndexGranularity & computed_index_granularity) override;
-
-    bool isStoredOnDisk() const override { return false; }
-    bool isStoredOnRemoteDisk() const override { return false; }
-    bool isStoredOnRemoteDiskWithZeroCopySupport() const override { return false; }
-    bool hasColumnFiles(const NameAndTypePair & column) const override { return !!getColumnPosition(column.getNameInStorage()); }
-    std::optional<String> getFileNameForColumn(const NameAndTypePair & /* column */) const override { return ""; }
-    void renameTo(const String & new_relative_path, bool remove_new_dir_if_exists) override;
-    DataPartStoragePtr makeCloneInDetached(const String & prefix, const StorageMetadataPtr & metadata_snapshot,
-                                           const DiskTransactionPtr & disk_transaction) const override;
-    std::optional<time_t> getColumnModificationTime(const String & /* column_name */) const override { return {}; }
-
-    MutableDataPartStoragePtr flushToDisk(const String & new_relative_path, const StorageMetadataPtr & metadata_snapshot) const;
-
-    /// Returns hash of parts's block
-    Checksum calculateBlockChecksum() const;
-
-    mutable Block block;
-
-private:
-    mutable std::condition_variable is_merged;
-
-    /// Calculates uncompressed sizes in memory.
-    void calculateEachColumnSizes(ColumnSizeByName & each_columns_size, ColumnSize & total_size) const override;
-};
-
-using DataPartInMemoryPtr = std::shared_ptr<const MergeTreeDataPartInMemory>;
-using MutableDataPartInMemoryPtr = std::shared_ptr<MergeTreeDataPartInMemory>;
-
-DataPartInMemoryPtr asInMemoryPart(const MergeTreeDataPartPtr & part);
-
-}
diff --git a/src/Storages/MergeTree/MergeTreeDataPartType.h b/src/Storages/MergeTree/MergeTreeDataPartType.h
index 5096ee86db1..64f3abee391 100644
--- a/src/Storages/MergeTree/MergeTreeDataPartType.h
+++ b/src/Storages/MergeTree/MergeTreeDataPartType.h
@@ -42,9 +42,6 @@ public:
         /// Data of all columns is stored in one file. Marks are also stored in single file.
         Compact,
 
-        /// Format with buffering data in RAM. Obsolete - new parts cannot be created in this format.
-        InMemory,
-
         Unknown,
     };
 
diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.h b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.h
index ae40eb03649..f5ff323563d 100644
--- a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.h
+++ b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.h
@@ -1,5 +1,8 @@
 #pragma once
+
 #include <Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h>
+#include <Formats/MarkInCompressedFile.h>
+
 
 namespace DB
 {
diff --git a/src/Storages/MergeTree/MergeTreeDataWriter.cpp b/src/Storages/MergeTree/MergeTreeDataWriter.cpp
index ebf887f5e9e..fdac16ae19a 100644
--- a/src/Storages/MergeTree/MergeTreeDataWriter.cpp
+++ b/src/Storages/MergeTree/MergeTreeDataWriter.cpp
@@ -648,18 +648,11 @@ MergeTreeDataWriter::TemporaryPart MergeTreeDataWriter::writeProjectionPartImpl(
     const auto & metadata_snapshot = projection.metadata;
 
     MergeTreeDataPartType part_type;
-    if (parent_part->getType() == MergeTreeDataPartType::InMemory)
-    {
-        part_type = MergeTreeDataPartType::InMemory;
-    }
-    else
-    {
-        /// Size of part would not be greater than block.bytes() + epsilon
-        size_t expected_size = block.bytes();
-        // just check if there is enough space on parent volume
-        data.reserveSpace(expected_size, parent_part->getDataPartStorage());
-        part_type = data.choosePartFormatOnDisk(expected_size, block.rows()).part_type;
-    }
+    /// Size of part would not be greater than block.bytes() + epsilon
+    size_t expected_size = block.bytes();
+    // just check if there is enough space on parent volume
+    data.reserveSpace(expected_size, parent_part->getDataPartStorage());
+    part_type = data.choosePartFormatOnDisk(expected_size, block.rows()).part_type;
 
     auto new_data_part = parent_part->getProjectionPartBuilder(part_name, is_temp).withPartType(part_type).build();
     auto projection_part_storage = new_data_part->getDataPartStoragePtr();
diff --git a/src/Storages/MergeTree/MergeTreeIndexGranularityInfo.cpp b/src/Storages/MergeTree/MergeTreeIndexGranularityInfo.cpp
index da89d52a9ff..031a0c08c26 100644
--- a/src/Storages/MergeTree/MergeTreeIndexGranularityInfo.cpp
+++ b/src/Storages/MergeTree/MergeTreeIndexGranularityInfo.cpp
@@ -81,8 +81,6 @@ std::string MarkType::getFileExtension() const
             return res + "2";
         case MergeTreeDataPartType::Compact:
             return res + "3";
-        case MergeTreeDataPartType::InMemory:
-            return "";
         case MergeTreeDataPartType::Unknown:
             throw Exception(ErrorCodes::LOGICAL_ERROR, "Unknown data part type");
     }
@@ -126,8 +124,6 @@ size_t MergeTreeIndexGranularityInfo::getMarkSizeInBytes(size_t columns_num) con
         return mark_type.adaptive ? getAdaptiveMrkSizeWide() : getNonAdaptiveMrkSizeWide();
     else if (mark_type.part_type == MergeTreeDataPartType::Compact)
         return getAdaptiveMrkSizeCompact(columns_num);
-    else if (mark_type.part_type == MergeTreeDataPartType::InMemory)
-        return 0;
     else
         throw Exception(ErrorCodes::UNKNOWN_PART_TYPE, "Unknown part type");
 }
diff --git a/src/Storages/MergeTree/MergeTreeReaderInMemory.cpp b/src/Storages/MergeTree/MergeTreeReaderInMemory.cpp
deleted file mode 100644
index 91fc8966a7a..00000000000
--- a/src/Storages/MergeTree/MergeTreeReaderInMemory.cpp
+++ /dev/null
@@ -1,119 +0,0 @@
-#include <Storages/MergeTree/MergeTreeReaderInMemory.h>
-#include <Storages/MergeTree/MergeTreeDataPartInMemory.h>
-#include <Interpreters/getColumnFromBlock.h>
-#include <DataTypes/DataTypeArray.h>
-#include <DataTypes/NestedUtils.h>
-#include <Columns/ColumnArray.h>
-
-namespace DB
-{
-
-namespace ErrorCodes
-{
-    extern const int CANNOT_READ_ALL_DATA;
-    extern const int ARGUMENT_OUT_OF_BOUND;
-}
-
-
-MergeTreeReaderInMemory::MergeTreeReaderInMemory(
-    MergeTreeDataPartInfoForReaderPtr data_part_info_for_read_,
-    DataPartInMemoryPtr data_part_,
-    NamesAndTypesList columns_,
-    const VirtualFields & virtual_fields_,
-    const StorageSnapshotPtr & storage_snapshot_,
-    MarkRanges mark_ranges_,
-    MergeTreeReaderSettings settings_)
-    : IMergeTreeReader(
-        data_part_info_for_read_,
-        columns_,
-        virtual_fields_,
-        storage_snapshot_,
-        nullptr,
-        nullptr,
-        mark_ranges_,
-        settings_,
-        {})
-    , part_in_memory(std::move(data_part_))
-{
-    for (const auto & column_to_read : columns_to_read)
-    {
-        /// If array of Nested column is missing in part,
-        /// we have to read its offsets if they exist.
-        if (typeid_cast<const DataTypeArray *>(column_to_read.type.get())
-            && !tryGetColumnFromBlock(part_in_memory->block, column_to_read))
-        {
-            if (auto offsets_position = findColumnForOffsets(column_to_read))
-            {
-                positions_for_offsets[column_to_read.name] = *data_part_info_for_read->getColumnPosition(offsets_position->first);
-                partially_read_columns.insert(column_to_read.name);
-            }
-        }
-    }
-}
-
-size_t MergeTreeReaderInMemory::readRows(
-    size_t from_mark, size_t /* current_task_last_mark */, bool continue_reading, size_t max_rows_to_read, Columns & res_columns)
-{
-    if (!continue_reading)
-        total_rows_read = 0;
-
-    size_t total_marks = data_part_info_for_read->getIndexGranularity().getMarksCount();
-    if (from_mark >= total_marks)
-        throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "Mark {} is out of bound. Max mark: {}",
-            toString(from_mark), toString(total_marks));
-
-    size_t num_columns = res_columns.size();
-    checkNumberOfColumns(num_columns);
-
-    size_t part_rows = part_in_memory->block.rows();
-    if (total_rows_read >= part_rows)
-        throw Exception(ErrorCodes::CANNOT_READ_ALL_DATA, "Cannot read data in MergeTreeReaderInMemory. "
-            "Rows already read: {}. Rows in part: {}", total_rows_read, part_rows);
-
-    size_t rows_to_read = std::min(max_rows_to_read, part_rows - total_rows_read);
-    for (size_t i = 0; i < num_columns; ++i)
-    {
-        const auto & column_to_read = columns_to_read[i];
-
-        /// Copy offsets, if array of Nested column is missing in part.
-        auto offsets_it = positions_for_offsets.find(column_to_read.name);
-        if (offsets_it != positions_for_offsets.end() && !column_to_read.isSubcolumn())
-        {
-            const auto & source_offsets = assert_cast<const ColumnArray &>(
-                *part_in_memory->block.getByPosition(offsets_it->second).column).getOffsets();
-
-            if (res_columns[i] == nullptr)
-                res_columns[i] = column_to_read.type->createColumn();
-
-            auto mutable_column = res_columns[i]->assumeMutable();
-            auto & res_offstes = assert_cast<ColumnArray &>(*mutable_column).getOffsets();
-            size_t start_offset = total_rows_read ? source_offsets[total_rows_read - 1] : 0;
-            for (size_t row = 0; row < rows_to_read; ++row)
-                res_offstes.push_back(source_offsets[total_rows_read + row] - start_offset);
-
-            res_columns[i] = std::move(mutable_column);
-        }
-        else if (part_in_memory->hasColumnFiles(column_to_read))
-        {
-            auto block_column = getColumnFromBlock(part_in_memory->block, column_to_read);
-            if (rows_to_read == part_rows)
-            {
-                res_columns[i] = block_column;
-            }
-            else
-            {
-                if (res_columns[i] == nullptr)
-                    res_columns[i] = column_to_read.type->createColumn();
-
-                auto mutable_column = res_columns[i]->assumeMutable();
-                mutable_column->insertRangeFrom(*block_column, total_rows_read, rows_to_read);
-                res_columns[i] = std::move(mutable_column);
-            }
-        }
-    }
-
-    total_rows_read += rows_to_read;
-    return rows_to_read;
-}
-
-}
diff --git a/src/Storages/MergeTree/MergeTreeReaderInMemory.h b/src/Storages/MergeTree/MergeTreeReaderInMemory.h
deleted file mode 100644
index cc1e2e9e4e2..00000000000
--- a/src/Storages/MergeTree/MergeTreeReaderInMemory.h
+++ /dev/null
@@ -1,40 +0,0 @@
-#pragma once
-
-#include <Core/NamesAndTypes.h>
-#include <Storages/MergeTree/IMergeTreeReader.h>
-
-
-namespace DB
-{
-
-class MergeTreeDataPartInMemory;
-using DataPartInMemoryPtr = std::shared_ptr<const MergeTreeDataPartInMemory>;
-
-/// Reader for InMemory parts
-class MergeTreeReaderInMemory : public IMergeTreeReader
-{
-public:
-    MergeTreeReaderInMemory(
-        MergeTreeDataPartInfoForReaderPtr data_part_info_for_read_,
-        DataPartInMemoryPtr data_part_,
-        NamesAndTypesList columns_,
-        const VirtualFields & virtual_fields_,
-        const StorageSnapshotPtr & storage_snapshot_,
-        MarkRanges mark_ranges_,
-        MergeTreeReaderSettings settings_);
-
-    /// Return the number of rows has been read or zero if there is no columns to read.
-    /// If continue_reading is true, continue reading from last state, otherwise seek to from_mark
-    size_t readRows(size_t from_mark, size_t current_tasl_last_mark,
-                    bool continue_reading, size_t max_rows_to_read, Columns & res_columns) override;
-
-    bool canReadIncompleteGranules() const override { return true; }
-
-private:
-    size_t total_rows_read = 0;
-    DataPartInMemoryPtr part_in_memory;
-
-    std::unordered_map<String, size_t> positions_for_offsets;
-};
-
-}
diff --git a/src/Storages/MergeTree/MergeTreeSink.cpp b/src/Storages/MergeTree/MergeTreeSink.cpp
index 36816904a81..87b0a04d244 100644
--- a/src/Storages/MergeTree/MergeTreeSink.cpp
+++ b/src/Storages/MergeTree/MergeTreeSink.cpp
@@ -1,5 +1,4 @@
 #include <Storages/MergeTree/MergeTreeSink.h>
-#include <Storages/MergeTree/MergeTreeDataPartInMemory.h>
 #include <Storages/StorageMergeTree.h>
 #include <Interpreters/PartLog.h>
 #include <DataTypes/ObjectUtils.h>
diff --git a/src/Storages/MergeTree/MergeTreeWriteAheadLog.cpp b/src/Storages/MergeTree/MergeTreeWriteAheadLog.cpp
deleted file mode 100644
index 2236c1a9380..00000000000
--- a/src/Storages/MergeTree/MergeTreeWriteAheadLog.cpp
+++ /dev/null
@@ -1,336 +0,0 @@
-#include <Storages/MergeTree/MergeTreeWriteAheadLog.h>
-#include <Storages/MergeTree/MergeTreeDataPartInMemory.h>
-#include <Storages/MergeTree/MergeTreeData.h>
-#include <Storages/MergeTree/MergeTreeDataWriter.h>
-#include <Storages/MergeTree/MergedBlockOutputStream.h>
-#include <Storages/MergeTree/MergeTreeDataPartState.h>
-#include <Storages/MergeTree/DataPartStorageOnDiskFull.h>
-#include <IO/MemoryReadWriteBuffer.h>
-#include <IO/ReadHelpers.h>
-#include <IO/copyData.h>
-#include <Interpreters/Context.h>
-#include <Common/logger_useful.h>
-#include <Poco/JSON/JSON.h>
-#include <Poco/JSON/Object.h>
-#include <Poco/JSON/Stringifier.h>
-#include <Poco/JSON/Parser.h>
-#include <sys/time.h>
-
-namespace DB
-{
-
-namespace ErrorCodes
-{
-    extern const int UNKNOWN_FORMAT_VERSION;
-    extern const int CANNOT_READ_ALL_DATA;
-    extern const int BAD_DATA_PART_NAME;
-    extern const int CORRUPTED_DATA;
-}
-
-MergeTreeWriteAheadLog::MergeTreeWriteAheadLog(
-    MergeTreeData & storage_,
-    const DiskPtr & disk_,
-    const String & name_)
-    : storage(storage_)
-    , disk(disk_)
-    , name(name_)
-    , path(storage.getRelativeDataPath() + name_)
-    , pool(storage.getContext()->getSchedulePool())
-    , log(getLogger(storage.getLogName() + " (WriteAheadLog)"))
-{
-    init();
-    sync_task = pool.createTask("MergeTreeWriteAheadLog::sync", [this]
-    {
-        std::lock_guard lock(write_mutex);
-        out->sync();
-        sync_scheduled = false;
-        sync_cv.notify_all();
-    });
-}
-
-MergeTreeWriteAheadLog::~MergeTreeWriteAheadLog()
-{
-    try
-    {
-        shutdown();
-    }
-    catch (...)
-    {
-        tryLogCurrentException(__PRETTY_FUNCTION__);
-    }
-}
-
-
-void MergeTreeWriteAheadLog::dropAllWriteAheadLogs(DiskPtr disk_to_drop, std::string relative_data_path)
-{
-    std::vector<std::string> files;
-    disk_to_drop->listFiles(relative_data_path, files);
-    for (const auto & file : files)
-    {
-        if (file.starts_with(WAL_FILE_NAME))
-            disk_to_drop->removeFile(fs::path(relative_data_path) / file);
-    }
-}
-
-void MergeTreeWriteAheadLog::init()
-{
-    out = disk->writeFile(path, DBMS_DEFAULT_BUFFER_SIZE, WriteMode::Append);
-
-    /// Small hack: in NativeWriter header is used only in `getHeader` method.
-    /// To avoid complex logic of changing it during ALTERs we leave it empty.
-    block_out = std::make_unique<NativeWriter>(*out, 0, Block{});
-    min_block_number = std::numeric_limits<Int64>::max();
-    max_block_number = -1;
-    bytes_at_last_sync = 0;
-}
-
-void MergeTreeWriteAheadLog::dropPart(const String & part_name)
-{
-    std::unique_lock lock(write_mutex);
-
-    writeIntBinary(WAL_VERSION, *out);
-
-    ActionMetadata metadata{};
-    metadata.write(*out);
-
-    writeIntBinary(static_cast<UInt8>(ActionType::DROP_PART), *out);
-    writeStringBinary(part_name, *out);
-    out->next();
-}
-
-void MergeTreeWriteAheadLog::rotate(const std::unique_lock<std::mutex> &)
-{
-    String new_name = String(WAL_FILE_NAME) + "_"
-        + toString(min_block_number) + "_"
-        + toString(max_block_number) + WAL_FILE_EXTENSION;
-
-    /// Finalize stream before file rename
-    out->finalize();
-    disk->replaceFile(path, storage.getRelativeDataPath() + new_name);
-    init();
-}
-
-MergeTreeData::MutableDataPartsVector MergeTreeWriteAheadLog::restore(
-    const StorageMetadataPtr & metadata_snapshot,
-    ContextPtr context,
-    DataPartsLock & parts_lock,
-    bool readonly)
-{
-    std::unique_lock lock(write_mutex);
-
-    MergeTreeData::MutableDataPartsVector parts;
-    auto in = disk->readFile(path);
-    NativeReader block_in(*in, 0);
-    NameSet dropped_parts;
-
-    while (!in->eof())
-    {
-        MergeTreeData::MutableDataPartPtr part;
-        UInt8 version;
-        String part_name;
-        Block block;
-        ActionType action_type;
-
-        try
-        {
-            ActionMetadata metadata;
-
-            readIntBinary(version, *in);
-            if (version > 0)
-            {
-                metadata.read(*in);
-            }
-
-            readIntBinary(action_type, *in);
-            readStringBinary(part_name, *in);
-
-            if (action_type == ActionType::DROP_PART)
-            {
-                dropped_parts.insert(part_name);
-            }
-            else if (action_type == ActionType::ADD_PART)
-            {
-                auto single_disk_volume = std::make_shared<SingleDiskVolume>("volume_" + part_name, disk, 0);
-
-                part = storage.getDataPartBuilder(part_name, single_disk_volume, part_name)
-                    .withPartType(MergeTreeDataPartType::InMemory)
-                    .withPartStorageType(MergeTreeDataPartStorageType::Full)
-                    .build();
-
-                part->uuid = metadata.part_uuid;
-                block = block_in.read();
-
-                if (storage.getActiveContainingPart(part->info, MergeTreeDataPartState::Active, parts_lock))
-                    continue;
-            }
-            else
-            {
-                throw Exception(ErrorCodes::CORRUPTED_DATA, "Unknown action type: {}", toString(static_cast<UInt8>(action_type)));
-            }
-        }
-        catch (const Exception & e)
-        {
-            if (e.code() == ErrorCodes::CANNOT_READ_ALL_DATA
-                || e.code() == ErrorCodes::UNKNOWN_FORMAT_VERSION
-                || e.code() == ErrorCodes::BAD_DATA_PART_NAME
-                || e.code() == ErrorCodes::CORRUPTED_DATA)
-            {
-                LOG_WARNING(log, "WAL file '{}' is broken. {}", path, e.displayText());
-
-                /// If file is broken, do not write new parts to it.
-                /// But if it contains any part rotate and save them.
-                if (max_block_number == -1)
-                {
-                    if (!readonly)
-                        disk->removeFile(path);
-                }
-                else if (name == DEFAULT_WAL_FILE_NAME)
-                    rotate(lock);
-
-                break;
-            }
-            throw;
-        }
-
-        if (action_type == ActionType::ADD_PART)
-        {
-            MergedBlockOutputStream part_out(
-                part,
-                metadata_snapshot,
-                block.getNamesAndTypesList(),
-                {}, {},
-                CompressionCodecFactory::instance().get("NONE", {}),
-                NO_TRANSACTION_PTR);
-
-            part->minmax_idx->update(block, storage.getMinMaxColumnsNames(metadata_snapshot->getPartitionKey()));
-            part->partition.create(metadata_snapshot, block, 0, context);
-            part->setColumns(block.getNamesAndTypesList(), {}, metadata_snapshot->getMetadataVersion());
-            if (metadata_snapshot->hasSortingKey())
-                metadata_snapshot->getSortingKey().expression->execute(block);
-
-            part_out.write(block);
-
-            for (const auto & projection : metadata_snapshot->getProjections())
-            {
-                auto projection_block = projection.calculate(block, context);
-                auto temp_part = MergeTreeDataWriter::writeProjectionPart(storage, log, projection_block, projection, part.get());
-                temp_part.finalize();
-                if (projection_block.rows())
-                    part->addProjectionPart(projection.name, std::move(temp_part.part));
-            }
-
-            part_out.finalizePart(part, false);
-
-            min_block_number = std::min(min_block_number, part->info.min_block);
-            max_block_number = std::max(max_block_number, part->info.max_block);
-            parts.push_back(std::move(part));
-        }
-    }
-
-    MergeTreeData::MutableDataPartsVector result;
-    std::copy_if(parts.begin(), parts.end(), std::back_inserter(result),
-        [&dropped_parts](const auto & part) { return dropped_parts.count(part->name) == 0; });
-
-    /// All parts in WAL had been already committed into the disk -> clear the WAL
-    if (!readonly && result.empty())
-    {
-        LOG_DEBUG(log, "WAL file '{}' had been completely processed. Removing.", path);
-        disk->removeFile(path);
-        init();
-        return {};
-    }
-
-    return result;
-}
-
-void MergeTreeWriteAheadLog::shutdown()
-{
-    {
-        std::unique_lock lock(write_mutex);
-        if (shutdown_called)
-             return;
-
-        if (sync_scheduled)
-            sync_cv.wait(lock, [this] { return !sync_scheduled; });
-
-        shutdown_called = true;
-        out->finalize();
-        out.reset();
-    }
-
-    /// Do it without lock, otherwise inversion between pool lock and write_mutex is possible
-    sync_task->deactivate();
-}
-
-std::optional<MergeTreeWriteAheadLog::MinMaxBlockNumber>
-MergeTreeWriteAheadLog::tryParseMinMaxBlockNumber(const String & filename)
-{
-    Int64 min_block;
-    Int64 max_block;
-    ReadBufferFromString in(filename);
-    if (!checkString(WAL_FILE_NAME, in)
-        || !checkChar('_', in)
-        || !tryReadIntText(min_block, in)
-        || !checkChar('_', in)
-        || !tryReadIntText(max_block, in))
-    {
-        return {};
-    }
-
-    return std::make_pair(min_block, max_block);
-}
-
-String MergeTreeWriteAheadLog::ActionMetadata::toJSON() const
-{
-    Poco::JSON::Object json;
-
-    if (part_uuid != UUIDHelpers::Nil)
-        json.set(JSON_KEY_PART_UUID, toString(part_uuid));
-
-    std::ostringstream oss; // STYLE_CHECK_ALLOW_STD_STRING_STREAM
-    oss.exceptions(std::ios::failbit);
-    json.stringify(oss);
-
-    return oss.str();
-}
-
-void MergeTreeWriteAheadLog::ActionMetadata::fromJSON(const String & buf)
-{
-    Poco::JSON::Parser parser;
-    auto json = parser.parse(buf).extract<Poco::JSON::Object::Ptr>();
-
-    if (json->has(JSON_KEY_PART_UUID))
-        part_uuid = parseFromString<UUID>(json->getValue<std::string>(JSON_KEY_PART_UUID));
-}
-
-void MergeTreeWriteAheadLog::ActionMetadata::read(ReadBuffer & meta_in)
-{
-    readIntBinary(min_compatible_version, meta_in);
-    if (min_compatible_version > WAL_VERSION)
-        throw Exception(ErrorCodes::UNKNOWN_FORMAT_VERSION,
-                        "WAL metadata version {} is not compatible with this ClickHouse version",
-                        toString(min_compatible_version));
-
-    size_t metadata_size;
-    readVarUInt(metadata_size, meta_in);
-
-    if (metadata_size == 0)
-        return;
-
-    String buf(metadata_size, ' ');
-    meta_in.readStrict(buf.data(), metadata_size);
-
-    fromJSON(buf);
-}
-
-void MergeTreeWriteAheadLog::ActionMetadata::write(WriteBuffer & meta_out) const
-{
-    writeIntBinary(min_compatible_version, meta_out);
-
-    String ser_meta = toJSON();
-
-    writeVarUInt(static_cast<UInt32>(ser_meta.length()), meta_out);
-    writeString(ser_meta, meta_out);
-}
-
-}
diff --git a/src/Storages/MergeTree/MergeTreeWriteAheadLog.h b/src/Storages/MergeTree/MergeTreeWriteAheadLog.h
deleted file mode 100644
index 9550fa6ecee..00000000000
--- a/src/Storages/MergeTree/MergeTreeWriteAheadLog.h
+++ /dev/null
@@ -1,105 +0,0 @@
-#pragma once
-
-#include <Formats/NativeReader.h>
-#include <Formats/NativeWriter.h>
-#include <Core/BackgroundSchedulePool.h>
-#include <Disks/IDisk.h>
-#include <Storages/MergeTree/IMergeTreeDataPart.h>
-#include <Storages/MergeTree/MergeTreeDataPartInMemory.h>
-
-namespace DB
-{
-
-class MergeTreeData;
-struct DataPartsLock;
-
-/** WAL stores addditions and removals of data parts in in-memory format.
-  * Format of data in WAL:
-  * - version
-  * - type of action (ADD or DROP)
-  * - part name
-  * - part's block in Native format. (for ADD action)
-  */
-class MergeTreeWriteAheadLog
-{
-public:
-    /// Append-only enum. It is serialized to WAL
-    enum class ActionType : UInt8
-    {
-        ADD_PART = 0,
-        DROP_PART = 1,
-    };
-
-    struct ActionMetadata
-    {
-        /// The minimum version of WAL reader that can understand metadata written by current ClickHouse version.
-        /// This field must be increased when making backwards incompatible changes.
-        ///
-        /// The same approach can be used recursively inside metadata.
-        UInt8 min_compatible_version = 0;
-
-        /// Actual metadata.
-        UUID part_uuid = UUIDHelpers::Nil;
-
-        void write(WriteBuffer & meta_out) const;
-        void read(ReadBuffer & meta_in);
-
-    private:
-        static constexpr auto JSON_KEY_PART_UUID = "part_uuid";
-
-        String toJSON() const;
-        void fromJSON(const String & buf);
-    };
-
-    constexpr static UInt8 WAL_VERSION = 1;
-    constexpr static auto WAL_FILE_NAME = "wal";
-    constexpr static auto WAL_FILE_EXTENSION = ".bin";
-    constexpr static auto DEFAULT_WAL_FILE_NAME = "wal.bin";
-
-    MergeTreeWriteAheadLog(MergeTreeData & storage_, const DiskPtr & disk_,
-        const String & name = DEFAULT_WAL_FILE_NAME);
-
-    ~MergeTreeWriteAheadLog();
-
-    void dropPart(const String & part_name);
-    std::vector<MergeTreeMutableDataPartPtr> restore(
-        const StorageMetadataPtr & metadata_snapshot,
-        ContextPtr context,
-        DataPartsLock & parts_lock,
-        bool readonly);
-
-    using MinMaxBlockNumber = std::pair<Int64, Int64>;
-    static std::optional<MinMaxBlockNumber> tryParseMinMaxBlockNumber(const String & filename);
-    void shutdown();
-
-    /// Drop all write ahead logs from disk. Useful during table drop.
-    static void dropAllWriteAheadLogs(DiskPtr disk_to_drop, std::string relative_data_path);
-private:
-    void init();
-    void rotate(const std::unique_lock<std::mutex> & lock);
-
-    const MergeTreeData & storage;
-    DiskPtr disk;
-    String name;
-    String path;
-
-    std::unique_ptr<WriteBuffer> out;
-    std::unique_ptr<NativeWriter> block_out;
-
-    Int64 min_block_number = std::numeric_limits<Int64>::max();
-    Int64 max_block_number = -1;
-
-    BackgroundSchedulePool & pool;
-    BackgroundSchedulePoolTaskHolder sync_task;
-    std::condition_variable sync_cv;
-
-    size_t bytes_at_last_sync = 0;
-    bool sync_scheduled = false;
-    bool shutdown_called = false;
-
-    mutable std::mutex write_mutex;
-
-    LoggerPtr log;
-};
-
-}
diff --git a/src/Storages/MergeTree/MutateTask.cpp b/src/Storages/MergeTree/MutateTask.cpp
index a5b8a2a2a6d..150cc27c369 100644
--- a/src/Storages/MergeTree/MutateTask.cpp
+++ b/src/Storages/MergeTree/MutateTask.cpp
@@ -1235,14 +1235,8 @@ void PartMergerWriter::prepare()
 
     for (size_t i = 0, size = ctx->projections_to_build.size(); i < size; ++i)
     {
-        // If the parent part is an in-memory part, squash projection output into one block and
-        // build in-memory projection because we don't support merging into a new in-memory part.
-        // Otherwise we split the materialization into multiple stages similar to the process of
-        // INSERT SELECT query.
-        if (ctx->new_data_part->getType() == MergeTreeDataPartType::InMemory)
-            projection_squashes.emplace_back(0, 0);
-        else
-            projection_squashes.emplace_back(settings.min_insert_block_size_rows, settings.min_insert_block_size_bytes);
+        // We split the materialization into multiple stages similar to the process of INSERT SELECT query.
+        projection_squashes.emplace_back(settings.min_insert_block_size_rows, settings.min_insert_block_size_bytes);
     }
 }
 
diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeQueue.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeQueue.cpp
index 775b9afaee2..858eae4afd9 100644
--- a/src/Storages/MergeTree/ReplicatedMergeTreeQueue.cpp
+++ b/src/Storages/MergeTree/ReplicatedMergeTreeQueue.cpp
@@ -1350,10 +1350,7 @@ bool ReplicatedMergeTreeQueue::shouldExecuteLogEntry(
             auto part = data.getPartIfExists(name, {MergeTreeDataPartState::PreActive, MergeTreeDataPartState::Active, MergeTreeDataPartState::Outdated});
             if (part)
             {
-                if (auto part_in_memory = asInMemoryPart(part))
-                    sum_parts_size_in_bytes += part_in_memory->block.bytes();
-                else
-                    sum_parts_size_in_bytes += part->getBytesOnDisk();
+                sum_parts_size_in_bytes += part->getBytesOnDisk();
 
                 if (entry.type == LogEntry::MUTATE_PART && !storage.mutation_backoff_policy.partCanBeMutated(part->name))
                 {
diff --git a/src/Storages/MergeTree/checkDataPart.cpp b/src/Storages/MergeTree/checkDataPart.cpp
index c74063cfd4b..aea0020f89d 100644
--- a/src/Storages/MergeTree/checkDataPart.cpp
+++ b/src/Storages/MergeTree/checkDataPart.cpp
@@ -7,7 +7,6 @@
 #include <Storages/MergeTree/MergeTreeIndexGranularity.h>
 #include <Storages/MergeTree/checkDataPart.h>
 #include <Storages/MergeTree/MergeTreeDataPartCompact.h>
-#include <Storages/MergeTree/MergeTreeDataPartInMemory.h>
 #include <Storages/MergeTree/IDataPartStorage.h>
 #include <Interpreters/Cache/FileCache.h>
 #include <Interpreters/Cache/FileCacheFactory.h>
@@ -310,22 +309,11 @@ static IMergeTreeDataPart::Checksums checkDataPart(
     return checksums_data;
 }
 
-IMergeTreeDataPart::Checksums checkDataPartInMemory(const DataPartInMemoryPtr & data_part)
-{
-    IMergeTreeDataPart::Checksums data_checksums;
-    data_checksums.files["data.bin"] = data_part->calculateBlockChecksum();
-    data_part->checksums.checkEqual(data_checksums, true);
-    return data_checksums;
-}
-
 IMergeTreeDataPart::Checksums checkDataPart(
     MergeTreeData::DataPartPtr data_part,
     bool require_checksums,
     std::function<bool()> is_cancelled)
 {
-    if (auto part_in_memory = asInMemoryPart(data_part))
-        return checkDataPartInMemory(part_in_memory);
-
     /// If check of part has failed and it is stored on disk with cache
     /// try to drop cache and check it once again because maybe the cache
     /// is broken not the part itself.
diff --git a/src/Storages/MergeTree/examples/CMakeLists.txt b/src/Storages/MergeTree/examples/CMakeLists.txt
deleted file mode 100644
index 25bba7ae0b4..00000000000
--- a/src/Storages/MergeTree/examples/CMakeLists.txt
+++ /dev/null
@@ -1,2 +0,0 @@
-clickhouse_add_executable (wal_action_metadata wal_action_metadata.cpp)
-target_link_libraries (wal_action_metadata PRIVATE dbms)
diff --git a/src/Storages/MergeTree/examples/wal_action_metadata.cpp b/src/Storages/MergeTree/examples/wal_action_metadata.cpp
deleted file mode 100644
index 03c38c7a186..00000000000
--- a/src/Storages/MergeTree/examples/wal_action_metadata.cpp
+++ /dev/null
@@ -1,61 +0,0 @@
-#include <iostream>
-
-#include <IO/MemoryReadWriteBuffer.h>
-#include <Storages/MergeTree/MergeTreeWriteAheadLog.h>
-
-namespace DB
-{
-namespace ErrorCodes
-{
-    extern const int UNKNOWN_FORMAT_VERSION;
-}
-}
-
-int main(int, char **)
-{
-    try
-    {
-        {
-            std::cout << "test: dummy test" << std::endl;
-
-            DB::MergeTreeWriteAheadLog::ActionMetadata metadata_out;
-            DB::MemoryWriteBuffer buf{};
-
-            metadata_out.write(buf);
-            buf.finalize();
-
-            metadata_out.read(*buf.tryGetReadBuffer());
-        }
-
-        {
-            std::cout << "test: min compatibility" << std::endl;
-
-            DB::MergeTreeWriteAheadLog::ActionMetadata metadata_out;
-            metadata_out.min_compatible_version = DB::MergeTreeWriteAheadLog::WAL_VERSION + 1;
-            DB::MemoryWriteBuffer buf{};
-
-            metadata_out.write(buf);
-            buf.finalize();
-
-            try
-            {
-                metadata_out.read(*buf.tryGetReadBuffer());
-            }
-            catch (const DB::Exception & e)
-            {
-                if (e.code() != DB::ErrorCodes::UNKNOWN_FORMAT_VERSION)
-                {
-                    std::cerr << "Expected UNKNOWN_FORMAT_VERSION exception but got: "
-                        << e.what() << ", " << e.displayText() << std::endl;
-                }
-            }
-        }
-    }
-    catch (const DB::Exception & e)
-    {
-        std::cerr << e.what() << ", " << e.displayText() << std::endl;
-        return 1;
-    }
-
-    return 0;
-}
diff --git a/src/Storages/StorageJoin.cpp b/src/Storages/StorageJoin.cpp
index b122674466f..d12e5b1a20b 100644
--- a/src/Storages/StorageJoin.cpp
+++ b/src/Storages/StorageJoin.cpp
@@ -16,6 +16,7 @@
 #include <Common/Exception.h>
 #include <Core/ColumnsWithTypeAndName.h>
 #include <Interpreters/JoinUtils.h>
+#include <Formats/NativeWriter.h>
 
 #include <Compression/CompressedWriteBuffer.h>
 #include <Processors/ISource.h>
@@ -25,6 +26,7 @@
 #include <Poco/String.h>
 #include <filesystem>
 
+
 namespace fs = std::filesystem;
 
 namespace DB
diff --git a/src/Storages/StorageMemory.cpp b/src/Storages/StorageMemory.cpp
index cf23d17fc1d..7a8fb9feeda 100644
--- a/src/Storages/StorageMemory.cpp
+++ b/src/Storages/StorageMemory.cpp
@@ -19,6 +19,8 @@
 #include <Processors/QueryPlan/ReadFromMemoryStorageStep.h>
 #include <Processors/QueryPlan/QueryPlan.h>
 #include <Parsers/ASTCreateQuery.h>
+#include <Formats/NativeReader.h>
+#include <Formats/NativeWriter.h>
 
 #include <Common/FileChecker.h>
 #include <Compression/CompressedReadBuffer.h>
diff --git a/src/Storages/StorageMergeTree.cpp b/src/Storages/StorageMergeTree.cpp
index e15b308f084..663e7f435b7 100644
--- a/src/Storages/StorageMergeTree.cpp
+++ b/src/Storages/StorageMergeTree.cpp
@@ -37,7 +37,6 @@
 #include <Storages/AlterCommands.h>
 #include <Storages/PartitionCommands.h>
 #include <Storages/MergeTree/MergeTreeSink.h>
-#include <Storages/MergeTree/MergeTreeDataPartInMemory.h>
 #include <Storages/MergeTree/MergePlainMergeTreeTask.h>
 #include <Storages/MergeTree/PartitionPruner.h>
 #include <Storages/MergeTree/MergeList.h>
diff --git a/utils/check-style/aspell-ignore/en/aspell-dict.txt b/utils/check-style/aspell-ignore/en/aspell-dict.txt
index 8bf12247ef2..7db9d699176 100644
--- a/utils/check-style/aspell-ignore/en/aspell-dict.txt
+++ b/utils/check-style/aspell-ignore/en/aspell-dict.txt
@@ -685,7 +685,6 @@ PartsCommitted
 PartsCompact
 PartsDeleteOnDestroy
 PartsDeleting
-PartsInMemory
 PartsOutdated
 PartsPreActive
 PartsPreCommitted

From 24a248859d97f7faf78a1dc98a056332cde97401 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 10 Mar 2024 06:52:47 +0100
Subject: [PATCH 120/443] Remove unneeded file

---
 src/Compression/examples/CMakeLists.txt       |  3 -
 .../cached_compressed_read_buffer.cpp         | 79 -------------------
 .../examples/compressed_buffer.cpp            |  3 -
 3 files changed, 85 deletions(-)
 delete mode 100644 src/Compression/examples/cached_compressed_read_buffer.cpp

diff --git a/src/Compression/examples/CMakeLists.txt b/src/Compression/examples/CMakeLists.txt
index 7bf68e8845e..86fa5c3f78d 100644
--- a/src/Compression/examples/CMakeLists.txt
+++ b/src/Compression/examples/CMakeLists.txt
@@ -1,5 +1,2 @@
 clickhouse_add_executable (compressed_buffer compressed_buffer.cpp)
 target_link_libraries (compressed_buffer PRIVATE dbms)
-
-clickhouse_add_executable (cached_compressed_read_buffer cached_compressed_read_buffer.cpp)
-target_link_libraries (cached_compressed_read_buffer PRIVATE dbms)
diff --git a/src/Compression/examples/cached_compressed_read_buffer.cpp b/src/Compression/examples/cached_compressed_read_buffer.cpp
deleted file mode 100644
index a8e14ac7271..00000000000
--- a/src/Compression/examples/cached_compressed_read_buffer.cpp
+++ /dev/null
@@ -1,79 +0,0 @@
-#include <iostream>
-#include <iomanip>
-#include <limits>
-
-#include <Compression/CompressionFactory.h>
-#include <Compression/CachedCompressedReadBuffer.h>
-#include <IO/WriteBufferFromFile.h>
-#include <Disks/IO/createReadBufferFromFileBase.h>
-#include <IO/copyData.h>
-
-#include <Common/Stopwatch.h>
-
-
-int main(int argc, char ** argv)
-{
-    using namespace DB;
-
-    if (argc < 2)
-    {
-        std::cerr << "Usage: program path\n";
-        return 1;
-    }
-
-    try
-    {
-        UncompressedCache cache("SLRU", 1024, 0.5);
-        std::string path = argv[1];
-
-        std::cerr << std::fixed << std::setprecision(3);
-
-        size_t hits = 0;
-        size_t misses = 0;
-
-        {
-            Stopwatch watch;
-            CachedCompressedReadBuffer in(
-                path,
-                [&]()
-                {
-                    return createReadBufferFromFileBase(path, {});
-                },
-                &cache
-            );
-            WriteBufferFromFile out("/dev/null");
-            copyData(in, out);
-
-            std::cerr << "Elapsed: " << watch.elapsedSeconds() << std::endl;
-        }
-
-        cache.getStats(hits, misses);
-        std::cerr << "Hits: " << hits << ", misses: " << misses << std::endl;
-
-        {
-            Stopwatch watch;
-            CachedCompressedReadBuffer in(
-                path,
-                [&]()
-                {
-                    return createReadBufferFromFileBase(path, {});
-                },
-                &cache
-            );
-            WriteBufferFromFile out("/dev/null");
-            copyData(in, out);
-
-            std::cerr << "Elapsed: " << watch.elapsedSeconds() << std::endl;
-        }
-
-        cache.getStats(hits, misses);
-        std::cerr << "Hits: " << hits << ", misses: " << misses << std::endl;
-    }
-    catch (const Exception & e)
-    {
-        std::cerr << e.what() << ", " << e.displayText() << std::endl;
-        return 1;
-    }
-
-    return 0;
-}
diff --git a/src/Compression/examples/compressed_buffer.cpp b/src/Compression/examples/compressed_buffer.cpp
index 74646ff0f28..530f0938662 100644
--- a/src/Compression/examples/compressed_buffer.cpp
+++ b/src/Compression/examples/compressed_buffer.cpp
@@ -1,7 +1,4 @@
-#include <string>
-
 #include <iostream>
-#include <fstream>
 #include <iomanip>
 
 #include <Common/Stopwatch.h>

From a231f7de9ae7b86c41204c774dda40c3fbe73ac6 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 10 Mar 2024 06:55:09 +0100
Subject: [PATCH 121/443] Remove unneeded file

---
 src/Compression/examples/CMakeLists.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Compression/examples/CMakeLists.txt b/src/Compression/examples/CMakeLists.txt
index 86fa5c3f78d..a924075d0dc 100644
--- a/src/Compression/examples/CMakeLists.txt
+++ b/src/Compression/examples/CMakeLists.txt
@@ -1,2 +1,2 @@
 clickhouse_add_executable (compressed_buffer compressed_buffer.cpp)
-target_link_libraries (compressed_buffer PRIVATE dbms)
+target_link_libraries (compressed_buffer PRIVATE clickhouse_common_io)

From 69cab686e4e418c0e7cfa74dfea459671d14bc49 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 10 Mar 2024 10:03:31 +0100
Subject: [PATCH 122/443] Fix build of examples

---
 src/Compression/examples/CMakeLists.txt       |   2 +-
 src/Core/examples/CMakeLists.txt              |   3 -
 src/Core/examples/mysql_protocol.cpp          | 390 ------------------
 src/Interpreters/examples/CMakeLists.txt      |  22 +-
 src/Interpreters/examples/hash_map.cpp        |   1 -
 src/Interpreters/examples/hash_map_lookup.cpp |   2 -
 src/Processors/CMakeLists.txt                 |   4 -
 src/Processors/examples/CMakeLists.txt        |   4 -
 .../examples/comma_separated_streams.cpp      | 117 ------
 src/Processors/examples/test_in               |   8 -
 src/Storages/MergeTree/CMakeLists.txt         |   3 -
 .../MergeTree/examples/CMakeLists.txt         |   2 -
 .../examples/wal_action_metadata.cpp          |  61 ---
 src/Storages/examples/CMakeLists.txt          |   8 -
 src/Storages/examples/active_parts.py         |  41 --
 .../examples/async_read_buffer_from_hdfs.cpp  |  37 --
 ...get_abandonable_lock_in_all_partitions.cpp |  71 ----
 17 files changed, 12 insertions(+), 764 deletions(-)
 delete mode 100644 src/Core/examples/mysql_protocol.cpp
 delete mode 100644 src/Processors/examples/CMakeLists.txt
 delete mode 100644 src/Processors/examples/comma_separated_streams.cpp
 delete mode 100644 src/Processors/examples/test_in
 delete mode 100644 src/Storages/MergeTree/examples/CMakeLists.txt
 delete mode 100644 src/Storages/MergeTree/examples/wal_action_metadata.cpp
 delete mode 100644 src/Storages/examples/active_parts.py
 delete mode 100644 src/Storages/examples/async_read_buffer_from_hdfs.cpp
 delete mode 100644 src/Storages/examples/get_abandonable_lock_in_all_partitions.cpp

diff --git a/src/Compression/examples/CMakeLists.txt b/src/Compression/examples/CMakeLists.txt
index a924075d0dc..a7cc6bebf42 100644
--- a/src/Compression/examples/CMakeLists.txt
+++ b/src/Compression/examples/CMakeLists.txt
@@ -1,2 +1,2 @@
 clickhouse_add_executable (compressed_buffer compressed_buffer.cpp)
-target_link_libraries (compressed_buffer PRIVATE clickhouse_common_io)
+target_link_libraries (compressed_buffer PRIVATE clickhouse_common_io clickhouse_compression)
diff --git a/src/Core/examples/CMakeLists.txt b/src/Core/examples/CMakeLists.txt
index 2326eada96d..f30ee25491f 100644
--- a/src/Core/examples/CMakeLists.txt
+++ b/src/Core/examples/CMakeLists.txt
@@ -6,6 +6,3 @@ target_link_libraries (field PRIVATE dbms)
 
 clickhouse_add_executable (string_ref_hash string_ref_hash.cpp)
 target_link_libraries (string_ref_hash PRIVATE clickhouse_common_io)
-
-clickhouse_add_executable (mysql_protocol mysql_protocol.cpp)
-target_link_libraries (mysql_protocol PRIVATE dbms)
diff --git a/src/Core/examples/mysql_protocol.cpp b/src/Core/examples/mysql_protocol.cpp
deleted file mode 100644
index a6247418e87..00000000000
--- a/src/Core/examples/mysql_protocol.cpp
+++ /dev/null
@@ -1,390 +0,0 @@
-#include <string>
-
-#include <Core/MySQL/Authentication.h>
-#include <Core/MySQL/MySQLClient.h>
-#include <Core/MySQL/PacketsConnection.h>
-#include <Core/MySQL/PacketsGeneric.h>
-#include <Core/MySQL/PacketsProtocolText.h>
-#include <IO/ReadBufferFromString.h>
-#include <IO/WriteBufferFromString.h>
-#include <IO/WriteBufferFromOStream.h>
-
-#include <boost/program_options.hpp>
-
-int main(int argc, char ** argv)
-{
-    using namespace DB;
-    using namespace MySQLProtocol;
-    using namespace MySQLProtocol::Generic;
-    using namespace MySQLProtocol::Authentication;
-    using namespace MySQLProtocol::ConnectionPhase;
-    using namespace MySQLProtocol::ProtocolText;
-
-
-    uint8_t server_sequence_id = 1;
-    uint8_t client_sequence_id = 1;
-    String user = "default";
-    String password = "123";
-    String database;
-
-    UInt8 charset_utf8 = 33;
-    UInt32 max_packet_size = MAX_PACKET_LENGTH;
-    String mysql_native_password = "mysql_native_password";
-
-    UInt32 server_capability_flags = CLIENT_PROTOCOL_41 | CLIENT_SECURE_CONNECTION | CLIENT_PLUGIN_AUTH
-        | CLIENT_PLUGIN_AUTH_LENENC_CLIENT_DATA | CLIENT_CONNECT_WITH_DB | CLIENT_DEPRECATE_EOF;
-
-    UInt32 client_capability_flags = CLIENT_PROTOCOL_41 | CLIENT_PLUGIN_AUTH | CLIENT_SECURE_CONNECTION;
-
-    /// Handshake packet
-    {
-        /// 1. Greeting:
-        /// 1.1 Server writes greeting to client
-        std::string s0;
-        WriteBufferFromString out0(s0);
-
-        Handshake server_handshake(
-            server_capability_flags, -1, "ClickHouse", "mysql_native_password", "aaaaaaaaaaaaaaaaaaaaa", CharacterSet::utf8_general_ci);
-        server_handshake.writePayload(out0, server_sequence_id);
-
-        /// 1.2 Client reads the greeting
-        ReadBufferFromString in0(s0);
-        Handshake client_handshake;
-        client_handshake.readPayload(in0, client_sequence_id);
-
-        /// Check packet
-        ASSERT(server_handshake.capability_flags == client_handshake.capability_flags)
-        ASSERT(server_handshake.status_flags == client_handshake.status_flags)
-        ASSERT(server_handshake.server_version == client_handshake.server_version)
-        ASSERT(server_handshake.protocol_version == client_handshake.protocol_version)
-        ASSERT(server_handshake.auth_plugin_data.substr(0, 20) == client_handshake.auth_plugin_data)
-        ASSERT(server_handshake.auth_plugin_name == client_handshake.auth_plugin_name)
-
-        /// 2. Greeting Response:
-        std::string s1;
-        WriteBufferFromString out1(s1);
-
-        /// 2.1 Client writes to server
-        Native41 native41(password, client_handshake.auth_plugin_data);
-        String auth_plugin_data = native41.getAuthPluginData();
-        HandshakeResponse client_handshake_response(
-            client_capability_flags, max_packet_size, charset_utf8, user, database, auth_plugin_data, mysql_native_password);
-        client_handshake_response.writePayload(out1, client_sequence_id);
-
-        /// 2.2 Server reads the response
-        ReadBufferFromString in1(s1);
-        HandshakeResponse server_handshake_response;
-        server_handshake_response.readPayload(in1, server_sequence_id);
-
-        /// Check
-        ASSERT(server_handshake_response.capability_flags == client_handshake_response.capability_flags)
-        ASSERT(server_handshake_response.character_set == client_handshake_response.character_set)
-        ASSERT(server_handshake_response.username == client_handshake_response.username)
-        ASSERT(server_handshake_response.database == client_handshake_response.database)
-        ASSERT(server_handshake_response.auth_response == client_handshake_response.auth_response)
-        ASSERT(server_handshake_response.auth_plugin_name == client_handshake_response.auth_plugin_name)
-    }
-
-    /// OK Packet
-    {
-        // 1. Server writes packet
-        std::string s0;
-        WriteBufferFromString out0(s0);
-        OKPacket server(0x00, server_capability_flags, 0, 0, 0, "", "");
-        server.writePayload(out0, server_sequence_id);
-
-        // 2. Client reads packet
-        ReadBufferFromString in0(s0);
-        ResponsePacket client(server_capability_flags);
-        client.readPayload(in0, client_sequence_id);
-
-        // Check
-        ASSERT(client.getType() == PACKET_OK)
-        ASSERT(client.ok.header == server.header)
-        ASSERT(client.ok.status_flags == server.status_flags)
-        ASSERT(client.ok.capabilities == server.capabilities)
-    }
-
-    /// ERR Packet
-    {
-        // 1. Server writes packet
-        std::string s0;
-        WriteBufferFromString out0(s0);
-        ERRPacket server(123, "12345", "This is the error message");
-        server.writePayload(out0, server_sequence_id);
-
-        // 2. Client reads packet
-        ReadBufferFromString in0(s0);
-        ResponsePacket client(server_capability_flags);
-        client.readPayload(in0, client_sequence_id);
-
-        // Check
-        ASSERT(client.getType() == PACKET_ERR)
-        ASSERT(client.err.header == server.header)
-        ASSERT(client.err.error_code == server.error_code)
-        ASSERT(client.err.sql_state == server.sql_state)
-        ASSERT(client.err.error_message == server.error_message)
-    }
-
-    /// EOF Packet
-    {
-        // 1. Server writes packet
-        std::string s0;
-        WriteBufferFromString out0(s0);
-        EOFPacket server(1, 1);
-        server.writePayload(out0, server_sequence_id);
-
-        // 2. Client reads packet
-        ReadBufferFromString in0(s0);
-        ResponsePacket client(server_capability_flags);
-        client.readPayload(in0, client_sequence_id);
-
-        // Check
-        ASSERT(client.getType() == PACKET_EOF)
-        ASSERT(client.eof.header == server.header)
-        ASSERT(client.eof.warnings == server.warnings)
-        ASSERT(client.eof.status_flags == server.status_flags)
-    }
-
-    /// ColumnDefinition Packet
-    {
-        // 1. Server writes packet
-        std::string s0;
-        WriteBufferFromString out0(s0);
-        ColumnDefinition server("schema", "tbl", "org_tbl", "name", "org_name", 33, 0x00, MYSQL_TYPE_STRING, 0x00, 0x00);
-        server.writePayload(out0, server_sequence_id);
-
-        // 2. Client reads packet
-        ReadBufferFromString in0(s0);
-        ColumnDefinition client;
-        client.readPayload(in0, client_sequence_id);
-
-        // Check
-        ASSERT(client.column_type == server.column_type)
-        ASSERT(client.column_length == server.column_length)
-        ASSERT(client.next_length == server.next_length)
-        ASSERT(client.character_set == server.character_set)
-        ASSERT(client.decimals == server.decimals)
-        ASSERT(client.name == server.name)
-        ASSERT(client.org_name == server.org_name)
-        ASSERT(client.table == server.table)
-        ASSERT(client.org_table == server.org_table)
-        ASSERT(client.schema == server.schema)
-    }
-
-    /// GTID sets tests.
-   {
-        struct Testcase
-        {
-            String name;
-            String sets;
-            String want;
-        };
-
-        Testcase cases[] = {
-            {"gtid-sets-without-whitespace",
-             "2c5adab4-d64a-11e5-82df-ac162d72dac0:1-247743812,9f58c169-d121-11e7-835b-ac162db9c048:1-56060985:56060987-56061175:56061177-"
-             "56061224:56061226-75201528:75201530-75201755:75201757-75201983:75201985-75407550:75407552-75407604:75407606-75407661:"
-             "75407663-87889848:87889850-87889935:87889937-87890042:87890044-88391955:88391957-88392125:88392127-88392245:88392247-"
-             "88755771:88755773-88755826:88755828-88755921:88755923-100279047:100279049-100279126:100279128-100279247:100279249-121672430:"
-             "121672432-121672503:121672505-121672524:121672526-122946019:122946021-122946291:122946293-122946469:122946471-134313284:"
-             "134313286-134313415:134313417-134313648:134313650-136492728:136492730-136492784:136492786-136492904:136492906-145582402:"
-             "145582404-145582439:145582441-145582463:145582465-147455222:147455224-147455262:147455264-147455277:147455279-149319049:"
-             "149319051-149319261:149319263-150635915,a6d83ff6-bfcf-11e7-8c93-246e96158550:1-126618302",
-             "2c5adab4-d64a-11e5-82df-ac162d72dac0:1-247743812,9f58c169-d121-11e7-835b-ac162db9c048:1-56060985:56060987-56061175:56061177-"
-             "56061224:56061226-75201528:75201530-75201755:75201757-75201983:75201985-75407550:75407552-75407604:75407606-75407661:"
-             "75407663-87889848:87889850-87889935:87889937-87890042:87890044-88391955:88391957-88392125:88392127-88392245:88392247-"
-             "88755771:88755773-88755826:88755828-88755921:88755923-100279047:100279049-100279126:100279128-100279247:100279249-121672430:"
-             "121672432-121672503:121672505-121672524:121672526-122946019:122946021-122946291:122946293-122946469:122946471-134313284:"
-             "134313286-134313415:134313417-134313648:134313650-136492728:136492730-136492784:136492786-136492904:136492906-145582402:"
-             "145582404-145582439:145582441-145582463:145582465-147455222:147455224-147455262:147455264-147455277:147455279-149319049:"
-             "149319051-149319261:149319263-150635915,a6d83ff6-bfcf-11e7-8c93-246e96158550:1-126618302"},
-
-            {"gtid-sets-with-whitespace",
-             "2c5adab4-d64a-11e5-82df-ac162d72dac0:1-247743812, 9f58c169-d121-11e7-835b-ac162db9c048:1-56060985:56060987-56061175:56061177",
-             "2c5adab4-d64a-11e5-82df-ac162d72dac0:1-247743812,9f58c169-d121-11e7-835b-ac162db9c048:1-56060985:56060987-56061175:56061177"},
-
-            {"gtid-sets-single", "2c5adab4-d64a-11e5-82df-ac162d72dac0:1-247743812", "2c5adab4-d64a-11e5-82df-ac162d72dac0:1-247743812"}};
-
-        for (auto & tc : cases)
-        {
-            GTIDSets gtid_sets;
-            gtid_sets.parse(tc.sets);
-
-            String want = tc.want;
-            String got = gtid_sets.toString();
-            ASSERT(want == got)
-        }
-    }
-
-    {
-        struct Testcase
-        {
-            String name;
-            String gtid_sets;
-            String gtid_str;
-            String want;
-        };
-
-        Testcase cases[] = {
-            {"merge",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:1-2:4-7",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:3",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:1-7"},
-
-            {"merge-front",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:1-2:5-7",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:3",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:1-3:5-7"},
-
-            {"extend-interval",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:1-2:6-7",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:4",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:1-2:4:6-7"},
-
-            {"extend-interval",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:1-2:4:7-9",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:5",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:1-2:4-5:7-9"},
-
-            {"extend-interval",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:6-7",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:4",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:4:6-7"},
-
-            {"extend-interval",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:6-7",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:9",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:6-7:9"},
-
-            {"extend-interval",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:6-7",
-             "20662d71-9d91-11ea-bbc2-0242ac110003:9",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:6-7,20662d71-9d91-11ea-bbc2-0242ac110003:9"},
-
-            {"shrink-sequence",
-              "10662d71-9d91-11ea-bbc2-0242ac110003:1-3:4-5:7",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:6",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:1-7"},
-
-            {"shrink-sequence",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:1-3:4-5:10",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:8",
-             "10662d71-9d91-11ea-bbc2-0242ac110003:1-5:8:10"
-            }
-        };
-
-        for (auto & tc : cases)
-        {
-            GTIDSets gtid_sets;
-            gtid_sets.parse(tc.gtid_sets);
-            ASSERT(tc.gtid_sets == gtid_sets.toString())
-
-            GTIDSets gtid_sets1;
-            gtid_sets1.parse(tc.gtid_str);
-
-            GTID gtid;
-            gtid.uuid = gtid_sets1.sets[0].uuid;
-            gtid.seq_no = gtid_sets1.sets[0].intervals[0].start;
-            gtid_sets.update(gtid);
-
-            String want = tc.want;
-            String got = gtid_sets.toString();
-            ASSERT(want == got)
-        }
-    }
-
-    {
-        /// mysql_protocol --host=172.17.0.3 --user=root --password=123 --db=sbtest
-        try
-        {
-            boost::program_options::options_description desc("Allowed options");
-            desc.add_options()("host", boost::program_options::value<std::string>()->required(), "master host")(
-                "port", boost::program_options::value<std::int32_t>()->default_value(3306), "master port")(
-                "user", boost::program_options::value<std::string>()->default_value("root"), "master user")(
-                "password", boost::program_options::value<std::string>()->required(), "master password")(
-                "gtid", boost::program_options::value<std::string>()->default_value(""), "executed GTID sets")(
-                "db", boost::program_options::value<std::string>()->required(), "replicate do db")(
-                "binlog_checksum", boost::program_options::value<std::string>()->default_value("CRC32"), "master binlog_checksum");
-
-            boost::program_options::variables_map options;
-            boost::program_options::store(boost::program_options::parse_command_line(argc, argv, desc), options);
-            if (argc == 0)
-            {
-                return 1;
-            }
-
-            auto host = options.at("host").as<String>();
-            auto port = options.at("port").as<Int32>();
-            auto master_user = options.at("user").as<String>();
-            auto master_password = options.at("password").as<String>();
-            auto gtid_sets = options.at("gtid").as<String>();
-            auto replicate_db = options.at("db").as<String>();
-            auto binlog_checksum = options.at("binlog_checksum").as<String>();
-
-            std::cerr << "Master Host: " << host << ", Port: " << port << ", User: " << master_user << ", Password: " << master_password
-                      << ", Replicate DB: " << replicate_db << ", GTID: " << gtid_sets << std::endl;
-
-            UInt32 slave_id = 9004;
-            MySQLClient slave(host, port, master_user, master_password);
-
-            /// Connect to the master.
-            slave.connect();
-            slave.startBinlogDumpGTID(slave_id, replicate_db, {}, gtid_sets, binlog_checksum);
-
-            WriteBufferFromOStream cerr(std::cerr);
-
-            /// Read one binlog event on by one.
-            while (true)
-            {
-                auto event = slave.readOneBinlogEvent();
-                switch (event->type())
-                {
-                    case MYSQL_QUERY_EVENT: {
-                        auto binlog_event = std::static_pointer_cast<QueryEvent>(event);
-                        binlog_event->dump(cerr);
-
-                        Position pos = slave.getPosition();
-                        pos.dump(cerr);
-                        break;
-                    }
-                    case MYSQL_WRITE_ROWS_EVENT: {
-                        auto binlog_event = std::static_pointer_cast<WriteRowsEvent>(event);
-                        binlog_event->dump(cerr);
-
-                        Position pos = slave.getPosition();
-                        pos.dump(cerr);
-                        break;
-                    }
-                    case MYSQL_UPDATE_ROWS_EVENT: {
-                        auto binlog_event = std::static_pointer_cast<UpdateRowsEvent>(event);
-                        binlog_event->dump(cerr);
-
-                        Position pos = slave.getPosition();
-                        pos.dump(cerr);
-                        break;
-                    }
-                    case MYSQL_DELETE_ROWS_EVENT: {
-                        auto binlog_event = std::static_pointer_cast<DeleteRowsEvent>(event);
-                        binlog_event->dump(cerr);
-
-                        Position pos = slave.getPosition();
-                        pos.dump(cerr);
-                        break;
-                    }
-                    default:
-                        if (event->header.type != MySQLReplication::EventType::HEARTBEAT_EVENT)
-                        {
-                            event->dump(cerr);
-                        }
-                        break;
-                }
-            }
-        }
-        catch (const Exception & ex)
-        {
-            std::cerr << "Error: " << ex.message() << std::endl;
-            return 1;
-        }
-    }
-}
diff --git a/src/Interpreters/examples/CMakeLists.txt b/src/Interpreters/examples/CMakeLists.txt
index 11c219ff64e..8bb7f9eeb98 100644
--- a/src/Interpreters/examples/CMakeLists.txt
+++ b/src/Interpreters/examples/CMakeLists.txt
@@ -1,35 +1,35 @@
 clickhouse_add_executable (hash_map hash_map.cpp)
-target_link_libraries (hash_map PRIVATE dbms ch_contrib::sparsehash)
+target_link_libraries (hash_map PRIVATE dbms clickhouse_functions ch_contrib::sparsehash)
 
 clickhouse_add_executable (hash_map_lookup hash_map_lookup.cpp)
-target_link_libraries (hash_map_lookup PRIVATE dbms)
+target_link_libraries (hash_map_lookup PRIVATE clickhouse_common_io clickhouse_compression)
 
 clickhouse_add_executable (hash_map3 hash_map3.cpp)
-target_link_libraries (hash_map3 PRIVATE dbms ch_contrib::farmhash ch_contrib::metrohash)
+target_link_libraries (hash_map3 PRIVATE clickhouse_common_io clickhouse_compression ch_contrib::farmhash ch_contrib::metrohash)
 
 clickhouse_add_executable (hash_map_string hash_map_string.cpp)
-target_link_libraries (hash_map_string PRIVATE dbms ch_contrib::sparsehash)
+target_link_libraries (hash_map_string PRIVATE clickhouse_common_io clickhouse_compression ch_contrib::sparsehash)
 
 clickhouse_add_executable (hash_map_string_2 hash_map_string_2.cpp)
-target_link_libraries (hash_map_string_2 PRIVATE dbms)
+target_link_libraries (hash_map_string_2 PRIVATE clickhouse_common_io clickhouse_compression)
 
 clickhouse_add_executable (hash_map_string_3 hash_map_string_3.cpp)
-target_link_libraries (hash_map_string_3 PRIVATE dbms ch_contrib::farmhash ch_contrib::metrohash)
+target_link_libraries (hash_map_string_3 PRIVATE clickhouse_common_io clickhouse_compression ch_contrib::farmhash ch_contrib::metrohash)
 
 clickhouse_add_executable (hash_map_string_small hash_map_string_small.cpp)
-target_link_libraries (hash_map_string_small PRIVATE dbms ch_contrib::sparsehash)
+target_link_libraries (hash_map_string_small PRIVATE clickhouse_common_io clickhouse_compression ch_contrib::sparsehash)
 
 clickhouse_add_executable (string_hash_map string_hash_map.cpp)
-target_link_libraries (string_hash_map PRIVATE dbms ch_contrib::sparsehash)
+target_link_libraries (string_hash_map PRIVATE clickhouse_common_io clickhouse_compression ch_contrib::sparsehash)
 
 clickhouse_add_executable (string_hash_map_aggregation string_hash_map.cpp)
-target_link_libraries (string_hash_map_aggregation PRIVATE dbms)
+target_link_libraries (string_hash_map_aggregation PRIVATE clickhouse_common_io clickhouse_compression)
 
 clickhouse_add_executable (string_hash_set string_hash_set.cpp)
-target_link_libraries (string_hash_set PRIVATE dbms)
+target_link_libraries (string_hash_set PRIVATE clickhouse_common_io clickhouse_compression)
 
 clickhouse_add_executable (two_level_hash_map two_level_hash_map.cpp)
-target_link_libraries (two_level_hash_map PRIVATE dbms ch_contrib::sparsehash)
+target_link_libraries (two_level_hash_map PRIVATE clickhouse_common_io clickhouse_compression ch_contrib::sparsehash)
 
 clickhouse_add_executable (jit_example jit_example.cpp)
 target_link_libraries (jit_example PRIVATE dbms)
diff --git a/src/Interpreters/examples/hash_map.cpp b/src/Interpreters/examples/hash_map.cpp
index b55f174678e..0a91d00809f 100644
--- a/src/Interpreters/examples/hash_map.cpp
+++ b/src/Interpreters/examples/hash_map.cpp
@@ -15,7 +15,6 @@
 #include <IO/ReadBufferFromFile.h>
 #include <Compression/CompressedReadBuffer.h>
 #include <Common/HashTable/HashMap.h>
-#include <Common/HashTable/PackedHashMap.h>
 #include <AggregateFunctions/IAggregateFunction.h>
 #include <AggregateFunctions/AggregateFunctionFactory.h>
 #include <DataTypes/DataTypesNumber.h>
diff --git a/src/Interpreters/examples/hash_map_lookup.cpp b/src/Interpreters/examples/hash_map_lookup.cpp
index fd6b231cf73..829a234c537 100644
--- a/src/Interpreters/examples/hash_map_lookup.cpp
+++ b/src/Interpreters/examples/hash_map_lookup.cpp
@@ -8,10 +8,8 @@
 #define DBMS_HASH_MAP_DEBUG_RESIZES
 
 #include <Compression/CompressedReadBuffer.h>
-#include <base/types.h>
 #include <IO/ReadBufferFromFile.h>
 #include <IO/ReadHelpers.h>
-#include <Interpreters/AggregationCommon.h>
 #include <Common/HashTable/FixedHashMap.h>
 #include <Common/HashTable/HashMap.h>
 
diff --git a/src/Processors/CMakeLists.txt b/src/Processors/CMakeLists.txt
index 7e965188b4c..e69de29bb2d 100644
--- a/src/Processors/CMakeLists.txt
+++ b/src/Processors/CMakeLists.txt
@@ -1,4 +0,0 @@
-if (ENABLE_EXAMPLES)
-    add_subdirectory(examples)
-endif ()
-
diff --git a/src/Processors/examples/CMakeLists.txt b/src/Processors/examples/CMakeLists.txt
deleted file mode 100644
index 5d43a0d7d08..00000000000
--- a/src/Processors/examples/CMakeLists.txt
+++ /dev/null
@@ -1,4 +0,0 @@
-if (TARGET ch_contrib::hivemetastore)
-    clickhouse_add_executable (comma_separated_streams comma_separated_streams.cpp)
-    target_link_libraries (comma_separated_streams PRIVATE dbms)
-endif()
diff --git a/src/Processors/examples/comma_separated_streams.cpp b/src/Processors/examples/comma_separated_streams.cpp
deleted file mode 100644
index 2ec5564f346..00000000000
--- a/src/Processors/examples/comma_separated_streams.cpp
+++ /dev/null
@@ -1,117 +0,0 @@
-#include <string>
-#include <iostream>
-#include <fstream>
-
-#include <IO/ReadBufferFromFile.h>
-#include <IO/WriteBufferFromFile.h>
-#include <DataTypes/DataTypesNumber.h>
-#include <DataTypes/DataTypeString.h>
-#include <Processors/Formats/Impl/HiveTextRowInputFormat.h>
-#include <Processors/Formats/Impl/CSVRowOutputFormat.h>
-#include <Formats/FormatFactory.h>
-#include <QueryPipeline/QueryPipeline.h>
-#include <Processors/Executors/PullingPipelineExecutor.h>
-
-using namespace DB;
-
-int main()
-try
-{
-    Block sample;
-    {
-        // a
-        ColumnWithTypeAndName col;
-        col.name = "a";
-        col.type = std::make_shared<DataTypeInt64>();
-        sample.insert(std::move(col));
-    }
-    {
-        // b
-        ColumnWithTypeAndName col;
-        col.name = "b";
-        col.type = std::make_shared<DataTypeFloat64>();
-        sample.insert(std::move(col));
-    }
-    {
-        // c
-        ColumnWithTypeAndName col;
-        col.name = "c";
-        col.type = std::make_shared<DataTypeInt64>();
-        sample.insert(std::move(col));
-    }
-    {
-        // d
-        ColumnWithTypeAndName col;
-        col.name = "d";
-        col.type = std::make_shared<DataTypeString>();
-        sample.insert(std::move(col));
-    }
-    {
-        // e
-        ColumnWithTypeAndName col;
-        col.name = "e";
-        col.type = std::make_shared<DataTypeString>();
-        sample.insert(std::move(col));
-    }
-    {
-        // f
-        ColumnWithTypeAndName col;
-        col.name = "f";
-        col.type = std::make_shared<DataTypeString>();
-        sample.insert(std::move(col));
-    }
-    {
-        // g
-        ColumnWithTypeAndName col;
-        col.name = "g";
-        col.type = std::make_shared<DataTypeInt64>();
-        sample.insert(std::move(col));
-    }
-    {
-        // h
-        ColumnWithTypeAndName col;
-        col.name = "h";
-        col.type = std::make_shared<DataTypeInt64>();
-        sample.insert(std::move(col));
-    }
-
-
-    ReadBufferFromFile in_buf("test_in");
-    WriteBufferFromFile out_buf("test_out");
-
-    FormatSettings format_settings;
-    format_settings.with_names_use_header = true;
-    format_settings.skip_unknown_fields = true;
-    format_settings.csv.delimiter = '\x01';
-    format_settings.hive_text.input_field_names =
-    {
-        "d",
-        "e",
-        "f",
-        "a",
-        "b",
-        "c",
-        "g",
-        "h",
-        "i",
-        "j",
-    };
-
-    RowInputFormatParams in_params{DEFAULT_INSERT_BLOCK_SIZE};
-    InputFormatPtr input_format = std::make_shared<HiveTextRowInputFormat>(sample, in_buf, in_params, format_settings);
-    auto pipeline = QueryPipeline(std::move(input_format));
-    auto reader = std::make_unique<PullingPipelineExecutor>(pipeline);
-
-    OutputFormatPtr output_format = std::make_shared<CSVRowOutputFormat>(out_buf, sample, true, true, format_settings);
-    Block res;
-    while (reader->pull(res))
-    {
-        output_format->write(res);
-    }
-    return 0;
-}
-catch (...)
-{
-    std::cerr << getCurrentExceptionMessage(true) << '\n';
-    return 1;
-}
diff --git a/src/Processors/examples/test_in b/src/Processors/examples/test_in
deleted file mode 100644
index c7df97a26a6..00000000000
--- a/src/Processors/examples/test_in
+++ /dev/null
@@ -1,8 +0,0 @@
-2021-09-14
JP
all
2
0.0
2
0
0
-2021-09-14
CI
all
2
0.0
1
0
0
-2021-09-14
JM
all
4
0.25
4
1
1
-2021-09-14
MM
all
31
0.1935483870967742
27
6
6
-2021-09-14
TZ
Android
3
0.3333333333333333
3
1
1
-2021-09-14
SG
all
8
0.25
4
1
2
-2021-09-14
PY
all
1
1.0
0
0
1
-2021-09-14
MX
all
1
0.0
1
0
0
diff --git a/src/Storages/MergeTree/CMakeLists.txt b/src/Storages/MergeTree/CMakeLists.txt
index 390835f17ae..e69de29bb2d 100644
--- a/src/Storages/MergeTree/CMakeLists.txt
+++ b/src/Storages/MergeTree/CMakeLists.txt
@@ -1,3 +0,0 @@
-if(ENABLE_EXAMPLES)
-    add_subdirectory(examples)
-endif()
diff --git a/src/Storages/MergeTree/examples/CMakeLists.txt b/src/Storages/MergeTree/examples/CMakeLists.txt
deleted file mode 100644
index 25bba7ae0b4..00000000000
--- a/src/Storages/MergeTree/examples/CMakeLists.txt
+++ /dev/null
@@ -1,2 +0,0 @@
-clickhouse_add_executable (wal_action_metadata wal_action_metadata.cpp)
-target_link_libraries (wal_action_metadata PRIVATE dbms)
diff --git a/src/Storages/MergeTree/examples/wal_action_metadata.cpp b/src/Storages/MergeTree/examples/wal_action_metadata.cpp
deleted file mode 100644
index 03c38c7a186..00000000000
--- a/src/Storages/MergeTree/examples/wal_action_metadata.cpp
+++ /dev/null
@@ -1,61 +0,0 @@
-#include <iostream>
-
-#include <IO/MemoryReadWriteBuffer.h>
-#include <Storages/MergeTree/MergeTreeWriteAheadLog.h>
-
-namespace DB
-{
-namespace ErrorCodes
-{
-    extern const int UNKNOWN_FORMAT_VERSION;
-}
-}
-
-int main(int, char **)
-{
-    try
-    {
-        {
-            std::cout << "test: dummy test" << std::endl;
-
-            DB::MergeTreeWriteAheadLog::ActionMetadata metadata_out;
-            DB::MemoryWriteBuffer buf{};
-
-            metadata_out.write(buf);
-            buf.finalize();
-
-            metadata_out.read(*buf.tryGetReadBuffer());
-        }
-
-        {
-            std::cout << "test: min compatibility" << std::endl;
-
-            DB::MergeTreeWriteAheadLog::ActionMetadata metadata_out;
-            metadata_out.min_compatible_version = DB::MergeTreeWriteAheadLog::WAL_VERSION + 1;
-            DB::MemoryWriteBuffer buf{};
-
-            metadata_out.write(buf);
-            buf.finalize();
-
-            try
-            {
-                metadata_out.read(*buf.tryGetReadBuffer());
-            }
-            catch (const DB::Exception & e)
-            {
-                if (e.code() != DB::ErrorCodes::UNKNOWN_FORMAT_VERSION)
-                {
-                    std::cerr << "Expected UNKNOWN_FORMAT_VERSION exception but got: "
-                        << e.what() << ", " << e.displayText() << std::endl;
-                }
-            }
-        }
-    }
-    catch (const DB::Exception & e)
-    {
-        std::cerr << e.what() << ", " << e.displayText() << std::endl;
-        return 1;
-    }
-
-    return 0;
-}
diff --git a/src/Storages/examples/CMakeLists.txt b/src/Storages/examples/CMakeLists.txt
index 225337d8ec8..cddfc9404d4 100644
--- a/src/Storages/examples/CMakeLists.txt
+++ b/src/Storages/examples/CMakeLists.txt
@@ -6,11 +6,3 @@ target_link_libraries (merge_selector2 PRIVATE dbms)
 
 clickhouse_add_executable (get_current_inserts_in_replicated get_current_inserts_in_replicated.cpp)
 target_link_libraries (get_current_inserts_in_replicated PRIVATE dbms clickhouse_common_config clickhouse_common_zookeeper string_utils)
-
-clickhouse_add_executable (get_abandonable_lock_in_all_partitions get_abandonable_lock_in_all_partitions.cpp)
-target_link_libraries (get_abandonable_lock_in_all_partitions PRIVATE dbms clickhouse_common_config clickhouse_common_zookeeper)
-
-if (TARGET ch_contrib::hdfs)
-    clickhouse_add_executable (async_read_buffer_from_hdfs async_read_buffer_from_hdfs.cpp)
-    target_link_libraries (async_read_buffer_from_hdfs PRIVATE dbms ch_contrib::hdfs)
-endif ()
diff --git a/src/Storages/examples/active_parts.py b/src/Storages/examples/active_parts.py
deleted file mode 100644
index d82c5ca96bf..00000000000
--- a/src/Storages/examples/active_parts.py
+++ /dev/null
@@ -1,41 +0,0 @@
-#!/usr/bin/python
-# coding=UTF-8
-
-# Displays a list of active parts - parts that are not overlapped by any other part.
-# Usage: `ls /var/lib/clickhouse/data/merge/visits | active_parts.py`
-
-import sys
-import re
-
-parts = {}
-for s in sys.stdin.read().split():
-    m = re.match(
-        "^([0-9]{6})[0-9]{2}_([0-9]{6})[0-9]{2}_([0-9]+)_([0-9]+)_([0-9]+)$", s
-    )
-    if m == None:
-        continue
-    m1 = m.group(1)
-    m2 = m.group(2)
-    i1 = int(m.group(3))
-    i2 = int(m.group(4))
-    l = int(m.group(5))
-    if m1 != m2:
-        raise Exception("not in single month: " + s)
-    if m1 not in parts:
-        parts[m1] = []
-    parts[m1].append((i1, i2, l, s))
-
-for m, ps in sorted(parts.items()):
-    ps.sort(key=lambda i1_i2_l_s: (i1_i2_l_s[0], -i1_i2_l_s[1], -i1_i2_l_s[2]))
-    (x2, y2, l2, s2) = (-1, -1, -1, -1)
-    for x1, y1, l1, s1 in ps:
-        if x1 >= x2 and y1 <= y2 and l1 < l2 and (x1, y1) != (x2, y2):  # 2 contains 1
-            pass
-        elif x1 > y2:  # 1 is to the right of 2
-            if x1 != y2 + 1 and y2 != -1:
-                print()  # to see the missing numbers
-            (x2, y2, l2, s2) = (x1, y1, l1, s1)
-            print(s1)
-        else:
-            raise Exception("invalid parts intersection: " + s1 + " and " + s2)
-    print()
diff --git a/src/Storages/examples/async_read_buffer_from_hdfs.cpp b/src/Storages/examples/async_read_buffer_from_hdfs.cpp
deleted file mode 100644
index 4f6aed8ef65..00000000000
--- a/src/Storages/examples/async_read_buffer_from_hdfs.cpp
+++ /dev/null
@@ -1,37 +0,0 @@
-#include <memory>
-#include <string>
-#include <filesystem>
-
-#include <IO/WriteHelpers.h>
-#include <IO/copyData.h>
-#include <IO/WriteBufferFromString.h>
-#include <Disks/ObjectStorages/IObjectStorage.h>
-#include <Disks/IO/getThreadPoolReader.h>
-#include <Interpreters/Context.h>
-#include <Common/Config/ConfigProcessor.h>
-#include <Storages/HDFS/AsynchronousReadBufferFromHDFS.h>
-
-int main()
-{
-    using namespace DB;
-    namespace fs = std::filesystem;
-
-    String config_path = "/path/to/config/file";
-    ConfigProcessor config_processor(config_path, false, true);
-    config_processor.setConfigPath(fs::path(config_path).parent_path());
-    auto loaded_config = config_processor.loadConfig(false);
-    auto * config = loaded_config.configuration.duplicate();
-
-    String hdfs_namenode_url = "hdfs://namenode:port/";
-    String path = "/path/to/hdfs/file";
-    ReadSettings settings = {};
-    auto in = std::make_unique<ReadBufferFromHDFS>(hdfs_namenode_url, path, *config, settings);
-    auto & reader = getThreadPoolReader(FilesystemReaderType::ASYNCHRONOUS_REMOTE_FS_READER);
-    AsynchronousReadBufferFromHDFS buf(reader, {}, std::move(in));
-
-    String output;
-    WriteBufferFromString out(output);
-    copyData(buf, out);
-    std::cout << "output:" << output << std::endl;
-    return 0;
-}
diff --git a/src/Storages/examples/get_abandonable_lock_in_all_partitions.cpp b/src/Storages/examples/get_abandonable_lock_in_all_partitions.cpp
deleted file mode 100644
index 4607d68f02d..00000000000
--- a/src/Storages/examples/get_abandonable_lock_in_all_partitions.cpp
+++ /dev/null
@@ -1,71 +0,0 @@
-#include <Common/Config/ConfigProcessor.h>
-#include <Common/ZooKeeper/ZooKeeper.h>
-#include <Common/Exception.h>
-#include <Common/Stopwatch.h>
-#include <Storages/MergeTree/EphemeralLockInZooKeeper.h>
-
-#include <base/scope_guard.h>
-
-#include <iostream>
-
-
-using namespace DB;
-
-/// This test is useful for assessing the performance of acquiring block numbers in all partitions (and there
-/// can be ~1000 of them). This is needed when creating a mutation entry for a ReplicatedMergeTree table.
-int main(int argc, char ** argv)
-try
-{
-    if (argc != 3)
-    {
-        std::cerr << "usage: " << argv[0] << " <zookeeper_config> <path_to_table>" << std::endl;
-        return 3;
-    }
-
-    ConfigProcessor processor(argv[1], false, true);
-    auto config = processor.loadConfig().configuration;
-    String root_path = argv[2];
-
-    auto zk = zkutil::ZooKeeper::createWithoutKillingPreviousSessions(*config, zkutil::getZooKeeperConfigName(*config), nullptr);
-
-    String temp_path = root_path + "/temp";
-    String blocks_path = root_path + "/block_numbers";
-
-    Stopwatch total_timer;
-    Stopwatch timer;
-
-    EphemeralLocksInAllPartitions locks(blocks_path, "test_lock-", temp_path, *zk);
-
-    std::cerr << "Locked, elapsed: " << timer.elapsedSeconds() << std::endl;
-    for (const auto & lock : locks.getLocks())
-        std::cout << lock.partition_id << " " << lock.number << std::endl;
-    timer.restart();
-
-    locks.unlock();
-    std::cerr << "Abandoned, elapsed: " << timer.elapsedSeconds() << std::endl;
-
-    std::cerr << "Total elapsed: " << total_timer.elapsedSeconds() << std::endl;
-
-    return 0;
-}
-catch (const Exception & e)
-{
-    std::cerr << e.what() << ", " << e.displayText() << ": " << std::endl
-              << e.getStackTraceString() << std::endl;
-    throw;
-}
-catch (Poco::Exception & e)
-{
-    std::cerr << "Exception: " << e.displayText() << std::endl;
-    throw;
-}
-catch (std::exception & e)
-{
-    std::cerr << "std::exception: " << e.what() << std::endl;
-    throw;
-}
-catch (...)
-{
-    std::cerr << "Some exception" << std::endl;
-    throw;
-}

From dbb6dad921be5ee24b222d29f375d31aa70028f8 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 10 Mar 2024 10:04:40 +0100
Subject: [PATCH 123/443] Fix style

---
 src/Storages/MergeTree/DataPartsExchange.cpp | 2 --
 1 file changed, 2 deletions(-)

diff --git a/src/Storages/MergeTree/DataPartsExchange.cpp b/src/Storages/MergeTree/DataPartsExchange.cpp
index cc2967003a4..296c11f4673 100644
--- a/src/Storages/MergeTree/DataPartsExchange.cpp
+++ b/src/Storages/MergeTree/DataPartsExchange.cpp
@@ -43,10 +43,8 @@ namespace ErrorCodes
     extern const int CANNOT_WRITE_TO_OSTREAM;
     extern const int CHECKSUM_DOESNT_MATCH;
     extern const int INSECURE_PATH;
-    extern const int CORRUPTED_DATA;
     extern const int LOGICAL_ERROR;
     extern const int S3_ERROR;
-    extern const int INCORRECT_PART_TYPE;
     extern const int ZERO_COPY_REPLICATION_ERROR;
 }
 

From d340d9cfaf58e63392c3725c55fb9e447bda6185 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 10 Mar 2024 10:05:07 +0100
Subject: [PATCH 124/443] Fix build

---
 src/Storages/MergeTree/CMakeLists.txt | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/src/Storages/MergeTree/CMakeLists.txt b/src/Storages/MergeTree/CMakeLists.txt
index 390835f17ae..e69de29bb2d 100644
--- a/src/Storages/MergeTree/CMakeLists.txt
+++ b/src/Storages/MergeTree/CMakeLists.txt
@@ -1,3 +0,0 @@
-if(ENABLE_EXAMPLES)
-    add_subdirectory(examples)
-endif()

From 6a0eb7a3cdc01f0ea827228b5e21be65951bc07b Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 10 Mar 2024 14:42:58 +0300
Subject: [PATCH 125/443] Update ColumnSizeEstimator.h

---
 src/Storages/MergeTree/ColumnSizeEstimator.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Storages/MergeTree/ColumnSizeEstimator.h b/src/Storages/MergeTree/ColumnSizeEstimator.h
index a9c4f75ed6b..e512415ff05 100644
--- a/src/Storages/MergeTree/ColumnSizeEstimator.h
+++ b/src/Storages/MergeTree/ColumnSizeEstimator.h
@@ -9,7 +9,7 @@ namespace DB
 /* Allow to compute more accurate progress statistics */
 class ColumnSizeEstimator
 {
-    using ColumnToSize = std::map<String, size_t>;
+    using ColumnToSize = std::map<String, UInt64>;
     ColumnToSize map;
 public:
 

From 1d8b95bdcd9c0b63c615d42b6242fa5dd0d1314c Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 10 Mar 2024 14:43:18 +0300
Subject: [PATCH 126/443] Update MergeTask.cpp

---
 src/Storages/MergeTree/MergeTask.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Storages/MergeTree/MergeTask.cpp b/src/Storages/MergeTree/MergeTask.cpp
index 77bc680eae2..5587e3884c8 100644
--- a/src/Storages/MergeTree/MergeTask.cpp
+++ b/src/Storages/MergeTree/MergeTask.cpp
@@ -309,7 +309,7 @@ bool MergeTask::ExecuteAndFinalizeHorizontalPart::prepare()
             ctx->rows_sources_uncompressed_write_buf = ctx->tmp_disk->createRawStream();
             ctx->rows_sources_write_buf = std::make_unique<CompressedWriteBuffer>(*ctx->rows_sources_uncompressed_write_buf);
 
-            std::map<String, size_t> local_merged_column_to_size;
+            std::map<String, UInt64> local_merged_column_to_size;
             for (const MergeTreeData::DataPartPtr & part : global_ctx->future_part->parts)
                 part->accumulateColumnSizes(local_merged_column_to_size);
 

From 2cc071e0e4c7486368e9dca8b2b68c0e1ed2af11 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 10 Mar 2024 12:48:15 +0100
Subject: [PATCH 127/443] Fix error

---
 docker/test/fuzzer/run-fuzzer.sh              |  2 +-
 src/Core/Field.h                              |  6 +++---
 src/Core/Types.h                              | 15 ++++++---------
 .../MergeTree/MergeTreeDataSelectExecutor.cpp | 11 ++++++-----
 .../0_stateless/03008_index_small.reference   |  2 ++
 .../queries/0_stateless/03008_index_small.sql | 19 +++++++++++++++++++
 6 files changed, 37 insertions(+), 18 deletions(-)
 create mode 100644 tests/queries/0_stateless/03008_index_small.reference
 create mode 100644 tests/queries/0_stateless/03008_index_small.sql

diff --git a/docker/test/fuzzer/run-fuzzer.sh b/docker/test/fuzzer/run-fuzzer.sh
index dc1b35b3a21..63471c288dd 100755
--- a/docker/test/fuzzer/run-fuzzer.sh
+++ b/docker/test/fuzzer/run-fuzzer.sh
@@ -343,7 +343,7 @@ quit
         # which is confusing.
         task_exit_code=$fuzzer_exit_code
         echo "failure" > status.txt
-        echo "Achtung!" > description.txt
+        echo "Let op!" > description.txt
         echo "Fuzzer went wrong with error code: ($fuzzer_exit_code). Its process died somehow when the server stayed alive. The server log probably won't tell you much so try to find information in other files." >>description.txt
         { rg -ao "Found error:.*" fuzzer.log || rg -ao "Exception:.*" fuzzer.log; } | tail -1 >>description.txt
     fi
diff --git a/src/Core/Field.h b/src/Core/Field.h
index 414874310a6..43f175cb476 100644
--- a/src/Core/Field.h
+++ b/src/Core/Field.h
@@ -497,7 +497,7 @@ public:
 
         switch (which)
         {
-            case Types::Null:    return false;
+            case Types::Null:    return get<Null>() < rhs.get<Null>();
             case Types::Bool:    [[fallthrough]];
             case Types::UInt64:  return get<UInt64>()  < rhs.get<UInt64>();
             case Types::UInt128: return get<UInt128>() < rhs.get<UInt128>();
@@ -541,7 +541,7 @@ public:
 
         switch (which)
         {
-            case Types::Null:    return true;
+            case Types::Null:    return get<Null>() <= rhs.get<Null>();
             case Types::Bool: [[fallthrough]];
             case Types::UInt64:  return get<UInt64>()  <= rhs.get<UInt64>();
             case Types::UInt128: return get<UInt128>() <= rhs.get<UInt128>();
@@ -590,7 +590,7 @@ public:
 
         switch (which)
         {
-            case Types::Null: return true;
+            case Types::Null: return get<Null>() == rhs.get<Null>();
             case Types::Bool: [[fallthrough]];
             case Types::UInt64: return get<UInt64>() == rhs.get<UInt64>();
             case Types::Int64:   return get<Int64>() == rhs.get<Int64>();
diff --git a/src/Core/Types.h b/src/Core/Types.h
index 74e18e9494d..2a930d09873 100644
--- a/src/Core/Types.h
+++ b/src/Core/Types.h
@@ -23,9 +23,9 @@ struct Null
 {
     enum class Value
     {
-        Null,
-        PositiveInfinity,
-        NegativeInfinity,
+        NegativeInfinity = -1,
+        Null = 0,
+        PositiveInfinity = 1,
     };
 
     Value value{Value::Null};
@@ -34,15 +34,12 @@ struct Null
     bool isPositiveInfinity() const { return value == Value::PositiveInfinity; }
     bool isNegativeInfinity() const { return value == Value::NegativeInfinity; }
 
-    bool operator==(const Null & other) const
+    auto operator<=>(const Null & other) const
     {
-        return value == other.value;
+        return static_cast<int>(value) <=> static_cast<int>(other.value);
     }
 
-    bool operator!=(const Null & other) const
-    {
-        return !(*this == other);
-    }
+    bool operator==(const Null &) const = default;
 };
 
 using UInt128 = ::UInt128;
diff --git a/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp b/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp
index 93b9ad60e3c..ef679b61a79 100644
--- a/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp
+++ b/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp
@@ -1130,6 +1130,7 @@ MarkRanges MergeTreeDataSelectExecutor::markRangesFromPKRange(
     bool part_offset_condition_exact_range
         = !part_offset_condition || part_offset_condition->alwaysUnknownOrTrue() || part_offset_condition->matchesExactContinuousRange();
     const String & part_name = part->isProjectionPart() ? fmt::format("{}.{}", part->name, part->getParentPart()->name) : part->name;
+
     if (!key_condition_exact_range || !part_offset_condition_exact_range)
     {
         // Do exclusion search, where we drop ranges that do not match
@@ -1144,10 +1145,10 @@ MarkRanges MergeTreeDataSelectExecutor::markRangesFromPKRange(
             part->index_granularity_info.index_granularity_bytes);
 
         /** There will always be disjoint suspicious segments on the stack, the leftmost one at the top (back).
-        * At each step, take the left segment and check if it fits.
-        * If fits, split it into smaller ones and put them on the stack. If not, discard it.
-        * If the segment is already of one mark length, add it to response and discard it.
-        */
+          * At each step, take the left segment and check if it fits.
+          * If fits, split it into smaller ones and put them on the stack. If not, discard it.
+          * If the segment is already of one mark length, add it to response and discard it.
+          */
         std::vector<MarkRange> ranges_stack = { {0, marks_count} };
 
         size_t steps = 0;
@@ -1157,7 +1158,7 @@ MarkRanges MergeTreeDataSelectExecutor::markRangesFromPKRange(
             MarkRange range = ranges_stack.back();
             ranges_stack.pop_back();
 
-            steps++;
+            ++steps;
 
             if (!may_be_true_in_range(range))
                 continue;
diff --git a/tests/queries/0_stateless/03008_index_small.reference b/tests/queries/0_stateless/03008_index_small.reference
new file mode 100644
index 00000000000..a5c8806279f
--- /dev/null
+++ b/tests/queries/0_stateless/03008_index_small.reference
@@ -0,0 +1,2 @@
+3
+3
diff --git a/tests/queries/0_stateless/03008_index_small.sql b/tests/queries/0_stateless/03008_index_small.sql
new file mode 100644
index 00000000000..4eb2db3abc3
--- /dev/null
+++ b/tests/queries/0_stateless/03008_index_small.sql
@@ -0,0 +1,19 @@
+DROP TABLE IF EXISTS test;
+
+CREATE TABLE test (a UInt8, b UInt8) ENGINE = MergeTree ORDER BY (a, b)
+SETTINGS index_granularity = 1, primary_key_ratio_of_unique_prefix_values_to_skip_suffix_columns = 0.01;
+
+SET optimize_move_to_prewhere = 0;
+
+INSERT INTO test
+SELECT number DIV 2, number
+FROM numbers(3);
+
+SELECT count() FROM test WHERE b >= 0;
+
+DETACH TABLE test;
+ATTACH TABLE test;
+
+SELECT count() FROM test WHERE b >= 0;
+
+-- DROP TABLE test;

From adf056b54ad35638245c1a5c30bc57c4cb17fc6c Mon Sep 17 00:00:00 2001
From: Dan Wu <wudanzy@google.com>
Date: Sun, 10 Mar 2024 20:42:41 +0800
Subject: [PATCH 128/443] Update check-large-objects.sh to be language neutral

The previous implementation uses `total` keyword to filter out lines that contains the total size of objects under a directory. But when the OS uses other language, it fails to filter out that line, so the script would fail.

This implementation changes the script to match only lines that contains object file by their extension. This implementation would be language neutral.
---
 utils/check-style/check-large-objects.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/utils/check-style/check-large-objects.sh b/utils/check-style/check-large-objects.sh
index 5c1276e5732..04f4e0c3171 100755
--- a/utils/check-style/check-large-objects.sh
+++ b/utils/check-style/check-large-objects.sh
@@ -12,7 +12,7 @@ TU_EXCLUDES=(
     Aggregator
 )
 
-if find $1 -name '*.o' | xargs wc -c | grep -v total | sort -rn | awk '{ if ($1 > 50000000) print }' \
+if find $1 -name '*.o' | xargs wc -c | grep --regexp='.o$' | sort -rn | awk '{ if ($1 > 50000000) print }' \
     | grep -v -f <(printf "%s\n" "${TU_EXCLUDES[@]}")
 then
     echo "^ It's not allowed to have so large translation units."

From 349894fc2b73fad70910c325e0de053a56ec1bc7 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Fri, 8 Mar 2024 11:27:09 +0000
Subject: [PATCH 129/443] Fixes #61051

---
 docs/en/operations/settings/settings.md       | 12 +++++
 .../functions/string-search-functions.md      | 18 ++++++-
 src/Core/Settings.h                           |  1 +
 src/Core/SettingsChangesHistory.h             |  1 +
 src/Functions/FunctionsStringSearch.h         | 53 ++++++++++++++-----
 .../array/FunctionsMapMiscellaneous.cpp       |  1 +
 src/Functions/like.cpp                        |  2 +-
 src/Functions/locate.cpp                      | 34 ++++++++++++
 src/Functions/position.cpp                    |  1 -
 .../0_stateless/00765_locate.reference        |  7 +++
 tests/queries/0_stateless/00765_locate.sql    | 15 ++++++
 .../00765_sql_compatibility_aliases.reference |  1 -
 .../00765_sql_compatibility_aliases.sql       |  1 -
 ..._case_insensitive_function_names.reference |  1 -
 ...malize_case_insensitive_function_names.sql |  2 +-
 15 files changed, 130 insertions(+), 20 deletions(-)
 create mode 100644 src/Functions/locate.cpp
 create mode 100644 tests/queries/0_stateless/00765_locate.reference
 create mode 100644 tests/queries/0_stateless/00765_locate.sql

diff --git a/docs/en/operations/settings/settings.md b/docs/en/operations/settings/settings.md
index 622644a1543..e75435669fb 100644
--- a/docs/en/operations/settings/settings.md
+++ b/docs/en/operations/settings/settings.md
@@ -4336,6 +4336,18 @@ Possible values:
 
 Default value: `0`.
 
+
+## function_locate_has_mysql_compatible_argument_order {#function-locate-has-mysql-compatible-argument-order}
+
+Controls the order of arguments in function [locate](../../sql-reference/functions/string-search-functions.md#locate).
+
+Possible values:
+
+- 0 — Function `locate` accepts arguments `(haystack, needle[, start_pos])`.
+- 1 — Function `locate` accepts arguments `(needle, haystack, [, start_pos])` (MySQL-compatible behavior)
+
+Default value: `1`.
+
 ## date_time_overflow_behavior {#date_time_overflow_behavior}
 
 Defines the behavior when [Date](../../sql-reference/data-types/date.md), [Date32](../../sql-reference/data-types/date32.md), [DateTime](../../sql-reference/data-types/datetime.md), [DateTime64](../../sql-reference/data-types/datetime64.md) or integers are converted into Date, Date32, DateTime or DateTime64 but the value cannot be represented in the result type.
diff --git a/docs/en/sql-reference/functions/string-search-functions.md b/docs/en/sql-reference/functions/string-search-functions.md
index 22f879c62ae..1b03f220db2 100644
--- a/docs/en/sql-reference/functions/string-search-functions.md
+++ b/docs/en/sql-reference/functions/string-search-functions.md
@@ -30,7 +30,6 @@ position(haystack, needle[, start_pos])
 
 Alias:
 - `position(needle IN haystack)`
-- `locate(haystack, needle[, start_pos])`.
 
 **Arguments**
 
@@ -49,7 +48,7 @@ If substring `needle` is empty, these rules apply:
 - if `start_pos >= 1` and `start_pos <= length(haystack) + 1`: return `start_pos`
 - otherwise: return `0`
 
-The same rules also apply to functions `positionCaseInsensitive`, `positionUTF8` and `positionCaseInsensitiveUTF8`
+The same rules also apply to functions `locate`, `positionCaseInsensitive`, `positionUTF8` and `positionCaseInsensitiveUTF8`.
 
 Type: `Integer`.
 
@@ -114,6 +113,21 @@ SELECT
 └─────────────────────┴────────────────────────┴────────────────────────┴────────────────────────┴────────────────────────┴────────────────────────┴────────────────────────┘
 ```
 
+## locate
+
+Like [position](#position) but with arguments `haystack` and `locate` switched.
+
+The behavior of this function depends on the ClickHouse version:
+- in versions < v24.3, `locate` was an alias of function `position` and accepted arguments `(haystack, needle[, start_pos])`.
+- in versions >= 24.3,, `locate` is an individual function (for better compatibility with MySQL) and accepts arguments `(needle, haystack[, start_pos])`. The previous behavior
+  can be restored using setting [function_locate_has_mysql_compatible_argument_order = false](../../operations/settings/settings.md#function-locate-has-mysql-compatible-argument-order);
+
+**Syntax**
+
+``` sql
+locate(needle, haystack[, start_pos])
+```
+
 ## positionCaseInsensitive
 
 Like [position](#position) but searches case-insensitively.
diff --git a/src/Core/Settings.h b/src/Core/Settings.h
index c8bdb515baf..185fa99e062 100644
--- a/src/Core/Settings.h
+++ b/src/Core/Settings.h
@@ -175,6 +175,7 @@ class IColumn;
     M(Bool, enable_positional_arguments, true, "Enable positional arguments in ORDER BY, GROUP BY and LIMIT BY", 0) \
     M(Bool, enable_extended_results_for_datetime_functions, false, "Enable date functions like toLastDayOfMonth return Date32 results (instead of Date results) for Date32/DateTime64 arguments.", 0) \
     M(Bool, allow_nonconst_timezone_arguments, false, "Allow non-const timezone arguments in certain time-related functions like toTimeZone(), fromUnixTimestamp*(), snowflakeToDateTime*()", 0) \
+    M(Bool, function_locate_has_mysql_compatible_argument_order, true, "Function locate() has arguments (needle, haystack[, start_pos]) like in MySQL instead of (haystack, needle[, start_pos]) like function position()", 0) \
     \
     M(Bool, group_by_use_nulls, false, "Treat columns mentioned in ROLLUP, CUBE or GROUPING SETS as Nullable", 0) \
     \
diff --git a/src/Core/SettingsChangesHistory.h b/src/Core/SettingsChangesHistory.h
index 2f1da7935e6..1c5e8ab3818 100644
--- a/src/Core/SettingsChangesHistory.h
+++ b/src/Core/SettingsChangesHistory.h
@@ -92,6 +92,7 @@ static std::map<ClickHouseVersion, SettingsChangesHistory::SettingsChanges> sett
               {"input_format_json_use_string_type_for_ambiguous_paths_in_named_tuples_inference_from_objects", false, false, "Allow to use String type for ambiguous paths during named tuple inference from JSON objects"},
               {"throw_if_deduplication_in_dependent_materialized_views_enabled_with_async_insert", false, true, "Deduplication is dependent materialized view cannot work together with async inserts."},
               {"parallel_replicas_allow_in_with_subquery", false, true, "If true, subquery for IN will be executed on every follower replica"},
+              {"function_locate_has_mysql_compatible_argument_order", false, true, "Increase compatibility with MySQL's locate function."},
               }},
     {"24.2", {{"allow_suspicious_variant_types", true, false, "Don't allow creating Variant type with suspicious variants by default"},
               {"validate_experimental_and_suspicious_types_inside_nested_types", false, true, "Validate usage of experimental and suspicious types inside nested types"},
diff --git a/src/Functions/FunctionsStringSearch.h b/src/Functions/FunctionsStringSearch.h
index 41b476ccc56..53d99198134 100644
--- a/src/Functions/FunctionsStringSearch.h
+++ b/src/Functions/FunctionsStringSearch.h
@@ -22,13 +22,13 @@ namespace DB
   * positionCaseInsensitive(haystack, needle)
   * positionCaseInsensitiveUTF8(haystack, needle)
   *
-  * like(haystack, pattern)        - search by the regular expression LIKE; Returns 0 or 1. Case-insensitive, but only for Latin.
-  * notLike(haystack, pattern)
+  * like(haystack, needle)        - search by the regular expression LIKE; Returns 0 or 1. Case-insensitive, but only for Latin.
+  * notLike(haystack, needle)
   *
-  * ilike(haystack, pattern) - like 'like' but case-insensitive
-  * notIlike(haystack, pattern)
+  * ilike(haystack, needle) - like 'like' but case-insensitive
+  * notIlike(haystack, needle)
   *
-  * match(haystack, pattern)       - search by regular expression re2; Returns 0 or 1.
+  * match(haystack, needle)       - search by regular expression re2; Returns 0 or 1.
   *
   * countSubstrings(haystack, needle) -- count number of occurrences of needle in haystack.
   * countSubstringsCaseInsensitive(haystack, needle)
@@ -53,7 +53,7 @@ namespace DB
   * - the first subpattern, if the regexp has a subpattern;
   * - the zero subpattern (the match part, otherwise);
   * - if not match - an empty string.
-  * extract(haystack, pattern)
+  * extract(haystack, needle)
   */
 
 namespace ErrorCodes
@@ -69,13 +69,39 @@ enum class ExecutionErrorPolicy
     Throw
 };
 
-template <typename Impl, ExecutionErrorPolicy execution_error_policy = ExecutionErrorPolicy::Throw>
+enum class HaystackNeedleOrderIsConfigurable
+{
+    No,     /// function arguments are always: (haystack, needle[, position])
+    Yes     /// depending on a setting, the function arguments are (haystack, needle[, position]) or (needle, haystack[, position])
+};
+
+template <typename Impl,
+         ExecutionErrorPolicy execution_error_policy = ExecutionErrorPolicy::Throw,
+         HaystackNeedleOrderIsConfigurable haystack_needle_order_is_configurable = HaystackNeedleOrderIsConfigurable::No>
 class FunctionsStringSearch : public IFunction
 {
+private:
+    enum class ArgumentOrder
+    {
+        HaystackNeedle,
+        NeedleHaystack
+    };
+
+    ArgumentOrder argument_order = ArgumentOrder::HaystackNeedle;
+
 public:
     static constexpr auto name = Impl::name;
 
-    static FunctionPtr create(ContextPtr) { return std::make_shared<FunctionsStringSearch>(); }
+    static FunctionPtr create(ContextPtr context) { return std::make_shared<FunctionsStringSearch>(context); }
+
+    explicit FunctionsStringSearch([[maybe_unused]] ContextPtr context)
+    {
+        if constexpr (haystack_needle_order_is_configurable == HaystackNeedleOrderIsConfigurable::Yes)
+        {
+            if (context->getSettingsRef().function_locate_has_mysql_compatible_argument_order)
+                argument_order = ArgumentOrder::NeedleHaystack;
+        }
+    }
 
     String getName() const override { return name; }
 
@@ -105,13 +131,16 @@ public:
                 "Number of arguments for function {} doesn't match: passed {}, should be 2 or 3",
                 getName(), arguments.size());
 
-        if (!isStringOrFixedString(arguments[0]))
+        const auto & haystack_type = (argument_order == ArgumentOrder::HaystackNeedle) ? arguments[0] : arguments[1];
+        const auto & needle_type = (argument_order == ArgumentOrder::HaystackNeedle) ? arguments[1] : arguments[0];
+
+        if (!isStringOrFixedString(haystack_type))
             throw Exception(
                 ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT,
                 "Illegal type {} of argument of function {}",
                 arguments[0]->getName(), getName());
 
-        if (!isString(arguments[1]))
+        if (!isString(needle_type))
             throw Exception(
                 ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT,
                 "Illegal type {} of argument of function {}",
@@ -135,8 +164,8 @@ public:
 
     ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t /*input_rows_count*/) const override
     {
-        const ColumnPtr & column_haystack = arguments[0].column;
-        const ColumnPtr & column_needle = arguments[1].column;
+        const ColumnPtr & column_haystack = (argument_order == ArgumentOrder::HaystackNeedle) ? arguments[0].column : arguments[1].column;
+        const ColumnPtr & column_needle = (argument_order == ArgumentOrder::HaystackNeedle) ? arguments[1].column : arguments[0].column;
 
         ColumnPtr column_start_pos = nullptr;
         if (arguments.size() >= 3)
diff --git a/src/Functions/array/FunctionsMapMiscellaneous.cpp b/src/Functions/array/FunctionsMapMiscellaneous.cpp
index 157f2fa8a26..d92bfcf0bc6 100644
--- a/src/Functions/array/FunctionsMapMiscellaneous.cpp
+++ b/src/Functions/array/FunctionsMapMiscellaneous.cpp
@@ -213,6 +213,7 @@ struct MapToSubcolumnAdapter : public MapAdapterBase<MapToSubcolumnAdapter<Name,
 class FunctionMapKeyLike : public IFunction
 {
 public:
+    FunctionMapKeyLike() : impl(/*context*/ nullptr) {} /// nullptr because getting a context here is hard and FunctionLike doesn't need context
     String getName() const override { return "mapKeyLike"; }
     size_t getNumberOfArguments() const override { return 3; }
     bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return false; }
diff --git a/src/Functions/like.cpp b/src/Functions/like.cpp
index 5a86e37a92d..de907380265 100644
--- a/src/Functions/like.cpp
+++ b/src/Functions/like.cpp
@@ -1,5 +1,5 @@
-#include "FunctionFactory.h"
 #include "like.h"
+#include "FunctionFactory.h"
 
 
 namespace DB
diff --git a/src/Functions/locate.cpp b/src/Functions/locate.cpp
new file mode 100644
index 00000000000..9a70fbb2d72
--- /dev/null
+++ b/src/Functions/locate.cpp
@@ -0,0 +1,34 @@
+#include "FunctionsStringSearch.h"
+#include "FunctionFactory.h"
+#include "PositionImpl.h"
+
+
+namespace DB
+{
+namespace
+{
+
+struct NameLocate
+{
+    static constexpr auto name = "locate";
+};
+
+using FunctionLocate = FunctionsStringSearch<PositionImpl<NameLocate, PositionCaseSensitiveASCII>, ExecutionErrorPolicy::Throw, HaystackNeedleOrderIsConfigurable::Yes>;
+
+}
+
+REGISTER_FUNCTION(Locate)
+{
+    FunctionDocumentation::Description doc_description = "Like function `position` but with arguments `haystack` and `locate` switched. The behavior of this function depends on the ClickHouse version: In versions < v24.3, `locate` was an alias of function `position` and accepted arguments `(haystack, needle[, start_pos])`. In versions >= 24.3,, `locate` is an individual function (for better compatibility with MySQL) and accepts arguments `(needle, haystack[, start_pos])`. The previous behaviorcan be restored using setting `function_locate_has_mysql_compatible_argument_order = false`.";
+    FunctionDocumentation::Syntax doc_syntax = "location(needle, haystack[, start_pos])";
+    FunctionDocumentation::Arguments doc_arguments = {{"needle", "Substring to be searched (String)"},
+                                                      {"haystack", "String in which the search is performed (String)."},
+                                                      {"start_pos", "Position (1-based) in `haystack` at which the search starts (UInt*)."}};
+    FunctionDocumentation::ReturnedValue doc_returned_value = "Starting position in bytes and counting from 1, if the substring was found. 0, if the substring was not found.";
+    FunctionDocumentation::Examples doc_examples = {{"Example", "SELECT locate('abcabc', 'ca');", "3"}};
+    FunctionDocumentation::Categories doc_categories = {"String search"};
+
+
+    factory.registerFunction<FunctionLocate>({doc_description, doc_syntax, doc_arguments, doc_returned_value, doc_examples, doc_categories}, FunctionFactory::CaseInsensitive);
+}
+}
diff --git a/src/Functions/position.cpp b/src/Functions/position.cpp
index 409a593b44c..29a5db2eb24 100644
--- a/src/Functions/position.cpp
+++ b/src/Functions/position.cpp
@@ -20,6 +20,5 @@ using FunctionPosition = FunctionsStringSearch<PositionImpl<NamePosition, Positi
 REGISTER_FUNCTION(Position)
 {
     factory.registerFunction<FunctionPosition>({}, FunctionFactory::CaseInsensitive);
-    factory.registerAlias("locate", NamePosition::name, FunctionFactory::CaseInsensitive);
 }
 }
diff --git a/tests/queries/0_stateless/00765_locate.reference b/tests/queries/0_stateless/00765_locate.reference
new file mode 100644
index 00000000000..3b066c0d68b
--- /dev/null
+++ b/tests/queries/0_stateless/00765_locate.reference
@@ -0,0 +1,7 @@
+-- negative tests
+-- test mysql compatibility setting
+0
+0
+3
+-- the function name needs to be case-insensitive for historical reasons
+0
diff --git a/tests/queries/0_stateless/00765_locate.sql b/tests/queries/0_stateless/00765_locate.sql
new file mode 100644
index 00000000000..3467ebd4249
--- /dev/null
+++ b/tests/queries/0_stateless/00765_locate.sql
@@ -0,0 +1,15 @@
+SET send_logs_level = 'fatal';
+
+SELECT '-- negative tests';
+SELECT locate(); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH }
+SELECT locate(1, 'abc'); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT }
+SELECT locate('abc', 1); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT }
+SELECT locate('abc', 'abc', 'abc'); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT }
+
+SELECT '-- test mysql compatibility setting';
+SELECT locate('abcabc', 'ca');
+SELECT locate('abcabc', 'ca') SETTINGS function_locate_has_mysql_compatible_argument_order = true;
+SELECT locate('abcabc', 'ca') SETTINGS function_locate_has_mysql_compatible_argument_order = false;
+
+SELECT '-- the function name needs to be case-insensitive for historical reasons';
+SELECT LoCaTe('abcabc', 'ca');
diff --git a/tests/queries/0_stateless/00765_sql_compatibility_aliases.reference b/tests/queries/0_stateless/00765_sql_compatibility_aliases.reference
index 285b9a62d20..6d31168c2b7 100644
--- a/tests/queries/0_stateless/00765_sql_compatibility_aliases.reference
+++ b/tests/queries/0_stateless/00765_sql_compatibility_aliases.reference
@@ -4,7 +4,6 @@ foo
 FOO
 baz
 zzz
-2
 fo
 oo
 o
diff --git a/tests/queries/0_stateless/00765_sql_compatibility_aliases.sql b/tests/queries/0_stateless/00765_sql_compatibility_aliases.sql
index da0eb9bea6d..995aaef9ea5 100644
--- a/tests/queries/0_stateless/00765_sql_compatibility_aliases.sql
+++ b/tests/queries/0_stateless/00765_sql_compatibility_aliases.sql
@@ -6,7 +6,6 @@ select LOWER('Foo');
 select UPPER('Foo');
 select REPLACE('bar', 'r', 'z');
 select REGEXP_REPLACE('bar', '.', 'z');
-select Locate('foo', 'o');
 select SUBSTRING('foo', 1, 2);
 select Substr('foo', 2);
 select mid('foo', 3);
diff --git a/tests/queries/0_stateless/01705_normalize_case_insensitive_function_names.reference b/tests/queries/0_stateless/01705_normalize_case_insensitive_function_names.reference
index de5a62159ef..76dd04b0e42 100644
--- a/tests/queries/0_stateless/01705_normalize_case_insensitive_function_names.reference
+++ b/tests/queries/0_stateless/01705_normalize_case_insensitive_function_names.reference
@@ -26,7 +26,6 @@ SELECT
     least(1),
     length('1'),
     log(1),
-    position('1', '1'),
     log(1),
     log10(1),
     log2(1),
diff --git a/tests/queries/0_stateless/01705_normalize_case_insensitive_function_names.sql b/tests/queries/0_stateless/01705_normalize_case_insensitive_function_names.sql
index dda2e045e76..d8054b9757c 100644
--- a/tests/queries/0_stateless/01705_normalize_case_insensitive_function_names.sql
+++ b/tests/queries/0_stateless/01705_normalize_case_insensitive_function_names.sql
@@ -1 +1 @@
-EXPLAIN SYNTAX SELECT CAST(1 AS INT), CEIL(1), CEILING(1), CHAR(49), CHAR_LENGTH('1'), CHARACTER_LENGTH('1'), COALESCE(1), CONCAT('1', '1'), CORR(1, 1), COS(1), COUNT(1), COVAR_POP(1, 1), COVAR_SAMP(1, 1), DATABASE(), SCHEMA(), DATEDIFF('DAY', toDate('2020-10-24'), toDate('2019-10-24')), EXP(1), FLATTEN([[1]]), FLOOR(1), FQDN(), GREATEST(1), IF(1, 1, 1), IFNULL(1, 1), LCASE('A'), LEAST(1), LENGTH('1'), LN(1), LOCATE('1', '1'), LOG(1), LOG10(1), LOG2(1), LOWER('A'), MAX(1), MID('123', 1, 1), MIN(1), MOD(1, 1), NOT(1), NOW(), NOW64(), NULLIF(1, 1), PI(), POSITION('123', '2'), POW(1, 1), POWER(1, 1), RAND(), REPLACE('1', '1', '2'), REVERSE('123'), ROUND(1), SIN(1), SQRT(1), STDDEV_POP(1), STDDEV_SAMP(1), SUBSTR('123', 2), SUBSTRING('123', 2), SUM(1), TAN(1), TANH(1), TRUNC(1), TRUNCATE(1), UCASE('A'), UPPER('A'), USER(), VAR_POP(1), VAR_SAMP(1), WEEK(toDate('2020-10-24')), YEARWEEK(toDate('2020-10-24')) format TSVRaw;
+EXPLAIN SYNTAX SELECT CAST(1 AS INT), CEIL(1), CEILING(1), CHAR(49), CHAR_LENGTH('1'), CHARACTER_LENGTH('1'), COALESCE(1), CONCAT('1', '1'), CORR(1, 1), COS(1), COUNT(1), COVAR_POP(1, 1), COVAR_SAMP(1, 1), DATABASE(), SCHEMA(), DATEDIFF('DAY', toDate('2020-10-24'), toDate('2019-10-24')), EXP(1), FLATTEN([[1]]), FLOOR(1), FQDN(), GREATEST(1), IF(1, 1, 1), IFNULL(1, 1), LCASE('A'), LEAST(1), LENGTH('1'), LN(1), LOG(1), LOG10(1), LOG2(1), LOWER('A'), MAX(1), MID('123', 1, 1), MIN(1), MOD(1, 1), NOT(1), NOW(), NOW64(), NULLIF(1, 1), PI(), POSITION('123', '2'), POW(1, 1), POWER(1, 1), RAND(), REPLACE('1', '1', '2'), REVERSE('123'), ROUND(1), SIN(1), SQRT(1), STDDEV_POP(1), STDDEV_SAMP(1), SUBSTR('123', 2), SUBSTRING('123', 2), SUM(1), TAN(1), TANH(1), TRUNC(1), TRUNCATE(1), UCASE('A'), UPPER('A'), USER(), VAR_POP(1), VAR_SAMP(1), WEEK(toDate('2020-10-24')), YEARWEEK(toDate('2020-10-24')) format TSVRaw;

From 0b63cb237a5a72c96bbc3d4cf52ab70a5d2ad2aa Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Sun, 10 Mar 2024 12:05:32 +0100
Subject: [PATCH 130/443] Fix

---
 src/Storages/System/StorageSystemDisks.cpp | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/src/Storages/System/StorageSystemDisks.cpp b/src/Storages/System/StorageSystemDisks.cpp
index f67d4f7acd0..eecc889f86b 100644
--- a/src/Storages/System/StorageSystemDisks.cpp
+++ b/src/Storages/System/StorageSystemDisks.cpp
@@ -73,9 +73,9 @@ Pipe StorageSystemDisks::read(
         col_unreserved->insert(disk_ptr->getUnreservedSpace().value_or(std::numeric_limits<UInt64>::max()));
         col_keep->insert(disk_ptr->getKeepingFreeSpace());
         auto data_source_description = disk_ptr->getDataSourceDescription();
-        col_type->insert(data_source_description.type);
-        col_object_storage_type->insert(data_source_description.object_storage_type);
-        col_metadata_type->insert(data_source_description.metadata_type);
+        col_type->insert(magic_enum::enum_name(data_source_description.type));
+        col_object_storage_type->insert(magic_enum::enum_name(data_source_description.object_storage_type));
+        col_metadata_type->insert(magic_enum::enum_name(data_source_description.metadata_type));
         col_is_encrypted->insert(data_source_description.is_encrypted);
         col_is_read_only->insert(disk_ptr->isReadOnly());
         col_is_write_once->insert(disk_ptr->isWriteOnce());

From 04d885b570325b1e66bf0c7c13d4732b4d9340ec Mon Sep 17 00:00:00 2001
From: alesapin <alesapin@gmail.com>
Date: Sun, 10 Mar 2024 16:42:40 +0100
Subject: [PATCH 131/443] Throw memory limit exceptions to avoid OOM in some
 places

---
 src/Processors/Executors/ExecutingGraph.h   | 7 ++++++-
 src/Processors/Executors/ExecutorTasks.h    | 5 ++++-
 src/Processors/Executors/PipelineExecutor.h | 7 ++++++-
 3 files changed, 16 insertions(+), 3 deletions(-)

diff --git a/src/Processors/Executors/ExecutingGraph.h b/src/Processors/Executors/ExecutingGraph.h
index 1a7659d3e08..e6d41321edd 100644
--- a/src/Processors/Executors/ExecutingGraph.h
+++ b/src/Processors/Executors/ExecutingGraph.h
@@ -3,6 +3,7 @@
 #include <Processors/Port.h>
 #include <Processors/IProcessor.h>
 #include <Common/SharedMutex.h>
+#include <Common/AllocatorWithMemoryTracking.h>
 #include <mutex>
 #include <queue>
 #include <stack>
@@ -117,7 +118,11 @@ public:
         }
     };
 
-    using Queue = std::queue<Node *>;
+    /// This queue can grow a lot and lead to OOM. That is why we use non-default
+    /// allocator for container which throws exceptions in operator new
+    using DequeWithMemoryTracker = std::deque<ExecutingGraph::Node *, AllocatorWithMemoryTracking<ExecutingGraph::Node *>>;
+    using Queue = std::queue<ExecutingGraph::Node *, DequeWithMemoryTracker>;
+
     using NodePtr = std::unique_ptr<Node>;
     using Nodes = std::vector<NodePtr>;
     Nodes nodes;
diff --git a/src/Processors/Executors/ExecutorTasks.h b/src/Processors/Executors/ExecutorTasks.h
index d35f8de94d1..202ca253c6c 100644
--- a/src/Processors/Executors/ExecutorTasks.h
+++ b/src/Processors/Executors/ExecutorTasks.h
@@ -47,7 +47,10 @@ class ExecutorTasks
 
 public:
     using Stack = std::stack<UInt64>;
-    using Queue = std::queue<ExecutingGraph::Node *>;
+    /// This queue can grow a lot and lead to OOM. That is why we use non-default
+    /// allocator for container which throws exceptions in operator new
+    using DequeWithMemoryTracker = std::deque<ExecutingGraph::Node *, AllocatorWithMemoryTracking<ExecutingGraph::Node *>>;
+    using Queue = std::queue<ExecutingGraph::Node *, DequeWithMemoryTracker>;
 
     void finish();
     bool isFinished() const { return finished; }
diff --git a/src/Processors/Executors/PipelineExecutor.h b/src/Processors/Executors/PipelineExecutor.h
index cb74b524163..03f0f7f1a0a 100644
--- a/src/Processors/Executors/PipelineExecutor.h
+++ b/src/Processors/Executors/PipelineExecutor.h
@@ -5,7 +5,9 @@
 #include <Common/EventCounter.h>
 #include <Common/ThreadPool_fwd.h>
 #include <Common/ConcurrencyControl.h>
+#include <Common/AllocatorWithMemoryTracking.h>
 
+#include <deque>
 #include <queue>
 #include <mutex>
 #include <memory>
@@ -90,7 +92,10 @@ private:
 
     ReadProgressCallbackPtr read_progress_callback;
 
-    using Queue = std::queue<ExecutingGraph::Node *>;
+    /// This queue can grow a lot and lead to OOM. That is why we use non-default
+    /// allocator for container which throws exceptions in operator new
+    using DequeWithMemoryTracker = std::deque<ExecutingGraph::Node *, AllocatorWithMemoryTracking<ExecutingGraph::Node *>>;
+    using Queue = std::queue<ExecutingGraph::Node *, DequeWithMemoryTracker>;
 
     void initializeExecution(size_t num_threads, bool concurrency_control); /// Initialize executor contexts and task_queue.
     void finalizeExecution(); /// Check all processors are finished.

From c1caeaa81483e33c666679fb1ce20d6a7da154c3 Mon Sep 17 00:00:00 2001
From: Han Fei <hanfei19910905@gmail.com>
Date: Sun, 10 Mar 2024 20:51:54 +0100
Subject: [PATCH 132/443] fix build

---
 src/Common/CgroupsMemoryUsageObserver.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Common/CgroupsMemoryUsageObserver.h b/src/Common/CgroupsMemoryUsageObserver.h
index 639433b5016..fc0d847af4f 100644
--- a/src/Common/CgroupsMemoryUsageObserver.h
+++ b/src/Common/CgroupsMemoryUsageObserver.h
@@ -91,7 +91,7 @@ public:
 
     void setLimits(uint64_t, uint64_t) {}
     size_t readMemoryUsage() { return 0; }
-    void startThread();
+    void startThread() {}
     size_t getHardLimit() { return 0; }
     size_t getSoftLimit() { return 0; }
     void setOnMemoryLimitUpdate(std::function<void()>) {}

From d2b7fb03b12c5ce3ec5a377f3872483639e8c4fe Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 10 Mar 2024 23:36:03 +0100
Subject: [PATCH 133/443] Fix localization in check-large-objects

---
 utils/check-style/check-large-objects.sh | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/utils/check-style/check-large-objects.sh b/utils/check-style/check-large-objects.sh
index 4eb9190512f..3e2a385bdd0 100755
--- a/utils/check-style/check-large-objects.sh
+++ b/utils/check-style/check-large-objects.sh
@@ -1,5 +1,7 @@
 #!/usr/bin/env bash
 
+export LC_ALL=C # The "total" should be printed without localization
+
 # Check that there are no new translation units compiled to an object file larger than a certain size.
 
 TU_EXCLUDES=(

From fc9efeddf77dc76981ccbad263566911994d33a2 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Mon, 11 Mar 2024 01:42:44 +0300
Subject: [PATCH 134/443] Update check-large-objects.sh

---
 utils/check-style/check-large-objects.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/utils/check-style/check-large-objects.sh b/utils/check-style/check-large-objects.sh
index 04f4e0c3171..5faacae84b9 100755
--- a/utils/check-style/check-large-objects.sh
+++ b/utils/check-style/check-large-objects.sh
@@ -12,7 +12,7 @@ TU_EXCLUDES=(
     Aggregator
 )
 
-if find $1 -name '*.o' | xargs wc -c | grep --regexp='.o$' | sort -rn | awk '{ if ($1 > 50000000) print }' \
+if find $1 -name '*.o' | xargs wc -c | grep --regexp='\.o$' | sort -rn | awk '{ if ($1 > 50000000) print }' \
     | grep -v -f <(printf "%s\n" "${TU_EXCLUDES[@]}")
 then
     echo "^ It's not allowed to have so large translation units."

From 2ea4de78dd3e5e25d43bc834ce5e05da630472ac Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Mon, 11 Mar 2024 00:00:13 +0100
Subject: [PATCH 135/443] Update the test

---
 .../0_stateless/02998_primary_key_skip_columns.reference        | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/queries/0_stateless/02998_primary_key_skip_columns.reference b/tests/queries/0_stateless/02998_primary_key_skip_columns.reference
index 08ccdb83b11..f3c4a4f40fe 100644
--- a/tests/queries/0_stateless/02998_primary_key_skip_columns.reference
+++ b/tests/queries/0_stateless/02998_primary_key_skip_columns.reference
@@ -10,7 +10,7 @@ Key size: 	2400000
 100000
 14954
 798
-15907
+15908
 108
 120
 2334

From 92734843fb9aff0c82871f352a19da7b190ba2f4 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Mon, 11 Mar 2024 02:42:44 +0300
Subject: [PATCH 136/443] Update 03008_index_small.sql

---
 tests/queries/0_stateless/03008_index_small.sql | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/queries/0_stateless/03008_index_small.sql b/tests/queries/0_stateless/03008_index_small.sql
index 4eb2db3abc3..72213ed4fc9 100644
--- a/tests/queries/0_stateless/03008_index_small.sql
+++ b/tests/queries/0_stateless/03008_index_small.sql
@@ -16,4 +16,4 @@ ATTACH TABLE test;
 
 SELECT count() FROM test WHERE b >= 0;
 
--- DROP TABLE test;
+DROP TABLE test;

From b354d07b829a94a43cf6f3867585efac389088c1 Mon Sep 17 00:00:00 2001
From: taiyang-li <654010905@qq.com>
Date: Mon, 11 Mar 2024 14:59:22 +0800
Subject: [PATCH 137/443] remove break

---
 src/Processors/Transforms/FilterTransform.cpp | 1 -
 1 file changed, 1 deletion(-)

diff --git a/src/Processors/Transforms/FilterTransform.cpp b/src/Processors/Transforms/FilterTransform.cpp
index 0f2509c7510..b3be9246f43 100644
--- a/src/Processors/Transforms/FilterTransform.cpp
+++ b/src/Processors/Transforms/FilterTransform.cpp
@@ -337,7 +337,6 @@ void FilterTransform::doTransform(Chunk & chunk)
                 min_size_in_memory = size_in_memory;
                 first_non_constant_column = i;
             }
-            break;
         }
     }
     (void)min_size_in_memory; /// Suppress error of clang-analyzer-deadcode.DeadStores

From cdb9c962f998d218bda99e69edbaabbd0e1a9433 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 9 Mar 2024 09:22:14 +0100
Subject: [PATCH 138/443] Loosen

---
 CMakeLists.txt | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/CMakeLists.txt b/CMakeLists.txt
index b55e9810361..9ffb4789dc9 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -61,8 +61,8 @@ if (ENABLE_CHECK_HEAVY_BUILDS)
     # set CPU time limit to 1000 seconds
     set (RLIMIT_CPU 1000)
 
-    # -fsanitize=memory is too heavy
-    if (SANITIZE STREQUAL "memory")
+    # -fsanitize=memory and address are too heavy
+    if (SANITIZE)
        set (RLIMIT_DATA 10000000000) # 10G
     endif()
 

From 1c82f8707d76821cbe006ed3f9be84822677cd9b Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Mon, 11 Mar 2024 08:38:03 +0100
Subject: [PATCH 139/443] Fix tidy

---
 src/Functions/FunctionsConversion.cpp | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index d2d72558500..94fd960a99e 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -1056,7 +1056,7 @@ struct ConvertImpl<DataTypeEnum<FieldType>, DataTypeNumber<FieldType>, Name, Con
     }
 };
 
-static inline ColumnUInt8::MutablePtr copyNullMap(ColumnPtr col)
+inline ColumnUInt8::MutablePtr copyNullMap(ColumnPtr col)
 {
     ColumnUInt8::MutablePtr null_map = nullptr;
     if (const auto * col_null = checkAndGetColumn<ColumnNullable>(col.get()))
@@ -1984,7 +1984,7 @@ struct NameParseDateTimeBestEffortOrZero;
 struct NameParseDateTimeBestEffortOrNull;
 
 template<typename Name, typename ToDataType>
-static inline bool isDateTime64(const ColumnsWithTypeAndName & arguments)
+inline bool isDateTime64(const ColumnsWithTypeAndName & arguments)
 {
     if constexpr (std::is_same_v<ToDataType, DataTypeDateTime64>)
         return true;
@@ -3391,7 +3391,7 @@ arguments, result_type, input_rows_count); \
             case IntervalKind::Kind::INTERVAL_KIND: \
                 return createFunctionAdaptor(FunctionConvert<DataTypeInterval, NameToInterval##INTERVAL_KIND, PositiveMonotonicity>::create(), from_type);
 
-    static WrapperType createIntervalWrapper(const DataTypePtr & from_type, IntervalKind kind)
+    static WrapperType createIntervalWrapper(const DataTypePtr & from_type, IntervalKind::Kind kind)
     {
         switch (kind)
         {
@@ -3994,7 +3994,7 @@ arguments, result_type, input_rows_count); \
         {
             return [is_nullable = to_type->hasNullableSubcolumns()] (ColumnsWithTypeAndName & arguments, const DataTypePtr & , const ColumnNullable * , size_t) -> ColumnPtr
             {
-                auto & column_object = assert_cast<const ColumnObject &>(*arguments.front().column);
+                const auto & column_object = assert_cast<const ColumnObject &>(*arguments.front().column);
                 auto res = ColumnObject::create(is_nullable);
                 for (size_t i = 0; i < column_object.size(); i++)
                     res->insert(column_object[i]);

From d16c6780c7b5c6b7180e0744c55856b1e5b8285d Mon Sep 17 00:00:00 2001
From: johnnymatthews <9611008+johnnymatthews@users.noreply.github.com>
Date: Thu, 25 Jan 2024 11:50:18 -0400
Subject: [PATCH 140/443] Adds basic emptyArrayUIntX docs.

---
 .../functions/array-functions.md              | 174 +++++++++++++++++-
 1 file changed, 172 insertions(+), 2 deletions(-)

diff --git a/docs/en/sql-reference/functions/array-functions.md b/docs/en/sql-reference/functions/array-functions.md
index 2120b675c73..6dea8e89b6a 100644
--- a/docs/en/sql-reference/functions/array-functions.md
+++ b/docs/en/sql-reference/functions/array-functions.md
@@ -19,7 +19,7 @@ empty([x])
 An array is considered empty if it does not contain any elements.
 
 :::note
-Can be optimized by enabling the [optimize_functions_to_subcolumns](../../operations/settings/settings.md#optimize-functions-to-subcolumns) setting. With `optimize_functions_to_subcolumns = 1` the function reads only [size0](../../sql-reference/data-types/array.md#array-size) subcolumn instead of reading and processing the whole array column. The query `SELECT empty(arr) FROM TABLE;` transforms to `SELECT arr.size0 = 0 FROM TABLE;`.
+Can be optimized by enabling the [`optimize_functions_to_subcolumns` setting](../../operations/settings/settings.md#optimize-functions-to-subcolumns). With `optimize_functions_to_subcolumns = 1` the function reads only [size0](../../sql-reference/data-types/array.md#array-size) subcolumn instead of reading and processing the whole array column. The query `SELECT empty(arr) FROM TABLE;` transforms to `SELECT arr.size0 = 0 FROM TABLE;`.
 :::
 
 The function also works for [strings](string-functions.md#empty) or [UUID](uuid-functions.md#empty).
@@ -104,7 +104,177 @@ Can be optimized by enabling the [optimize_functions_to_subcolumns](../../operat
 
 Alias: `OCTET_LENGTH`
 
-## emptyArrayUInt8, emptyArrayUInt16, emptyArrayUInt32, emptyArrayUInt64
+## emptyArrayUInt8
+
+Accepts zero arguments and returns an empty UInt8 array.
+
+**Syntax**
+
+```sql
+emptyArrayUInt8()
+```
+
+**Arguments**
+
+None.
+
+**Returned value**
+
+An empty array.
+
+**Implementation details**
+
+None.
+
+**Examples**
+
+Query:
+
+```sql
+CREATE TABLE users (uid Int16, name String, age Int16, contacts Array(UInt8)) ENGINE=Memory;
+
+INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayUInt8());
+SELECT * FROM users;
+
+ALTER TABLE users UPDATE contacts = arrayPushBack(contacts, 255) WHERE uid = 1231;
+SELECT * FROM users;
+```
+
+Result:
+
+```response
+1234	Alex	33	[]
+1234	Alex	33	[255]
+```
+
+## emptyArrayUInt16
+
+Accepts zero arguments and returns an empty UInt16 array.
+
+**Syntax**
+
+```sql
+emptyArrayUInt16()
+```
+
+**Arguments**
+
+None.
+
+**Returned value**
+
+An empty array.
+
+**Implementation details**
+
+None.
+
+**Examples**
+
+Query:
+
+```sql
+CREATE TABLE users (uid Int16, name String, age Int16, contacts Array(UInt16)) ENGINE=Memory;
+
+INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayUInt16());
+SELECT * FROM users;
+
+ALTER TABLE users UPDATE contacts = arrayPushBack(contacts, 65535) WHERE uid = 1231;
+SELECT * FROM users;
+```
+
+Result:
+
+```response
+1231	Alex    33	[]
+1231	Alex    33	[65535]
+```
+
+## emptyArrayUInt32
+
+Accepts zero arguments and returns an empty UInt32 array.
+
+**Syntax**
+
+```sql
+emptyArrayUInt32()
+```
+
+**Arguments**
+
+None.
+
+**Returned value**
+
+An empty array.
+
+**Implementation details**
+
+None.
+
+**Examples**
+
+Query:
+
+```sql
+CREATE TABLE users (uid Int16, name String, age Int16, contacts Array(UInt32)) ENGINE=Memory;
+
+INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayUInt32());
+SELECT * FROM users;
+
+ALTER TABLE users UPDATE contacts = arrayPushBack(contacts, 4294967295) WHERE uid = 1231;
+SELECT * FROM users;
+```
+
+Result:
+
+```response
+1231	Alex	33	[]
+1231	Alex	33	[4294967295]
+```
+
+## emptyArrayUInt64
+
+Accepts zero arguments and returns an empty UInt64 array.
+
+**Syntax**
+
+```sql
+emptyArrayUInt64()
+```
+
+**Arguments**
+
+None.
+
+**Returned value**
+
+An empty array.
+
+**Implementation details**
+
+None.
+
+**Examples**
+
+Query:
+
+```sql
+CREATE TABLE users (uid Int16, name String, age Int16, contacts Array(UInt64)) ENGINE=Memory;
+
+INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayUInt64());
+SELECT * FROM users;
+
+ALTER TABLE users UPDATE contacts = arrayPushBack(contacts, 18446744073709551615) WHERE uid = 1231;
+SELECT * FROM users;
+```
+
+Result:
+
+```response
+1231	Alex	33	[]
+1231	Alex	33	[18446744073709551615]
+```
 
 ## emptyArrayInt8, emptyArrayInt16, emptyArrayInt32, emptyArrayInt64
 

From 515cb59b00ac358a95f65ab7b76b8349bc6e02e7 Mon Sep 17 00:00:00 2001
From: johnnymatthews <9611008+johnnymatthews@users.noreply.github.com>
Date: Thu, 25 Jan 2024 12:29:23 -0400
Subject: [PATCH 141/443] Adds emptyArrayIntX docs.

---
 .../functions/array-functions.md              | 172 +++++++++++++++++-
 1 file changed, 171 insertions(+), 1 deletion(-)

diff --git a/docs/en/sql-reference/functions/array-functions.md b/docs/en/sql-reference/functions/array-functions.md
index 6dea8e89b6a..ebeac4196d7 100644
--- a/docs/en/sql-reference/functions/array-functions.md
+++ b/docs/en/sql-reference/functions/array-functions.md
@@ -276,7 +276,177 @@ Result:
 1231	Alex	33	[18446744073709551615]
 ```
 
-## emptyArrayInt8, emptyArrayInt16, emptyArrayInt32, emptyArrayInt64
+## emptyArrayInt8
+
+Accepts zero arguments and returns an empty Int8 array.
+
+**Syntax**
+
+```sql
+emptyArrayInt8()
+```
+
+**Arguments**
+
+None.
+
+**Returned value**
+
+An empty array.
+
+**Implementation details**
+
+None.
+
+**Examples**
+
+Query:
+
+```sql
+CREATE TABLE users (uid Int16, name String, age Int16, contacts Array(Int8)) ENGINE=Memory;
+
+INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayInt8());
+SELECT * FROM users;
+
+ALTER TABLE users UPDATE contacts = arrayPushBack(contacts, 127) WHERE uid = 1231;
+SELECT * FROM users;
+```
+
+Result:
+
+```response
+1234	Alex	33	[]
+1234	Alex	33	[127]
+```
+
+## emptyArrayInt16
+
+Accepts zero arguments and returns an empty Int16 array.
+
+**Syntax**
+
+```sql
+emptyArrayInt16()
+```
+
+**Arguments**
+
+None.
+
+**Returned value**
+
+An empty array.
+
+**Implementation details**
+
+None.
+
+**Examples**
+
+Query:
+
+```sql
+CREATE TABLE users (uid Int16, name String, age Int16, contacts Array(Int16)) ENGINE=Memory;
+
+INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayInt16());
+SELECT * FROM users;
+
+ALTER TABLE users UPDATE contacts = arrayPushBack(contacts, 32767) WHERE uid = 1231;
+SELECT * FROM users;
+```
+
+Result:
+
+```response
+1231	Alex    33	[]
+1231	Alex    33	[32767]
+```
+
+## emptyArrayInt32
+
+Accepts zero arguments and returns an empty Int32 array.
+
+**Syntax**
+
+```sql
+emptyArrayInt32()
+```
+
+**Arguments**
+
+None.
+
+**Returned value**
+
+An empty array.
+
+**Implementation details**
+
+None.
+
+**Examples**
+
+Query:
+
+```sql
+CREATE TABLE users (uid Int16, name String, age Int16, contacts Array(Int32)) ENGINE=Memory;
+
+INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayInt32());
+SELECT * FROM users;
+
+ALTER TABLE users UPDATE contacts = arrayPushBack(contacts, 2147483647) WHERE uid = 1231;
+SELECT * FROM users;
+```
+
+Result:
+
+```response
+1231	Alex	33	[]
+1231	Alex	33	[2147483647]
+```
+
+## emptyArrayInt64
+
+Accepts zero arguments and returns an empty Int64 array.
+
+**Syntax**
+
+```sql
+emptyArrayInt64()
+```
+
+**Arguments**
+
+None.
+
+**Returned value**
+
+An empty array.
+
+**Implementation details**
+
+None.
+
+**Examples**
+
+Query:
+
+```sql
+CREATE TABLE users (uid Int16, name String, age Int16, contacts Array(Int64)) ENGINE=Memory;
+
+INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayInt64());
+SELECT * FROM users;
+
+ALTER TABLE users UPDATE contacts = arrayPushBack(contacts, 9223372036854775807) WHERE uid = 1231;
+SELECT * FROM users;
+```
+
+Result:
+
+```response
+1231	Alex	33	[]
+1231	Alex	33	[9223372036854775807]
+```
 
 ## emptyArrayFloat32, emptyArrayFloat64
 

From c3895c3546b66d3d1b8eac0e3e9377d4da9a1f29 Mon Sep 17 00:00:00 2001
From: johnnymatthews <9611008+johnnymatthews@users.noreply.github.com>
Date: Thu, 25 Jan 2024 14:41:32 -0400
Subject: [PATCH 142/443] Adds emptyArrayFloat32/64 docs.

---
 .../functions/array-functions.md              | 86 ++++++++++++++++++-
 1 file changed, 85 insertions(+), 1 deletion(-)

diff --git a/docs/en/sql-reference/functions/array-functions.md b/docs/en/sql-reference/functions/array-functions.md
index ebeac4196d7..471dd0ff3f5 100644
--- a/docs/en/sql-reference/functions/array-functions.md
+++ b/docs/en/sql-reference/functions/array-functions.md
@@ -448,7 +448,91 @@ Result:
 1231	Alex	33	[9223372036854775807]
 ```
 
-## emptyArrayFloat32, emptyArrayFloat64
+## emptyArrayFloat32 
+
+Returns an empty Float32 array. Accepts zero arguments.
+
+**Syntax**
+
+```sql
+emptyArrayFloat32()
+```
+
+**Arguments**
+
+None.
+
+**Returned value**
+
+An empty array.
+
+**Implementation details**
+
+None.
+
+**Examples**
+
+Query:
+
+```sql
+CREATE TABLE users (uid Int16, name String, age Int16, contacts Array(Int32)) ENGINE=Memory;
+
+INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayFloat32());
+SELECT * FROM users;
+
+ALTER TABLE users UPDATE contacts = arrayPushBack(contacts, 3.4e+38) WHERE uid = 1231;
+SELECT * FROM users;
+```
+
+Result:
+
+```response
+1231	Alex	33	[]
+1231	Alex	33	[-2147483648]
+```
+
+## emptyArrayFloat64
+
+Accepts zero arguments and returns an empty Float64 array.
+
+**Syntax**
+
+```sql
+emptyArrayFloat64()
+```
+
+**Arguments**
+
+None.
+
+**Returned value**
+
+An empty array.
+
+**Implementation details**
+
+None.
+
+**Examples**
+
+Query:
+
+```sql
+CREATE TABLE users (uid Int16, name String, age Int16, contacts Array(Int)) ENGINE=Memory;
+
+INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayFloat64());
+SELECT * FROM users;
+
+ALTER TABLE users UPDATE contacts = arrayPushBack(contacts, 1.999999999e+9) WHERE uid = 1231;
+SELECT * FROM users;
+```
+
+Result:
+
+```response
+1231	Alex	33	[]
+1231	Alex	33	[1999999999]
+```
 
 ## emptyArrayDate, emptyArrayDateTime
 

From ade0f9f75fcf0d8e7394655a1938a16ed86c85a1 Mon Sep 17 00:00:00 2001
From: johnnymatthews <9611008+johnnymatthews@users.noreply.github.com>
Date: Mon, 29 Jan 2024 14:21:58 -0400
Subject: [PATCH 143/443] Adds docs for emptyArrayDate and emptyArrayDateTime.

---
 .../functions/array-functions.md              | 123 +++++++++++++++++-
 1 file changed, 122 insertions(+), 1 deletion(-)

diff --git a/docs/en/sql-reference/functions/array-functions.md b/docs/en/sql-reference/functions/array-functions.md
index 471dd0ff3f5..d811112a861 100644
--- a/docs/en/sql-reference/functions/array-functions.md
+++ b/docs/en/sql-reference/functions/array-functions.md
@@ -536,9 +536,130 @@ Result:
 
 ## emptyArrayDate, emptyArrayDateTime
 
+## emptyArrayDate
+
+Accepts zero arguments and returns an empty Date array.
+
+**Syntax**
+
+```sql
+emptyArrayDate()
+```
+
+**Arguments**
+
+None.
+
+**Returned value**
+
+An empty array.
+
+**Implementation details**
+
+None.
+
+**Examples**
+
+Query:
+
+```sql
+CREATE TABLE users (uid Int16, name String, age Int16, userDates Array(Date)) ENGINE=Memory;
+
+INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayDate());
+SELECT * FROM users;
+
+ALTER TABLE users UPDATE userDates = arrayPushBack(userDates, today()) WHERE uid = 1231;
+ALTER TABLE users UPDATE userDates = arrayPushBack(userDates, addDays(today(), 1)) WHERE uid = 1231;
+SELECT * FROM users;
+```
+
+## emptyArrayDateTime
+
+Accepts zero arguments and returns an empty DateTime array.
+
+**Syntax**
+
+```sql
+emptyArrayDateTime()
+```
+
+**Arguments**
+
+None.
+
+**Returned value**
+
+An empty array.
+
+**Implementation details**
+
+None.
+
+**Examples**
+
+Query:
+
+```sql
+CREATE TABLE users (uid Int16, name String, age Int16, userDates Array(DateTime)) ENGINE=Memory;
+
+INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayDateTime());
+SELECT * FROM users;
+
+ALTER TABLE users UPDATE userDates = arrayPushBack(userDates, now()) WHERE uid = 1231;
+ALTER TABLE users UPDATE userDates = arrayPushBack(userDates, addDays(now(), 1)) WHERE uid = 1231;
+SELECT * FROM users;
+```
+
+Result:
+
+```response
+1231	Alex	33	[]
+1231	Alex	33	['2024-01-29 18:08:01','2024-01-30 18:08:01']
+```
+
 ## emptyArrayString
 
-Accepts zero arguments and returns an empty array of the appropriate type.
+Accepts zero arguments and returns an empty String array.
+
+**Syntax**
+
+```sql
+emptyArrayString()
+```
+
+**Arguments**
+
+None.
+
+**Returned value**
+
+An empty array.
+
+**Implementation details**
+
+None.
+
+**Examples**
+
+Query:
+
+```sql
+CREATE TABLE users (uid Int16, name String, age Int16, aliases Array(String)) ENGINE=Memory;
+
+INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayString());
+SELECT * FROM users;
+
+ALTER TABLE users UPDATE aliases = arrayPushBack(aliases, 'Al') WHERE uid = 1231;
+ALTER TABLE users UPDATE aliases = arrayPushBack(aliases, 'Little Alex') WHERE uid = 1231;
+SELECT * FROM users;
+```
+
+Result:
+
+```response
+1231	Alex	33	[]
+1231	Alex	33	['Al','Little Alex']
+```
 
 ## emptyArrayToSingle
 

From e7270069a487be62e067e1f158ad0d745280622b Mon Sep 17 00:00:00 2001
From: Johnny <9611008+johnnymatthews@users.noreply.github.com>
Date: Mon, 29 Jan 2024 14:26:20 -0400
Subject: [PATCH 144/443] Removes dupliated h2 header.

---
 docs/en/sql-reference/functions/array-functions.md | 2 --
 1 file changed, 2 deletions(-)

diff --git a/docs/en/sql-reference/functions/array-functions.md b/docs/en/sql-reference/functions/array-functions.md
index d811112a861..d3054d25106 100644
--- a/docs/en/sql-reference/functions/array-functions.md
+++ b/docs/en/sql-reference/functions/array-functions.md
@@ -534,8 +534,6 @@ Result:
 1231	Alex	33	[1999999999]
 ```
 
-## emptyArrayDate, emptyArrayDateTime
-
 ## emptyArrayDate
 
 Accepts zero arguments and returns an empty Date array.

From 3c60e829710ffb72c8571ef664ad146cfcd3f266 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Mon, 11 Mar 2024 07:39:08 +0000
Subject: [PATCH 145/443] Incorporate review feedback

---
 .../functions/array-functions.md              | 222 +++---------------
 1 file changed, 39 insertions(+), 183 deletions(-)

diff --git a/docs/en/sql-reference/functions/array-functions.md b/docs/en/sql-reference/functions/array-functions.md
index d3054d25106..67a4c026851 100644
--- a/docs/en/sql-reference/functions/array-functions.md
+++ b/docs/en/sql-reference/functions/array-functions.md
@@ -106,7 +106,7 @@ Alias: `OCTET_LENGTH`
 
 ## emptyArrayUInt8
 
-Accepts zero arguments and returns an empty UInt8 array.
+Returns an empty UInt8 array.
 
 **Syntax**
 
@@ -122,34 +122,23 @@ None.
 
 An empty array.
 
-**Implementation details**
-
-None.
-
 **Examples**
 
 Query:
 
 ```sql
-CREATE TABLE users (uid Int16, name String, age Int16, contacts Array(UInt8)) ENGINE=Memory;
-
-INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayUInt8());
-SELECT * FROM users;
-
-ALTER TABLE users UPDATE contacts = arrayPushBack(contacts, 255) WHERE uid = 1231;
-SELECT * FROM users;
+SELECT emptyArrayUInt8();
 ```
 
 Result:
 
 ```response
-1234	Alex	33	[]
-1234	Alex	33	[255]
+[]
 ```
 
 ## emptyArrayUInt16
 
-Accepts zero arguments and returns an empty UInt16 array.
+Returns an empty UInt16 array.
 
 **Syntax**
 
@@ -165,34 +154,24 @@ None.
 
 An empty array.
 
-**Implementation details**
-
-None.
-
 **Examples**
 
 Query:
 
 ```sql
-CREATE TABLE users (uid Int16, name String, age Int16, contacts Array(UInt16)) ENGINE=Memory;
+SELECT emptyArrayUInt16();
 
-INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayUInt16());
-SELECT * FROM users;
-
-ALTER TABLE users UPDATE contacts = arrayPushBack(contacts, 65535) WHERE uid = 1231;
-SELECT * FROM users;
 ```
 
 Result:
 
 ```response
-1231	Alex    33	[]
-1231	Alex    33	[65535]
+[]
 ```
 
 ## emptyArrayUInt32
 
-Accepts zero arguments and returns an empty UInt32 array.
+Returns an empty UInt32 array.
 
 **Syntax**
 
@@ -208,34 +187,23 @@ None.
 
 An empty array.
 
-**Implementation details**
-
-None.
-
 **Examples**
 
 Query:
 
 ```sql
-CREATE TABLE users (uid Int16, name String, age Int16, contacts Array(UInt32)) ENGINE=Memory;
-
-INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayUInt32());
-SELECT * FROM users;
-
-ALTER TABLE users UPDATE contacts = arrayPushBack(contacts, 4294967295) WHERE uid = 1231;
-SELECT * FROM users;
+SELECT emptyArrayUInt32();
 ```
 
 Result:
 
 ```response
-1231	Alex	33	[]
-1231	Alex	33	[4294967295]
+[]
 ```
 
 ## emptyArrayUInt64
 
-Accepts zero arguments and returns an empty UInt64 array.
+Returns an empty UInt64 array.
 
 **Syntax**
 
@@ -251,34 +219,23 @@ None.
 
 An empty array.
 
-**Implementation details**
-
-None.
-
 **Examples**
 
 Query:
 
 ```sql
-CREATE TABLE users (uid Int16, name String, age Int16, contacts Array(UInt64)) ENGINE=Memory;
-
-INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayUInt64());
-SELECT * FROM users;
-
-ALTER TABLE users UPDATE contacts = arrayPushBack(contacts, 18446744073709551615) WHERE uid = 1231;
-SELECT * FROM users;
+SELECT emptyArrayUInt64();
 ```
 
 Result:
 
 ```response
-1231	Alex	33	[]
-1231	Alex	33	[18446744073709551615]
+[]
 ```
 
 ## emptyArrayInt8
 
-Accepts zero arguments and returns an empty Int8 array.
+Returns an empty Int8 array.
 
 **Syntax**
 
@@ -294,34 +251,23 @@ None.
 
 An empty array.
 
-**Implementation details**
-
-None.
-
 **Examples**
 
 Query:
 
 ```sql
-CREATE TABLE users (uid Int16, name String, age Int16, contacts Array(Int8)) ENGINE=Memory;
-
-INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayInt8());
-SELECT * FROM users;
-
-ALTER TABLE users UPDATE contacts = arrayPushBack(contacts, 127) WHERE uid = 1231;
-SELECT * FROM users;
+SELECT emptyArrayInt8();
 ```
 
 Result:
 
 ```response
-1234	Alex	33	[]
-1234	Alex	33	[127]
+[]
 ```
 
 ## emptyArrayInt16
 
-Accepts zero arguments and returns an empty Int16 array.
+Returns an empty Int16 array.
 
 **Syntax**
 
@@ -337,34 +283,23 @@ None.
 
 An empty array.
 
-**Implementation details**
-
-None.
-
 **Examples**
 
 Query:
 
 ```sql
-CREATE TABLE users (uid Int16, name String, age Int16, contacts Array(Int16)) ENGINE=Memory;
-
-INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayInt16());
-SELECT * FROM users;
-
-ALTER TABLE users UPDATE contacts = arrayPushBack(contacts, 32767) WHERE uid = 1231;
-SELECT * FROM users;
+SELECT emptyArrayInt16();
 ```
 
 Result:
 
 ```response
-1231	Alex    33	[]
-1231	Alex    33	[32767]
+[]
 ```
 
 ## emptyArrayInt32
 
-Accepts zero arguments and returns an empty Int32 array.
+Returns an empty Int32 array.
 
 **Syntax**
 
@@ -380,34 +315,23 @@ None.
 
 An empty array.
 
-**Implementation details**
-
-None.
-
 **Examples**
 
 Query:
 
 ```sql
-CREATE TABLE users (uid Int16, name String, age Int16, contacts Array(Int32)) ENGINE=Memory;
-
-INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayInt32());
-SELECT * FROM users;
-
-ALTER TABLE users UPDATE contacts = arrayPushBack(contacts, 2147483647) WHERE uid = 1231;
-SELECT * FROM users;
+SELECT emptyArrayInt32();
 ```
 
 Result:
 
 ```response
-1231	Alex	33	[]
-1231	Alex	33	[2147483647]
+[]
 ```
 
 ## emptyArrayInt64
 
-Accepts zero arguments and returns an empty Int64 array.
+Returns an empty Int64 array.
 
 **Syntax**
 
@@ -423,34 +347,23 @@ None.
 
 An empty array.
 
-**Implementation details**
-
-None.
-
 **Examples**
 
 Query:
 
 ```sql
-CREATE TABLE users (uid Int16, name String, age Int16, contacts Array(Int64)) ENGINE=Memory;
-
-INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayInt64());
-SELECT * FROM users;
-
-ALTER TABLE users UPDATE contacts = arrayPushBack(contacts, 9223372036854775807) WHERE uid = 1231;
-SELECT * FROM users;
+SELECT emptyArrayInt64();
 ```
 
 Result:
 
 ```response
-1231	Alex	33	[]
-1231	Alex	33	[9223372036854775807]
+[]
 ```
 
 ## emptyArrayFloat32 
 
-Returns an empty Float32 array. Accepts zero arguments.
+Returns an empty Float32 array.
 
 **Syntax**
 
@@ -466,34 +379,23 @@ None.
 
 An empty array.
 
-**Implementation details**
-
-None.
-
 **Examples**
 
 Query:
 
 ```sql
-CREATE TABLE users (uid Int16, name String, age Int16, contacts Array(Int32)) ENGINE=Memory;
-
-INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayFloat32());
-SELECT * FROM users;
-
-ALTER TABLE users UPDATE contacts = arrayPushBack(contacts, 3.4e+38) WHERE uid = 1231;
-SELECT * FROM users;
+SELECT emptyArrayFloat32();
 ```
 
 Result:
 
 ```response
-1231	Alex	33	[]
-1231	Alex	33	[-2147483648]
+[]
 ```
 
 ## emptyArrayFloat64
 
-Accepts zero arguments and returns an empty Float64 array.
+Returns an empty Float64 array.
 
 **Syntax**
 
@@ -509,34 +411,23 @@ None.
 
 An empty array.
 
-**Implementation details**
-
-None.
-
 **Examples**
 
 Query:
 
 ```sql
-CREATE TABLE users (uid Int16, name String, age Int16, contacts Array(Int)) ENGINE=Memory;
-
-INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayFloat64());
-SELECT * FROM users;
-
-ALTER TABLE users UPDATE contacts = arrayPushBack(contacts, 1.999999999e+9) WHERE uid = 1231;
-SELECT * FROM users;
+SELECT emptyArrayFloat64();
 ```
 
 Result:
 
 ```response
-1231	Alex	33	[]
-1231	Alex	33	[1999999999]
+[]
 ```
 
 ## emptyArrayDate
 
-Accepts zero arguments and returns an empty Date array.
+Returns an empty Date array.
 
 **Syntax**
 
@@ -552,33 +443,22 @@ None.
 
 An empty array.
 
-**Implementation details**
-
-None.
-
 **Examples**
 
 Query:
 
 ```sql
-CREATE TABLE users (uid Int16, name String, age Int16, userDates Array(Date)) ENGINE=Memory;
-
-INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayDate());
-SELECT * FROM users;
-
-ALTER TABLE users UPDATE userDates = arrayPushBack(userDates, today()) WHERE uid = 1231;
-ALTER TABLE users UPDATE userDates = arrayPushBack(userDates, addDays(today(), 1)) WHERE uid = 1231;
-SELECT * FROM users;
+SELECT emptyArrayDate();
 ```
 
 ## emptyArrayDateTime
 
-Accepts zero arguments and returns an empty DateTime array.
+Returns an empty DateTime array.
 
 **Syntax**
 
 ```sql
-emptyArrayDateTime()
+[]
 ```
 
 **Arguments**
@@ -589,35 +469,23 @@ None.
 
 An empty array.
 
-**Implementation details**
-
-None.
-
 **Examples**
 
 Query:
 
 ```sql
-CREATE TABLE users (uid Int16, name String, age Int16, userDates Array(DateTime)) ENGINE=Memory;
-
-INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayDateTime());
-SELECT * FROM users;
-
-ALTER TABLE users UPDATE userDates = arrayPushBack(userDates, now()) WHERE uid = 1231;
-ALTER TABLE users UPDATE userDates = arrayPushBack(userDates, addDays(now(), 1)) WHERE uid = 1231;
-SELECT * FROM users;
+SELECT emptyArrayDateTime();
 ```
 
 Result:
 
 ```response
-1231	Alex	33	[]
-1231	Alex	33	['2024-01-29 18:08:01','2024-01-30 18:08:01']
+[]
 ```
 
 ## emptyArrayString
 
-Accepts zero arguments and returns an empty String array.
+Returns an empty String array.
 
 **Syntax**
 
@@ -633,30 +501,18 @@ None.
 
 An empty array.
 
-**Implementation details**
-
-None.
-
 **Examples**
 
 Query:
 
 ```sql
-CREATE TABLE users (uid Int16, name String, age Int16, aliases Array(String)) ENGINE=Memory;
-
-INSERT INTO users VALUES (1231, 'Alex', 33, emptyArrayString());
-SELECT * FROM users;
-
-ALTER TABLE users UPDATE aliases = arrayPushBack(aliases, 'Al') WHERE uid = 1231;
-ALTER TABLE users UPDATE aliases = arrayPushBack(aliases, 'Little Alex') WHERE uid = 1231;
-SELECT * FROM users;
+SELECT emptyArrayString();
 ```
 
 Result:
 
 ```response
-1231	Alex	33	[]
-1231	Alex	33	['Al','Little Alex']
+[]
 ```
 
 ## emptyArrayToSingle

From 27099cedcb974d704e240cedf4da547d50debb19 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Mon, 11 Mar 2024 08:34:01 +0000
Subject: [PATCH 146/443] Some fixups

---
 .../functions/random-functions.md             | 84 +++++--------------
 .../functions/type-conversion-functions.md    | 16 ++--
 .../sql-reference/functions/url-functions.md  |  6 +-
 3 files changed, 34 insertions(+), 72 deletions(-)

diff --git a/docs/en/sql-reference/functions/random-functions.md b/docs/en/sql-reference/functions/random-functions.md
index b745d2833d3..2d7752ed022 100644
--- a/docs/en/sql-reference/functions/random-functions.md
+++ b/docs/en/sql-reference/functions/random-functions.md
@@ -20,7 +20,9 @@ The random numbers are generated by non-cryptographic algorithms.
 
 ## rand
 
-Returns a random UInt32 number.
+Returns a random UInt32 number with uniform distribution.
+
+Uses a linear congruential generator with an initial state obtained from the system, which means that while it appears random, it's not truly random and can be predictable if the initial state is known. For scenarios where true randomness is crucial, consider using alternative methods like system-level calls or integrating with external libraries.
 
 ### Syntax
 
@@ -28,18 +30,16 @@ Returns a random UInt32 number.
 rand()
 ```
 
-### Parameters
+Alias: `rand32`
+
+### Arguments
 
 None.
 
-### Output
+### Returned value
 
 Returns a number of type UInt32.
 
-### Implementation details
-
-Uses a linear congruential generator.
-
 ### Example
 
 ```sql
@@ -47,44 +47,12 @@ SELECT rand();
 ```
 
 ```response
-1569354847
+1569354847 -- Note: The actual output will be a random number, not the specific number shown in the example
 ```
 
-**Note:** The actual output will be a random number, not the specific number shown in the example.
-
-## rand32
-
-Returns a random 32-bit unsigned integer (UInt32) number.
-
-### Syntax
-
-```sql
-rand32()
-```
-
-### Parameters
-
-None.
-
-### Output
-
-Returns a number of type UInt32, evenly distributed across the range of all possible UInt32 values.
-
-### Example
-
-```sql
-SELECT rand32();
-```
-
-```response
-2754546224
-```
-
-**Note:** The actual output will be a random number, not the specific number shown in the example.
-
 ## rand64
 
-Returns a random 64-bit unsigned integer (UInt64) number.
+Returns a random UInt64 integer (UInt64) number
 
 ### Syntax
 
@@ -92,17 +60,15 @@ Returns a random 64-bit unsigned integer (UInt64) number.
 rand64()
 ```
 
-### Parameters
+### Arguments
 
 None.
 
-### Output
+### Returned value
 
-Returns a number of type UInt64, evenly distributed across the range of all possible UInt64 values.
+Returns a number UInt64 number with uniform distribution.
 
-### Implementation details
-
-The `rand64` function uses a linear congruential generator, which means that while it appears random, it's not truly random and can be predictable if the initial state is known. For scenarios where true randomness is crucial, consider using alternative methods like system-level calls or integrating with external libraries.
+Uses a linear congruential generator with an initial state obtained from the system, which means that while it appears random, it's not truly random and can be predictable if the initial state is known. For scenarios where true randomness is crucial, consider using alternative methods like system-level calls or integrating with external libraries.
 
 ### Example
 
@@ -111,14 +77,12 @@ SELECT rand64();
 ```
 
 ```response
-15030268859237645412
+15030268859237645412 -- Note: The actual output will be a random number, not the specific number shown in the example.
 ```
 
-**Note:** The actual output will be a random number, not the specific number shown in the example.
-
 ## randCanonical
 
-Returns a random floating-point number of type Float64, evenly distributed within the closed interval.
+Returns a random Float64 number.
 
 ### Syntax
 
@@ -126,11 +90,11 @@ Returns a random floating-point number of type Float64, evenly distributed withi
 randCanonical()
 ```
 
-### Parameters
+### Arguments
 
 None.
 
-### Output
+### Returned value
 
 Returns a Float64 value between 0 (inclusive) and 1 (exclusive).
 
@@ -141,14 +105,12 @@ SELECT randCanonical();
 ```
 
 ```response
-0.3452178901234567
+0.3452178901234567 - Note: The actual output will be a random Float64 number between 0 and 1, not the specific number shown in the example.
 ```
 
-**Note:** The actual output will be a random decimal number between 0 and 1, not the specific number shown in the example.
-
 ## randConstant
 
-Generates a single constant column filled with a random value. Unlike `rand`, `randConstant` ensures the same random value appears in every row of the generated column, making it useful for scenarios requiring a consistent random seed across rows in a single query.
+Generates a single constant column filled with a random value. Unlike `rand`, this function ensures the same random value appears in every row of the generated column, making it useful for scenarios requiring a consistent random seed across rows in a single query.
 
 ### Syntax
 
@@ -156,11 +118,11 @@ Generates a single constant column filled with a random value. Unlike `rand`, `r
 randConstant([x]);
 ```
 
-### Parameters
+### Arguments
 
 - **[x] (Optional):** An optional expression that influences the generated random value. Even if provided, the resulting value will still be constant within the same query execution. Different queries using the same expression will likely generate different constant values.
 
-### Output
+### Returned value
 
 Returns a column of type UInt32 containing the same random value in each row.
 
@@ -200,12 +162,12 @@ Returns a random Float64 drawn uniformly from interval [`min`, `max`].
 randUniform(min, max)
 ```
 
-### Parameters
+### Arguments
 
 - `min` - `Float64` - left boundary of the range,
 - `max` - `Float64` - right boundary of the range.
 
-### Output
+### Returned value
 
 A random number of type [Float64](/docs/en/sql-reference/data-types/float.md).
 
diff --git a/docs/en/sql-reference/functions/type-conversion-functions.md b/docs/en/sql-reference/functions/type-conversion-functions.md
index 37d4ac30648..ea08ffa50e7 100644
--- a/docs/en/sql-reference/functions/type-conversion-functions.md
+++ b/docs/en/sql-reference/functions/type-conversion-functions.md
@@ -968,7 +968,7 @@ Converts a numeric value to String with the number of fractional digits in the o
 toDecimalString(number, scale)
 ```
 
-**Parameters**
+**Arguments**
 
 - `number` — Value to be represented as String, [Int, UInt](/docs/en/sql-reference/data-types/int-uint.md), [Float](/docs/en/sql-reference/data-types/float.md), [Decimal](/docs/en/sql-reference/data-types/decimal.md),
 - `scale` — Number of fractional digits, [UInt8](/docs/en/sql-reference/data-types/int-uint.md).
@@ -1261,7 +1261,7 @@ Converts input value `x` to the specified data type `T`. Always returns [Nullabl
 accurateCastOrNull(x, T)
 ```
 
-**Parameters**
+**Arguments**
 
 - `x` — Input value.
 - `T` — The name of the returned data type.
@@ -1314,7 +1314,7 @@ Converts input value `x` to the specified data type `T`. Returns default type va
 accurateCastOrDefault(x, T)
 ```
 
-**Parameters**
+**Arguments**
 
 - `x` — Input value.
 - `T` — The name of the returned data type.
@@ -1675,7 +1675,7 @@ Same as [parseDateTimeBestEffort](#parsedatetimebesteffort) function but also pa
 parseDateTime64BestEffort(time_string [, precision [, time_zone]])
 ```
 
-**Parameters**
+**Arguments**
 
 - `time_string` — String containing a date or date with time to convert. [String](/docs/en/sql-reference/data-types/string.md).
 - `precision` — Required precision. `3` — for milliseconds, `6` — for microseconds. Default — `3`. Optional. [UInt8](/docs/en/sql-reference/data-types/int-uint.md).
@@ -1990,7 +1990,7 @@ Extracts the timestamp component of a [Snowflake ID](https://en.wikipedia.org/wi
 snowflakeToDateTime(value[, time_zone])
 ```
 
-**Parameters**
+**Arguments**
 
 - `value` — Snowflake ID. [Int64](/docs/en/sql-reference/data-types/int-uint.md).
 - `time_zone` — [Timezone](/docs/en/operations/server-configuration-parameters/settings.md/#server_configuration_parameters-timezone). The function parses `time_string` according to the timezone. Optional. [String](/docs/en/sql-reference/data-types/string.md).
@@ -2026,7 +2026,7 @@ Extracts the timestamp component of a [Snowflake ID](https://en.wikipedia.org/wi
 snowflakeToDateTime64(value[, time_zone])
 ```
 
-**Parameters**
+**Arguments**
 
 - `value` — Snowflake ID. [Int64](/docs/en/sql-reference/data-types/int-uint.md).
 - `time_zone` — [Timezone](/docs/en/operations/server-configuration-parameters/settings.md/#server_configuration_parameters-timezone). The function parses `time_string` according to the timezone. Optional. [String](/docs/en/sql-reference/data-types/string.md).
@@ -2062,7 +2062,7 @@ Converts a [DateTime](/docs/en/sql-reference/data-types/datetime.md) value to th
 dateTimeToSnowflake(value)
 ```
 
-**Parameters**
+**Arguments**
 
 - `value` — Date with time. [DateTime](/docs/en/sql-reference/data-types/datetime.md).
 
@@ -2096,7 +2096,7 @@ Convert a [DateTime64](/docs/en/sql-reference/data-types/datetime64.md) to the f
 dateTime64ToSnowflake(value)
 ```
 
-**Parameters**
+**Arguments**
 
 - `value` — Date with time. [DateTime64](/docs/en/sql-reference/data-types/datetime64.md).
 
diff --git a/docs/en/sql-reference/functions/url-functions.md b/docs/en/sql-reference/functions/url-functions.md
index f6871c86c4f..ac81815b47f 100644
--- a/docs/en/sql-reference/functions/url-functions.md
+++ b/docs/en/sql-reference/functions/url-functions.md
@@ -155,7 +155,7 @@ Configuration example:
 cutToFirstSignificantSubdomain(URL, TLD)
 ```
 
-**Parameters**
+**Arguments**
 
 - `URL` — URL. [String](../../sql-reference/data-types/string.md).
 - `TLD` — Custom TLD list name. [String](../../sql-reference/data-types/string.md).
@@ -209,7 +209,7 @@ Configuration example:
 cutToFirstSignificantSubdomainCustomWithWWW(URL, TLD)
 ```
 
-**Parameters**
+**Arguments**
 
 - `URL` — URL. [String](../../sql-reference/data-types/string.md).
 - `TLD` — Custom TLD list name. [String](../../sql-reference/data-types/string.md).
@@ -263,7 +263,7 @@ Configuration example:
 firstSignificantSubdomainCustom(URL, TLD)
 ```
 
-**Parameters**
+**Arguments**
 
 - `URL` — URL. [String](../../sql-reference/data-types/string.md).
 - `TLD` — Custom TLD list name. [String](../../sql-reference/data-types/string.md).

From 318fc7efcf197d4b9c23ee118b16556c37424174 Mon Sep 17 00:00:00 2001
From: Azat Khuzhin <a3at.mail@gmail.com>
Date: Sun, 10 Mar 2024 23:57:37 +0100
Subject: [PATCH 147/443] Fix
 test_distributed_directory_monitor_split_batch_on_failure flakienss

The memory limit had been changed in #52439

CI: https://s3.amazonaws.com/clickhouse-test-reports/61132/04d885b570325b1e66bf0c7c13d4732b4d9340ec/integration_tests__tsan__[2_6].html
Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>
---
 .../test.py                                              | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/tests/integration/test_distributed_directory_monitor_split_batch_on_failure/test.py b/tests/integration/test_distributed_directory_monitor_split_batch_on_failure/test.py
index 0ad00c39a3b..43728ff600f 100644
--- a/tests/integration/test_distributed_directory_monitor_split_batch_on_failure/test.py
+++ b/tests/integration/test_distributed_directory_monitor_split_batch_on_failure/test.py
@@ -67,7 +67,7 @@ def test_distributed_background_insert_split_batch_on_failure_OFF(started_cluste
                 f"insert into dist select number/100, number from system.numbers limit {limit} offset {limit*i}",
                 settings={
                     # max_memory_usage is the limit for the batch on the remote node
-                    # (local query should not be affected since 30MB is enough for 100K rows)
+                    # (local query should not be affected since 20MB is enough for 100K rows)
                     "max_memory_usage": "20Mi",
                     "max_untracked_memory": "0",
                 },
@@ -76,7 +76,8 @@ def test_distributed_background_insert_split_batch_on_failure_OFF(started_cluste
         if setting == "background_insert_batch" and setting_value == 1:
             with pytest.raises(
                 QueryRuntimeException,
-                match=r"DB::Exception: Received from.*Memory limit \(for query\) exceeded: .*while pushing to view default\.mv",
+                # no DOTALL in pytest.raises, use '(.|\n)'
+                match=r"DB::Exception: Received from.*Memory limit \(for query\) exceeded: (.|\n)*While sending a batch",
             ):
                 node2.query("system flush distributed dist")
             assert int(node2.query("select count() from dist_data")) == 0
@@ -94,8 +95,8 @@ def test_distributed_background_insert_split_batch_on_failure_ON(started_cluster
                 f"insert into dist select number/100, number from system.numbers limit {limit} offset {limit*i}",
                 settings={
                     # max_memory_usage is the limit for the batch on the remote node
-                    # (local query should not be affected since 30MB is enough for 100K rows)
-                    "max_memory_usage": "30Mi",
+                    # (local query should not be affected since 20MB is enough for 100K rows)
+                    "max_memory_usage": "20Mi",
                     "max_untracked_memory": "0",
                 },
             )

From 49b78e0a0c6b7729045ce78874756b69193775a5 Mon Sep 17 00:00:00 2001
From: Max Kainov <max.kainov@clickhouse.com>
Date: Mon, 11 Mar 2024 08:45:58 +0000
Subject: [PATCH 148/443] CI: check cpp and py files style on changes only 
 #do_not_test

---
 tests/ci/report.py                            |  3 ++
 tests/ci/style_check.py                       | 50 ++++++++++++++-----
 .../{check_cpp_docs.sh => check_cpp.sh}       | 34 +++++++------
 utils/check-style/check_docs.sh               | 20 ++++++++
 utils/check-style/check_py.sh                 | 17 ++++---
 .../check-style/process_style_check_result.py | 11 ++--
 6 files changed, 96 insertions(+), 39 deletions(-)
 rename utils/check-style/{check_cpp_docs.sh => check_cpp.sh} (55%)
 create mode 100755 utils/check-style/check_docs.sh

diff --git a/tests/ci/report.py b/tests/ci/report.py
index 669409d1729..a3c9b53637a 100644
--- a/tests/ci/report.py
+++ b/tests/ci/report.py
@@ -288,6 +288,9 @@ class JobReport:
     # if False no GH commit status will be created by CI
     need_commit_status: bool = True
 
+    def __post_init__(self):
+        assert self.status in (SUCCESS, ERROR, FAILURE, PENDING)
+
     @classmethod
     def exist(cls) -> bool:
         return JOB_REPORT_FILE.is_file()
diff --git a/tests/ci/style_check.py b/tests/ci/style_check.py
index d0565e136d3..9f113b6e6f9 100644
--- a/tests/ci/style_check.py
+++ b/tests/ci/style_check.py
@@ -4,13 +4,14 @@ from concurrent.futures import ProcessPoolExecutor
 import csv
 import logging
 import os
+import shutil
 import subprocess
 import sys
 from pathlib import Path
 from typing import List, Tuple
 
 from docker_images_helper import get_docker_image, pull_image
-from env_helper import REPO_COPY, TEMP_PATH
+from env_helper import CI, REPO_COPY, TEMP_PATH
 from git_helper import GIT_PREFIX, git_runner
 from pr_info import PRInfo
 from report import ERROR, FAILURE, SUCCESS, JobReport, TestResults, read_test_results
@@ -126,34 +127,57 @@ def main():
 
     repo_path = Path(REPO_COPY)
     temp_path = Path(TEMP_PATH)
+    if temp_path.is_dir():
+        shutil.rmtree(temp_path)
     temp_path.mkdir(parents=True, exist_ok=True)
 
-    # pr_info = PRInfo()
+    pr_info = PRInfo()
 
     IMAGE_NAME = "clickhouse/style-test"
     image = pull_image(get_docker_image(IMAGE_NAME))
-    cmd_1 = (
+    cmd_cpp = (
         f"docker run -u $(id -u ${{USER}}):$(id -g ${{USER}}) --cap-add=SYS_PTRACE "
         f"--volume={repo_path}:/ClickHouse --volume={temp_path}:/test_output "
         f"--entrypoint= -w/ClickHouse/utils/check-style "
-        f"{image} ./check_cpp_docs.sh"
+        f"{image} ./check_cpp.sh"
     )
-    cmd_2 = (
+    cmd_py = (
         f"docker run -u $(id -u ${{USER}}):$(id -g ${{USER}}) --cap-add=SYS_PTRACE "
         f"--volume={repo_path}:/ClickHouse --volume={temp_path}:/test_output "
         f"--entrypoint= -w/ClickHouse/utils/check-style "
         f"{image} ./check_py.sh"
     )
-    logging.info("Is going to run the command: %s", cmd_1)
-    logging.info("Is going to run the command: %s", cmd_2)
+    cmd_docs = (
+        f"docker run -u $(id -u ${{USER}}):$(id -g ${{USER}}) --cap-add=SYS_PTRACE "
+        f"--volume={repo_path}:/ClickHouse --volume={temp_path}:/test_output "
+        f"--entrypoint= -w/ClickHouse/utils/check-style "
+        f"{image} ./check_docs.sh"
+    )
 
     with ProcessPoolExecutor(max_workers=2) as executor:
-        # Submit commands for execution in parallel
-        future1 = executor.submit(subprocess.run, cmd_1, shell=True)
-        future2 = executor.submit(subprocess.run, cmd_2, shell=True)
-        # Wait for both commands to complete
-        _ = future1.result()
-        _ = future2.result()
+        logging.info("Run docs files check: %s", cmd_docs)
+        future = executor.submit(subprocess.run, cmd_docs, shell=True)
+        # Parallelization  does not make it faster - run subsequently
+        _ = future.result()
+
+        run_cppcheck = True
+        run_pycheck = True
+        if CI and pr_info.number > 0:
+            pr_info.fetch_changed_files()
+            if not any(file.endswith(".py") for file in pr_info.changed_files):
+                run_pycheck = False
+            if all(file.endswith(".py") for file in pr_info.changed_files):
+                run_cppcheck = False
+
+        if run_cppcheck:
+            logging.info("Run source files check: %s", cmd_cpp)
+            future1 = executor.submit(subprocess.run, cmd_cpp, shell=True)
+            _ = future1.result()
+
+        if run_pycheck:
+            logging.info("Run py files check: %s", cmd_py)
+            future2 = executor.submit(subprocess.run, cmd_py, shell=True)
+            _ = future2.result()
 
     # if args.push:
     #     checkout_head(pr_info)
diff --git a/utils/check-style/check_cpp_docs.sh b/utils/check-style/check_cpp.sh
similarity index 55%
rename from utils/check-style/check_cpp_docs.sh
rename to utils/check-style/check_cpp.sh
index 7ad3cede758..ea90d79418c 100755
--- a/utils/check-style/check_cpp_docs.sh
+++ b/utils/check-style/check_cpp.sh
@@ -4,31 +4,35 @@
 
 cd /ClickHouse/utils/check-style || echo -e "failure\tRepo not found" > /test_output/check_status.tsv
 
+start_total=`date +%s`
+
 # FIXME: 30 sec to wait
 # echo "Check duplicates" | ts
 # ./check-duplicate-includes.sh |& tee /test_output/duplicate_includes_output.txt
 
-echo "Check style" | ts
+start=`date +%s`
 ./check-style -n              |& tee /test_output/style_output.txt
-echo "Check typos" | ts
-./check-typos                 |& tee /test_output/typos_output.txt
-echo "Check docs spelling" | ts
-./check-doc-aspell            |& tee /test_output/docs_spelling_output.txt
-echo "Check whitespaces" | ts
+runtime=$((`date +%s`-start))
+echo "Check style. Done. $runtime seconds."
+
+start=`date +%s`
 ./check-whitespaces -n        |& tee /test_output/whitespaces_output.txt
-echo "Check workflows" | ts
+runtime=$((`date +%s`-start))
+echo "Check whitespaces. Done. $runtime seconds."
+
+start=`date +%s`
 ./check-workflows             |& tee /test_output/workflows_output.txt
-echo "Check submodules" | ts
+runtime=$((`date +%s`-start))
+echo "Check workflows. Done. $runtime seconds."
+
+start=`date +%s`
 ./check-submodules            |& tee /test_output/submodules_output.txt
-echo "Check style. Done" | ts
+runtime=$((`date +%s`-start))
+echo "Check submodules. Done. $runtime seconds."
 
 # FIXME: 6 min to wait
 # echo "Check shell scripts with shellcheck" | ts
 # ./shellcheck-run.sh           |& tee /test_output/shellcheck_output.txt
 
-
-# FIXME: move out
-# /process_style_check_result.py || echo -e "failure\tCannot parse results" > /test_output/check_status.tsv
-# echo "Check help for changelog generator works" | ts
-# cd ../changelog || exit 1
-# ./changelog.py -h 2>/dev/null 1>&2
+runtime=$((`date +%s`-start_total))
+echo "Check style total. Done. $runtime seconds."
diff --git a/utils/check-style/check_docs.sh b/utils/check-style/check_docs.sh
new file mode 100755
index 00000000000..78b8a402ea0
--- /dev/null
+++ b/utils/check-style/check_docs.sh
@@ -0,0 +1,20 @@
+#!/bin/bash
+
+# yaml check is not the best one
+
+cd /ClickHouse/utils/check-style || echo -e "failure\tRepo not found" > /test_output/check_status.tsv
+
+start_total=`date +%s`
+
+start=`date +%s`
+./check-typos                 |& tee /test_output/typos_output.txt
+runtime=$((`date +%s`-start))
+echo "Check typos. Done. $runtime seconds."
+
+start=`date +%s`
+./check-doc-aspell            |& tee /test_output/docs_spelling_output.txt
+runtime=$((`date +%s`-start))
+echo "Check docs spelling. Done. $runtime seconds."
+
+runtime=$((`date +%s`-start_total))
+echo "Check Docs, total. Done. $runtime seconds."
diff --git a/utils/check-style/check_py.sh b/utils/check-style/check_py.sh
index 48c02013734..b729cd78124 100755
--- a/utils/check-style/check_py.sh
+++ b/utils/check-style/check_py.sh
@@ -1,17 +1,22 @@
 #!/bin/bash
 
-# yaml check is not the best one
-
 cd /ClickHouse/utils/check-style || echo -e "failure\tRepo not found" > /test_output/check_status.tsv
 
+start_total=`date +%s`
+
 # FIXME: 1 min to wait + head checkout
 # echo "Check python formatting with black" | ts
 # ./check-black -n              |& tee /test_output/black_output.txt
 
-echo "Check pylint" | ts
+start=`date +%s`
 ./check-pylint -n               |& tee /test_output/pylint_output.txt
-echo "Check pylint. Done" | ts
+runtime=$((`date +%s`-start))
+echo "Check pylint. Done. $runtime seconds."
 
-echo "Check python type hinting with mypy" | ts
+start=`date +%s`
 ./check-mypy -n               |& tee /test_output/mypy_output.txt
-echo "Check python type hinting with mypy. Done" | ts
+runtime=$((`date +%s`-start))
+echo "Check python type hinting with mypy. Done. $runtime seconds."
+
+runtime=$((`date +%s`-start_total))
+echo "Check python total. Done. $runtime seconds."
diff --git a/utils/check-style/process_style_check_result.py b/utils/check-style/process_style_check_result.py
index 7980c01dd37..e620d85b9d0 100755
--- a/utils/check-style/process_style_check_result.py
+++ b/utils/check-style/process_style_check_result.py
@@ -13,11 +13,11 @@ def process_result(result_folder):
     description = ""
     test_results = []
     checks = (
-        #"duplicate includes",
-        #"shellcheck",
+        # "duplicate includes",
+        # "shellcheck",
         "style",
         "pylint",
-        #"black",
+        # "black",
         "mypy",
         "typos",
         "whitespaces",
@@ -30,8 +30,7 @@ def process_result(result_folder):
         out_file = name.replace(" ", "_") + "_output.txt"
         full_path = os.path.join(result_folder, out_file)
         if not os.path.exists(full_path):
-            logging.info("No %s check log on path %s", name, full_path)
-            return "exception", f"No {name} check log", []
+            test_results.append((f"Check {name}", "SKIPPED"))
         elif os.stat(full_path).st_size != 0:
             description += f"Check {name} failed. "
             test_results.append((f"Check {name}", "FAIL"))
@@ -42,6 +41,8 @@ def process_result(result_folder):
     if not description:
         description += "Style check success"
 
+    assert test_results, "No single style-check output found"
+
     return status, description, test_results
 
 

From 61c3d917ae82bb6ae14cea3a4a7c36c19f6d3b22 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <benjamin.antal@clickhouse.com>
Date: Mon, 11 Mar 2024 10:33:09 +0000
Subject: [PATCH 149/443] Use `boost::algorithm::join`

---
 src/Common/FunctionDocumentation.cpp | 12 +++---------
 1 file changed, 3 insertions(+), 9 deletions(-)

diff --git a/src/Common/FunctionDocumentation.cpp b/src/Common/FunctionDocumentation.cpp
index 0dc5b48f9d1..7b554539a4f 100644
--- a/src/Common/FunctionDocumentation.cpp
+++ b/src/Common/FunctionDocumentation.cpp
@@ -1,5 +1,7 @@
 #include <Common/FunctionDocumentation.h>
 
+#include <boost/algorithm/string.hpp>
+
 namespace DB
 {
 
@@ -31,15 +33,7 @@ std::string FunctionDocumentation::examplesAsString() const
 
 std::string FunctionDocumentation::categoriesAsString() const
 {
-    if (categories.empty())
-        return "";
-
-    auto it = categories.begin();
-    std::string res = *it;
-    ++it;
-    for (; it != categories.end(); ++it)
-        res += ", " + *it;
-    return res;
+    return boost::algorithm::join(categories, ", ");
 }
 
 }

From ecb11005e34948a365555da1e2271e9da31d5074 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Mon, 11 Mar 2024 10:56:26 +0000
Subject: [PATCH 150/443] List clang-tidy checks as list instead of a string

The string representation will be deprecated at some point:
https://reviews.llvm.org/D147876
---
 .clang-tidy | 195 ++++++++++++++++++++++++++--------------------------
 1 file changed, 97 insertions(+), 98 deletions(-)

diff --git a/.clang-tidy b/.clang-tidy
index 0dacf813c7e..4aeb38ca409 100644
--- a/.clang-tidy
+++ b/.clang-tidy
@@ -5,128 +5,127 @@
 # a) the new check is not controversial (this includes many checks in readability-* and google-*) or
 # b) too noisy (checks with > 100 new warnings are considered noisy, this includes e.g. cppcoreguidelines-*).
 
-# TODO: Once clang(-tidy) 17 is the minimum, we can convert this list to YAML
-# See https://releases.llvm.org/17.0.1/tools/clang/tools/extra/docs/ReleaseNotes.html#improvements-to-clang-tidy
-
 # TODO Let clang-tidy check headers in further directories
 #      --> HeaderFilterRegex: '^.*/(src|base|programs|utils)/.*(h|hpp)$'
 HeaderFilterRegex: '^.*/(base|programs|utils)/.*(h|hpp)$'
 
-Checks: '*,
-    -abseil-*,
+Checks: [
+  '*',
 
-    -altera-*,
+  '-abseil-*',
 
-    -android-*,
+  '-altera-*',
 
-    -bugprone-assignment-in-if-condition,
-    -bugprone-branch-clone,
-    -bugprone-easily-swappable-parameters,
-    -bugprone-exception-escape,
-    -bugprone-implicit-widening-of-multiplication-result,
-    -bugprone-narrowing-conversions,
-    -bugprone-not-null-terminated-result,
-    -bugprone-reserved-identifier, # useful but too slow, TODO retry when https://reviews.llvm.org/rG1c282052624f9d0bd273bde0b47b30c96699c6c7 is merged
-    -bugprone-unchecked-optional-access,
+  '-android-*',
 
-    -cert-dcl16-c,
-    -cert-dcl37-c,
-    -cert-dcl51-cpp,
-    -cert-err58-cpp,
-    -cert-msc32-c,
-    -cert-msc51-cpp,
-    -cert-oop54-cpp,
-    -cert-oop57-cpp,
+  '-bugprone-assignment-in-if-condition',
+  '-bugprone-branch-clone',
+  '-bugprone-easily-swappable-parameters',
+  '-bugprone-exception-escape',
+  '-bugprone-implicit-widening-of-multiplication-result',
+  '-bugprone-narrowing-conversions',
+  '-bugprone-not-null-terminated-result',
+  '-bugprone-reserved-identifier', # useful but too slow, TODO retry when https://reviews.llvm.org/rG1c282052624f9d0bd273bde0b47b30c96699c6c7 is merged
+  '-bugprone-unchecked-optional-access',
 
-    -clang-analyzer-unix.Malloc,
+  '-cert-dcl16-c',
+  '-cert-dcl37-c',
+  '-cert-dcl51-cpp',
+  '-cert-err58-cpp',
+  '-cert-msc32-c',
+  '-cert-msc51-cpp',
+  '-cert-oop54-cpp',
+  '-cert-oop57-cpp',
 
-    -cppcoreguidelines-*, # impractical in a codebase as large as ClickHouse, also slow
+  '-clang-analyzer-unix.Malloc',
 
-    -darwin-*,
+  '-cppcoreguidelines-*', # impractical in a codebase as large as ClickHouse, also slow
 
-    -fuchsia-*,
+  '-darwin-*',
 
-    -google-build-using-namespace,
-    -google-readability-braces-around-statements,
-    -google-readability-casting,
-    -google-readability-function-size,
-    -google-readability-namespace-comments,
-    -google-readability-todo,
+  '-fuchsia-*',
 
-    -hicpp-avoid-c-arrays,
-    -hicpp-avoid-goto,
-    -hicpp-braces-around-statements,
-    -hicpp-explicit-conversions,
-    -hicpp-function-size,
-    -hicpp-member-init,
-    -hicpp-move-const-arg,
-    -hicpp-multiway-paths-covered,
-    -hicpp-named-parameter,
-    -hicpp-no-array-decay,
-    -hicpp-no-assembler,
-    -hicpp-no-malloc,
-    -hicpp-signed-bitwise,
-    -hicpp-special-member-functions,
-    -hicpp-uppercase-literal-suffix,
-    -hicpp-use-auto,
-    -hicpp-use-emplace,
-    -hicpp-vararg,
+  '-google-build-using-namespace',
+  '-google-readability-braces-around-statements',
+  '-google-readability-casting',
+  '-google-readability-function-size',
+  '-google-readability-namespace-comments',
+  '-google-readability-todo',
 
-    -linuxkernel-*,
+  '-hicpp-avoid-c-arrays',
+  '-hicpp-avoid-goto',
+  '-hicpp-braces-around-statements',
+  '-hicpp-explicit-conversions',
+  '-hicpp-function-size',
+  '-hicpp-member-init',
+  '-hicpp-move-const-arg',
+  '-hicpp-multiway-paths-covered',
+  '-hicpp-named-parameter',
+  '-hicpp-no-array-decay',
+  '-hicpp-no-assembler',
+  '-hicpp-no-malloc',
+  '-hicpp-signed-bitwise',
+  '-hicpp-special-member-functions',
+  '-hicpp-uppercase-literal-suffix',
+  '-hicpp-use-auto',
+  '-hicpp-use-emplace',
+  '-hicpp-vararg',
 
-    -llvm-*,
+  '-linuxkernel-*',
 
-    -llvmlibc-*,
+  '-llvm-*',
 
-    -openmp-*,
+  '-llvmlibc-*',
 
-    -misc-const-correctness,
-    -misc-include-cleaner, # useful but far too many occurrences
-    -misc-no-recursion,
-    -misc-non-private-member-variables-in-classes,
-    -misc-confusable-identifiers, # useful but slooow
-    -misc-use-anonymous-namespace,
+  '-openmp-*',
 
-    -modernize-avoid-c-arrays,
-    -modernize-concat-nested-namespaces,
-    -modernize-macro-to-enum,
-    -modernize-pass-by-value,
-    -modernize-return-braced-init-list,
-    -modernize-use-auto,
-    -modernize-use-default-member-init,
-    -modernize-use-emplace,
-    -modernize-use-nodiscard,
-    -modernize-use-override,
-    -modernize-use-trailing-return-type,
+  '-misc-const-correctness',
+  '-misc-include-cleaner', # useful but far too many occurrences
+  '-misc-no-recursion',
+  '-misc-non-private-member-variables-in-classes',
+  '-misc-confusable-identifiers', # useful but slooo
+  '-misc-use-anonymous-namespace',
 
-    -performance-inefficient-string-concatenation,
-    -performance-no-int-to-ptr,
-    -performance-avoid-endl,
-    -performance-unnecessary-value-param,
+  '-modernize-avoid-c-arrays',
+  '-modernize-concat-nested-namespaces',
+  '-modernize-macro-to-enum',
+  '-modernize-pass-by-value',
+  '-modernize-return-braced-init-list',
+  '-modernize-use-auto',
+  '-modernize-use-default-member-init',
+  '-modernize-use-emplace',
+  '-modernize-use-nodiscard',
+  '-modernize-use-override',
+  '-modernize-use-trailing-return-type',
 
-    -portability-simd-intrinsics,
+  '-performance-inefficient-string-concatenation',
+  '-performance-no-int-to-ptr',
+  '-performance-avoid-endl',
+  '-performance-unnecessary-value-param',
 
-    -readability-avoid-unconditional-preprocessor-if,
-    -readability-braces-around-statements,
-    -readability-convert-member-functions-to-static,
-    -readability-else-after-return,
-    -readability-function-cognitive-complexity,
-    -readability-function-size,
-    -readability-identifier-length,
-    -readability-identifier-naming, # useful but too slow
-    -readability-implicit-bool-conversion,
-    -readability-isolate-declaration,
-    -readability-magic-numbers,
-    -readability-named-parameter,
-    -readability-redundant-declaration,
-    -readability-simplify-boolean-expr,
-    -readability-static-accessed-through-instance,
-    -readability-suspicious-call-argument,
-    -readability-uppercase-literal-suffix,
-    -readability-use-anyofallof,
+  '-portability-simd-intrinsics',
 
-    -zircon-*,
-'
+  '-readability-avoid-unconditional-preprocessor-if',
+  '-readability-braces-around-statements',
+  '-readability-convert-member-functions-to-static',
+  '-readability-else-after-return',
+  '-readability-function-cognitive-complexity',
+  '-readability-function-size',
+  '-readability-identifier-length',
+  '-readability-identifier-naming', # useful but too slow
+  '-readability-implicit-bool-conversion',
+  '-readability-isolate-declaration',
+  '-readability-magic-numbers',
+  '-readability-named-parameter',
+  '-readability-redundant-declaration',
+  '-readability-simplify-boolean-expr',
+  '-readability-static-accessed-through-instance',
+  '-readability-suspicious-call-argument',
+  '-readability-uppercase-literal-suffix',
+  '-readability-use-anyofallof',
+
+  '-zircon-*'
+]
 
 WarningsAsErrors: '*'
 

From e5e84419aff0f559bc545737bfdc0518a732f7ff Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Sun, 10 Mar 2024 14:29:18 +0000
Subject: [PATCH 151/443] Fix clang-tidy-s

---
 contrib/libmetrohash/src/metrohash128.h          |  3 +++
 src/Access/AccessControl.h                       | 10 +++++-----
 src/Access/IAccessStorage.cpp                    |  2 +-
 src/Access/IAccessStorage.h                      |  2 +-
 src/Common/Arena.h                               |  4 +---
 src/Common/DNSResolver.cpp                       |  2 +-
 src/Common/DNSResolver.h                         |  2 +-
 src/Common/DateLUTImpl.h                         |  2 +-
 src/Common/MultiVersion.h                        |  4 ++--
 src/Common/PODArray.h                            |  6 +++---
 src/Common/SipHash.h                             |  2 +-
 src/Common/TransactionID.h                       |  2 +-
 src/Common/ZooKeeper/IKeeper.cpp                 |  8 ++++----
 src/Common/ZooKeeper/IKeeper.h                   | 16 ++++++++--------
 src/Common/logger_useful.h                       | 16 ++++++++--------
 src/Core/PostgreSQL/insertPostgreSQLValue.cpp    |  4 ++--
 src/Core/PostgreSQL/insertPostgreSQLValue.h      |  4 ++--
 src/Core/Settings.h                              |  2 ++
 src/Dictionaries/CacheDictionary.cpp             |  4 ++--
 src/Dictionaries/CacheDictionary.h               |  2 +-
 .../GeodataProviders/IHierarchiesProvider.h      |  2 +-
 src/Dictionaries/RegExpTreeDictionary.cpp        |  2 +-
 src/Dictionaries/RegExpTreeDictionary.h          |  2 +-
 src/Functions/IFunction.h                        |  4 ----
 src/IO/ReadSettings.h                            |  1 +
 src/Interpreters/AsynchronousInsertQueue.cpp     |  2 +-
 src/Interpreters/AsynchronousInsertQueue.h       |  2 +-
 src/Interpreters/Context.h                       |  4 ++--
 src/Interpreters/IExternalLoadable.h             |  2 +-
 src/Interpreters/ProcessList.h                   |  2 +-
 src/Processors/Chunk.h                           |  2 +-
 .../Algorithms/AggregatingSortedAlgorithm.cpp    |  2 +-
 src/Processors/Port.h                            |  2 +-
 src/Processors/TTL/TTLUpdateInfoAlgorithm.cpp    |  4 ++--
 src/Processors/TTL/TTLUpdateInfoAlgorithm.h      |  4 ++--
 src/Storages/StorageInMemoryMetadata.h           |  4 ++--
 36 files changed, 69 insertions(+), 69 deletions(-)

diff --git a/contrib/libmetrohash/src/metrohash128.h b/contrib/libmetrohash/src/metrohash128.h
index 2dbb6ca5a8a..f507c917caf 100644
--- a/contrib/libmetrohash/src/metrohash128.h
+++ b/contrib/libmetrohash/src/metrohash128.h
@@ -17,6 +17,8 @@
 #ifndef METROHASH_METROHASH_128_H
 #define METROHASH_METROHASH_128_H
 
+// NOLINTBEGIN(readability-avoid-const-params-in-decls)
+
 #include <stdint.h>
 
 class MetroHash128
@@ -68,5 +70,6 @@ private:
 void metrohash128_1(const uint8_t * key, uint64_t len, uint32_t seed, uint8_t * out);
 void metrohash128_2(const uint8_t * key, uint64_t len, uint32_t seed, uint8_t * out);
 
+// NOLINTEND(readability-avoid-const-params-in-decls)
 
 #endif // #ifndef METROHASH_METROHASH_128_H
diff --git a/src/Access/AccessControl.h b/src/Access/AccessControl.h
index 55ea4e4f717..1af74e02fb7 100644
--- a/src/Access/AccessControl.h
+++ b/src/Access/AccessControl.h
@@ -133,20 +133,20 @@ public:
     /// This function also enables custom prefixes to be used.
     void setCustomSettingsPrefixes(const Strings & prefixes);
     void setCustomSettingsPrefixes(const String & comma_separated_prefixes);
-    bool isSettingNameAllowed(const std::string_view name) const;
-    void checkSettingNameIsAllowed(const std::string_view name) const;
+    bool isSettingNameAllowed(std::string_view name) const;
+    void checkSettingNameIsAllowed(std::string_view name) const;
 
     /// Allows implicit user creation without password (by default it's allowed).
     /// In other words, allow 'CREATE USER' queries without 'IDENTIFIED WITH' clause.
-    void setImplicitNoPasswordAllowed(const bool allow_implicit_no_password_);
+    void setImplicitNoPasswordAllowed(bool allow_implicit_no_password_);
     bool isImplicitNoPasswordAllowed() const;
 
     /// Allows users without password (by default it's allowed).
-    void setNoPasswordAllowed(const bool allow_no_password_);
+    void setNoPasswordAllowed(bool allow_no_password_);
     bool isNoPasswordAllowed() const;
 
     /// Allows users with plaintext password (by default it's allowed).
-    void setPlaintextPasswordAllowed(const bool allow_plaintext_password_);
+    void setPlaintextPasswordAllowed(bool allow_plaintext_password_);
     bool isPlaintextPasswordAllowed() const;
 
     /// Default password type when the user does not specify it.
diff --git a/src/Access/IAccessStorage.cpp b/src/Access/IAccessStorage.cpp
index fbe9e231002..1d6b8d99cd5 100644
--- a/src/Access/IAccessStorage.cpp
+++ b/src/Access/IAccessStorage.cpp
@@ -616,7 +616,7 @@ UUID IAccessStorage::generateRandomID()
 }
 
 
-void IAccessStorage::clearConflictsInEntitiesList(std::vector<std::pair<UUID, AccessEntityPtr>> & entities, const LoggerPtr log_)
+void IAccessStorage::clearConflictsInEntitiesList(std::vector<std::pair<UUID, AccessEntityPtr>> & entities, LoggerPtr log_)
 {
     std::unordered_map<UUID, size_t> positions_by_id;
     std::unordered_map<std::string_view, size_t> positions_by_type_and_name[static_cast<size_t>(AccessEntityType::MAX)];
diff --git a/src/Access/IAccessStorage.h b/src/Access/IAccessStorage.h
index ebb5a39cdf0..ad78bf92e02 100644
--- a/src/Access/IAccessStorage.h
+++ b/src/Access/IAccessStorage.h
@@ -228,7 +228,7 @@ protected:
     static UUID generateRandomID();
     LoggerPtr getLogger() const;
     static String formatEntityTypeWithName(AccessEntityType type, const String & name) { return AccessEntityTypeInfo::get(type).formatEntityNameWithType(name); }
-    static void clearConflictsInEntitiesList(std::vector<std::pair<UUID, AccessEntityPtr>> & entities, const LoggerPtr log_);
+    static void clearConflictsInEntitiesList(std::vector<std::pair<UUID, AccessEntityPtr>> & entities, LoggerPtr log_);
     [[noreturn]] void throwNotFound(const UUID & id) const;
     [[noreturn]] void throwNotFound(AccessEntityType type, const String & name) const;
     [[noreturn]] static void throwBadCast(const UUID & id, AccessEntityType type, const String & name, AccessEntityType required_type);
diff --git a/src/Common/Arena.h b/src/Common/Arena.h
index cb26397844b..ba5b9ea9205 100644
--- a/src/Common/Arena.h
+++ b/src/Common/Arena.h
@@ -47,9 +47,7 @@ private:
 
         std::unique_ptr<MemoryChunk> prev;
 
-        MemoryChunk()
-        {
-        }
+        MemoryChunk() = default;
 
         void swap(MemoryChunk & other)
         {
diff --git a/src/Common/DNSResolver.cpp b/src/Common/DNSResolver.cpp
index e36e1483da8..4b577a251af 100644
--- a/src/Common/DNSResolver.cpp
+++ b/src/Common/DNSResolver.cpp
@@ -297,7 +297,7 @@ void DNSResolver::setDisableCacheFlag(bool is_disabled)
     impl->disable_cache = is_disabled;
 }
 
-void DNSResolver::setCacheMaxEntries(const UInt64 cache_max_entries)
+void DNSResolver::setCacheMaxEntries(UInt64 cache_max_entries)
 {
     impl->cache_address.setMaxSizeInBytes(cache_max_entries);
     impl->cache_host.setMaxSizeInBytes(cache_max_entries);
diff --git a/src/Common/DNSResolver.h b/src/Common/DNSResolver.h
index e3030e51a96..1ddd9d3b991 100644
--- a/src/Common/DNSResolver.h
+++ b/src/Common/DNSResolver.h
@@ -56,7 +56,7 @@ public:
     void setDisableCacheFlag(bool is_disabled = true);
 
     /// Set a limit of entries in cache
-    void setCacheMaxEntries(const UInt64 cache_max_entries);
+    void setCacheMaxEntries(UInt64 cache_max_entries);
 
     /// Drops all caches
     void dropCache();
diff --git a/src/Common/DateLUTImpl.h b/src/Common/DateLUTImpl.h
index 7bf66c0504a..4087e77d588 100644
--- a/src/Common/DateLUTImpl.h
+++ b/src/Common/DateLUTImpl.h
@@ -255,7 +255,7 @@ private:
 
     static LUTIndex toLUTIndex(ExtendedDayNum d)
     {
-        return normalizeLUTIndex(static_cast<Int64>(d + daynum_offset_epoch));
+        return normalizeLUTIndex(static_cast<Int64>(d) + daynum_offset_epoch);
     }
 
     LUTIndex toLUTIndex(Time t) const
diff --git a/src/Common/MultiVersion.h b/src/Common/MultiVersion.h
index 8f488f9fcbc..680e224f869 100644
--- a/src/Common/MultiVersion.h
+++ b/src/Common/MultiVersion.h
@@ -41,9 +41,9 @@ public:
     }
 
     /// There is no copy constructor because only one MultiVersion should own the same object.
-    MultiVersion(MultiVersion && src) { *this = std::move(src); }
+    MultiVersion(MultiVersion && src) { *this = std::move(src); } /// NOLINT
 
-    MultiVersion & operator=(MultiVersion && src)
+    MultiVersion & operator=(MultiVersion && src) /// NOLINT
     {
         if (this != &src)
         {
diff --git a/src/Common/PODArray.h b/src/Common/PODArray.h
index 1a4047a2588..af863e01fb2 100644
--- a/src/Common/PODArray.h
+++ b/src/Common/PODArray.h
@@ -25,7 +25,7 @@
   */
 template <typename T, typename U>
 constexpr bool memcpy_can_be_used_for_assignment = std::is_same_v<T, U>
-    || (std::is_integral_v<T> && std::is_integral_v<U> && sizeof(T) == sizeof(U));
+    || (std::is_integral_v<T> && std::is_integral_v<U> && sizeof(T) == sizeof(U)); /// NOLINT(misc-redundant-expression)
 
 namespace DB
 {
@@ -558,7 +558,7 @@ public:
     }
 
     template <typename... TAllocatorParams>
-    void swap(PODArray & rhs, TAllocatorParams &&... allocator_params)
+    void swap(PODArray & rhs, TAllocatorParams &&... allocator_params) /// NOLINT(performance-noexcept-swap)
     {
 #ifndef NDEBUG
         this->unprotect();
@@ -756,7 +756,7 @@ public:
 };
 
 template <typename T, size_t initial_bytes, typename TAllocator, size_t pad_right_, size_t pad_left_>
-void swap(PODArray<T, initial_bytes, TAllocator, pad_right_, pad_left_> & lhs, PODArray<T, initial_bytes, TAllocator, pad_right_, pad_left_> & rhs)
+void swap(PODArray<T, initial_bytes, TAllocator, pad_right_, pad_left_> & lhs, PODArray<T, initial_bytes, TAllocator, pad_right_, pad_left_> & rhs) /// NOLINT
 {
     lhs.swap(rhs);
 }
diff --git a/src/Common/SipHash.h b/src/Common/SipHash.h
index 729fb76a573..c89ee2c9d90 100644
--- a/src/Common/SipHash.h
+++ b/src/Common/SipHash.h
@@ -149,7 +149,7 @@ public:
 
         /// Pad the remainder, which is missing up to an 8-byte word.
         current_word = 0;
-        switch (end - data)
+        switch (end - data) /// NOLINT(bugprone-switch-missing-default-case)
         {
             case 7: current_bytes[CURRENT_BYTES_IDX(6)] = data[6]; [[fallthrough]];
             case 6: current_bytes[CURRENT_BYTES_IDX(5)] = data[5]; [[fallthrough]];
diff --git a/src/Common/TransactionID.h b/src/Common/TransactionID.h
index 3ab86f7589c..97d0072bc14 100644
--- a/src/Common/TransactionID.h
+++ b/src/Common/TransactionID.h
@@ -16,7 +16,7 @@ class MergeTreeTransaction;
 /// or transaction object is not needed and not passed intentionally.
 #ifndef NO_TRANSACTION_PTR
 #define NO_TRANSACTION_PTR std::shared_ptr<MergeTreeTransaction>(nullptr)
-#define NO_TRANSACTION_RAW static_cast<MergeTreeTransaction *>(nullptr)
+#define NO_TRANSACTION_RAW static_cast<MergeTreeTransaction *>(nullptr) /// NOLINT(bugprone-macro-parentheses)
 #endif
 
 /// Commit Sequence Number
diff --git a/src/Common/ZooKeeper/IKeeper.cpp b/src/Common/ZooKeeper/IKeeper.cpp
index 6c47ea68b84..7d2602bde1e 100644
--- a/src/Common/ZooKeeper/IKeeper.cpp
+++ b/src/Common/ZooKeeper/IKeeper.cpp
@@ -23,7 +23,7 @@ namespace ProfileEvents
 namespace Coordination
 {
 
-void Exception::incrementErrorMetrics(const Error code_)
+void Exception::incrementErrorMetrics(Error code_)
 {
     if (Coordination::isUserError(code_))
         ProfileEvents::increment(ProfileEvents::ZooKeeperUserExceptions);
@@ -33,14 +33,14 @@ void Exception::incrementErrorMetrics(const Error code_)
         ProfileEvents::increment(ProfileEvents::ZooKeeperOtherExceptions);
 }
 
-Exception::Exception(const std::string & msg, const Error code_, int)
+Exception::Exception(const std::string & msg, Error code_, int)
     : DB::Exception(msg, DB::ErrorCodes::KEEPER_EXCEPTION)
     , code(code_)
 {
     incrementErrorMetrics(code);
 }
 
-Exception::Exception(PreformattedMessage && msg, const Error code_)
+Exception::Exception(PreformattedMessage && msg, Error code_)
     : DB::Exception(std::move(msg), DB::ErrorCodes::KEEPER_EXCEPTION)
     , code(code_)
 {
@@ -48,7 +48,7 @@ Exception::Exception(PreformattedMessage && msg, const Error code_)
     incrementErrorMetrics(code);
 }
 
-Exception::Exception(const Error code_)
+Exception::Exception(Error code_)
     : Exception(code_, "Coordination error: {}", errorMessage(code_))
 {
 }
diff --git a/src/Common/ZooKeeper/IKeeper.h b/src/Common/ZooKeeper/IKeeper.h
index c7b902ea03a..ec49c94808e 100644
--- a/src/Common/ZooKeeper/IKeeper.h
+++ b/src/Common/ZooKeeper/IKeeper.h
@@ -466,13 +466,13 @@ class Exception : public DB::Exception
 {
 private:
     /// Delegate constructor, used to minimize repetition; last parameter used for overload resolution.
-    Exception(const std::string & msg, const Error code_, int); /// NOLINT
-    Exception(PreformattedMessage && msg, const Error code_);
+    Exception(const std::string & msg, Error code_, int); /// NOLINT
+    Exception(PreformattedMessage && msg, Error code_);
 
     /// Message must be a compile-time constant
     template <typename T>
     requires std::is_convertible_v<T, String>
-    Exception(T && message, const Error code_) : DB::Exception(std::forward<T>(message), DB::ErrorCodes::KEEPER_EXCEPTION, /* remote_= */ false), code(code_)
+    Exception(T && message, Error code_) : DB::Exception(std::forward<T>(message), DB::ErrorCodes::KEEPER_EXCEPTION, /* remote_= */ false), code(code_)
     {
         incrementErrorMetrics(code);
     }
@@ -480,23 +480,23 @@ private:
     static void incrementErrorMetrics(Error code_);
 
 public:
-    explicit Exception(const Error code_); /// NOLINT
+    explicit Exception(Error code_); /// NOLINT
     Exception(const Exception & exc);
 
     template <typename... Args>
-    Exception(const Error code_, FormatStringHelper<Args...> fmt, Args &&... args)
+    Exception(Error code_, FormatStringHelper<Args...> fmt, Args &&... args)
         : DB::Exception(DB::ErrorCodes::KEEPER_EXCEPTION, std::move(fmt), std::forward<Args>(args)...)
         , code(code_)
     {
         incrementErrorMetrics(code);
     }
 
-    inline static Exception createDeprecated(const std::string & msg, const Error code_)
+    inline static Exception createDeprecated(const std::string & msg, Error code_)
     {
         return Exception(msg, code_, 0);
     }
 
-    inline static Exception fromPath(const Error code_, const std::string & path)
+    inline static Exception fromPath(Error code_, const std::string & path)
     {
         return Exception(code_, "Coordination error: {}, path {}", errorMessage(code_), path);
     }
@@ -504,7 +504,7 @@ public:
     /// Message must be a compile-time constant
     template <typename T>
     requires std::is_convertible_v<T, String>
-    inline static Exception fromMessage(const Error code_, T && message)
+    inline static Exception fromMessage(Error code_, T && message)
     {
         return Exception(std::forward<T>(message), code_);
     }
diff --git a/src/Common/logger_useful.h b/src/Common/logger_useful.h
index 8e78e93e198..013b35e695e 100644
--- a/src/Common/logger_useful.h
+++ b/src/Common/logger_useful.h
@@ -19,14 +19,14 @@ namespace Poco { class Logger; }
 
 using LogSeriesLimiterPtr = std::shared_ptr<LogSeriesLimiter>;
 
-namespace
+namespace impl
 {
-    [[maybe_unused]] LoggerPtr getLoggerHelper(const LoggerPtr & logger) { return logger; }
-    [[maybe_unused]] LoggerPtr getLoggerHelper(const AtomicLogger & logger) { return logger.load(); }
-    [[maybe_unused]] const ::Poco::Logger * getLoggerHelper(const ::Poco::Logger * logger) { return logger; }
-    [[maybe_unused]] std::unique_ptr<LogToStrImpl> getLoggerHelper(std::unique_ptr<LogToStrImpl> && logger) { return logger; }
-    [[maybe_unused]] std::unique_ptr<LogFrequencyLimiterIml> getLoggerHelper(std::unique_ptr<LogFrequencyLimiterIml> && logger) { return logger; }
-    [[maybe_unused]] LogSeriesLimiterPtr getLoggerHelper(LogSeriesLimiterPtr & logger) { return logger; }
+    [[maybe_unused]] inline LoggerPtr getLoggerHelper(const LoggerPtr & logger) { return logger; }
+    [[maybe_unused]] inline LoggerPtr getLoggerHelper(const AtomicLogger & logger) { return logger.load(); }
+    [[maybe_unused]] inline const ::Poco::Logger * getLoggerHelper(const ::Poco::Logger * logger) { return logger; }
+    [[maybe_unused]] inline std::unique_ptr<LogToStrImpl> getLoggerHelper(std::unique_ptr<LogToStrImpl> && logger) { return logger; }
+    [[maybe_unused]] inline std::unique_ptr<LogFrequencyLimiterIml> getLoggerHelper(std::unique_ptr<LogFrequencyLimiterIml> && logger) { return logger; }
+    [[maybe_unused]] inline LogSeriesLimiterPtr getLoggerHelper(LogSeriesLimiterPtr & logger) { return logger; }
 }
 
 #define LOG_IMPL_FIRST_ARG(X, ...) X
@@ -65,7 +65,7 @@ namespace
 
 #define LOG_IMPL(logger, priority, PRIORITY, ...) do                                                                \
 {                                                                                                                   \
-    auto _logger = ::getLoggerHelper(logger);                                                                             \
+    auto _logger = ::impl::getLoggerHelper(logger);                                                                 \
     const bool _is_clients_log = (DB::CurrentThread::getGroup() != nullptr) &&                                      \
         (DB::CurrentThread::get().getClientLogsLevel() >= (priority));                                              \
     if (!_is_clients_log && !_logger->is((PRIORITY)))                                                               \
diff --git a/src/Core/PostgreSQL/insertPostgreSQLValue.cpp b/src/Core/PostgreSQL/insertPostgreSQLValue.cpp
index aa60bdee28a..b507b300769 100644
--- a/src/Core/PostgreSQL/insertPostgreSQLValue.cpp
+++ b/src/Core/PostgreSQL/insertPostgreSQLValue.cpp
@@ -36,7 +36,7 @@ void insertDefaultPostgreSQLValue(IColumn & column, const IColumn & sample_colum
 
 void insertPostgreSQLValue(
         IColumn & column, std::string_view value,
-        const ExternalResultDescription::ValueType type, const DataTypePtr data_type,
+        ExternalResultDescription::ValueType type, DataTypePtr data_type,
         const std::unordered_map<size_t, PostgreSQLArrayInfo> & array_info, size_t idx)
 {
     switch (type)
@@ -170,7 +170,7 @@ void insertPostgreSQLValue(
 
 
 void preparePostgreSQLArrayInfo(
-        std::unordered_map<size_t, PostgreSQLArrayInfo> & array_info, size_t column_idx, const DataTypePtr data_type)
+        std::unordered_map<size_t, PostgreSQLArrayInfo> & array_info, size_t column_idx, DataTypePtr data_type)
 {
     const auto * array_type = typeid_cast<const DataTypeArray *>(data_type.get());
     auto nested = array_type->getNestedType();
diff --git a/src/Core/PostgreSQL/insertPostgreSQLValue.h b/src/Core/PostgreSQL/insertPostgreSQLValue.h
index 3bc83292b96..bfb85422aa1 100644
--- a/src/Core/PostgreSQL/insertPostgreSQLValue.h
+++ b/src/Core/PostgreSQL/insertPostgreSQLValue.h
@@ -22,11 +22,11 @@ struct PostgreSQLArrayInfo
 
 void insertPostgreSQLValue(
         IColumn & column, std::string_view value,
-        const ExternalResultDescription::ValueType type, const DataTypePtr data_type,
+        ExternalResultDescription::ValueType type, DataTypePtr data_type,
         const std::unordered_map<size_t, PostgreSQLArrayInfo> & array_info, size_t idx);
 
 void preparePostgreSQLArrayInfo(
-        std::unordered_map<size_t, PostgreSQLArrayInfo> & array_info, size_t column_idx, const DataTypePtr data_type);
+        std::unordered_map<size_t, PostgreSQLArrayInfo> & array_info, size_t column_idx, DataTypePtr data_type);
 
 void insertDefaultPostgreSQLValue(IColumn & column, const IColumn & sample_column);
 
diff --git a/src/Core/Settings.h b/src/Core/Settings.h
index d70a6cf51c5..c41db9d2141 100644
--- a/src/Core/Settings.h
+++ b/src/Core/Settings.h
@@ -1192,6 +1192,7 @@ class IColumn;
     FORMAT_FACTORY_SETTINGS(M, ALIAS)  \
     OBSOLETE_FORMAT_SETTINGS(M, ALIAS) \
 
+/// NOLINTNEXTLINE(clang-analyzer-optin.performance.Padding)
 DECLARE_SETTINGS_TRAITS_ALLOW_CUSTOM_SETTINGS(SettingsTraits, LIST_OF_SETTINGS)
 
 
@@ -1236,6 +1237,7 @@ private:
 /*
  * User-specified file format settings for File and URL engines.
  */
+/// NOLINTNEXTLINE(clang-analyzer-optin.performance.Padding)
 DECLARE_SETTINGS_TRAITS(FormatFactorySettingsTraits, LIST_OF_ALL_FORMAT_SETTINGS)
 
 struct FormatFactorySettings : public BaseSettings<FormatFactorySettingsTraits>
diff --git a/src/Dictionaries/CacheDictionary.cpp b/src/Dictionaries/CacheDictionary.cpp
index 8444042db9e..6e9b09f8919 100644
--- a/src/Dictionaries/CacheDictionary.cpp
+++ b/src/Dictionaries/CacheDictionary.cpp
@@ -423,7 +423,7 @@ MutableColumns CacheDictionary<dictionary_key_type>::aggregateColumnsInOrderOfKe
     const DictionaryStorageFetchRequest & request,
     const MutableColumns & fetched_columns,
     const PaddedPODArray<KeyState> & key_index_to_state,
-    IColumn::Filter * const default_mask) const
+    IColumn::Filter * default_mask) const
 {
     MutableColumns aggregated_columns = request.makeAttributesResultColumns();
 
@@ -473,7 +473,7 @@ MutableColumns CacheDictionary<dictionary_key_type>::aggregateColumns(
         const PaddedPODArray<KeyState> & key_index_to_fetched_columns_from_storage_result,
         const MutableColumns & fetched_columns_during_update,
         const HashMap<KeyType, size_t> & found_keys_to_fetched_columns_during_update_index,
-        IColumn::Filter * const default_mask) const
+        IColumn::Filter * default_mask) const
 {
     /**
     * Aggregation of columns fetched from storage and from source during update.
diff --git a/src/Dictionaries/CacheDictionary.h b/src/Dictionaries/CacheDictionary.h
index 8897fb40fa9..c02fb91c60e 100644
--- a/src/Dictionaries/CacheDictionary.h
+++ b/src/Dictionaries/CacheDictionary.h
@@ -162,7 +162,7 @@ private:
         const DictionaryStorageFetchRequest & request,
         const MutableColumns & fetched_columns,
         const PaddedPODArray<KeyState> & key_index_to_state,
-        IColumn::Filter * const default_mask = nullptr) const;
+        IColumn::Filter * default_mask = nullptr) const;
 
     MutableColumns aggregateColumns(
         const PaddedPODArray<KeyType> & keys,
diff --git a/src/Dictionaries/Embedded/GeodataProviders/IHierarchiesProvider.h b/src/Dictionaries/Embedded/GeodataProviders/IHierarchiesProvider.h
index 68ab0fdca2d..a4b88127786 100644
--- a/src/Dictionaries/Embedded/GeodataProviders/IHierarchiesProvider.h
+++ b/src/Dictionaries/Embedded/GeodataProviders/IHierarchiesProvider.h
@@ -14,7 +14,7 @@ class IRegionsHierarchyReader
 public:
     virtual bool readNext(RegionEntry & entry) = 0;
 
-    virtual ~IRegionsHierarchyReader() {}
+    virtual ~IRegionsHierarchyReader() = default;
 };
 
 using IRegionsHierarchyReaderPtr = std::unique_ptr<IRegionsHierarchyReader>;
diff --git a/src/Dictionaries/RegExpTreeDictionary.cpp b/src/Dictionaries/RegExpTreeDictionary.cpp
index 4d82aa9ca0e..1f5c2d6d2c7 100644
--- a/src/Dictionaries/RegExpTreeDictionary.cpp
+++ b/src/Dictionaries/RegExpTreeDictionary.cpp
@@ -568,7 +568,7 @@ bool RegExpTreeDictionary::setAttributesShortCircuit(
     const String & data,
     std::unordered_set<UInt64> & visited_nodes,
     const std::unordered_map<String, const DictionaryAttribute &> & attributes,
-    std::unordered_set<String> * const defaults) const
+    std::unordered_set<String> * defaults) const
 {
     if (visited_nodes.contains(id))
         return attributes_to_set.attributesFull() == attributes.size();
diff --git a/src/Dictionaries/RegExpTreeDictionary.h b/src/Dictionaries/RegExpTreeDictionary.h
index 9e14abb49d0..d6bc90ef651 100644
--- a/src/Dictionaries/RegExpTreeDictionary.h
+++ b/src/Dictionaries/RegExpTreeDictionary.h
@@ -210,7 +210,7 @@ private:
         const String & data,
         std::unordered_set<UInt64> & visited_nodes,
         const std::unordered_map<String, const DictionaryAttribute &> & attributes,
-        std::unordered_set<String> * const defaults) const;
+        std::unordered_set<String> * defaults) const;
 
     struct RegexTreeNode;
     using RegexTreeNodePtr = std::shared_ptr<RegexTreeNode>;
diff --git a/src/Functions/IFunction.h b/src/Functions/IFunction.h
index 05aa08e2ad7..9b7cdf12d57 100644
--- a/src/Functions/IFunction.h
+++ b/src/Functions/IFunction.h
@@ -13,10 +13,6 @@
 
 #include <memory>
 
-#if USE_EMBEDDED_COMPILER
-#    include <Core/ValuesWithType.h>
-#endif
-
 /// This file contains user interface for functions.
 
 namespace llvm
diff --git a/src/IO/ReadSettings.h b/src/IO/ReadSettings.h
index 38904df4403..5c401c0c8d9 100644
--- a/src/IO/ReadSettings.h
+++ b/src/IO/ReadSettings.h
@@ -63,6 +63,7 @@ enum class RemoteFSReadMethod
 class MMappedFileCache;
 class PageCache;
 
+/// NOLINTNEXTLINE(clang-analyzer-optin.performance.Padding)
 struct ReadSettings
 {
     /// Method to use reading from local filesystem.
diff --git a/src/Interpreters/AsynchronousInsertQueue.cpp b/src/Interpreters/AsynchronousInsertQueue.cpp
index 7d56dbabe3c..9327f31b6ff 100644
--- a/src/Interpreters/AsynchronousInsertQueue.cpp
+++ b/src/Interpreters/AsynchronousInsertQueue.cpp
@@ -905,7 +905,7 @@ Chunk AsynchronousInsertQueue::processEntriesWithParsing(
     const InsertDataPtr & data,
     const Block & header,
     const ContextPtr & insert_context,
-    const LoggerPtr logger,
+    LoggerPtr logger,
     LogFunc && add_to_async_insert_log)
 {
     size_t total_rows = 0;
diff --git a/src/Interpreters/AsynchronousInsertQueue.h b/src/Interpreters/AsynchronousInsertQueue.h
index f60b3d343fb..5076701d0b0 100644
--- a/src/Interpreters/AsynchronousInsertQueue.h
+++ b/src/Interpreters/AsynchronousInsertQueue.h
@@ -265,7 +265,7 @@ private:
         const InsertDataPtr & data,
         const Block & header,
         const ContextPtr & insert_context,
-        const LoggerPtr logger,
+        LoggerPtr logger,
         LogFunc && add_to_async_insert_log);
 
     template <typename LogFunc>
diff --git a/src/Interpreters/Context.h b/src/Interpreters/Context.h
index c8aa3604a6f..43df8d6adf2 100644
--- a/src/Interpreters/Context.h
+++ b/src/Interpreters/Context.h
@@ -330,7 +330,7 @@ protected:
             return *this;
         }
 
-        void swap(QueryAccessInfo & rhs)
+        void swap(QueryAccessInfo & rhs) noexcept
         {
             std::swap(databases, rhs.databases);
             std::swap(tables, rhs.tables);
@@ -680,7 +680,7 @@ public:
     void addSpecialScalar(const String & name, const Block & block);
 
     const QueryAccessInfo & getQueryAccessInfo() const { return *getQueryAccessInfoPtr(); }
-    const QueryAccessInfoPtr getQueryAccessInfoPtr() const { return query_access_info; }
+    QueryAccessInfoPtr getQueryAccessInfoPtr() const { return query_access_info; }
     void setQueryAccessInfo(QueryAccessInfoPtr other) { query_access_info = other; }
 
     void addQueryAccessInfo(
diff --git a/src/Interpreters/IExternalLoadable.h b/src/Interpreters/IExternalLoadable.h
index 3c004508b0a..47031778876 100644
--- a/src/Interpreters/IExternalLoadable.h
+++ b/src/Interpreters/IExternalLoadable.h
@@ -23,7 +23,7 @@ struct ExternalLoadableLifetime
     UInt64 max_sec = 0;
 
     ExternalLoadableLifetime(const Poco::Util::AbstractConfiguration & config, const std::string & config_prefix);
-    ExternalLoadableLifetime() {}
+    ExternalLoadableLifetime() = default;
 };
 
 /// Get delay before trying to load again after error.
diff --git a/src/Interpreters/ProcessList.h b/src/Interpreters/ProcessList.h
index 1c253f562e8..ad47041c762 100644
--- a/src/Interpreters/ProcessList.h
+++ b/src/Interpreters/ProcessList.h
@@ -318,7 +318,7 @@ public:
     ~ProcessListEntry();
 
     QueryStatusPtr getQueryStatus() { return *it; }
-    const QueryStatusPtr getQueryStatus() const { return *it; }
+    QueryStatusPtr getQueryStatus() const { return *it; }
 };
 
 
diff --git a/src/Processors/Chunk.h b/src/Processors/Chunk.h
index 9a7d6bc294d..4f753798eaa 100644
--- a/src/Processors/Chunk.h
+++ b/src/Processors/Chunk.h
@@ -59,7 +59,7 @@ public:
 
     Chunk clone() const;
 
-    void swap(Chunk & other)
+    void swap(Chunk & other) noexcept
     {
         columns.swap(other.columns);
         chunk_info.swap(other.chunk_info);
diff --git a/src/Processors/Merges/Algorithms/AggregatingSortedAlgorithm.cpp b/src/Processors/Merges/Algorithms/AggregatingSortedAlgorithm.cpp
index d2d2434c477..3bd0b532d90 100644
--- a/src/Processors/Merges/Algorithms/AggregatingSortedAlgorithm.cpp
+++ b/src/Processors/Merges/Algorithms/AggregatingSortedAlgorithm.cpp
@@ -126,7 +126,7 @@ static void postprocessChunk(Chunk & chunk, const AggregatingSortedAlgorithm::Co
 
 
 AggregatingSortedAlgorithm::SimpleAggregateDescription::SimpleAggregateDescription(
-    AggregateFunctionPtr function_, const size_t column_number_,
+    AggregateFunctionPtr function_, size_t column_number_,
     DataTypePtr nested_type_, DataTypePtr real_type_)
     : function(std::move(function_)), column_number(column_number_)
     , nested_type(std::move(nested_type_)), real_type(std::move(real_type_))
diff --git a/src/Processors/Port.h b/src/Processors/Port.h
index 67af2f041aa..f3c7bbb5fee 100644
--- a/src/Processors/Port.h
+++ b/src/Processors/Port.h
@@ -110,7 +110,7 @@ protected:
                 return result;
             }
 
-            uintptr_t ALWAYS_INLINE swap(std::atomic<Data *> & value, std::uintptr_t flags, std::uintptr_t mask)
+            uintptr_t ALWAYS_INLINE swap(std::atomic<Data *> & value, std::uintptr_t flags, std::uintptr_t mask) /// NOLINT
             {
                 Data * expected = nullptr;
                 Data * desired = getPtr(flags | getUInt(data));
diff --git a/src/Processors/TTL/TTLUpdateInfoAlgorithm.cpp b/src/Processors/TTL/TTLUpdateInfoAlgorithm.cpp
index b7cddf3c165..13d3030bbb8 100644
--- a/src/Processors/TTL/TTLUpdateInfoAlgorithm.cpp
+++ b/src/Processors/TTL/TTLUpdateInfoAlgorithm.cpp
@@ -6,8 +6,8 @@ namespace DB
 TTLUpdateInfoAlgorithm::TTLUpdateInfoAlgorithm(
     const TTLExpressions & ttl_expressions_,
     const TTLDescription & description_,
-    const TTLUpdateField ttl_update_field_,
-    const String ttl_update_key_,
+    TTLUpdateField ttl_update_field_,
+    String ttl_update_key_,
     const TTLInfo & old_ttl_info_,
     time_t current_time_,
     bool force_)
diff --git a/src/Processors/TTL/TTLUpdateInfoAlgorithm.h b/src/Processors/TTL/TTLUpdateInfoAlgorithm.h
index 0cf31765aef..b6aee6f7cb0 100644
--- a/src/Processors/TTL/TTLUpdateInfoAlgorithm.h
+++ b/src/Processors/TTL/TTLUpdateInfoAlgorithm.h
@@ -22,8 +22,8 @@ public:
     TTLUpdateInfoAlgorithm(
         const TTLExpressions & ttl_expressions_,
         const TTLDescription & description_,
-        const TTLUpdateField ttl_update_field_,
-        const String ttl_update_key_,
+        TTLUpdateField ttl_update_field_,
+        String ttl_update_key_,
         const TTLInfo & old_ttl_info_,
         time_t current_time_, bool force_
     );
diff --git a/src/Storages/StorageInMemoryMetadata.h b/src/Storages/StorageInMemoryMetadata.h
index 2823aba1224..69cd3422a7d 100644
--- a/src/Storages/StorageInMemoryMetadata.h
+++ b/src/Storages/StorageInMemoryMetadata.h
@@ -72,8 +72,8 @@ struct StorageInMemoryMetadata
     StorageInMemoryMetadata(const StorageInMemoryMetadata & other);
     StorageInMemoryMetadata & operator=(const StorageInMemoryMetadata & other);
 
-    StorageInMemoryMetadata(StorageInMemoryMetadata && other) = default;
-    StorageInMemoryMetadata & operator=(StorageInMemoryMetadata && other) = default;
+    StorageInMemoryMetadata(StorageInMemoryMetadata && other) = default; /// NOLINT
+    StorageInMemoryMetadata & operator=(StorageInMemoryMetadata && other) = default; /// NOLINT
 
     /// NOTE: Thread unsafe part. You should not modify same StorageInMemoryMetadata
     /// structure from different threads. It should be used as MultiVersion

From 90b27432a26c0a5204e09ff5ff5f2ae8df3055af Mon Sep 17 00:00:00 2001
From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com>
Date: Mon, 11 Mar 2024 12:18:58 +0100
Subject: [PATCH 152/443] Update test.py

---
 tests/integration/test_backup_restore_s3/test.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/integration/test_backup_restore_s3/test.py b/tests/integration/test_backup_restore_s3/test.py
index 452a9143067..f3f4837c317 100644
--- a/tests/integration/test_backup_restore_s3/test.py
+++ b/tests/integration/test_backup_restore_s3/test.py
@@ -130,7 +130,7 @@ def check_system_tables(backup_query_id=None):
         if disk
     ]
     expected_disks = (
-        ("default", "local", "", ""),
+        ("default", "local", "None", "None"),
         ("disk_s3", "object_storage", "s3", "local"),
         ("disk_s3_cache", "object_storage", "s3", "local"),
         ("disk_s3_other_bucket", "object_storage", "s3", "local"),

From 57f6263f67dd91e624003199295c840a228947a0 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Mon, 11 Mar 2024 12:31:40 +0100
Subject: [PATCH 153/443] Lock contention fix

---
 src/Common/ProfileEvents.cpp           |  1 +
 src/Interpreters/Cache/FileCache.cpp   | 12 +++++++++---
 src/Interpreters/Cache/FileCache.h     |  2 +-
 src/Interpreters/Cache/FileCache_fwd.h |  1 +
 src/Interpreters/Cache/Guards.h        | 15 ++++++++++++---
 5 files changed, 24 insertions(+), 7 deletions(-)

diff --git a/src/Common/ProfileEvents.cpp b/src/Common/ProfileEvents.cpp
index c1ac3d08245..ab1a16a3edf 100644
--- a/src/Common/ProfileEvents.cpp
+++ b/src/Common/ProfileEvents.cpp
@@ -476,6 +476,7 @@ The server successfully detected this situation and will download merged part fr
     M(FileSegmentRemoveMicroseconds, "File segment remove() time") \
     M(FileSegmentHolderCompleteMicroseconds, "File segments holder complete() time") \
     M(FileSegmentFailToIncreasePriority, "Number of times the priority was not increased due to a high contention on the cache lock") \
+    M(FilesystemCacheFailToReserveSpaceBecauseOfLockContention, "Number of times space reservation was skipped due to a high contention on the cache lock") \
     M(FilesystemCacheHoldFileSegments, "Filesystem cache file segments count, which were hold") \
     M(FilesystemCacheUnusedHoldFileSegments, "Filesystem cache file segments count, which were hold, but not used (because of seek or LIMIT n, etc)") \
     \
diff --git a/src/Interpreters/Cache/FileCache.cpp b/src/Interpreters/Cache/FileCache.cpp
index 9c705ddc27c..5650b9ce44e 100644
--- a/src/Interpreters/Cache/FileCache.cpp
+++ b/src/Interpreters/Cache/FileCache.cpp
@@ -27,6 +27,7 @@ namespace ProfileEvents
     extern const Event FilesystemCacheReserveMicroseconds;
     extern const Event FilesystemCacheGetOrSetMicroseconds;
     extern const Event FilesystemCacheGetMicroseconds;
+    extern const Event FilesystemCacheFailToReserveSpaceBecauseOfLockContention;
 }
 
 namespace DB
@@ -188,9 +189,9 @@ CacheGuard::Lock FileCache::lockCache() const
     return cache_guard.lock();
 }
 
-CacheGuard::Lock FileCache::tryLockCache() const
+CacheGuard::Lock FileCache::tryLockCache(std::optional<std::chrono::milliseconds> acquire_timeout) const
 {
-    return cache_guard.tryLock();
+    return acquire_timeout.has_value() ? cache_guard.tryLockFor(acquire_timeout.value()) : cache_guard.tryLock();
 }
 
 FileSegments FileCache::getImpl(const LockedKey & locked_key, const FileSegment::Range & range, size_t file_segments_limit) const
@@ -781,7 +782,12 @@ bool FileCache::tryReserve(
     ProfileEventTimeIncrement<Microseconds> watch(ProfileEvents::FilesystemCacheReserveMicroseconds);
 
     assertInitialized();
-    auto cache_lock = lockCache();
+    auto cache_lock = tryLockCache(std::chrono::milliseconds(FILECACHE_TRY_RESERVE_LOCK_TIMEOUT_MILLISECONDS));
+    if (!cache_lock)
+    {
+        ProfileEvents::increment(ProfileEvents::FilesystemCacheFailToReserveSpaceBecauseOfLockContention);
+        return false;
+    }
 
     LOG_TEST(
         log, "Trying to reserve space ({} bytes) for {}:{}, current usage {}/{}",
diff --git a/src/Interpreters/Cache/FileCache.h b/src/Interpreters/Cache/FileCache.h
index 5b665ad0271..7434b2ac78a 100644
--- a/src/Interpreters/Cache/FileCache.h
+++ b/src/Interpreters/Cache/FileCache.h
@@ -173,7 +173,7 @@ public:
     void deactivateBackgroundOperations();
 
     CacheGuard::Lock lockCache() const;
-    CacheGuard::Lock tryLockCache() const;
+    CacheGuard::Lock tryLockCache(std::optional<std::chrono::milliseconds> acquire_timeout = std::nullopt) const;
 
     std::vector<FileSegment::Info> sync();
 
diff --git a/src/Interpreters/Cache/FileCache_fwd.h b/src/Interpreters/Cache/FileCache_fwd.h
index 06261b19db7..eaed279e7fd 100644
--- a/src/Interpreters/Cache/FileCache_fwd.h
+++ b/src/Interpreters/Cache/FileCache_fwd.h
@@ -12,6 +12,7 @@ static constexpr int FILECACHE_DEFAULT_LOAD_METADATA_THREADS = 16;
 static constexpr int FILECACHE_DEFAULT_MAX_ELEMENTS = 10000000;
 static constexpr int FILECACHE_DEFAULT_HITS_THRESHOLD = 0;
 static constexpr size_t FILECACHE_BYPASS_THRESHOLD = 256 * 1024 * 1024;
+static constexpr size_t FILECACHE_TRY_RESERVE_LOCK_TIMEOUT_MILLISECONDS = 1000; /// 1 sec.
 
 class FileCache;
 using FileCachePtr = std::shared_ptr<FileCache>;
diff --git a/src/Interpreters/Cache/Guards.h b/src/Interpreters/Cache/Guards.h
index 5729620d82f..0ac7cb80483 100644
--- a/src/Interpreters/Cache/Guards.h
+++ b/src/Interpreters/Cache/Guards.h
@@ -61,17 +61,26 @@ namespace DB
  */
 struct CacheGuard : private boost::noncopyable
 {
+    using Mutex = std::timed_mutex;
     /// struct is used (not keyword `using`) to make CacheGuard::Lock non-interchangable with other guards locks
     /// so, we wouldn't be able to pass CacheGuard::Lock to a function which accepts KeyGuard::Lock, for example
-    struct Lock : public std::unique_lock<std::mutex>
+    struct Lock : public std::unique_lock<Mutex>
     {
-        using Base = std::unique_lock<std::mutex>;
+        using Base = std::unique_lock<Mutex>;
         using Base::Base;
     };
 
     Lock lock() { return Lock(mutex); }
+
     Lock tryLock() { return Lock(mutex, std::try_to_lock); }
-    std::mutex mutex;
+
+    Lock tryLockFor(const std::chrono::milliseconds & acquire_timeout)
+    {
+        return Lock(mutex, std::chrono::duration<double, std::milli>(acquire_timeout));
+    }
+
+private:
+    Mutex mutex;
 };
 
 /**

From c0689f3760c738dc1f73cf58c1c9de12b4c096a7 Mon Sep 17 00:00:00 2001
From: vdimir <vdimir@clickhouse.com>
Date: Thu, 7 Mar 2024 11:24:34 +0000
Subject: [PATCH 154/443] Fix ASTRenameQuery::clone

---
 src/Parsers/ASTRenameQuery.h       |  1 +
 src/Parsers/tests/gtest_Parser.cpp | 38 ++++++++++++++++++++++++++----
 2 files changed, 35 insertions(+), 4 deletions(-)

diff --git a/src/Parsers/ASTRenameQuery.h b/src/Parsers/ASTRenameQuery.h
index 73d12be094a..582060ab34a 100644
--- a/src/Parsers/ASTRenameQuery.h
+++ b/src/Parsers/ASTRenameQuery.h
@@ -60,6 +60,7 @@ public:
     ASTPtr clone() const override
     {
         auto res = std::make_shared<ASTRenameQuery>(*this);
+        res->cloneChildren();
         cloneOutputOptions(*res);
         return res;
     }
diff --git a/src/Parsers/tests/gtest_Parser.cpp b/src/Parsers/tests/gtest_Parser.cpp
index 36d2deae8d7..19947cd38cc 100644
--- a/src/Parsers/tests/gtest_Parser.cpp
+++ b/src/Parsers/tests/gtest_Parser.cpp
@@ -9,6 +9,7 @@
 #include <Parsers/ParserAlterQuery.h>
 #include <Parsers/ParserCreateQuery.h>
 #include <Parsers/ParserOptimizeQuery.h>
+#include <Parsers/ParserRenameQuery.h>
 #include <Parsers/ParserQueryWithOutput.h>
 #include <Parsers/ParserAttachAccessEntity.h>
 #include <Parsers/formatAST.h>
@@ -62,10 +63,29 @@ TEST_P(ParserTest, parseQuery)
             if (std::string("CREATE USER or ALTER USER query") != parser->getName()
                     && std::string("ATTACH access entity query") != parser->getName())
             {
-                WriteBufferFromOwnString buf;
-                formatAST(*ast->clone(), buf, false, false);
-                String formatted_ast = buf.str();
-                EXPECT_EQ(expected_ast, formatted_ast);
+                ASTPtr ast_clone = ast->clone();
+                {
+                    WriteBufferFromOwnString buf;
+                    formatAST(*ast_clone, buf, false, false);
+                    String formatted_ast = buf.str();
+                    EXPECT_EQ(expected_ast, formatted_ast);
+                }
+
+
+                ASTPtr ast_clone2 = ast_clone->clone();
+                /// Break `ast_clone2`, it should not affect `ast_clone` if `clone()` implemented properly
+                for (auto & child : ast_clone2->children)
+                {
+                    if (auto * identifier = dynamic_cast<ASTIdentifier *>(child.get()))
+                        identifier->setShortName("new_name");
+                }
+
+                {
+                    WriteBufferFromOwnString buf;
+                    formatAST(*ast_clone, buf, false, false);
+                    String formatted_ast = buf.str();
+                    EXPECT_EQ(expected_ast, formatted_ast);
+                }
             }
             else
             {
@@ -299,6 +319,16 @@ INSTANTIATE_TEST_SUITE_P(ParserAttachUserQuery, ParserTest,
         }
 })));
 
+INSTANTIATE_TEST_SUITE_P(ParserRenameQuery, ParserTest,
+    ::testing::Combine(
+        ::testing::Values(std::make_shared<ParserRenameQuery>()),
+        ::testing::ValuesIn(std::initializer_list<ParserTestCase>{
+        {
+            "RENAME TABLE eligible_test TO eligible_test2",
+            "RENAME TABLE eligible_test TO eligible_test2"
+        }
+})));
+
 INSTANTIATE_TEST_SUITE_P(ParserKQLQuery, ParserKQLTest,
     ::testing::Combine(
         ::testing::Values(std::make_shared<ParserKQLQuery>()),

From 38b8e7ac7e87b0efcea7560735697de0583e0f17 Mon Sep 17 00:00:00 2001
From: vdimir <vdimir@clickhouse.com>
Date: Fri, 8 Mar 2024 11:54:20 +0000
Subject: [PATCH 155/443] Maintain list of ASTRenameQuery children

---
 src/Interpreters/AddDefaultDatabaseVisitor.h  |  8 +---
 src/Interpreters/InterpreterCreateQuery.cpp   |  3 +-
 src/Interpreters/InterpreterRenameQuery.cpp   |  8 ++--
 .../MySQL/InterpretersMySQLDDLQuery.cpp       | 25 ++++++------
 src/Interpreters/SystemLog.cpp                |  3 +-
 src/Parsers/ASTRenameQuery.h                  | 38 ++++++++++++++++++-
 src/Parsers/ParserRenameQuery.cpp             | 26 ++++---------
 src/Storages/StorageMaterializedView.cpp      |  5 ++-
 8 files changed, 69 insertions(+), 47 deletions(-)

diff --git a/src/Interpreters/AddDefaultDatabaseVisitor.h b/src/Interpreters/AddDefaultDatabaseVisitor.h
index b977a73d461..e36f22e8ba1 100644
--- a/src/Interpreters/AddDefaultDatabaseVisitor.h
+++ b/src/Interpreters/AddDefaultDatabaseVisitor.h
@@ -275,13 +275,7 @@ private:
         if (only_replace_current_database_function)
             return;
 
-        for (ASTRenameQuery::Element & elem : node.elements)
-        {
-            if (!elem.from.database)
-                elem.from.database = std::make_shared<ASTIdentifier>(database_name);
-            if (!elem.to.database)
-                elem.to.database = std::make_shared<ASTIdentifier>(database_name);
-        }
+        node.setDatabaseIfNotExists(database_name);
     }
 
     void visitDDL(ASTAlterQuery & node, ASTPtr &) const
diff --git a/src/Interpreters/InterpreterCreateQuery.cpp b/src/Interpreters/InterpreterCreateQuery.cpp
index d05d8b8deb1..edd7452c130 100644
--- a/src/Interpreters/InterpreterCreateQuery.cpp
+++ b/src/Interpreters/InterpreterCreateQuery.cpp
@@ -1612,7 +1612,6 @@ BlockIO InterpreterCreateQuery::doCreateOrReplaceTable(ASTCreateQuery & create,
         executeTrivialBlockIO(fill_io, getContext());
 
         /// Replace target table with created one
-        auto ast_rename = std::make_shared<ASTRenameQuery>();
         ASTRenameQuery::Element elem
         {
             ASTRenameQuery::Table
@@ -1627,7 +1626,7 @@ BlockIO InterpreterCreateQuery::doCreateOrReplaceTable(ASTCreateQuery & create,
             }
         };
 
-        ast_rename->elements.push_back(std::move(elem));
+        auto ast_rename = std::make_shared<ASTRenameQuery>(ASTRenameQuery::Elements{std::move(elem)});
         ast_rename->dictionary = create.is_dictionary;
         if (create.create_or_replace)
         {
diff --git a/src/Interpreters/InterpreterRenameQuery.cpp b/src/Interpreters/InterpreterRenameQuery.cpp
index 52001fdcaf4..06b6ebc9cbb 100644
--- a/src/Interpreters/InterpreterRenameQuery.cpp
+++ b/src/Interpreters/InterpreterRenameQuery.cpp
@@ -47,12 +47,12 @@ BlockIO InterpreterRenameQuery::execute()
       */
 
     RenameDescriptions descriptions;
-    descriptions.reserve(rename.elements.size());
+    descriptions.reserve(rename.getElements().size());
 
     /// Don't allow to drop tables (that we are renaming); don't allow to create tables in places where tables will be renamed.
     TableGuards table_guards;
 
-    for (const auto & elem : rename.elements)
+    for (const auto & elem : rename.getElements())
     {
         descriptions.emplace_back(elem, current_database);
         const auto & description = descriptions.back();
@@ -186,7 +186,7 @@ AccessRightsElements InterpreterRenameQuery::getRequiredAccess(InterpreterRename
 {
     AccessRightsElements required_access;
     const auto & rename = query_ptr->as<const ASTRenameQuery &>();
-    for (const auto & elem : rename.elements)
+    for (const auto & elem : rename.getElements())
     {
         if (type == RenameType::RenameTable)
         {
@@ -214,7 +214,7 @@ AccessRightsElements InterpreterRenameQuery::getRequiredAccess(InterpreterRename
 void InterpreterRenameQuery::extendQueryLogElemImpl(QueryLogElement & elem, const ASTPtr & ast, ContextPtr) const
 {
     const auto & rename = ast->as<const ASTRenameQuery &>();
-    for (const auto & element : rename.elements)
+    for (const auto & element : rename.getElements())
     {
         {
             String database = backQuoteIfNeed(!element.from.database ? getContext()->getCurrentDatabase() : element.from.getDatabase());
diff --git a/src/Interpreters/MySQL/InterpretersMySQLDDLQuery.cpp b/src/Interpreters/MySQL/InterpretersMySQLDDLQuery.cpp
index 107b435ded4..fd7ffca2872 100644
--- a/src/Interpreters/MySQL/InterpretersMySQLDDLQuery.cpp
+++ b/src/Interpreters/MySQL/InterpretersMySQLDDLQuery.cpp
@@ -579,7 +579,7 @@ ASTs InterpreterRenameImpl::getRewrittenQueries(
     const InterpreterRenameImpl::TQuery & rename_query, ContextPtr context, const String & mapped_to_database, const String & mysql_database)
 {
     ASTRenameQuery::Elements elements;
-    for (const auto & rename_element : rename_query.elements)
+    for (const auto & rename_element : rename_query.getElements())
     {
         const auto & to_database = resolveDatabase(rename_element.to.getDatabase(), mysql_database, mapped_to_database, context);
         const auto & from_database = resolveDatabase(rename_element.from.getDatabase(), mysql_database, mapped_to_database, context);
@@ -600,8 +600,7 @@ ASTs InterpreterRenameImpl::getRewrittenQueries(
     if (elements.empty())
         return ASTs{};
 
-    auto rewritten_query = std::make_shared<ASTRenameQuery>();
-    rewritten_query->elements = elements;
+    auto rewritten_query = std::make_shared<ASTRenameQuery>(std::move(elements));
     return ASTs{rewritten_query};
 }
 
@@ -616,7 +615,8 @@ ASTs InterpreterAlterImpl::getRewrittenQueries(
         return {};
 
     auto rewritten_alter_query = std::make_shared<ASTAlterQuery>();
-    auto rewritten_rename_query = std::make_shared<ASTRenameQuery>();
+    ASTRenameQuery::Elements rename_elements;
+
     rewritten_alter_query->setDatabase(mapped_to_database);
     rewritten_alter_query->setTable(alter_query.table);
     rewritten_alter_query->alter_object = ASTAlterQuery::AlterObjectType::TABLE;
@@ -749,13 +749,13 @@ ASTs InterpreterAlterImpl::getRewrittenQueries(
 
             /// For ALTER TABLE table_name RENAME TO new_table_name_1, RENAME TO new_table_name_2;
             /// We just need to generate RENAME TABLE table_name TO new_table_name_2;
-            if (rewritten_rename_query->elements.empty())
-                rewritten_rename_query->elements.push_back(ASTRenameQuery::Element());
+            if (rename_elements.empty())
+                rename_elements.push_back(ASTRenameQuery::Element());
 
-            rewritten_rename_query->elements.back().from.database = std::make_shared<ASTIdentifier>(mapped_to_database);
-            rewritten_rename_query->elements.back().from.table = std::make_shared<ASTIdentifier>(alter_query.table);
-            rewritten_rename_query->elements.back().to.database = std::make_shared<ASTIdentifier>(mapped_to_database);
-            rewritten_rename_query->elements.back().to.table = std::make_shared<ASTIdentifier>(alter_command->new_table_name);
+            rename_elements.back().from.database = std::make_shared<ASTIdentifier>(mapped_to_database);
+            rename_elements.back().from.table = std::make_shared<ASTIdentifier>(alter_query.table);
+            rename_elements.back().to.database = std::make_shared<ASTIdentifier>(mapped_to_database);
+            rename_elements.back().to.table = std::make_shared<ASTIdentifier>(alter_command->new_table_name);
         }
     }
 
@@ -765,8 +765,11 @@ ASTs InterpreterAlterImpl::getRewrittenQueries(
     if (!rewritten_alter_query->command_list->children.empty())
         rewritten_queries.push_back(rewritten_alter_query);
 
-    if (!rewritten_rename_query->elements.empty())
+    if (!rename_elements.empty())
+    {
+        auto rewritten_rename_query = std::make_shared<ASTRenameQuery>(std::move(rename_elements));
         rewritten_queries.push_back(rewritten_rename_query);
+    }
 
     return rewritten_queries;
 }
diff --git a/src/Interpreters/SystemLog.cpp b/src/Interpreters/SystemLog.cpp
index 6580dc3e9b7..a74b5c67726 100644
--- a/src/Interpreters/SystemLog.cpp
+++ b/src/Interpreters/SystemLog.cpp
@@ -563,7 +563,6 @@ void SystemLog<LogElement>::prepareTable()
                 {table_id.database_name, table_id.table_name + "_" + toString(suffix)}, getContext()))
                 ++suffix;
 
-            auto rename = std::make_shared<ASTRenameQuery>();
             ASTRenameQuery::Element elem
             {
                 ASTRenameQuery::Table
@@ -586,7 +585,7 @@ void SystemLog<LogElement>::prepareTable()
                 old_create_query,
                 create_query);
 
-            rename->elements.emplace_back(std::move(elem));
+            auto rename = std::make_shared<ASTRenameQuery>(ASTRenameQuery::Elements{std::move(elem)});
 
             ActionLock merges_lock;
             if (DatabaseCatalog::instance().getDatabase(table_id.database_name)->getUUID() == UUIDHelpers::Nil)
diff --git a/src/Parsers/ASTRenameQuery.h b/src/Parsers/ASTRenameQuery.h
index 582060ab34a..beaf93c4761 100644
--- a/src/Parsers/ASTRenameQuery.h
+++ b/src/Parsers/ASTRenameQuery.h
@@ -45,7 +45,6 @@ public:
     };
 
     using Elements = std::vector<Element>;
-    Elements elements;
 
     bool exchange{false};   /// For EXCHANGE TABLES
     bool database{false};   /// For RENAME DATABASE
@@ -54,6 +53,41 @@ public:
     /// Special flag for CREATE OR REPLACE. Do not throw if the second table does not exist.
     bool rename_if_cannot_exchange{false};
 
+    explicit ASTRenameQuery(Elements elements_ = {})
+        : elements(std::move(elements_))
+    {
+        for (const auto & elem : elements)
+        {
+            if (elem.from.database)
+                children.push_back(elem.from.database);
+            if (elem.from.table)
+                children.push_back(elem.from.table);
+            if (elem.to.database)
+                children.push_back(elem.to.database);
+            if (elem.to.table)
+                children.push_back(elem.to.table);
+        }
+    }
+
+    void setDatabaseIfNotExists(const String & database_name)
+    {
+        for (auto & elem : elements)
+        {
+            if (!elem.from.database)
+            {
+                elem.from.database = std::make_shared<ASTIdentifier>(database_name);
+                children.push_back(elem.from.database);
+            }
+            if (!elem.to.database)
+            {
+                elem.to.database = std::make_shared<ASTIdentifier>(database_name);
+                children.push_back(elem.to.database);
+            }
+        }
+    }
+
+    const Elements & getElements() const { return elements; }
+
     /** Get the text that identifies this element. */
     String getID(char) const override { return "Rename"; }
 
@@ -146,6 +180,8 @@ protected:
 
         formatOnCluster(settings);
     }
+
+    Elements elements;
 };
 
 }
diff --git a/src/Parsers/ParserRenameQuery.cpp b/src/Parsers/ParserRenameQuery.cpp
index 27f2ed1cd22..f9d29108ed6 100644
--- a/src/Parsers/ParserRenameQuery.cpp
+++ b/src/Parsers/ParserRenameQuery.cpp
@@ -44,15 +44,14 @@ bool ParserRenameQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
             if (!ASTQueryWithOnCluster::parse(pos, cluster_str, expected))
                 return false;
         }
+        ASTRenameQuery::Elements rename_elements;
+        rename_elements.emplace_back();
+        rename_elements.back().if_exists = if_exists;
+        rename_elements.back().from.database = from_db;
+        rename_elements.back().to.database = to_db;
 
-        auto query = std::make_shared<ASTRenameQuery>();
+        auto query = std::make_shared<ASTRenameQuery>(std::move(rename_elements));
         query->database = true;
-        query->elements.emplace({});
-        query->elements.front().if_exists = if_exists;
-        query->elements.front().from.database = from_db;
-        query->elements.front().to.database = to_db;
-        query->children.push_back(std::move(from_db));
-        query->children.push_back(std::move(to_db));
         query->cluster = cluster_str;
         node = query;
         return true;
@@ -75,9 +74,8 @@ bool ParserRenameQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 
     const auto ignore_delim = [&] { return exchange ? s_and.ignore(pos) : s_to.ignore(pos); };
 
-    auto query = std::make_shared<ASTRenameQuery>();
 
-    ASTRenameQuery::Elements & elements = query->elements;
+    ASTRenameQuery::Elements elements;
 
     while (true)
     {
@@ -93,15 +91,6 @@ bool ParserRenameQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
             || !ignore_delim()
             || !parseDatabaseAndTableAsAST(pos, expected, ref.to.database, ref.to.table))
             return false;
-
-        if (ref.from.database)
-            query->children.push_back(ref.from.database);
-        if (ref.from.table)
-            query->children.push_back(ref.from.table);
-        if (ref.to.database)
-            query->children.push_back(ref.to.database);
-        if (ref.to.table)
-            query->children.push_back(ref.to.table);
     }
 
     String cluster_str;
@@ -111,6 +100,7 @@ bool ParserRenameQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
             return false;
     }
 
+    auto query = std::make_shared<ASTRenameQuery>(std::move(elements));
     query->cluster = cluster_str;
     query->exchange = exchange;
     query->dictionary = dictionary;
diff --git a/src/Storages/StorageMaterializedView.cpp b/src/Storages/StorageMaterializedView.cpp
index 02cba1cf753..9958d65819b 100644
--- a/src/Storages/StorageMaterializedView.cpp
+++ b/src/Storages/StorageMaterializedView.cpp
@@ -465,8 +465,8 @@ void StorageMaterializedView::renameInMemory(const StorageID & new_table_id)
     if (!from_atomic_to_atomic_database && has_inner_table && tryGetTargetTable())
     {
         auto new_target_table_name = generateInnerTableName(new_table_id);
-        auto rename = std::make_shared<ASTRenameQuery>();
 
+        ASTRenameQuery::Elements rename_elements;
         assert(inner_table_id.database_name == old_table_id.database_name);
 
         ASTRenameQuery::Element elem
@@ -482,8 +482,9 @@ void StorageMaterializedView::renameInMemory(const StorageID & new_table_id)
                 std::make_shared<ASTIdentifier>(new_target_table_name)
             }
         };
-        rename->elements.emplace_back(std::move(elem));
+        rename_elements.emplace_back(std::move(elem));
 
+        auto rename = std::make_shared<ASTRenameQuery>(std::move(rename_elements));
         InterpreterRenameQuery(rename, getContext()).execute();
         updateTargetTableId(new_table_id.database_name, new_target_table_name);
     }

From 59c6311ead26e48f861e27d19d58deffe4c6d622 Mon Sep 17 00:00:00 2001
From: Max Kainov <max.kainov@clickhouse.com>
Date: Mon, 11 Mar 2024 09:55:13 +0000
Subject: [PATCH 156/443] improve report  #do_not_test

---
 utils/check-style/process_style_check_result.py | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/utils/check-style/process_style_check_result.py b/utils/check-style/process_style_check_result.py
index e620d85b9d0..b043aa548d7 100755
--- a/utils/check-style/process_style_check_result.py
+++ b/utils/check-style/process_style_check_result.py
@@ -32,8 +32,13 @@ def process_result(result_folder):
         if not os.path.exists(full_path):
             test_results.append((f"Check {name}", "SKIPPED"))
         elif os.stat(full_path).st_size != 0:
+            with open(full_path, 'r') as file:
+                lines = file.readlines()
+                if len(lines) > 100:
+                    lines = lines[:100] + ['====TRIMMED====']
+                content = "\n".join(lines)
             description += f"Check {name} failed. "
-            test_results.append((f"Check {name}", "FAIL"))
+            test_results.append((f"Check {name}", "FAIL", None, content))
             status = "failure"
         else:
             test_results.append((f"Check {name}", "OK"))

From 8f2ff2ccd833dd1cc839922c59a90360f2b882c7 Mon Sep 17 00:00:00 2001
From: "Mikhail f. Shiryaev" <felixoid@clickhouse.com>
Date: Mon, 11 Mar 2024 13:07:39 +0100
Subject: [PATCH 157/443] Enable black back

---
 tests/ci/style_check.py                        | 18 +++++++++---------
 utils/check-style/check_py.sh                  |  4 ++--
 .../check-style/process_style_check_result.py  | 10 +++++-----
 3 files changed, 16 insertions(+), 16 deletions(-)

diff --git a/tests/ci/style_check.py b/tests/ci/style_check.py
index 9f113b6e6f9..7fb5d22a3ce 100644
--- a/tests/ci/style_check.py
+++ b/tests/ci/style_check.py
@@ -1,12 +1,12 @@
 #!/usr/bin/env python3
 import argparse
-from concurrent.futures import ProcessPoolExecutor
 import csv
 import logging
 import os
 import shutil
 import subprocess
 import sys
+from concurrent.futures import ProcessPoolExecutor
 from pathlib import Path
 from typing import List, Tuple
 
@@ -121,7 +121,7 @@ def checkout_last_ref(pr_info: PRInfo) -> None:
 def main():
     logging.basicConfig(level=logging.INFO)
     logging.getLogger("git_helper").setLevel(logging.DEBUG)
-    # args = parse_args()
+    args = parse_args()
 
     stopwatch = Stopwatch()
 
@@ -141,12 +141,14 @@ def main():
         f"--entrypoint= -w/ClickHouse/utils/check-style "
         f"{image} ./check_cpp.sh"
     )
+
     cmd_py = (
         f"docker run -u $(id -u ${{USER}}):$(id -g ${{USER}}) --cap-add=SYS_PTRACE "
         f"--volume={repo_path}:/ClickHouse --volume={temp_path}:/test_output "
         f"--entrypoint= -w/ClickHouse/utils/check-style "
         f"{image} ./check_py.sh"
     )
+
     cmd_docs = (
         f"docker run -u $(id -u ${{USER}}):$(id -g ${{USER}}) --cap-add=SYS_PTRACE "
         f"--volume={repo_path}:/ClickHouse --volume={temp_path}:/test_output "
@@ -175,12 +177,14 @@ def main():
             _ = future1.result()
 
         if run_pycheck:
+            if args.push:
+                checkout_head(pr_info)
             logging.info("Run py files check: %s", cmd_py)
             future2 = executor.submit(subprocess.run, cmd_py, shell=True)
             _ = future2.result()
-
-    # if args.push:
-    #     checkout_head(pr_info)
+            if args.push:
+                commit_push_staged(pr_info)
+                checkout_last_ref(pr_info)
 
     subprocess.check_call(
         f"python3 ../../utils/check-style/process_style_check_result.py --in-results-dir {temp_path} "
@@ -189,10 +193,6 @@ def main():
         shell=True,
     )
 
-    # if args.push:
-    #     commit_push_staged(pr_info)
-    #     checkout_last_ref(pr_info)
-
     state, description, test_results, additional_files = process_result(temp_path)
 
     JobReport(
diff --git a/utils/check-style/check_py.sh b/utils/check-style/check_py.sh
index b729cd78124..13f4e754ed3 100755
--- a/utils/check-style/check_py.sh
+++ b/utils/check-style/check_py.sh
@@ -5,8 +5,8 @@ cd /ClickHouse/utils/check-style || echo -e "failure\tRepo not found" > /test_ou
 start_total=`date +%s`
 
 # FIXME: 1 min to wait + head checkout
-# echo "Check python formatting with black" | ts
-# ./check-black -n              |& tee /test_output/black_output.txt
+echo "Check python formatting with black" | ts
+./check-black -n              |& tee /test_output/black_output.txt
 
 start=`date +%s`
 ./check-pylint -n               |& tee /test_output/pylint_output.txt
diff --git a/utils/check-style/process_style_check_result.py b/utils/check-style/process_style_check_result.py
index b043aa548d7..8c9837b4fc6 100755
--- a/utils/check-style/process_style_check_result.py
+++ b/utils/check-style/process_style_check_result.py
@@ -1,9 +1,9 @@
 #!/usr/bin/env python3
 
-import os
-import logging
 import argparse
 import csv
+import logging
+import os
 
 
 # TODO: add typing and log files to the fourth column, think about launching
@@ -17,7 +17,7 @@ def process_result(result_folder):
         # "shellcheck",
         "style",
         "pylint",
-        # "black",
+        "black",
         "mypy",
         "typos",
         "whitespaces",
@@ -32,10 +32,10 @@ def process_result(result_folder):
         if not os.path.exists(full_path):
             test_results.append((f"Check {name}", "SKIPPED"))
         elif os.stat(full_path).st_size != 0:
-            with open(full_path, 'r') as file:
+            with open(full_path, "r") as file:
                 lines = file.readlines()
                 if len(lines) > 100:
-                    lines = lines[:100] + ['====TRIMMED====']
+                    lines = lines[:100] + ["====TRIMMED===="]
                 content = "\n".join(lines)
             description += f"Check {name} failed. "
             test_results.append((f"Check {name}", "FAIL", None, content))

From 1ea68265b50a8c8ded253e392181b3b81df72815 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Mon, 11 Mar 2024 12:46:30 +0000
Subject: [PATCH 158/443] Revert "Adds makeDateTime64 function."

This reverts commit bd194aab41401492c5d628269df53e68243a1211.
---
 .../functions/date-time-functions.md          | 137 ++++--------------
 .../functions/other-functions.md              |  65 +--------
 2 files changed, 32 insertions(+), 170 deletions(-)

diff --git a/docs/en/sql-reference/functions/date-time-functions.md b/docs/en/sql-reference/functions/date-time-functions.md
index 12f0c996ce7..41503abfa2f 100644
--- a/docs/en/sql-reference/functions/date-time-functions.md
+++ b/docs/en/sql-reference/functions/date-time-functions.md
@@ -26,115 +26,66 @@ SELECT
 
 ## makeDate
 
-Creates a [Date](../../sql-reference/data-types/date.md) from either one of the following sets of arguments:
+Creates a [Date](../../sql-reference/data-types/date.md)
+- from a year, month and day argument, or
+- from a year and day of year argument.
 
-- a year, month, and day.
-- a year and day of year.
+**Syntax**
 
-### Syntax
-
-Using a year, month, and day:
-
-```sql
-makeDate(year, month, day)
+``` sql
+makeDate(year, month, day);
+makeDate(year, day_of_year);
 ```
 
-Using a year and day of year:
+Alias:
+- `MAKEDATE(year, month, day);`
+- `MAKEDATE(year, day_of_year);`
 
-```sql
-makeDate(year, day_of_year)
-```
-
-### Arguments
+**Arguments**
 
 - `year` — Year. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md).
 - `month` — Month. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md).
 - `day` — Day. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md).
 - `day_of_year` — Day of the year. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md).
 
-### Returned values
+**Returned value**
 
-A date created from the arguments.
+- A date created from the arguments.
 
 Type: [Date](../../sql-reference/data-types/date.md).
 
-### Examples
+**Example**
 
 Create a Date from a year, month and day:
 
-```sql
+``` sql
 SELECT makeDate(2023, 2, 28) AS Date;
 ```
 
-```response
-2023-02-28
+Result:
+
+``` text
+┌───────date─┐
+│ 2023-02-28 │
+└────────────┘
 ```
 
-Create a Date from a year and day of year:
+Create a Date from a year and day of year argument:
 
 ``` sql
 SELECT makeDate(2023, 42) AS Date;
 ```
 
-```response
-2023-02-11
-```
+Result:
 
+``` text
+┌───────date─┐
+│ 2023-02-11 │
+└────────────┘
+```
 ## makeDate32
 
-Creates a date of type [Date32](../../sql-reference/data-types/date32.md) from either one of the following sets of arguments:
-
-- a year, month, and day.
-- a year and day of year.
-
-### Syntax
-
-Using a year, month, and day:
-
-```sql
-makeDate32(year, month, day)
-```
-
-Using a year and day of year:
-
-```sql
-makeDate32(year, day_of_year)
-```
-
-### Arguments
-
-- `year` — Year. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md).
-- `month` — Month. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md).
-- `day` — Day. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md).
-- `day_of_year` — Day of the year. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md).
-
-### Returned values
-
-A date created from the arguments.
-
-Type: [Date32](../../sql-reference/data-types/date32.md).
-
-### Examples
-
-Create a date from a year, month, and day:
-
-```sql
-SELECT makeDate32(2024, 1, 1);
-```
-
-```response
-2024-01-01
-```
-
-Create a Date from a year and day of year:
-
-``` sql
-SELECT makeDate32(2024, 100);
-```
-
-```response
-2024-04-09
-```
+Like [makeDate](#makeDate) but produces a [Date32](../../sql-reference/data-types/date32.md).
 
 ## makeDateTime
 
@@ -178,38 +129,12 @@ Result:
 
 ## makeDateTime64
 
-Create a [DateTime64](../../sql-reference/data-types/datetime64.md) data type value from its components (year, month, day, hour, minute, second, and optionally, subsecond precision).
-
-The DateTime64 data type stores both the date and time components in a single 64-bit integer value. The precision of the time component is configurable, allowing you to store time values with subsecond precision up to nanoseconds.
+Like [makeDateTime](#makedatetime) but produces a [DateTime64](../../sql-reference/data-types/datetime64.md).
 
 **Syntax**
 
-```sql
-makeDateTime64(year, month, day, hour, minute, second[, fraction[, precision]])
-```
-
-**Arguments**
-
-- `year` — [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md). The year component (0-9999).
-- `month` — Month. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md). The month component (1-12).
-- `day` — Day. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md). The day component (1-31).
-- `hour` — Hour. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md). The hour component (0-23).
-- `minute` — Minute. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md). The minute component (0-59).
-- `second` — Second. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md). The second component (0-59).
-- `subsecond_precision` (optional) [Integer](../../sql-reference/data-types/int-uint.md): The precision of the subsecond component (0-9, where 0 means no subsecond precision, and 9 means nanosecond precision).
-
-**Returned value**
-
-A date and time element of type [DateTime64](../../sql-reference/data-types/datetime64.md) with created from the supplied arguments.
-
-**Example**
-
 ``` sql
-SELECT makeDateTime64(2023, 5, 15, 10, 30, 45, 779, 5);
-```
-
-```response
-2023-05-15 10:30:45.00779
+makeDateTime64(year, month, day, hour, minute, second[, fraction[, precision[, timezone]]])
 ```
 
 ## timestamp
diff --git a/docs/en/sql-reference/functions/other-functions.md b/docs/en/sql-reference/functions/other-functions.md
index 10ceedad9aa..739b688a0d2 100644
--- a/docs/en/sql-reference/functions/other-functions.md
+++ b/docs/en/sql-reference/functions/other-functions.md
@@ -1866,7 +1866,7 @@ As you can see, `runningAccumulate` merges states for each group of rows separat
 
 ## joinGet
 
-Allows you to extract data from a specific column in a Join table, similar to how you would access a value from a dictionary.
+The function lets you extract data from the table the same way as from a [dictionary](../../sql-reference/dictionaries/index.md).
 
 Gets the data from [Join](../../engines/table-engines/special/join.md#creating-a-table) tables using the specified join key.
 
@@ -1927,69 +1927,6 @@ Result:
 └──────────────────────────────────────────────────┘
 ```
 
-## joinGetOrNull
-
-Allows you to extract data from a specific column in a Join table, similar to how you would access a value from a dictionary.
-
-Gets the data from [Join](../../engines/table-engines/special/join.md#creating-a-table) tables using the specified join key.
-
-Only supports tables created with the `ENGINE = Join(ANY, LEFT, <join_keys>)` statement.
-
-### Syntax
-
-```sql
-joinGet(join_storage_table_name, `value_column`, join_keys)
-```
-
-### Parameters
-
-- `join_storage_table_name` — an [identifier](../../sql-reference/syntax.md#syntax-identifiers) indicating where the search is performed. The identifier is searched in the default database (see setting `default_database` in the config file). To override the default database, use `USE db_name` or specify the database and the table through the separator `db_name.db_table` as in the example.
-- `value_column` — name of the column of the table that contains required data.
-- `join_keys` — list of keys.
-
-### Returned value
-
-Returns a list of values corresponded to list of keys.
-
-If certain does not exist in source table then `0` or `null` will be returned based on [join_use_nulls](../../operations/settings/settings.md#join_use_nulls) setting.
-
-More info about `join_use_nulls` in [Join operation](../../engines/table-engines/special/join.md).
-
-**Example**
-
-Input table:
-
-``` sql
-CREATE DATABASE db_test
-CREATE TABLE db_test.id_val(`id` UInt32, `val` UInt32) ENGINE = Join(ANY, LEFT, id) SETTINGS join_use_nulls = 1
-INSERT INTO db_test.id_val VALUES (1,11)(2,12)(4,13)
-```
-
-``` text
-┌─id─┬─val─┐
-│  4 │  13 │
-│  2 │  12 │
-│  1 │  11 │
-└────┴─────┘
-```
-
-Query:
-
-``` sql
-SELECT joinGet(db_test.id_val, 'val', toUInt32(number)) from numbers(4) SETTINGS join_use_nulls = 1
-```
-
-Result:
-
-``` text
-┌─joinGet(db_test.id_val, 'val', toUInt32(number))─┐
-│                                                0 │
-│                                               11 │
-│                                               12 │
-│                                                0 │
-└──────────────────────────────────────────────────┘
-```
-
 ## catboostEvaluate(path_to_model, feature_1, feature_2, …, feature_n)
 
 :::note

From a7350299396d5ba3f2322584195554a7d946562f Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Mon, 11 Mar 2024 12:50:54 +0000
Subject: [PATCH 159/443] Fix tests

---
 src/Common/DateLUTImpl.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Common/DateLUTImpl.h b/src/Common/DateLUTImpl.h
index 4087e77d588..082127e717c 100644
--- a/src/Common/DateLUTImpl.h
+++ b/src/Common/DateLUTImpl.h
@@ -255,7 +255,7 @@ private:
 
     static LUTIndex toLUTIndex(ExtendedDayNum d)
     {
-        return normalizeLUTIndex(static_cast<Int64>(d) + daynum_offset_epoch);
+        return normalizeLUTIndex(static_cast<Int64>(d + daynum_offset_epoch)); /// NOLINT
     }
 
     LUTIndex toLUTIndex(Time t) const

From 802bae9661a6f22a8c42a6f88f9816e3497d9355 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Mon, 11 Mar 2024 12:54:34 +0000
Subject: [PATCH 160/443] GCC --> clang pragmas

---
 base/base/coverage.cpp                             | 2 +-
 base/base/sort.h                                   | 6 +++---
 programs/client/Client.cpp                         | 4 ++--
 programs/copier/ClusterCopierApp.cpp               | 4 ++--
 programs/extract-from-config/ExtractFromConfig.cpp | 4 ++--
 programs/format/Format.cpp                         | 4 ++--
 programs/local/LocalServer.cpp                     | 4 ++--
 programs/obfuscator/Obfuscator.cpp                 | 4 ++--
 src/Common/SSH/Wrappers.cpp                        | 8 ++++----
 src/Functions/GatherUtils/Sources.h                | 8 ++++----
 10 files changed, 24 insertions(+), 24 deletions(-)

diff --git a/base/base/coverage.cpp b/base/base/coverage.cpp
index 99b897c4571..d96b3ea1e9a 100644
--- a/base/base/coverage.cpp
+++ b/base/base/coverage.cpp
@@ -1,7 +1,7 @@
 #include "coverage.h"
 #include <sys/mman.h>
 
-#pragma GCC diagnostic ignored "-Wreserved-identifier"
+#pragma clang diagnostic ignored "-Wreserved-identifier"
 
 
 /// WITH_COVERAGE enables the default implementation of code coverage,
diff --git a/base/base/sort.h b/base/base/sort.h
index 99bf8a0830e..e46c388d185 100644
--- a/base/base/sort.h
+++ b/base/base/sort.h
@@ -59,8 +59,8 @@ using ComparatorWrapper = Comparator;
 
 #endif
 
-#pragma GCC diagnostic push
-#pragma GCC diagnostic ignored "-Wold-style-cast"
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wold-style-cast"
 
 #include <miniselect/floyd_rivest_select.h>
 
@@ -115,7 +115,7 @@ void partial_sort(RandomIt first, RandomIt middle, RandomIt last)
     ::partial_sort(first, middle, last, comparator());
 }
 
-#pragma GCC diagnostic pop
+#pragma clang diagnostic pop
 
 template <typename RandomIt, typename Compare>
 void sort(RandomIt first, RandomIt last, Compare compare)
diff --git a/programs/client/Client.cpp b/programs/client/Client.cpp
index a2bd6b6016a..1c7e57dac76 100644
--- a/programs/client/Client.cpp
+++ b/programs/client/Client.cpp
@@ -1377,8 +1377,8 @@ void Client::readArguments(
 }
 
 
-#pragma GCC diagnostic ignored "-Wunused-function"
-#pragma GCC diagnostic ignored "-Wmissing-declarations"
+#pragma clang diagnostic ignored "-Wunused-function"
+#pragma clang diagnostic ignored "-Wmissing-declarations"
 
 int mainEntryClickHouseClient(int argc, char ** argv)
 {
diff --git a/programs/copier/ClusterCopierApp.cpp b/programs/copier/ClusterCopierApp.cpp
index fdf07dec61a..ed748a17a55 100644
--- a/programs/copier/ClusterCopierApp.cpp
+++ b/programs/copier/ClusterCopierApp.cpp
@@ -232,8 +232,8 @@ int ClusterCopierApp::main(const std::vector<std::string> &)
 
 }
 
-#pragma GCC diagnostic ignored "-Wunused-function"
-#pragma GCC diagnostic ignored "-Wmissing-declarations"
+#pragma clang diagnostic ignored "-Wunused-function"
+#pragma clang diagnostic ignored "-Wmissing-declarations"
 
 int mainEntryClickHouseClusterCopier(int argc, char ** argv)
 {
diff --git a/programs/extract-from-config/ExtractFromConfig.cpp b/programs/extract-from-config/ExtractFromConfig.cpp
index 56041ee382f..61d451664e3 100644
--- a/programs/extract-from-config/ExtractFromConfig.cpp
+++ b/programs/extract-from-config/ExtractFromConfig.cpp
@@ -109,8 +109,8 @@ static std::vector<std::string> extractFromConfig(
     return {configuration->getString(key)};
 }
 
-#pragma GCC diagnostic ignored "-Wunused-function"
-#pragma GCC diagnostic ignored "-Wmissing-declarations"
+#pragma clang diagnostic ignored "-Wunused-function"
+#pragma clang diagnostic ignored "-Wmissing-declarations"
 
 int mainEntryClickHouseExtractFromConfig(int argc, char ** argv)
 {
diff --git a/programs/format/Format.cpp b/programs/format/Format.cpp
index a1c51565ae3..c92106e2f90 100644
--- a/programs/format/Format.cpp
+++ b/programs/format/Format.cpp
@@ -70,8 +70,8 @@ void skipSpacesAndComments(const char*& pos, const char* end, bool print_comment
 
 }
 
-#pragma GCC diagnostic ignored "-Wunused-function"
-#pragma GCC diagnostic ignored "-Wmissing-declarations"
+#pragma clang diagnostic ignored "-Wunused-function"
+#pragma clang diagnostic ignored "-Wmissing-declarations"
 
 extern const char * auto_time_zones[];
 
diff --git a/programs/local/LocalServer.cpp b/programs/local/LocalServer.cpp
index 68f0e52ce08..99639d5e604 100644
--- a/programs/local/LocalServer.cpp
+++ b/programs/local/LocalServer.cpp
@@ -944,8 +944,8 @@ void LocalServer::readArguments(int argc, char ** argv, Arguments & common_argum
 
 }
 
-#pragma GCC diagnostic ignored "-Wunused-function"
-#pragma GCC diagnostic ignored "-Wmissing-declarations"
+#pragma clang diagnostic ignored "-Wunused-function"
+#pragma clang diagnostic ignored "-Wmissing-declarations"
 
 int mainEntryClickHouseLocal(int argc, char ** argv)
 {
diff --git a/programs/obfuscator/Obfuscator.cpp b/programs/obfuscator/Obfuscator.cpp
index 242e995e466..317d93aaf0c 100644
--- a/programs/obfuscator/Obfuscator.cpp
+++ b/programs/obfuscator/Obfuscator.cpp
@@ -1204,8 +1204,8 @@ public:
 
 }
 
-#pragma GCC diagnostic ignored "-Wunused-function"
-#pragma GCC diagnostic ignored "-Wmissing-declarations"
+#pragma clang diagnostic ignored "-Wunused-function"
+#clang GCC diagnostic ignored "-Wmissing-declarations"
 
 int mainEntryClickHouseObfuscator(int argc, char ** argv)
 try
diff --git a/src/Common/SSH/Wrappers.cpp b/src/Common/SSH/Wrappers.cpp
index 463338dbe3f..a9b9f758c6e 100644
--- a/src/Common/SSH/Wrappers.cpp
+++ b/src/Common/SSH/Wrappers.cpp
@@ -2,13 +2,13 @@
 # if USE_SSH
 #    include <stdexcept>
 
-#    pragma GCC diagnostic push
-#    pragma GCC diagnostic ignored "-Wreserved-macro-identifier"
-#    pragma GCC diagnostic ignored "-Wreserved-identifier"
+#    pragma clang diagnostic push
+#    pragma clang diagnostic ignored "-Wreserved-macro-identifier"
+#    pragma clang diagnostic ignored "-Wreserved-identifier"
 
 #    include <libssh/libssh.h>
 
-#    pragma GCC diagnostic pop
+#    pragma clang diagnostic pop
 
 namespace
 {
diff --git a/src/Functions/GatherUtils/Sources.h b/src/Functions/GatherUtils/Sources.h
index 222f9f19168..80fb9ce3900 100644
--- a/src/Functions/GatherUtils/Sources.h
+++ b/src/Functions/GatherUtils/Sources.h
@@ -140,9 +140,9 @@ struct NumericArraySource : public ArraySourceImpl<NumericArraySource<T>>
 
 
 /// The methods can be virtual or not depending on the template parameter. See IStringSource.
-#pragma GCC diagnostic push
-#pragma GCC diagnostic ignored "-Wsuggest-override"
-#pragma GCC diagnostic ignored "-Wsuggest-destructor-override"
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wsuggest-override"
+#pragma clang diagnostic ignored "-Wsuggest-destructor-override"
 
 template <typename Base>
 struct ConstSource : public Base
@@ -231,7 +231,7 @@ struct ConstSource : public Base
     }
 };
 
-#pragma GCC diagnostic pop
+#pragma clang diagnostic pop
 
 struct StringSource
 {

From a7568ddbfaf107e9dafa6fa83c9d5f747a0e7153 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Mon, 11 Mar 2024 12:58:20 +0000
Subject: [PATCH 161/443] Fix spelling

---
 utils/check-style/aspell-ignore/en/aspell-dict.txt | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/utils/check-style/aspell-ignore/en/aspell-dict.txt b/utils/check-style/aspell-ignore/en/aspell-dict.txt
index 3614bcb7452..473907b45ae 100644
--- a/utils/check-style/aspell-ignore/en/aspell-dict.txt
+++ b/utils/check-style/aspell-ignore/en/aspell-dict.txt
@@ -1846,6 +1846,7 @@ linearized
 lineasstring
 linefeeds
 lineorder
+linestring
 linux
 llvm
 loadDefaultCAFile
@@ -2204,7 +2205,9 @@ rankCorr
 rapidjson
 rawblob
 readWKTMultiPolygon
+readWKTPoint
 readWKTPolygon
+readWKTRing
 readahead
 readline
 readme

From c4f29af8ec0927c09b9d12d83767607020915c30 Mon Sep 17 00:00:00 2001
From: Max Kainov <max.kainov@clickhouse.com>
Date: Mon, 11 Mar 2024 13:35:05 +0000
Subject: [PATCH 162/443] CI: fix nightly job issue

 #do_not_test
---
 tests/ci/ci.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tests/ci/ci.py b/tests/ci/ci.py
index 1eec9a6771b..9d57f161be3 100644
--- a/tests/ci/ci.py
+++ b/tests/ci/ci.py
@@ -1756,7 +1756,8 @@ def main() -> int:
         result["build"] = build_digest
         result["docs"] = docs_digest
         result["ci_flags"] = ci_flags
-        result["stages_data"] = _generate_ci_stage_config(jobs_data)
+        if not args.skip_jobs:
+            result["stages_data"] = _generate_ci_stage_config(jobs_data)
         result["jobs_data"] = jobs_data
         result["docker_data"] = docker_data
     ### CONFIGURE action: end

From 19d8256fa83a4e8353dcad372067085ec8f0828d Mon Sep 17 00:00:00 2001
From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com>
Date: Mon, 11 Mar 2024 14:44:19 +0100
Subject: [PATCH 163/443] Update test.py

---
 tests/integration/test_backup_restore_s3/test.py | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/tests/integration/test_backup_restore_s3/test.py b/tests/integration/test_backup_restore_s3/test.py
index f3f4837c317..d65fc1f09d6 100644
--- a/tests/integration/test_backup_restore_s3/test.py
+++ b/tests/integration/test_backup_restore_s3/test.py
@@ -130,11 +130,11 @@ def check_system_tables(backup_query_id=None):
         if disk
     ]
     expected_disks = (
-        ("default", "local", "None", "None"),
-        ("disk_s3", "object_storage", "s3", "local"),
-        ("disk_s3_cache", "object_storage", "s3", "local"),
-        ("disk_s3_other_bucket", "object_storage", "s3", "local"),
-        ("disk_s3_plain", "object_storage", "s3", "plain"),
+        ("default", "Local", "None", "None"),
+        ("disk_s3", "ObjectStorage", "S3", "Local"),
+        ("disk_s3_cache", "ObjectStorage", "S3", "Local"),
+        ("disk_s3_other_bucket", "ObjectStorage", "S3", "Local"),
+        ("disk_s3_plain", "ObjectStorage", "S3", "Plain"),
     )
     assert len(expected_disks) == len(disks)
     for expected_disk in expected_disks:

From 9bada70f45654495a30e394d94a374a862c24fb5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Mon, 11 Mar 2024 14:52:32 +0100
Subject: [PATCH 164/443] Remove a bunch of transitive dependencies

---
 src/Backups/BackupCoordinationRemote.cpp     |  2 ++
 src/Formats/ReadSchemaUtils.cpp              |  9 +++++----
 src/Interpreters/DatabaseCatalog.h           |  7 +++----
 src/Interpreters/GraceHashJoin.cpp           | 15 ++++++---------
 src/Interpreters/TemporaryDataOnDisk.h       |  6 +++---
 src/Planner/PlannerExpressionAnalysis.cpp    |  2 ++
 src/Processors/QueryPlan/AggregatingStep.cpp |  1 +
 src/Processors/QueryPlan/CubeStep.cpp        |  7 ++++---
 src/Processors/QueryPlan/SortingStep.cpp     |  1 +
 9 files changed, 27 insertions(+), 23 deletions(-)

diff --git a/src/Backups/BackupCoordinationRemote.cpp b/src/Backups/BackupCoordinationRemote.cpp
index 9c509858b2a..ec652f20069 100644
--- a/src/Backups/BackupCoordinationRemote.cpp
+++ b/src/Backups/BackupCoordinationRemote.cpp
@@ -14,6 +14,8 @@
 #include <IO/WriteBufferFromString.h>
 #include <IO/WriteHelpers.h>
 
+#include <boost/algorithm/string.hpp>
+
 
 namespace DB
 {
diff --git a/src/Formats/ReadSchemaUtils.cpp b/src/Formats/ReadSchemaUtils.cpp
index 5badf4301bf..b05b768899b 100644
--- a/src/Formats/ReadSchemaUtils.cpp
+++ b/src/Formats/ReadSchemaUtils.cpp
@@ -1,10 +1,11 @@
 #include <Formats/ReadSchemaUtils.h>
-#include <Interpreters/Context.h>
-#include <Processors/Formats/ISchemaReader.h>
-#include <Common/assert_cast.h>
-#include <IO/WithFileSize.h>
 #include <IO/EmptyReadBuffer.h>
 #include <IO/PeekableReadBuffer.h>
+#include <IO/WithFileSize.h>
+#include <Interpreters/Context.h>
+#include <Processors/Formats/ISchemaReader.h>
+#include <Storages/IStorage.h>
+#include <Common/assert_cast.h>
 
 namespace DB
 {
diff --git a/src/Interpreters/DatabaseCatalog.h b/src/Interpreters/DatabaseCatalog.h
index 4fe114cc493..6f05a3cea0f 100644
--- a/src/Interpreters/DatabaseCatalog.h
+++ b/src/Interpreters/DatabaseCatalog.h
@@ -1,15 +1,14 @@
 #pragma once
 
 #include <Core/UUID.h>
+#include <Databases/IDatabase.h>
+#include <Databases/TablesDependencyGraph.h>
 #include <Interpreters/Context_fwd.h>
 #include <Interpreters/StorageID.h>
-#include <Databases/TablesDependencyGraph.h>
 #include <Parsers/IAST_fwd.h>
 #include <Storages/IStorage_fwd.h>
-#include "Common/NamePrompter.h"
+#include <Common/NamePrompter.h>
 #include <Common/SharedMutex.h>
-#include "Storages/IStorage.h"
-#include "Databases/IDatabase.h"
 
 #include <boost/noncopyable.hpp>
 #include <Poco/Logger.h>
diff --git a/src/Interpreters/GraceHashJoin.cpp b/src/Interpreters/GraceHashJoin.cpp
index 5fb92a68a29..53d1f48c291 100644
--- a/src/Interpreters/GraceHashJoin.cpp
+++ b/src/Interpreters/GraceHashJoin.cpp
@@ -1,21 +1,18 @@
+#include <Compression/CompressedWriteBuffer.h>
+#include <Formats/NativeWriter.h>
+#include <Formats/formatBlock.h>
+#include <Interpreters/Context.h>
 #include <Interpreters/GraceHashJoin.h>
 #include <Interpreters/HashJoin.h>
 #include <Interpreters/TableJoin.h>
-
-#include <Formats/NativeWriter.h>
 #include <Interpreters/TemporaryDataOnDisk.h>
-
-#include <Compression/CompressedWriteBuffer.h>
+#include <base/FnTraits.h>
 #include <Common/formatReadable.h>
 #include <Common/logger_useful.h>
 #include <Common/thread_local_rng.h>
 
-#include <base/FnTraits.h>
-#include <fmt/format.h>
-
-#include <Formats/formatBlock.h>
-
 #include <numeric>
+#include <fmt/format.h>
 
 
 namespace CurrentMetrics
diff --git a/src/Interpreters/TemporaryDataOnDisk.h b/src/Interpreters/TemporaryDataOnDisk.h
index e57d9130369..8b0649be1b1 100644
--- a/src/Interpreters/TemporaryDataOnDisk.h
+++ b/src/Interpreters/TemporaryDataOnDisk.h
@@ -2,11 +2,11 @@
 
 #include <boost/noncopyable.hpp>
 
-#include <Interpreters/Context.h>
-#include <Disks/TemporaryFileOnDisk.h>
+#include <Core/Block.h>
 #include <Disks/IVolume.h>
-#include <Common/CurrentMetrics.h>
+#include <Disks/TemporaryFileOnDisk.h>
 #include <Interpreters/Cache/FileSegment.h>
+#include <Common/CurrentMetrics.h>
 
 
 namespace CurrentMetrics
diff --git a/src/Planner/PlannerExpressionAnalysis.cpp b/src/Planner/PlannerExpressionAnalysis.cpp
index 52001eb27c5..30d90a68072 100644
--- a/src/Planner/PlannerExpressionAnalysis.cpp
+++ b/src/Planner/PlannerExpressionAnalysis.cpp
@@ -3,6 +3,8 @@
 #include <DataTypes/DataTypesNumber.h>
 #include <DataTypes/DataTypeNullable.h>
 
+#include <Interpreters/Context.h>
+
 #include <Analyzer/FunctionNode.h>
 #include <Analyzer/ConstantNode.h>
 #include <Analyzer/WindowNode.h>
diff --git a/src/Processors/QueryPlan/AggregatingStep.cpp b/src/Processors/QueryPlan/AggregatingStep.cpp
index f374a7b7b10..a76bacdd97b 100644
--- a/src/Processors/QueryPlan/AggregatingStep.cpp
+++ b/src/Processors/QueryPlan/AggregatingStep.cpp
@@ -8,6 +8,7 @@
 #include <IO/Operators.h>
 #include <Interpreters/Aggregator.h>
 #include <Interpreters/Context.h>
+#include <Interpreters/ExpressionActions.h>
 #include <Processors/Merges/AggregatingSortedTransform.h>
 #include <Processors/Merges/FinishAggregatingInOrderTransform.h>
 #include <Processors/QueryPlan/AggregatingStep.h>
diff --git a/src/Processors/QueryPlan/CubeStep.cpp b/src/Processors/QueryPlan/CubeStep.cpp
index 0c632c346c7..d010a3327a6 100644
--- a/src/Processors/QueryPlan/CubeStep.cpp
+++ b/src/Processors/QueryPlan/CubeStep.cpp
@@ -1,10 +1,11 @@
+#include <DataTypes/DataTypesNumber.h>
+#include <Functions/FunctionFactory.h>
+#include <Interpreters/ExpressionActions.h>
+#include <Processors/QueryPlan/AggregatingStep.h>
 #include <Processors/QueryPlan/CubeStep.h>
 #include <Processors/Transforms/CubeTransform.h>
 #include <Processors/Transforms/ExpressionTransform.h>
-#include <Processors/QueryPlan/AggregatingStep.h>
 #include <QueryPipeline/QueryPipelineBuilder.h>
-#include <DataTypes/DataTypesNumber.h>
-#include <Functions/FunctionFactory.h>
 
 namespace DB
 {
diff --git a/src/Processors/QueryPlan/SortingStep.cpp b/src/Processors/QueryPlan/SortingStep.cpp
index 641b9036d4c..d0491cb4b82 100644
--- a/src/Processors/QueryPlan/SortingStep.cpp
+++ b/src/Processors/QueryPlan/SortingStep.cpp
@@ -1,6 +1,7 @@
 #include <memory>
 #include <stdexcept>
 #include <IO/Operators.h>
+#include <Interpreters/Context.h>
 #include <Processors/Merges/MergingSortedTransform.h>
 #include <Processors/QueryPlan/SortingStep.h>
 #include <Processors/Transforms/FinishSortingTransform.h>

From 91de3825171eefb8f828c2907181b6a5e6b0f017 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <benjamin.antal@clickhouse.com>
Date: Mon, 11 Mar 2024 14:00:01 +0000
Subject: [PATCH 165/443] Spit `DictionaryPipelineExecutor` into separate file

---
 src/Dictionaries/CacheDictionary.cpp          |  2 +-
 .../DictionaryPipelineExecutor.cpp            | 42 +++++++++++++++++++
 src/Dictionaries/DictionaryPipelineExecutor.h | 27 ++++++++++++
 src/Dictionaries/DictionarySourceHelpers.cpp  | 29 -------------
 src/Dictionaries/DictionarySourceHelpers.h    | 17 --------
 src/Dictionaries/FlatDictionary.cpp           |  2 +-
 src/Dictionaries/HashedArrayDictionary.cpp    |  1 +
 src/Dictionaries/HashedDictionary.h           |  2 +-
 src/Dictionaries/IPAddressDictionary.cpp      |  1 +
 src/Dictionaries/PolygonDictionary.cpp        |  3 +-
 src/Dictionaries/RangeHashedDictionary.h      |  6 +--
 src/Dictionaries/RegExpTreeDictionary.cpp     |  1 +
 .../registerRangeHashedDictionary.cpp         |  5 ++-
 13 files changed, 82 insertions(+), 56 deletions(-)
 create mode 100644 src/Dictionaries/DictionaryPipelineExecutor.cpp
 create mode 100644 src/Dictionaries/DictionaryPipelineExecutor.h

diff --git a/src/Dictionaries/CacheDictionary.cpp b/src/Dictionaries/CacheDictionary.cpp
index 8444042db9e..b136d5ebd71 100644
--- a/src/Dictionaries/CacheDictionary.cpp
+++ b/src/Dictionaries/CacheDictionary.cpp
@@ -9,7 +9,7 @@
 #include <Common/ProfilingScopedRWLock.h>
 
 #include <Dictionaries/DictionarySource.h>
-#include <Dictionaries/DictionarySourceHelpers.h>
+#include <Dictionaries/DictionaryPipelineExecutor.h>
 #include <Dictionaries/HierarchyDictionariesUtils.h>
 
 #include <QueryPipeline/QueryPipelineBuilder.h>
diff --git a/src/Dictionaries/DictionaryPipelineExecutor.cpp b/src/Dictionaries/DictionaryPipelineExecutor.cpp
new file mode 100644
index 00000000000..30d1ab95f53
--- /dev/null
+++ b/src/Dictionaries/DictionaryPipelineExecutor.cpp
@@ -0,0 +1,42 @@
+#include <Dictionaries/DictionaryPipelineExecutor.h>
+
+#include <Core/Block.h>
+#include <Processors/Executors/PullingAsyncPipelineExecutor.h>
+#include <Processors/Executors/PullingPipelineExecutor.h>
+#include <QueryPipeline/QueryPipeline.h>
+
+namespace DB
+{
+
+namespace ErrorCodes
+{
+extern const int LOGICAL_ERROR;
+}
+
+DictionaryPipelineExecutor::DictionaryPipelineExecutor(QueryPipeline & pipeline_, bool async)
+    : async_executor(async ? std::make_unique<PullingAsyncPipelineExecutor>(pipeline_) : nullptr)
+    , executor(async ? nullptr : std::make_unique<PullingPipelineExecutor>(pipeline_))
+{
+}
+
+bool DictionaryPipelineExecutor::pull(Block & block)
+{
+    if (async_executor)
+    {
+        while (true)
+        {
+            bool has_data = async_executor->pull(block);
+            if (has_data && !block)
+                continue;
+            return has_data;
+        }
+    }
+    else if (executor)
+        return executor->pull(block);
+    else
+        throw Exception(ErrorCodes::LOGICAL_ERROR, "DictionaryPipelineExecutor is not initialized");
+}
+
+DictionaryPipelineExecutor::~DictionaryPipelineExecutor() = default;
+
+}
diff --git a/src/Dictionaries/DictionaryPipelineExecutor.h b/src/Dictionaries/DictionaryPipelineExecutor.h
new file mode 100644
index 00000000000..601213e5039
--- /dev/null
+++ b/src/Dictionaries/DictionaryPipelineExecutor.h
@@ -0,0 +1,27 @@
+#pragma once
+
+#include <memory>
+
+namespace DB
+{
+
+class Block;
+class QueryPipeline;
+class PullingAsyncPipelineExecutor;
+class PullingPipelineExecutor;
+
+/// Wrapper for `Pulling(Async)PipelineExecutor` to dynamically dispatch calls to the right executor
+class DictionaryPipelineExecutor
+{
+public:
+    DictionaryPipelineExecutor(QueryPipeline & pipeline_, bool async);
+    bool pull(Block & block);
+
+    ~DictionaryPipelineExecutor();
+
+private:
+    std::unique_ptr<PullingAsyncPipelineExecutor> async_executor;
+    std::unique_ptr<PullingPipelineExecutor> executor;
+};
+
+}
diff --git a/src/Dictionaries/DictionarySourceHelpers.cpp b/src/Dictionaries/DictionarySourceHelpers.cpp
index d9a4d9ccbcf..f0e1bc4109a 100644
--- a/src/Dictionaries/DictionarySourceHelpers.cpp
+++ b/src/Dictionaries/DictionarySourceHelpers.cpp
@@ -9,15 +9,11 @@
 #include <Poco/Util/AbstractConfiguration.h>
 #include <Common/SettingsChanges.h>
 
-#include <Processors/Executors/PullingPipelineExecutor.h>
-#include <Processors/Executors/PullingAsyncPipelineExecutor.h>
-
 namespace DB
 {
 
 namespace ErrorCodes
 {
-    extern const int LOGICAL_ERROR;
     extern const int SIZES_OF_COLUMNS_DOESNT_MATCH;
 }
 
@@ -135,29 +131,4 @@ String TransformWithAdditionalColumns::getName() const
     return "TransformWithAdditionalColumns";
 }
 
-DictionaryPipelineExecutor::DictionaryPipelineExecutor(QueryPipeline & pipeline_, bool async)
-    : async_executor(async ? std::make_unique<PullingAsyncPipelineExecutor>(pipeline_) : nullptr)
-    , executor(async ? nullptr : std::make_unique<PullingPipelineExecutor>(pipeline_))
-{}
-
-bool DictionaryPipelineExecutor::pull(Block & block)
-{
-    if (async_executor)
-    {
-        while (true)
-        {
-            bool has_data = async_executor->pull(block);
-            if (has_data && !block)
-                continue;
-            return has_data;
-        }
-    }
-    else if (executor)
-        return executor->pull(block);
-    else
-        throw Exception(ErrorCodes::LOGICAL_ERROR, "DictionaryPipelineExecutor is not initialized");
-}
-
-DictionaryPipelineExecutor::~DictionaryPipelineExecutor() = default;
-
 }
diff --git a/src/Dictionaries/DictionarySourceHelpers.h b/src/Dictionaries/DictionarySourceHelpers.h
index a545b5cdac7..39c6e7b3c42 100644
--- a/src/Dictionaries/DictionarySourceHelpers.h
+++ b/src/Dictionaries/DictionarySourceHelpers.h
@@ -16,10 +16,6 @@ namespace DB
 struct DictionaryStructure;
 class SettingsChanges;
 
-class PullingPipelineExecutor;
-class PullingAsyncPipelineExecutor;
-class QueryPipeline;
-
 /// For simple key
 
 Block blockForIds(
@@ -55,17 +51,4 @@ private:
     size_t current_range_index = 0;
 };
 
-/// Wrapper for `Pulling(Async)PipelineExecutor` to dynamically dispatch calls to the right executor
-class DictionaryPipelineExecutor
-{
-public:
-    DictionaryPipelineExecutor(QueryPipeline & pipeline_, bool async);
-    bool pull(Block & block);
-
-    ~DictionaryPipelineExecutor();
-private:
-    std::unique_ptr<PullingAsyncPipelineExecutor> async_executor;
-    std::unique_ptr<PullingPipelineExecutor> executor;
-};
-
 }
diff --git a/src/Dictionaries/FlatDictionary.cpp b/src/Dictionaries/FlatDictionary.cpp
index efb7d0a176c..fc58ff525bd 100644
--- a/src/Dictionaries/FlatDictionary.cpp
+++ b/src/Dictionaries/FlatDictionary.cpp
@@ -15,7 +15,7 @@
 #include <QueryPipeline/QueryPipelineBuilder.h>
 
 #include <Dictionaries/DictionarySource.h>
-#include <Dictionaries/DictionarySourceHelpers.h>
+#include <Dictionaries/DictionaryPipelineExecutor.h>
 #include <Dictionaries/DictionaryFactory.h>
 #include <Dictionaries/HierarchyDictionariesUtils.h>
 
diff --git a/src/Dictionaries/HashedArrayDictionary.cpp b/src/Dictionaries/HashedArrayDictionary.cpp
index d09f402143e..2420c07277c 100644
--- a/src/Dictionaries/HashedArrayDictionary.cpp
+++ b/src/Dictionaries/HashedArrayDictionary.cpp
@@ -10,6 +10,7 @@
 #include <Dictionaries/ClickHouseDictionarySource.h>
 #include <Dictionaries/DictionarySource.h>
 #include <Dictionaries/DictionarySourceHelpers.h>
+#include <Dictionaries/DictionaryPipelineExecutor.h>
 #include <Dictionaries/DictionaryFactory.h>
 #include <Dictionaries/HierarchyDictionariesUtils.h>
 
diff --git a/src/Dictionaries/HashedDictionary.h b/src/Dictionaries/HashedDictionary.h
index b3b8cc56868..46a0af487f5 100644
--- a/src/Dictionaries/HashedDictionary.h
+++ b/src/Dictionaries/HashedDictionary.h
@@ -6,7 +6,7 @@
 #include <Dictionaries/DictionaryHelpers.h>
 #include <Dictionaries/ClickHouseDictionarySource.h>
 #include <Dictionaries/DictionarySource.h>
-#include <Dictionaries/DictionarySourceHelpers.h>
+#include <Dictionaries/DictionaryPipelineExecutor.h>
 #include <Dictionaries/HierarchyDictionariesUtils.h>
 #include <Dictionaries/HashedDictionaryCollectionType.h>
 #include <Dictionaries/HashedDictionaryCollectionTraits.h>
diff --git a/src/Dictionaries/IPAddressDictionary.cpp b/src/Dictionaries/IPAddressDictionary.cpp
index e1c9572e607..1bc6d16c932 100644
--- a/src/Dictionaries/IPAddressDictionary.cpp
+++ b/src/Dictionaries/IPAddressDictionary.cpp
@@ -19,6 +19,7 @@
 #include <Dictionaries/ClickHouseDictionarySource.h>
 #include <Dictionaries/DictionarySource.h>
 #include <Dictionaries/DictionarySourceHelpers.h>
+#include <Dictionaries/DictionaryPipelineExecutor.h>
 #include <Dictionaries/DictionaryFactory.h>
 #include <Functions/FunctionHelpers.h>
 
diff --git a/src/Dictionaries/PolygonDictionary.cpp b/src/Dictionaries/PolygonDictionary.cpp
index 485b48d9d81..1456a0db750 100644
--- a/src/Dictionaries/PolygonDictionary.cpp
+++ b/src/Dictionaries/PolygonDictionary.cpp
@@ -1,6 +1,5 @@
 #include "PolygonDictionary.h"
 
-#include <numeric>
 #include <cmath>
 
 #include <base/sort.h>
@@ -15,7 +14,7 @@
 #include <Processors/Sources/SourceFromSingleChunk.h>
 #include <Dictionaries/DictionaryFactory.h>
 #include <Dictionaries/DictionarySource.h>
-#include <Dictionaries/DictionarySourceHelpers.h>
+#include <Dictionaries/DictionaryPipelineExecutor.h>
 
 
 namespace DB
diff --git a/src/Dictionaries/RangeHashedDictionary.h b/src/Dictionaries/RangeHashedDictionary.h
index 1a6ee7e81d4..509b991b30c 100644
--- a/src/Dictionaries/RangeHashedDictionary.h
+++ b/src/Dictionaries/RangeHashedDictionary.h
@@ -15,6 +15,8 @@
 #include <Dictionaries/IDictionary.h>
 #include <Dictionaries/IDictionarySource.h>
 #include <Dictionaries/DictionaryHelpers.h>
+#include <Dictionaries/DictionarySource.h>
+#include <Dictionaries/DictionaryPipelineExecutor.h>
 
 #include <DataTypes/DataTypesNumber.h>
 #include <DataTypes/DataTypesDecimal.h>
@@ -29,10 +31,6 @@
 #include <Functions/FunctionHelpers.h>
 #include <Interpreters/castColumn.h>
 
-#include <Dictionaries/ClickHouseDictionarySource.h>
-#include <Dictionaries/DictionarySource.h>
-#include <Dictionaries/DictionarySourceHelpers.h>
-
 
 namespace DB
 {
diff --git a/src/Dictionaries/RegExpTreeDictionary.cpp b/src/Dictionaries/RegExpTreeDictionary.cpp
index 4d82aa9ca0e..8930074bbe0 100644
--- a/src/Dictionaries/RegExpTreeDictionary.cpp
+++ b/src/Dictionaries/RegExpTreeDictionary.cpp
@@ -27,6 +27,7 @@
 #include <Dictionaries/DictionaryHelpers.h>
 #include <Dictionaries/DictionaryStructure.h>
 #include <Dictionaries/DictionarySourceHelpers.h>
+#include <Dictionaries/DictionaryPipelineExecutor.h>
 #include <Dictionaries/RegExpTreeDictionary.h>
 #include <Dictionaries/YAMLRegExpTreeDictionarySource.h>
 
diff --git a/src/Dictionaries/registerRangeHashedDictionary.cpp b/src/Dictionaries/registerRangeHashedDictionary.cpp
index 4e20abfdb79..8123b811198 100644
--- a/src/Dictionaries/registerRangeHashedDictionary.cpp
+++ b/src/Dictionaries/registerRangeHashedDictionary.cpp
@@ -1,5 +1,8 @@
-#include "RangeHashedDictionary.h"
+#include <Dictionaries/RangeHashedDictionary.h>
+
 #include <Dictionaries/DictionarySource.h>
+#include <Dictionaries/ClickHouseDictionarySource.h>
+#include <Dictionaries/DictionarySourceHelpers.h>
 #include <Dictionaries/DictionaryFactory.h>
 
 namespace DB

From 8b5ccb4735365ef81af4debcc3180f296452268d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Mon, 11 Mar 2024 15:53:46 +0100
Subject: [PATCH 166/443] Remove one template

---
 src/Interpreters/Aggregator.cpp | 65 +++++++++++++--------------------
 src/Interpreters/Aggregator.h   |  4 +-
 2 files changed, 28 insertions(+), 41 deletions(-)

diff --git a/src/Interpreters/Aggregator.cpp b/src/Interpreters/Aggregator.cpp
index 80a98683867..69625dbd57d 100644
--- a/src/Interpreters/Aggregator.cpp
+++ b/src/Interpreters/Aggregator.cpp
@@ -2609,8 +2609,9 @@ void NO_INLINE Aggregator::mergeDataNullKey(
     }
 }
 
-template <typename Method, bool use_compiled_functions, bool prefetch, typename Table>
-void NO_INLINE Aggregator::mergeDataImpl(Table & table_dst, Table & table_src, Arena * arena) const
+template <typename Method, bool prefetch, typename Table>
+void NO_INLINE
+Aggregator::mergeDataImpl(Table & table_dst, Table & table_src, Arena * arena, bool use_compiled_functions [[maybe_unused]]) const
 {
     if constexpr (Method::low_cardinality_optimization || Method::one_key_nullable_optimization)
         mergeDataNullKey<Method, Table>(table_dst, table_src, arena);
@@ -2637,7 +2638,7 @@ void NO_INLINE Aggregator::mergeDataImpl(Table & table_dst, Table & table_src, A
     table_src.clearAndShrink();
 
 #if USE_EMBEDDED_COMPILER
-    if constexpr (use_compiled_functions)
+    if (use_compiled_functions)
     {
         const auto & compiled_functions = compiled_aggregate_functions_holder->compiled_aggregate_functions;
         compiled_functions.merge_aggregate_states_function(dst_places.data(), src_places.data(), dst_places.size());
@@ -2787,26 +2788,16 @@ void NO_INLINE Aggregator::mergeSingleLevelDataImpl(
 
         if (!no_more_keys)
         {
+            bool use_compiled_functions = false;
 #if USE_EMBEDDED_COMPILER
-            if (compiled_aggregate_functions_holder)
-            {
-                if (prefetch)
-                    mergeDataImpl<Method, true, true>(
-                        getDataVariant<Method>(*res).data, getDataVariant<Method>(current).data, res->aggregates_pool);
-                else
-                    mergeDataImpl<Method, true, false>(
-                        getDataVariant<Method>(*res).data, getDataVariant<Method>(current).data, res->aggregates_pool);
-            }
-            else
+            use_compiled_functions = compiled_aggregate_functions_holder != nullptr;
 #endif
-            {
-                if (prefetch)
-                    mergeDataImpl<Method, false, true>(
-                        getDataVariant<Method>(*res).data, getDataVariant<Method>(current).data, res->aggregates_pool);
-                else
-                    mergeDataImpl<Method, false, false>(
-                        getDataVariant<Method>(*res).data, getDataVariant<Method>(current).data, res->aggregates_pool);
-            }
+            if (prefetch)
+                mergeDataImpl<Method, true>(
+                    getDataVariant<Method>(*res).data, getDataVariant<Method>(current).data, res->aggregates_pool, use_compiled_functions);
+            else
+                mergeDataImpl<Method, false>(
+                    getDataVariant<Method>(*res).data, getDataVariant<Method>(current).data, res->aggregates_pool, use_compiled_functions);
         }
         else if (res->without_key)
         {
@@ -2851,26 +2842,22 @@ void NO_INLINE Aggregator::mergeBucketImpl(
             return;
 
         AggregatedDataVariants & current = *data[result_num];
+        bool use_compiled_functions = false;
 #if USE_EMBEDDED_COMPILER
-        if (compiled_aggregate_functions_holder)
-        {
-            if (prefetch)
-                mergeDataImpl<Method, true, true>(
-                    getDataVariant<Method>(*res).data.impls[bucket], getDataVariant<Method>(current).data.impls[bucket], arena);
-            else
-                mergeDataImpl<Method, true, false>(
-                    getDataVariant<Method>(*res).data.impls[bucket], getDataVariant<Method>(current).data.impls[bucket], arena);
-        }
-        else
+        use_compiled_functions = compiled_aggregate_functions_holder != nullptr;
 #endif
-        {
-            if (prefetch)
-                mergeDataImpl<Method, false, true>(
-                    getDataVariant<Method>(*res).data.impls[bucket], getDataVariant<Method>(current).data.impls[bucket], arena);
-            else
-                mergeDataImpl<Method, false, false>(
-                    getDataVariant<Method>(*res).data.impls[bucket], getDataVariant<Method>(current).data.impls[bucket], arena);
-        }
+        if (prefetch)
+            mergeDataImpl<Method, true>(
+                getDataVariant<Method>(*res).data.impls[bucket],
+                getDataVariant<Method>(current).data.impls[bucket],
+                arena,
+                use_compiled_functions);
+        else
+            mergeDataImpl<Method, false>(
+                getDataVariant<Method>(*res).data.impls[bucket],
+                getDataVariant<Method>(current).data.impls[bucket],
+                arena,
+                use_compiled_functions);
     }
 }
 
diff --git a/src/Interpreters/Aggregator.h b/src/Interpreters/Aggregator.h
index 375b8986101..67e82cdd784 100644
--- a/src/Interpreters/Aggregator.h
+++ b/src/Interpreters/Aggregator.h
@@ -1429,8 +1429,8 @@ private:
             Arena * arena) const;
 
     /// Merge data from hash table `src` into `dst`.
-    template <typename Method, bool use_compiled_functions, bool prefetch, typename Table>
-    void mergeDataImpl(Table & table_dst, Table & table_src, Arena * arena) const;
+    template <typename Method, bool prefetch, typename Table>
+    void mergeDataImpl(Table & table_dst, Table & table_src, Arena * arena, bool use_compiled_functions) const;
 
     /// Merge data from hash table `src` into `dst`, but only for keys that already exist in dst. In other cases, merge the data into `overflows`.
     template <typename Method, typename Table>

From 724cc903afb9283a8369a62a836d04eceae42e57 Mon Sep 17 00:00:00 2001
From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com>
Date: Mon, 11 Mar 2024 15:56:02 +0100
Subject: [PATCH 167/443] Restart CI

---
 tests/queries/0_stateless/03001_parallel_parsing_deadlock.sh | 1 -
 1 file changed, 1 deletion(-)

diff --git a/tests/queries/0_stateless/03001_parallel_parsing_deadlock.sh b/tests/queries/0_stateless/03001_parallel_parsing_deadlock.sh
index 1bf21dfc53b..6cd5c3b486c 100755
--- a/tests/queries/0_stateless/03001_parallel_parsing_deadlock.sh
+++ b/tests/queries/0_stateless/03001_parallel_parsing_deadlock.sh
@@ -9,4 +9,3 @@ DATA_FILE=$CLICKHOUSE_TEST_UNIQUE_NAME.csv
 $CLICKHOUSE_LOCAL -q "select number > 1000000 ? 'error' : toString(number) from numbers(2000000) format CSV" > $DATA_FILE
 $CLICKHOUSE_LOCAL -q "select * from file($DATA_FILE, CSV, 'x UInt64') format Null settings input_format_allow_errors_ratio=1"
 rm $DATA_FILE
-

From ecc30448baa1c6283f3f0f13c83cfd1bf4428b9b Mon Sep 17 00:00:00 2001
From: avogar <avogar@clickhouse.com>
Date: Mon, 11 Mar 2024 15:26:29 +0000
Subject: [PATCH 168/443] Fix filtering when
 optimize_use_implicit_projections=1

---
 .../optimizeUseAggregateProjection.cpp        |  4 ++
 src/Storages/VirtualColumnUtils.cpp           |  2 +-
 src/Storages/VirtualColumnUtils.h             |  3 +
 ...ions_non_deterministoc_functions.reference | 55 +++++++++++++++++++
 ...rojections_non_deterministoc_functions.sql | 28 ++++++++++
 5 files changed, 91 insertions(+), 1 deletion(-)
 create mode 100644 tests/queries/0_stateless/03008_filter_projections_non_deterministoc_functions.reference
 create mode 100644 tests/queries/0_stateless/03008_filter_projections_non_deterministoc_functions.sql

diff --git a/src/Processors/QueryPlan/Optimizations/optimizeUseAggregateProjection.cpp b/src/Processors/QueryPlan/Optimizations/optimizeUseAggregateProjection.cpp
index 91f4213ff43..b40fea47b3c 100644
--- a/src/Processors/QueryPlan/Optimizations/optimizeUseAggregateProjection.cpp
+++ b/src/Processors/QueryPlan/Optimizations/optimizeUseAggregateProjection.cpp
@@ -18,6 +18,7 @@
 
 #include <Common/logger_useful.h>
 #include <Storages/StorageDummy.h>
+#include <Storages/VirtualColumnUtils.h>
 #include <Planner/PlannerExpressionAnalysis.h>
 #include <Interpreters/InterpreterSelectQuery.h>
 #include <Interpreters/InterpreterSelectQueryAnalyzer.h>
@@ -464,6 +465,9 @@ AggregateProjectionCandidates getAggregateProjectionCandidates(
     // LOG_TRACE(getLogger("optimizeUseProjections"), "Query DAG: {}", dag.dag->dumpDAG());
 
     candidates.has_filter = dag.filter_node;
+    /// We can't use minmax projection if filter has non-deterministic functions.
+    if (dag.filter_node && !VirtualColumnUtils::isDeterministicInScopeOfQuery(dag.filter_node))
+        can_use_minmax_projection = false;
 
     if (can_use_minmax_projection)
     {
diff --git a/src/Storages/VirtualColumnUtils.cpp b/src/Storages/VirtualColumnUtils.cpp
index 897090223d6..c3ac27903c9 100644
--- a/src/Storages/VirtualColumnUtils.cpp
+++ b/src/Storages/VirtualColumnUtils.cpp
@@ -238,7 +238,7 @@ static bool canEvaluateSubtree(const ActionsDAG::Node * node, const Block & allo
     return true;
 }
 
-static bool isDeterministicInScopeOfQuery(const ActionsDAG::Node * node)
+bool isDeterministicInScopeOfQuery(const ActionsDAG::Node * node)
 {
     for (const auto * child : node->children)
     {
diff --git a/src/Storages/VirtualColumnUtils.h b/src/Storages/VirtualColumnUtils.h
index b5526fc5c7f..83494872cac 100644
--- a/src/Storages/VirtualColumnUtils.h
+++ b/src/Storages/VirtualColumnUtils.h
@@ -25,6 +25,9 @@ void filterBlockWithPredicate(const ActionsDAG::Node * predicate, Block & block,
 /// Just filters block. Block should contain all the required columns.
 void filterBlockWithDAG(ActionsDAGPtr dag, Block & block, ContextPtr context);
 
+/// Recursively checks if all functions used in DAG are deterministic in scope of query.
+bool isDeterministicInScopeOfQuery(const ActionsDAG::Node * node);
+
 /// Extract a part of predicate that can be evaluated using only columns from input_names.
 ActionsDAGPtr splitFilterDagForAllowedInputs(const ActionsDAG::Node * predicate, const Block * allowed_inputs);
 
diff --git a/tests/queries/0_stateless/03008_filter_projections_non_deterministoc_functions.reference b/tests/queries/0_stateless/03008_filter_projections_non_deterministoc_functions.reference
new file mode 100644
index 00000000000..8233925d609
--- /dev/null
+++ b/tests/queries/0_stateless/03008_filter_projections_non_deterministoc_functions.reference
@@ -0,0 +1,55 @@
+-- count
+100000	all_10_10_0
+100000	all_1_1_0
+100000	all_2_2_0
+100000	all_3_3_0
+100000	all_4_4_0
+100000	all_5_5_0
+100000	all_6_6_0
+100000	all_7_7_0
+100000	all_8_8_0
+100000	all_9_9_0
+-- rand()%2=0:
+1	all_10_10_0
+1	all_1_1_0
+1	all_2_2_0
+1	all_3_3_0
+1	all_4_4_0
+1	all_5_5_0
+1	all_6_6_0
+1	all_7_7_0
+1	all_8_8_0
+1	all_9_9_0
+-- optimize_use_implicit_projections=0
+1	all_10_10_0
+1	all_1_1_0
+1	all_2_2_0
+1	all_3_3_0
+1	all_4_4_0
+1	all_5_5_0
+1	all_6_6_0
+1	all_7_7_0
+1	all_8_8_0
+1	all_9_9_0
+-- optimize_trivial_count_query=0
+1	all_10_10_0
+1	all_1_1_0
+1	all_2_2_0
+1	all_3_3_0
+1	all_4_4_0
+1	all_5_5_0
+1	all_6_6_0
+1	all_7_7_0
+1	all_8_8_0
+1	all_9_9_0
+-- optimize_trivial_count_query=0, optimize_use_implicit_projections=0
+1	all_10_10_0
+1	all_1_1_0
+1	all_2_2_0
+1	all_3_3_0
+1	all_4_4_0
+1	all_5_5_0
+1	all_6_6_0
+1	all_7_7_0
+1	all_8_8_0
+1	all_9_9_0
diff --git a/tests/queries/0_stateless/03008_filter_projections_non_deterministoc_functions.sql b/tests/queries/0_stateless/03008_filter_projections_non_deterministoc_functions.sql
new file mode 100644
index 00000000000..3be9bc3982a
--- /dev/null
+++ b/tests/queries/0_stateless/03008_filter_projections_non_deterministoc_functions.sql
@@ -0,0 +1,28 @@
+create table test (number UInt64) engine=MergeTree order by number;
+system stop merges test;
+INSERT INTO test select number from numbers(100000);
+INSERT INTO test select number from numbers(100000);
+INSERT INTO test select number from numbers(100000);
+INSERT INTO test select number from numbers(100000);
+INSERT INTO test select number from numbers(100000);
+INSERT INTO test select number from numbers(100000);
+INSERT INTO test select number from numbers(100000);
+INSERT INTO test select number from numbers(100000);
+INSERT INTO test select number from numbers(100000);
+INSERT INTO test select number from numbers(100000);
+
+select '-- count';
+SELECT count(), _part FROM test GROUP BY _part ORDER BY _part;
+
+select '-- rand()%2=0:';
+SELECT count() > 0 AND count() < 100000, _part FROM test WHERE rand(1)%2=1 GROUP BY _part ORDER BY _part;
+
+select '-- optimize_use_implicit_projections=0';
+SELECT count() > 0 AND count() < 100000, _part FROM test WHERE rand(2)%2=1 GROUP BY _part ORDER BY _part settings optimize_use_implicit_projections=0;
+
+select '-- optimize_trivial_count_query=0';
+SELECT count() > 0 AND count() < 100000, _part FROM test WHERE rand(3)%2=1 GROUP BY _part ORDER BY _part settings optimize_trivial_count_query=0;
+
+select '-- optimize_trivial_count_query=0, optimize_use_implicit_projections=0';
+SELECT count() > 0 AND count() < 100000, _part FROM test WHERE rand(4)%2=1 GROUP BY _part ORDER BY _part settings optimize_trivial_count_query=0,optimize_use_implicit_projections=0;
+

From 879f7f2f8c862aae51ddc5a8faebb8d07b5d4493 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Mon, 11 Mar 2024 16:28:25 +0100
Subject: [PATCH 169/443] Remove more templates for JIT

---
 src/Interpreters/Aggregator.cpp | 185 ++++++++++++++------------------
 src/Interpreters/Aggregator.h   |  21 ++--
 2 files changed, 96 insertions(+), 110 deletions(-)

diff --git a/src/Interpreters/Aggregator.cpp b/src/Interpreters/Aggregator.cpp
index 69625dbd57d..c7ce3e46446 100644
--- a/src/Interpreters/Aggregator.cpp
+++ b/src/Interpreters/Aggregator.cpp
@@ -1111,6 +1111,7 @@ void NO_INLINE Aggregator::executeImpl(
     bool all_keys_are_const,
     AggregateDataPtr overflow_row) const
 {
+    bool use_compiled_functions = false;
     if (!no_more_keys)
     {
         /// Prefetching doesn't make sense for small hash tables, because they fit in caches entirely.
@@ -1118,33 +1119,47 @@ void NO_INLINE Aggregator::executeImpl(
             && (method.data.getBufferSizeInBytes() > min_bytes_for_prefetch);
 
 #if USE_EMBEDDED_COMPILER
-        if (compiled_aggregate_functions_holder && !hasSparseArguments(aggregate_instructions))
-        {
-            if (prefetch)
-                executeImplBatch<false, true, true>(
-                    method, state, aggregates_pool, row_begin, row_end, aggregate_instructions, all_keys_are_const, overflow_row);
-            else
-                executeImplBatch<false, true, false>(
-                    method, state, aggregates_pool, row_begin, row_end, aggregate_instructions, all_keys_are_const, overflow_row);
-        }
-        else
+        use_compiled_functions = compiled_aggregate_functions_holder && !hasSparseArguments(aggregate_instructions);
 #endif
-        {
-            if (prefetch)
-                executeImplBatch<false, false, true>(
-                    method, state, aggregates_pool, row_begin, row_end, aggregate_instructions, all_keys_are_const, overflow_row);
-            else
-                executeImplBatch<false, false, false>(
-                    method, state, aggregates_pool, row_begin, row_end, aggregate_instructions, all_keys_are_const, overflow_row);
-        }
+        if (prefetch)
+            executeImplBatch<false, true>(
+                method,
+                state,
+                aggregates_pool,
+                row_begin,
+                row_end,
+                aggregate_instructions,
+                all_keys_are_const,
+                use_compiled_functions,
+                overflow_row);
+        else
+            executeImplBatch<false, false>(
+                method,
+                state,
+                aggregates_pool,
+                row_begin,
+                row_end,
+                aggregate_instructions,
+                all_keys_are_const,
+                use_compiled_functions,
+                overflow_row);
     }
     else
     {
-        executeImplBatch<true, false, false>(method, state, aggregates_pool, row_begin, row_end, aggregate_instructions, all_keys_are_const, overflow_row);
+        executeImplBatch<true, false>(
+            method,
+            state,
+            aggregates_pool,
+            row_begin,
+            row_end,
+            aggregate_instructions,
+            all_keys_are_const,
+            use_compiled_functions,
+            overflow_row);
     }
 }
 
-template <bool no_more_keys, bool use_compiled_functions, bool prefetch, typename Method, typename State>
+template <bool no_more_keys, bool prefetch, typename Method, typename State>
 void NO_INLINE Aggregator::executeImplBatch(
     Method & method,
     State & state,
@@ -1153,6 +1168,7 @@ void NO_INLINE Aggregator::executeImplBatch(
     size_t row_end,
     AggregateFunctionInstruction * aggregate_instructions,
     bool all_keys_are_const,
+    bool use_compiled_functions [[maybe_unused]],
     AggregateDataPtr overflow_row) const
 {
     using KeyHolder = decltype(state.getKeyHolder(0, std::declval<Arena &>()));
@@ -1284,7 +1300,7 @@ void NO_INLINE Aggregator::executeImplBatch(
                 aggregate_data = aggregates_pool->alignedAlloc(total_size_of_aggregate_states, align_aggregate_states);
 
 #if USE_EMBEDDED_COMPILER
-                if constexpr (use_compiled_functions)
+                if (use_compiled_functions)
                 {
                     const auto & compiled_aggregate_functions = compiled_aggregate_functions_holder->compiled_aggregate_functions;
                     compiled_aggregate_functions.create_aggregate_states_function(aggregate_data);
@@ -1293,20 +1309,6 @@ void NO_INLINE Aggregator::executeImplBatch(
                         static constexpr bool skip_compiled_aggregate_functions = true;
                         createAggregateStates<skip_compiled_aggregate_functions>(aggregate_data);
                     }
-
-#if defined(MEMORY_SANITIZER)
-
-                    /// We compile only functions that do not allocate some data in Arena. Only store necessary state in AggregateData place.
-                    for (size_t aggregate_function_index = 0; aggregate_function_index < aggregate_functions.size(); ++aggregate_function_index)
-                    {
-                        if (!is_aggregate_function_compiled[aggregate_function_index])
-                            continue;
-
-                        auto aggregate_data_with_offset = aggregate_data + offsets_of_aggregate_states[aggregate_function_index];
-                        auto data_size = params.aggregates[aggregate_function_index].function->sizeOfData();
-                        __msan_unpoison(aggregate_data_with_offset, data_size);
-                    }
-#endif
                 }
                 else
 #endif
@@ -1339,7 +1341,7 @@ void NO_INLINE Aggregator::executeImplBatch(
     }
 
 #if USE_EMBEDDED_COMPILER
-    if constexpr (use_compiled_functions)
+    if (use_compiled_functions)
     {
         std::vector<ColumnData> columns_data;
 
@@ -1372,9 +1374,8 @@ void NO_INLINE Aggregator::executeImplBatch(
     for (size_t i = 0; i < aggregate_functions.size(); ++i)
     {
 #if USE_EMBEDDED_COMPILER
-        if constexpr (use_compiled_functions)
-            if (is_aggregate_function_compiled[i])
-                continue;
+        if (use_compiled_functions && is_aggregate_function_compiled[i])
+            continue;
 #endif
 
         AggregateFunctionInstruction * inst = aggregate_instructions + i;
@@ -1387,18 +1388,19 @@ void NO_INLINE Aggregator::executeImplBatch(
 }
 
 
-template <bool use_compiled_functions>
 void NO_INLINE Aggregator::executeWithoutKeyImpl(
     AggregatedDataWithoutKey & res,
-    size_t row_begin, size_t row_end,
+    size_t row_begin,
+    size_t row_end,
     AggregateFunctionInstruction * aggregate_instructions,
-    Arena * arena) const
+    Arena * arena,
+    bool use_compiled_functions [[maybe_unused]]) const
 {
     if (row_begin == row_end)
         return;
 
 #if USE_EMBEDDED_COMPILER
-    if constexpr (use_compiled_functions)
+    if (use_compiled_functions)
     {
         std::vector<ColumnData> columns_data;
 
@@ -1418,20 +1420,6 @@ void NO_INLINE Aggregator::executeWithoutKeyImpl(
 
         auto add_into_aggregate_states_function_single_place = compiled_aggregate_functions_holder->compiled_aggregate_functions.add_into_aggregate_states_function_single_place;
         add_into_aggregate_states_function_single_place(row_begin, row_end, columns_data.data(), res);
-
-#if defined(MEMORY_SANITIZER)
-
-        /// We compile only functions that do not allocate some data in Arena. Only store necessary state in AggregateData place.
-        for (size_t aggregate_function_index = 0; aggregate_function_index < aggregate_functions.size(); ++aggregate_function_index)
-        {
-            if (!is_aggregate_function_compiled[aggregate_function_index])
-                continue;
-
-            auto aggregate_data_with_offset = res + offsets_of_aggregate_states[aggregate_function_index];
-            auto data_size = params.aggregates[aggregate_function_index].function->sizeOfData();
-            __msan_unpoison(aggregate_data_with_offset, data_size);
-        }
-#endif
     }
 #endif
 
@@ -1439,13 +1427,10 @@ void NO_INLINE Aggregator::executeWithoutKeyImpl(
     for (size_t i = 0; i < aggregate_functions.size(); ++i)
     {
         AggregateFunctionInstruction * inst = aggregate_instructions + i;
-
 #if USE_EMBEDDED_COMPILER
-        if constexpr (use_compiled_functions)
-            if (is_aggregate_function_compiled[i])
-                continue;
+        if (use_compiled_functions && is_aggregate_function_compiled[i])
+            continue;
 #endif
-
         addBatchSinglePlace(row_begin, row_end, inst, res + inst->state_offset, arena);
     }
 }
@@ -1704,16 +1689,14 @@ bool Aggregator::executeOnBlock(Columns columns,
     if (result.type == AggregatedDataVariants::Type::without_key)
     {
         /// TODO: Enable compilation after investigation
-// #if USE_EMBEDDED_COMPILER
-//         if (compiled_aggregate_functions_holder)
-//         {
-//             executeWithoutKeyImpl<true>(result.without_key, row_begin, row_end, aggregate_functions_instructions.data(), result.aggregates_pool);
-//         }
-//         else
-// #endif
-        {
-            executeWithoutKeyImpl<false>(result.without_key, row_begin, row_end, aggregate_functions_instructions.data(), result.aggregates_pool);
-        }
+        bool use_compiled_functions = false;
+        executeWithoutKeyImpl(
+            result.without_key,
+            row_begin,
+            row_end,
+            aggregate_functions_instructions.data(),
+            result.aggregates_pool,
+            use_compiled_functions);
     }
     else
     {
@@ -1965,19 +1948,13 @@ Aggregator::convertToBlockImpl(Method & method, Table & data, Arena * arena, Are
 
     ConvertToBlockRes<return_single_block> res;
 
+    bool use_compiled_functions = false;
     if (final)
     {
 #if USE_EMBEDDED_COMPILER
-        if (compiled_aggregate_functions_holder)
-        {
-            static constexpr bool use_compiled_functions = !Method::low_cardinality_optimization;
-            res = convertToBlockImplFinal<Method, use_compiled_functions, return_single_block>(method, data, arena, aggregates_pools, rows);
-        }
-        else
+        use_compiled_functions = compiled_aggregate_functions_holder != nullptr && !Method::low_cardinality_optimization;
 #endif
-        {
-            res = convertToBlockImplFinal<Method, false, return_single_block>(method, data, arena, aggregates_pools, rows);
-        }
+        res = convertToBlockImplFinal<Method, return_single_block>(method, data, arena, aggregates_pools, use_compiled_functions, rows);
     }
     else
     {
@@ -2059,8 +2036,12 @@ inline void Aggregator::insertAggregatesIntoColumns(Mapped & mapped, MutableColu
 }
 
 
-template <bool use_compiled_functions>
-Block Aggregator::insertResultsIntoColumns(PaddedPODArray<AggregateDataPtr> & places, OutputBlockColumns && out_cols, Arena * arena, bool has_null_key_data [[maybe_unused]]) const
+Block Aggregator::insertResultsIntoColumns(
+    PaddedPODArray<AggregateDataPtr> & places,
+    OutputBlockColumns && out_cols,
+    Arena * arena,
+    bool has_null_key_data [[maybe_unused]],
+    bool use_compiled_functions [[maybe_unused]]) const
 {
     std::exception_ptr exception;
     size_t aggregate_functions_destroy_index = 0;
@@ -2068,7 +2049,7 @@ Block Aggregator::insertResultsIntoColumns(PaddedPODArray<AggregateDataPtr> & pl
     try
     {
 #if USE_EMBEDDED_COMPILER
-        if constexpr (use_compiled_functions)
+        if (use_compiled_functions)
         {
             /** For JIT compiled functions we need to resize columns before pass them into compiled code.
               * insert_aggregates_into_columns_function function does not throw exception.
@@ -2098,14 +2079,13 @@ Block Aggregator::insertResultsIntoColumns(PaddedPODArray<AggregateDataPtr> & pl
 
         for (; aggregate_functions_destroy_index < params.aggregates_size;)
         {
-            if constexpr (use_compiled_functions)
+#if USE_EMBEDDED_COMPILER
+            if (use_compiled_functions && is_aggregate_function_compiled[aggregate_functions_destroy_index])
             {
-                if (is_aggregate_function_compiled[aggregate_functions_destroy_index])
-                {
-                    ++aggregate_functions_destroy_index;
-                    continue;
-                }
+                ++aggregate_functions_destroy_index;
+                continue;
             }
+#endif
 
             auto & final_aggregate_column = out_cols.final_aggregate_columns[aggregate_functions_destroy_index];
             size_t offset = offsets_of_aggregate_states[aggregate_functions_destroy_index];
@@ -2127,14 +2107,13 @@ Block Aggregator::insertResultsIntoColumns(PaddedPODArray<AggregateDataPtr> & pl
 
     for (; aggregate_functions_destroy_index < params.aggregates_size; ++aggregate_functions_destroy_index)
     {
-        if constexpr (use_compiled_functions)
+#if USE_EMBEDDED_COMPILER
+        if (use_compiled_functions && is_aggregate_function_compiled[aggregate_functions_destroy_index])
         {
-            if (is_aggregate_function_compiled[aggregate_functions_destroy_index])
-            {
-                ++aggregate_functions_destroy_index;
-                continue;
-            }
+            ++aggregate_functions_destroy_index;
+            continue;
         }
+#endif
 
         size_t offset = offsets_of_aggregate_states[aggregate_functions_destroy_index];
         aggregate_functions[aggregate_functions_destroy_index]->destroyBatch(0, places.size(), places.data(), offset);
@@ -2146,9 +2125,9 @@ Block Aggregator::insertResultsIntoColumns(PaddedPODArray<AggregateDataPtr> & pl
     return finalizeBlock(params, getHeader(/* final */ true), std::move(out_cols), /* final */ true, places.size());
 }
 
-template <typename Method, bool use_compiled_functions, bool return_single_block, typename Table>
-Aggregator::ConvertToBlockRes<return_single_block> NO_INLINE
-Aggregator::convertToBlockImplFinal(Method & method, Table & data, Arena * arena, Arenas & aggregates_pools, size_t) const
+template <typename Method, bool return_single_block, typename Table>
+Aggregator::ConvertToBlockRes<return_single_block> NO_INLINE Aggregator::convertToBlockImplFinal(
+    Method & method, Table & data, Arena * arena, Arenas & aggregates_pools, bool use_compiled_functions [[maybe_unused]], size_t) const
 {
     /// +1 for nullKeyData, if `data` doesn't have it - not a problem, just some memory for one excessive row will be preallocated
     const size_t max_block_size = (return_single_block ? data.size() : std::min(params.max_block_size, data.size())) + 1;
@@ -2204,7 +2183,8 @@ Aggregator::convertToBlockImplFinal(Method & method, Table & data, Arena * arena
             {
                 if (places.size() >= max_block_size)
                 {
-                    res.emplace_back(insertResultsIntoColumns<use_compiled_functions>(places, std::move(out_cols.value()), arena, has_null_key_data));
+                    res.emplace_back(
+                        insertResultsIntoColumns(places, std::move(out_cols.value()), arena, has_null_key_data, use_compiled_functions));
                     places.clear();
                     out_cols.reset();
                     has_null_key_data = false;
@@ -2214,12 +2194,13 @@ Aggregator::convertToBlockImplFinal(Method & method, Table & data, Arena * arena
 
     if constexpr (return_single_block)
     {
-        return insertResultsIntoColumns<use_compiled_functions>(places, std::move(out_cols.value()), arena, has_null_key_data);
+        return insertResultsIntoColumns(places, std::move(out_cols.value()), arena, has_null_key_data, use_compiled_functions);
     }
     else
     {
         if (out_cols.has_value())
-            res.emplace_back(insertResultsIntoColumns<use_compiled_functions>(places, std::move(out_cols.value()), arena, has_null_key_data));
+            res.emplace_back(
+                insertResultsIntoColumns(places, std::move(out_cols.value()), arena, has_null_key_data, use_compiled_functions));
         return res;
     }
 }
diff --git a/src/Interpreters/Aggregator.h b/src/Interpreters/Aggregator.h
index 67e82cdd784..d7bbe5950a0 100644
--- a/src/Interpreters/Aggregator.h
+++ b/src/Interpreters/Aggregator.h
@@ -1395,7 +1395,7 @@ private:
         AggregateDataPtr overflow_row) const;
 
     /// Specialization for a particular value no_more_keys.
-    template <bool no_more_keys, bool use_compiled_functions, bool prefetch, typename Method, typename State>
+    template <bool no_more_keys, bool prefetch, typename Method, typename State>
     void executeImplBatch(
         Method & method,
         State & state,
@@ -1404,16 +1404,17 @@ private:
         size_t row_end,
         AggregateFunctionInstruction * aggregate_instructions,
         bool all_keys_are_const,
+        bool use_compiled_functions,
         AggregateDataPtr overflow_row) const;
 
     /// For case when there are no keys (all aggregate into one row).
-    template <bool use_compiled_functions>
     void executeWithoutKeyImpl(
         AggregatedDataWithoutKey & res,
         size_t row_begin,
         size_t row_end,
         AggregateFunctionInstruction * aggregate_instructions,
-        Arena * arena) const;
+        Arena * arena,
+        bool use_compiled_functions) const;
 
     template <typename Method>
     void writeToTemporaryFileImpl(
@@ -1467,12 +1468,16 @@ private:
         MutableColumns & final_aggregate_columns,
         Arena * arena) const;
 
-    template <bool use_compiled_functions>
-    Block insertResultsIntoColumns(PaddedPODArray<AggregateDataPtr> & places, OutputBlockColumns && out_cols, Arena * arena, bool has_null_key_data) const;
+    Block insertResultsIntoColumns(
+        PaddedPODArray<AggregateDataPtr> & places,
+        OutputBlockColumns && out_cols,
+        Arena * arena,
+        bool has_null_key_data,
+        bool use_compiled_functions) const;
 
-    template <typename Method, bool use_compiled_functions, bool return_single_block, typename Table>
-    ConvertToBlockRes<return_single_block>
-    convertToBlockImplFinal(Method & method, Table & data, Arena * arena, Arenas & aggregates_pools, size_t rows) const;
+    template <typename Method, bool return_single_block, typename Table>
+    ConvertToBlockRes<return_single_block> convertToBlockImplFinal(
+        Method & method, Table & data, Arena * arena, Arenas & aggregates_pools, bool use_compiled_functions, size_t rows) const;
 
     template <bool return_single_block, typename Method, typename Table>
     ConvertToBlockRes<return_single_block>

From 81b2a1f621d9bd64fde2c4e4f6a83c9b3b0c461a Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Mon, 11 Mar 2024 15:34:02 +0000
Subject: [PATCH 170/443] Fix build

---
 programs/obfuscator/Obfuscator.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/programs/obfuscator/Obfuscator.cpp b/programs/obfuscator/Obfuscator.cpp
index 317d93aaf0c..b2bf942af4e 100644
--- a/programs/obfuscator/Obfuscator.cpp
+++ b/programs/obfuscator/Obfuscator.cpp
@@ -1205,7 +1205,7 @@ public:
 }
 
 #pragma clang diagnostic ignored "-Wunused-function"
-#clang GCC diagnostic ignored "-Wmissing-declarations"
+#pragma clang diagnostic ignored "-Wmissing-declarations"
 
 int mainEntryClickHouseObfuscator(int argc, char ** argv)
 try

From 38f41ee311d0a36d194965e5815489a25c60e449 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Mon, 11 Mar 2024 16:55:30 +0100
Subject: [PATCH 171/443] Fix integration test

---
 tests/integration/test_disk_types/test.py              | 10 +++++-----
 .../test_endpoint_macro_substitution/test.py           | 10 +++++-----
 2 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/tests/integration/test_disk_types/test.py b/tests/integration/test_disk_types/test.py
index af482b97be3..86579911b3e 100644
--- a/tests/integration/test_disk_types/test.py
+++ b/tests/integration/test_disk_types/test.py
@@ -3,10 +3,10 @@ from helpers.cluster import ClickHouseCluster
 from helpers.test_tools import TSV
 
 disk_types = {
-    "default": "local",
-    "disk_s3": "s3",
-    "disk_hdfs": "hdfs",
-    "disk_encrypted": "s3",
+    "default": "Local",
+    "disk_s3": "S3",
+    "disk_hdfs": "HDFS",
+    "disk_encrypted": "S3",
 }
 
 
@@ -55,7 +55,7 @@ def test_different_types(cluster):
 def test_select_by_type(cluster):
     node = cluster.instances["node"]
     for name, disk_type in list(disk_types.items()):
-        if disk_type != "s3":
+        if disk_type != "S3":
             assert (
                 node.query(
                     "SELECT name FROM system.disks WHERE type='" + disk_type + "'"
diff --git a/tests/integration/test_endpoint_macro_substitution/test.py b/tests/integration/test_endpoint_macro_substitution/test.py
index 42a8ddbda84..bec3d9de0e3 100644
--- a/tests/integration/test_endpoint_macro_substitution/test.py
+++ b/tests/integration/test_endpoint_macro_substitution/test.py
@@ -4,10 +4,10 @@ from helpers.test_tools import TSV
 from pyhdfs import HdfsClient
 
 disk_types = {
-    "default": "local",
-    "disk_s3": "s3",
-    "disk_hdfs": "hdfs",
-    "disk_encrypted": "s3",
+    "default": "Local",
+    "disk_s3": "S3",
+    "disk_hdfs": "HDFS",
+    "disk_encrypted": "S3",
 }
 
 
@@ -63,7 +63,7 @@ def test_select_by_type(cluster):
     fs = HdfsClient(hosts=cluster.hdfs_ip)
 
     for name, disk_type in list(disk_types.items()):
-        if disk_type != "s3":
+        if disk_type != "S3":
             assert (
                 node.query(
                     "SELECT name FROM system.disks WHERE type='" + disk_type + "'"

From 5db08292455fb0c6f47fc0344382ab7cf3508e91 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Mon, 11 Mar 2024 17:20:53 +0100
Subject: [PATCH 172/443] Remove another template

---
 src/Interpreters/Aggregator.cpp | 55 ++++++++++++++-------------------
 src/Interpreters/Aggregator.h   |  3 +-
 2 files changed, 26 insertions(+), 32 deletions(-)

diff --git a/src/Interpreters/Aggregator.cpp b/src/Interpreters/Aggregator.cpp
index c7ce3e46446..a9578b5540f 100644
--- a/src/Interpreters/Aggregator.cpp
+++ b/src/Interpreters/Aggregator.cpp
@@ -2906,11 +2906,12 @@ ManyAggregatedDataVariants Aggregator::prepareVariantsToMerge(ManyAggregatedData
     return non_empty_data;
 }
 
-template <bool no_more_keys, typename State, typename Table>
+template <typename State, typename Table>
 void NO_INLINE Aggregator::mergeStreamsImplCase(
     Arena * aggregates_pool,
     State & state,
     Table & data,
+    bool no_more_keys,
     AggregateDataPtr overflow_row,
     size_t row_begin,
     size_t row_end,
@@ -2922,36 +2923,34 @@ void NO_INLINE Aggregator::mergeStreamsImplCase(
     if (!arena_for_keys)
         arena_for_keys = aggregates_pool;
 
-    for (size_t i = row_begin; i < row_end; ++i)
+    if (no_more_keys)
     {
-        AggregateDataPtr aggregate_data = nullptr;
-
-        if constexpr (!no_more_keys)
+        for (size_t i = row_begin; i < row_end; i++)
         {
-            auto emplace_result = state.emplaceKey(data, i, *arena_for_keys); // NOLINT
-            if (emplace_result.isInserted())
+            auto find_result = state.findKey(data, i, *arena_for_keys);
+            /// aggregate_date == nullptr means that the new key did not fit in the hash table because of no_more_keys.
+            AggregateDataPtr value = find_result.isFound() ? find_result.getMapped() : overflow_row;
+            places[i] = value;
+        }
+    }
+    else
+    {
+        for (size_t i = row_begin; i < row_end; i++)
+        {
+            auto emplace_result = state.emplaceKey(data, i, *arena_for_keys);
+            if (!emplace_result.isInserted())
+                places[i] = emplace_result.getMapped();
+            else
             {
                 emplace_result.setMapped(nullptr);
 
-                aggregate_data = aggregates_pool->alignedAlloc(total_size_of_aggregate_states, align_aggregate_states);
+                AggregateDataPtr aggregate_data = aggregates_pool->alignedAlloc(total_size_of_aggregate_states, align_aggregate_states);
                 createAggregateStates(aggregate_data);
 
                 emplace_result.setMapped(aggregate_data);
+                places[i] = aggregate_data;
             }
-            else
-                aggregate_data = emplace_result.getMapped();
         }
-        else
-        {
-            auto find_result = state.findKey(data, i, *arena_for_keys);
-            if (find_result.isFound())
-                aggregate_data = find_result.getMapped();
-        }
-
-        /// aggregate_date == nullptr means that the new key did not fit in the hash table because of no_more_keys.
-
-        AggregateDataPtr value = aggregate_data ? aggregate_data : overflow_row;
-        places[i] = value;
     }
 
     for (size_t j = 0; j < params.aggregates_size; ++j)
@@ -3005,22 +3004,16 @@ void NO_INLINE Aggregator::mergeStreamsImpl(
     if (use_cache)
     {
         typename Method::State state(key_columns, key_sizes, aggregation_state_cache);
-
-        if (!no_more_keys)
-            mergeStreamsImplCase<false>(aggregates_pool, state, data, overflow_row, row_begin, row_end, aggregate_columns_data, arena_for_keys);
-        else
-            mergeStreamsImplCase<true>(aggregates_pool, state, data, overflow_row, row_begin, row_end, aggregate_columns_data, arena_for_keys);
+        mergeStreamsImplCase(
+            aggregates_pool, state, data, no_more_keys, overflow_row, row_begin, row_end, aggregate_columns_data, arena_for_keys);
 
         consecutive_keys_cache_stats.update(row_end - row_begin, state.getCacheMissesSinceLastReset());
     }
     else
     {
         typename Method::StateNoCache state(key_columns, key_sizes, aggregation_state_cache);
-
-        if (!no_more_keys)
-            mergeStreamsImplCase<false>(aggregates_pool, state, data, overflow_row, row_begin, row_end, aggregate_columns_data, arena_for_keys);
-        else
-            mergeStreamsImplCase<true>(aggregates_pool, state, data, overflow_row, row_begin, row_end, aggregate_columns_data, arena_for_keys);
+        mergeStreamsImplCase(
+            aggregates_pool, state, data, no_more_keys, overflow_row, row_begin, row_end, aggregate_columns_data, arena_for_keys);
     }
 }
 
diff --git a/src/Interpreters/Aggregator.h b/src/Interpreters/Aggregator.h
index d7bbe5950a0..6c357623003 100644
--- a/src/Interpreters/Aggregator.h
+++ b/src/Interpreters/Aggregator.h
@@ -1513,11 +1513,12 @@ private:
         bool final,
         ThreadPool * thread_pool) const;
 
-    template <bool no_more_keys, typename State, typename Table>
+    template <typename State, typename Table>
     void mergeStreamsImplCase(
         Arena * aggregates_pool,
         State & state,
         Table & data,
+        bool no_more_keys,
         AggregateDataPtr overflow_row,
         size_t row_begin,
         size_t row_end,

From 74281b0a4e3d9328d01379129ace85e03c145fa4 Mon Sep 17 00:00:00 2001
From: avogar <avogar@clickhouse.com>
Date: Mon, 11 Mar 2024 16:42:45 +0000
Subject: [PATCH 173/443] Avoid some logical errors in experimantal Object type

---
 src/Columns/ColumnObject.cpp                  |  8 ++--
 src/Common/ErrorCodes.cpp                     |  1 +
 src/DataTypes/ObjectUtils.cpp                 | 41 ++++++++++++++-----
 .../Serializations/SerializationObject.cpp    |  6 +--
 4 files changed, 38 insertions(+), 18 deletions(-)

diff --git a/src/Columns/ColumnObject.cpp b/src/Columns/ColumnObject.cpp
index ac2ee309e87..25a62440629 100644
--- a/src/Columns/ColumnObject.cpp
+++ b/src/Columns/ColumnObject.cpp
@@ -20,12 +20,12 @@ namespace DB
 
 namespace ErrorCodes
 {
-    extern const int LOGICAL_ERROR;
     extern const int ILLEGAL_COLUMN;
     extern const int DUPLICATE_COLUMN;
     extern const int NUMBER_OF_DIMENSIONS_MISMATCHED;
     extern const int SIZES_OF_COLUMNS_DOESNT_MATCH;
     extern const int ARGUMENT_OUT_OF_BOUND;
+    extern const int EXPERIMENTAL_FEATURE_ERROR;
 }
 
 namespace
@@ -247,7 +247,7 @@ void ColumnObject::Subcolumn::checkTypes() const
         prefix_types.push_back(current_type);
         auto prefix_common_type = getLeastSupertype(prefix_types);
         if (!prefix_common_type->equals(*current_type))
-            throw Exception(ErrorCodes::LOGICAL_ERROR,
+            throw Exception(ErrorCodes::EXPERIMENTAL_FEATURE_ERROR,
                 "Data type {} of column at position {} cannot represent all columns from i-th prefix",
                 current_type->getName(), i);
     }
@@ -635,7 +635,7 @@ void ColumnObject::checkConsistency() const
     {
         if (num_rows != leaf->data.size())
         {
-            throw Exception(ErrorCodes::LOGICAL_ERROR, "Sizes of subcolumns are inconsistent in ColumnObject."
+            throw Exception(ErrorCodes::EXPERIMENTAL_FEATURE_ERROR, "Sizes of subcolumns are inconsistent in ColumnObject."
                 " Subcolumn '{}' has {} rows, but expected size is {}",
                 leaf->path.getPath(), leaf->data.size(), num_rows);
         }
@@ -919,7 +919,7 @@ void ColumnObject::addSubcolumn(const PathInData & key, size_t new_size)
 void ColumnObject::addNestedSubcolumn(const PathInData & key, const FieldInfo & field_info, size_t new_size)
 {
     if (!key.hasNested())
-        throw Exception(ErrorCodes::LOGICAL_ERROR,
+        throw Exception(ErrorCodes::EXPERIMENTAL_FEATURE_ERROR,
             "Cannot add Nested subcolumn, because path doesn't contain Nested");
 
     bool inserted = false;
diff --git a/src/Common/ErrorCodes.cpp b/src/Common/ErrorCodes.cpp
index 44463f7f437..9f2572cbfc6 100644
--- a/src/Common/ErrorCodes.cpp
+++ b/src/Common/ErrorCodes.cpp
@@ -598,6 +598,7 @@
     M(714, UNEXPECTED_CLUSTER) \
     M(715, CANNOT_DETECT_FORMAT) \
     M(716, CANNOT_FORGET_PARTITION) \
+    M(717, EXPERIMENTAL_FEATURE_ERROR) \
     \
     M(999, KEEPER_EXCEPTION) \
     M(1000, POCO_EXCEPTION) \
diff --git a/src/DataTypes/ObjectUtils.cpp b/src/DataTypes/ObjectUtils.cpp
index 23d29136c85..48392a614a5 100644
--- a/src/DataTypes/ObjectUtils.cpp
+++ b/src/DataTypes/ObjectUtils.cpp
@@ -28,9 +28,9 @@ namespace DB
 namespace ErrorCodes
 {
     extern const int TYPE_MISMATCH;
-    extern const int LOGICAL_ERROR;
     extern const int INCOMPATIBLE_COLUMNS;
     extern const int NOT_IMPLEMENTED;
+    extern const int EXPERIMENTAL_FEATURE_ERROR;
 }
 
 size_t getNumberOfDimensions(const IDataType & type)
@@ -92,7 +92,7 @@ ColumnPtr createArrayOfColumn(ColumnPtr column, size_t num_dimensions)
 Array createEmptyArrayField(size_t num_dimensions)
 {
     if (num_dimensions == 0)
-        throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot create array field with 0 dimensions");
+        throw Exception(ErrorCodes::EXPERIMENTAL_FEATURE_ERROR, "Cannot create array field with 0 dimensions");
 
     Array array;
     Array * current_array = &array;
@@ -231,7 +231,7 @@ static std::pair<ColumnPtr, DataTypePtr> recursivlyConvertDynamicColumnToTuple(
         };
     }
 
-    throw Exception(ErrorCodes::LOGICAL_ERROR, "Type {} unexpectedly has dynamic columns", type->getName());
+    throw Exception(ErrorCodes::EXPERIMENTAL_FEATURE_ERROR, "Type {} unexpectedly has dynamic columns", type->getName());
 }
 
 void convertDynamicColumnsToTuples(Block & block, const StorageSnapshotPtr & storage_snapshot)
@@ -247,7 +247,7 @@ void convertDynamicColumnsToTuples(Block & block, const StorageSnapshotPtr & sto
         GetColumnsOptions options(GetColumnsOptions::AllPhysical);
         auto storage_column = storage_snapshot->tryGetColumn(options, column.name);
         if (!storage_column)
-            throw Exception(ErrorCodes::LOGICAL_ERROR, "Column '{}' not found in storage", column.name);
+            throw Exception(ErrorCodes::EXPERIMENTAL_FEATURE_ERROR, "Column '{}' not found in storage", column.name);
 
         auto storage_column_concrete = storage_snapshot->getColumn(options.withExtendedObjects(), column.name);
 
@@ -315,7 +315,7 @@ static DataTypePtr getLeastCommonTypeForObject(const DataTypes & types, bool che
     {
         const auto * type_tuple = typeid_cast<const DataTypeTuple *>(type.get());
         if (!type_tuple)
-            throw Exception(ErrorCodes::LOGICAL_ERROR,
+            throw Exception(ErrorCodes::EXPERIMENTAL_FEATURE_ERROR,
                 "Least common type for object can be deduced only from tuples, but {} given", type->getName());
 
         auto [tuple_paths, tuple_types] = flattenTuple(type);
@@ -427,7 +427,7 @@ static DataTypePtr getLeastCommonTypeForDynamicColumnsImpl(
     if (const auto * type_tuple = typeid_cast<const DataTypeTuple *>(type_in_storage.get()))
         return getLeastCommonTypeForTuple(*type_tuple, concrete_types, check_ambiguos_paths);
 
-    throw Exception(ErrorCodes::LOGICAL_ERROR, "Type {} unexpectedly has dynamic columns", type_in_storage->getName());
+    throw Exception(ErrorCodes::EXPERIMENTAL_FEATURE_ERROR, "Type {} unexpectedly has dynamic columns", type_in_storage->getName());
 }
 
 DataTypePtr getLeastCommonTypeForDynamicColumns(
@@ -481,7 +481,7 @@ DataTypePtr createConcreteEmptyDynamicColumn(const DataTypePtr & type_in_storage
         return recreateTupleWithElements(*type_tuple, new_elements);
     }
 
-    throw Exception(ErrorCodes::LOGICAL_ERROR, "Type {} unexpectedly has dynamic columns", type_in_storage->getName());
+    throw Exception(ErrorCodes::EXPERIMENTAL_FEATURE_ERROR, "Type {} unexpectedly has dynamic columns", type_in_storage->getName());
 }
 
 bool hasDynamicSubcolumns(const ColumnsDescription & columns)
@@ -613,7 +613,7 @@ DataTypePtr reduceNumberOfDimensions(DataTypePtr type, size_t dimensions_to_redu
     {
         const auto * type_array = typeid_cast<const DataTypeArray *>(type.get());
         if (!type_array)
-            throw Exception(ErrorCodes::LOGICAL_ERROR, "Not enough dimensions to reduce");
+            throw Exception(ErrorCodes::EXPERIMENTAL_FEATURE_ERROR, "Not enough dimensions to reduce");
 
         type = type_array->getNestedType();
     }
@@ -627,7 +627,7 @@ ColumnPtr reduceNumberOfDimensions(ColumnPtr column, size_t dimensions_to_reduce
     {
         const auto * column_array = typeid_cast<const ColumnArray *>(column.get());
         if (!column_array)
-            throw Exception(ErrorCodes::LOGICAL_ERROR, "Not enough dimensions to reduce");
+            throw Exception(ErrorCodes::EXPERIMENTAL_FEATURE_ERROR, "Not enough dimensions to reduce");
 
         column = column_array->getDataPtr();
     }
@@ -705,6 +705,7 @@ ColumnWithTypeAndDimensions createTypeFromNode(const Node & node)
         size_t num_elements = tuple_columns.size();
         Columns tuple_elements_columns(num_elements);
         DataTypes tuple_elements_types(num_elements);
+        size_t last_offset = assert_cast<const ColumnArray::ColumnOffsets &>(*offsets_columns.back()).getData().back();
 
         /// Reduce extra array dimensions to get columns and types of Nested elements.
         for (size_t i = 0; i < num_elements; ++i)
@@ -712,6 +713,14 @@ ColumnWithTypeAndDimensions createTypeFromNode(const Node & node)
             assert(tuple_columns[i].array_dimensions == tuple_columns[0].array_dimensions);
             tuple_elements_columns[i] = reduceNumberOfDimensions(tuple_columns[i].column, tuple_columns[i].array_dimensions);
             tuple_elements_types[i] = reduceNumberOfDimensions(tuple_columns[i].type, tuple_columns[i].array_dimensions);
+            if (tuple_elements_columns[i]->size() != last_offset)
+                throw Exception(
+                    ErrorCodes::EXPERIMENTAL_FEATURE_ERROR,
+                    "Cannot create a type for subcolumn {} in Object data type: offsets_column has data inconsistent with nested_column. "
+                    "Data size: {}, last offset: {}",
+                    node.path.getPath(),
+                    tuple_elements_columns[i]->size(),
+                    last_offset);
         }
 
         auto result_column = ColumnArray::create(ColumnTuple::create(tuple_elements_columns), offsets_columns.back());
@@ -720,6 +729,16 @@ ColumnWithTypeAndDimensions createTypeFromNode(const Node & node)
         /// Recreate result Array type and Array column.
         for (auto it = offsets_columns.rbegin() + 1; it != offsets_columns.rend(); ++it)
         {
+            last_offset = assert_cast<const ColumnArray::ColumnOffsets &>((**it)).getData().back();
+            if (result_column->size() != last_offset)
+                throw Exception(
+                    ErrorCodes::EXPERIMENTAL_FEATURE_ERROR,
+                    "Cannot create a type for subcolumn {} in Object data type: offsets_column has data inconsistent with nested_column. "
+                    "Data size: {}, last offset: {}",
+                    node.path.getPath(),
+                    result_column->size(),
+                    last_offset);
+
             result_column = ColumnArray::create(result_column, *it);
             result_type = std::make_shared<DataTypeArray>(result_type);
         }
@@ -822,7 +841,7 @@ std::pair<ColumnPtr, DataTypePtr> unflattenTuple(
     assert(paths.size() == tuple_columns.size());
 
     if (paths.empty())
-        throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot unflatten empty Tuple");
+        throw Exception(ErrorCodes::EXPERIMENTAL_FEATURE_ERROR, "Cannot unflatten empty Tuple");
 
     /// We add all paths to the subcolumn tree and then create a type from it.
     /// The tree stores column, type and number of array dimensions
@@ -841,7 +860,7 @@ std::pair<ColumnPtr, DataTypePtr> unflattenTuple(
         tree.add(paths[i], [&](Node::Kind kind, bool exists) -> std::shared_ptr<Node>
             {
                 if (pos >= num_parts)
-                    throw Exception(ErrorCodes::LOGICAL_ERROR,
+                    throw Exception(ErrorCodes::EXPERIMENTAL_FEATURE_ERROR,
                         "Not enough name parts for path {}. Expected at least {}, got {}",
                             paths[i].getPath(), pos + 1, num_parts);
 
diff --git a/src/DataTypes/Serializations/SerializationObject.cpp b/src/DataTypes/Serializations/SerializationObject.cpp
index e6dc16ef5a0..1bdc29daebd 100644
--- a/src/DataTypes/Serializations/SerializationObject.cpp
+++ b/src/DataTypes/Serializations/SerializationObject.cpp
@@ -29,7 +29,7 @@ namespace ErrorCodes
     extern const int INCORRECT_DATA;
     extern const int CANNOT_READ_ALL_DATA;
     extern const int ARGUMENT_OUT_OF_BOUND;
-    extern const int LOGICAL_ERROR;
+    extern const int EXPERIMENTAL_FEATURE_ERROR;
 }
 
 template <typename Parser>
@@ -177,7 +177,7 @@ void SerializationObject<Parser>::serializeBinaryBulkStatePrefix(
     auto * stream = settings.getter(settings.path);
 
     if (!stream)
-        throw Exception(ErrorCodes::LOGICAL_ERROR, "Missing stream for kind of binary serialization");
+        throw Exception(ErrorCodes::EXPERIMENTAL_FEATURE_ERROR, "Missing stream for kind of binary serialization");
 
     auto [tuple_column, tuple_type] = unflattenObjectToTuple(column_object);
 
@@ -288,7 +288,7 @@ void SerializationObject<Parser>::serializeBinaryBulkWithMultipleStreams(
 
     if (!state_object->nested_type->equals(*tuple_type))
     {
-        throw Exception(ErrorCodes::LOGICAL_ERROR,
+        throw Exception(ErrorCodes::EXPERIMENTAL_FEATURE_ERROR,
             "Types of internal column of Object mismatched. Expected: {}, Got: {}",
             state_object->nested_type->getName(), tuple_type->getName());
     }

From 260c6387555b3823e56db135f80adaa5e0c300ab Mon Sep 17 00:00:00 2001
From: avogar <avogar@clickhouse.com>
Date: Mon, 11 Mar 2024 16:59:01 +0000
Subject: [PATCH 174/443] Replace forgotten logical error

---
 src/DataTypes/ObjectUtils.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/DataTypes/ObjectUtils.cpp b/src/DataTypes/ObjectUtils.cpp
index 48392a614a5..c565930f49a 100644
--- a/src/DataTypes/ObjectUtils.cpp
+++ b/src/DataTypes/ObjectUtils.cpp
@@ -653,7 +653,7 @@ ColumnWithTypeAndDimensions createTypeFromNode(const Node & node)
     auto collect_tuple_elemets = [](const auto & children)
     {
         if (children.empty())
-            throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot create type from empty Tuple or Nested node");
+            throw Exception(ErrorCodes::EXPERIMENTAL_FEATURE_ERROR, "Cannot create type from empty Tuple or Nested node");
 
         std::vector<std::tuple<String, ColumnWithTypeAndDimensions>> tuple_elements;
         tuple_elements.reserve(children.size());

From 16e01eb93ad449c61417dcaccd570439364b0714 Mon Sep 17 00:00:00 2001
From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com>
Date: Mon, 11 Mar 2024 18:05:51 +0100
Subject: [PATCH 175/443] Fix style

---
 src/Core/Settings.cpp | 1 +
 1 file changed, 1 insertion(+)

diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp
index fb456b46d89..8257b94cd9f 100644
--- a/src/Core/Settings.cpp
+++ b/src/Core/Settings.cpp
@@ -15,6 +15,7 @@ namespace ErrorCodes
     extern const int THERE_IS_NO_PROFILE;
     extern const int NO_ELEMENTS_IN_CONFIG;
     extern const int UNKNOWN_ELEMENT_IN_CONFIG;
+    extern const int BAD_ARGUMENTS;
 }
 
 IMPLEMENT_SETTINGS_TRAITS(SettingsTraits, LIST_OF_SETTINGS)

From f656a015385898602cb651b419b46927f99ab602 Mon Sep 17 00:00:00 2001
From: Max Kainov <max.kainov@clickhouse.com>
Date: Mon, 11 Mar 2024 16:39:13 +0000
Subject: [PATCH 176/443] CI: fix sync build issue with reuse  #do_not_test

---
 tests/ci/ci.py | 21 +++++++++++++++------
 1 file changed, 15 insertions(+), 6 deletions(-)

diff --git a/tests/ci/ci.py b/tests/ci/ci.py
index 9d57f161be3..cc4d0b11eef 100644
--- a/tests/ci/ci.py
+++ b/tests/ci/ci.py
@@ -140,7 +140,7 @@ class CiCache:
         self.s3 = s3
         self.job_digests = job_digests
         self.cache_s3_paths = {
-            job_type: f"{self._S3_CACHE_PREFIX}/{job_type.value}-{self.job_digests[self._get_reference_job_name(job_type)]}/"
+            job_type: f"{self._S3_CACHE_PREFIX}/{job_type.value}-{self._get_digest_for_job_type(self.job_digests, job_type)}/"
             for job_type in self.JobType
         }
         self.s3_record_prefixes = {
@@ -155,14 +155,23 @@ class CiCache:
         if not self._LOCAL_CACHE_PATH.exists():
             self._LOCAL_CACHE_PATH.mkdir(parents=True, exist_ok=True)
 
-    def _get_reference_job_name(self, job_type: JobType) -> str:
-        res = Build.PACKAGE_RELEASE
+    def _get_digest_for_job_type(
+        self, job_digests: Dict[str, str], job_type: JobType
+    ) -> str:
         if job_type == self.JobType.DOCS:
-            res = JobNames.DOCS_CHECK
+            res = job_digests[JobNames.DOCS_CHECK]
         elif job_type == self.JobType.SRCS:
-            res = Build.PACKAGE_RELEASE
+            # any build type job has the same digest - pick up Build.PACKAGE_RELEASE or Build.PACKAGE_ASAN as a failover
+            # Build.PACKAGE_RELEASE may not exist in the list if we have reduced CI pipeline
+            if Build.PACKAGE_RELEASE in job_digests:
+                res = job_digests[Build.PACKAGE_RELEASE]
+            elif Build.PACKAGE_ASAN in job_digests:
+                # failover, if failover does not work - fix it!
+                res = job_digests[Build.PACKAGE_ASAN]
+            else:
+                assert False, "BUG, no build job in digest' list"
         else:
-            assert False
+            assert False, "BUG, New JobType? - please update func"
         return res
 
     def _get_record_file_name(

From f973e405eeb4f28a6a937c26d19cad54acd00eb4 Mon Sep 17 00:00:00 2001
From: Max Kainov <max.kainov@clickhouse.com>
Date: Wed, 7 Feb 2024 19:36:28 +0000
Subject: [PATCH 177/443] CI: fixing ARM integration tests

 #do_not_test
---
 tests/ci/ci.py | 41 ++++++++++++++++++++++++++++++++++++-----
 1 file changed, 36 insertions(+), 5 deletions(-)

diff --git a/tests/ci/ci.py b/tests/ci/ci.py
index 9d57f161be3..898d23be843 100644
--- a/tests/ci/ci.py
+++ b/tests/ci/ci.py
@@ -1183,13 +1183,13 @@ def _configure_jobs(
 
         if batches_to_do:
             jobs_to_do.append(job)
+            jobs_params[job] = {
+                "batches": batches_to_do,
+                "num_batches": num_batches,
+            }
         elif add_to_skip:
             # treat job as being skipped only if it's controlled by digest
             jobs_to_skip.append(job)
-        jobs_params[job] = {
-            "batches": batches_to_do,
-            "num_batches": num_batches,
-        }
 
     if not pr_info.is_release_branch():
         # randomization bucket filtering (pick one random job from each bucket, for jobs with configured random_bucket property)
@@ -1268,6 +1268,33 @@ def _configure_jobs(
             jobs_to_do = list(
                 set(job for job in jobs_to_do_requested if job not in jobs_to_skip)
             )
+            # if requested job does not have params in jobs_params (it happens for "run_by_label" job)
+            #   we need to add params - otherwise it won't run as "batches" list will be empty
+            for job in jobs_to_do:
+                if job not in jobs_params:
+                    num_batches = CI_CONFIG.get_job_config(job).num_batches
+                    jobs_params[job] = {
+                        "batches": list(range(num_batches)),
+                        "num_batches": num_batches,
+                    }
+
+        requested_batches = set()
+        for token in commit_tokens:
+            if token.startswith("batch_"):
+                try:
+                    batches = [
+                        int(batch) for batch in token.removeprefix("batch_").split("_")
+                    ]
+                except Exception:
+                    print(f"ERROR: failed to parse commit tag [{token}]")
+                requested_batches.update(batches)
+        if requested_batches:
+            print(
+                f"NOTE: Only specific job batches were requested [{list(requested_batches)}]"
+            )
+            for job, params in jobs_params.items():
+                if params["num_batches"] > 1:
+                    params["batches"] = list(requested_batches)
 
     return {
         "digests": digests,
@@ -1372,7 +1399,11 @@ def _update_gh_statuses_action(indata: Dict, s3: S3Helper) -> None:
 def _fetch_commit_tokens(message: str) -> List[str]:
     pattern = r"#[\w-]+"
     matches = [match[1:] for match in re.findall(pattern, message)]
-    res = [match for match in matches if match in Labels or match.startswith("job_")]
+    res = [
+        match
+        for match in matches
+        if match in Labels or match.startswith("job_") or match.startswith("batch_")
+    ]
     return res
 
 

From 4b94bcd54ebd8713db8aefc453edaf15a1aa55ab Mon Sep 17 00:00:00 2001
From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com>
Date: Mon, 11 Mar 2024 18:41:17 +0100
Subject: [PATCH 178/443] Update ReadSettings.h

---
 src/IO/ReadSettings.h | 1 -
 1 file changed, 1 deletion(-)

diff --git a/src/IO/ReadSettings.h b/src/IO/ReadSettings.h
index fb8cbaf4a98..c0a63bf51b1 100644
--- a/src/IO/ReadSettings.h
+++ b/src/IO/ReadSettings.h
@@ -99,7 +99,6 @@ struct ReadSettings
     bool enable_filesystem_cache = true;
     bool read_from_filesystem_cache_if_exists_otherwise_bypass_cache = false;
     bool enable_filesystem_cache_log = false;
-    bool force_read_through_cache_merges = false;
     size_t filesystem_cache_segments_batch_size = 20;
 
     bool use_page_cache_for_disks_without_file_cache = false;

From fd68fed0470214d5c8734b0c89fbb779cf9b821f Mon Sep 17 00:00:00 2001
From: Han Fei <hanfei19910905@gmail.com>
Date: Mon, 11 Mar 2024 19:21:35 +0100
Subject: [PATCH 179/443] fix flaky 02949_ttl_group_by_bug

---
 tests/queries/0_stateless/02949_ttl_group_by_bug.sql | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/queries/0_stateless/02949_ttl_group_by_bug.sql b/tests/queries/0_stateless/02949_ttl_group_by_bug.sql
index 2888f6e7d66..a3d0794c897 100644
--- a/tests/queries/0_stateless/02949_ttl_group_by_bug.sql
+++ b/tests/queries/0_stateless/02949_ttl_group_by_bug.sql
@@ -2,7 +2,7 @@ DROP TABLE IF EXISTS ttl_group_by_bug;
 
 CREATE TABLE ttl_group_by_bug
 (key UInt32, ts DateTime, value UInt32, min_value UInt32 default value, max_value UInt32 default value)
-ENGINE = MergeTree() PARTITION BY toYYYYMM(ts)
+ENGINE = MergeTree()
 ORDER BY (key, toStartOfInterval(ts, toIntervalMinute(3)), ts)
 TTL ts + INTERVAL 5 MINUTE GROUP BY key, toStartOfInterval(ts, toIntervalMinute(3))
 SET value = sum(value), min_value = min(min_value), max_value = max(max_value),  ts=min(toStartOfInterval(ts, toIntervalMinute(3)));

From a1e5161cee50650a5c4e87ca60e7ed9eb61451b4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Mon, 11 Mar 2024 19:25:34 +0100
Subject: [PATCH 180/443] Disable sanitizers with
 02784_parallel_replicas_automatic_decision_join

---
 .../02784_parallel_replicas_automatic_decision_join.sh         | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tests/queries/0_stateless/02784_parallel_replicas_automatic_decision_join.sh b/tests/queries/0_stateless/02784_parallel_replicas_automatic_decision_join.sh
index ef3e6000903..801cd22b79f 100755
--- a/tests/queries/0_stateless/02784_parallel_replicas_automatic_decision_join.sh
+++ b/tests/queries/0_stateless/02784_parallel_replicas_automatic_decision_join.sh
@@ -1,5 +1,6 @@
 #!/usr/bin/env bash
-## Note: The analyzer doesn't support JOIN with parallel replicas yet
+# Tags: no-tsan, no-asan, no-msan
+# It's not clear why distributed aggregation is much slower with sanitizers (https://github.com/ClickHouse/ClickHouse/issues/60625)
 
 CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
 # shellcheck source=../shell_config.sh

From 55a82047613c607dedb592fed019d04455e8c8e8 Mon Sep 17 00:00:00 2001
From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com>
Date: Mon, 11 Mar 2024 19:43:30 +0100
Subject: [PATCH 181/443] Fix test

---
 .../0_stateless/03003_compatibility_setting_bad_value.sql       | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/queries/0_stateless/03003_compatibility_setting_bad_value.sql b/tests/queries/0_stateless/03003_compatibility_setting_bad_value.sql
index 9a6f4e7944a..48e98798c51 100644
--- a/tests/queries/0_stateless/03003_compatibility_setting_bad_value.sql
+++ b/tests/queries/0_stateless/03003_compatibility_setting_bad_value.sql
@@ -1,2 +1,2 @@
-select 42 settings compatibility=NULL;  -- {clientError BAD_GET}
+select 42 settings compatibility=NULL;  -- {clientError BAD_ARGUMENTS}
 

From 5a71636411cb358c94e58b7caac18c22104b0e1c Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Mon, 11 Mar 2024 19:44:52 +0100
Subject: [PATCH 182/443] Fxi

---
 tests/integration/test_disk_types/test.py     | 30 +++++++++++++++++++
 .../test_endpoint_macro_substitution/test.py  |  9 +++++-
 2 files changed, 38 insertions(+), 1 deletion(-)

diff --git a/tests/integration/test_disk_types/test.py b/tests/integration/test_disk_types/test.py
index 86579911b3e..5047cdc605e 100644
--- a/tests/integration/test_disk_types/test.py
+++ b/tests/integration/test_disk_types/test.py
@@ -50,6 +50,36 @@ def test_different_types(cluster):
             assert (
                 fields[encrypted_col_ix] == "0"
             ), f"{fields[name_col_ix]} expected to be non-encrypted!"
+def test_different_types(cluster):
+    node = cluster.instances["node"]
+    response = TSV.toMat(node.query("SELECT * FROM system.disks FORMAT TSVWithNames"))
+
+    assert len(response) > len(disk_types)  # at least one extra line for header
+
+    name_col_ix = response[0].index("name")
+    type_col_ix = response[0].index("type")
+    encrypted_col_ix = response[0].index("is_encrypted")
+
+    for fields in response[1:]:  # skip header
+        assert len(fields) >= 7
+        expected_disk_type = disk_types.get(fields[name_col_ix], "UNKNOWN")
+
+        if expected_disk_type != "Local":
+            disk_type = fields[response[0].index("object_storage_type")]
+        else:
+            disk_type = fields[type_col_ix]
+
+        assert (
+            expected_disk_type == disk_type
+        ), f"Wrong type ({fields[type_col_ix]}) for disk {fields[name_col_ix]}!"
+        if "encrypted" in fields[name_col_ix]:
+            assert (
+                fields[encrypted_col_ix] == "1"
+            ), f"{fields[name_col_ix]} expected to be encrypted!"
+        else:
+            assert (
+                fields[encrypted_col_ix] == "0"
+            ), f"{fields[name_col_ix]} expected to be non-encrypted!"
 
 
 def test_select_by_type(cluster):
diff --git a/tests/integration/test_endpoint_macro_substitution/test.py b/tests/integration/test_endpoint_macro_substitution/test.py
index bec3d9de0e3..e161d8e82ff 100644
--- a/tests/integration/test_endpoint_macro_substitution/test.py
+++ b/tests/integration/test_endpoint_macro_substitution/test.py
@@ -45,8 +45,15 @@ def test_different_types(cluster):
 
     for fields in response[1:]:  # skip header
         assert len(fields) >= 7
+        expected_disk_type = disk_types.get(fields[name_col_ix], "UNKNOWN")
+
+        if expected_disk_type != "Local":
+            disk_type = fields[response[0].index("object_storage_type")]
+        else:
+            disk_type = fields[type_col_ix]
+
         assert (
-            disk_types.get(fields[name_col_ix], "UNKNOWN") == fields[type_col_ix]
+            expected_disk_type == disk_type
         ), f"Wrong type ({fields[type_col_ix]}) for disk {fields[name_col_ix]}!"
         if "encrypted" in fields[name_col_ix]:
             assert (

From b7b83085d1722b192d802ffc3677ea6d0f03c85a Mon Sep 17 00:00:00 2001
From: Max Kainov <max.kainov@clickhouse.com>
Date: Mon, 11 Mar 2024 18:49:45 +0000
Subject: [PATCH 183/443] CI: ci.py hot style fix

 #do_not_test
---
 tests/ci/ci.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tests/ci/ci.py b/tests/ci/ci.py
index a699642691b..29906e6571f 100644
--- a/tests/ci/ci.py
+++ b/tests/ci/ci.py
@@ -55,6 +55,8 @@ from report import ERROR, SUCCESS, BuildResult, JobReport
 from s3_helper import S3Helper
 from version_helper import get_version_from_repo
 
+# pylint: disable=too-many-lines
+
 
 @dataclass
 class PendingState:

From 5a400c181b818cf3bbf371eb388a937ee816a66c Mon Sep 17 00:00:00 2001
From: Azat Khuzhin <a3at.mail@gmail.com>
Date: Mon, 11 Mar 2024 09:49:25 +0100
Subject: [PATCH 184/443] Fix llvm symbolizer on CI

In #61011 the whole toolchain installation had been removed from the
base image to reduce image sizes, and this is a good thing indeed.

However it also breaks the symbolizer for sanitizers, which makes
stacktraces unreadable, so let's fix this by getting back llvm package,
this should be OK, since it's size is not gigabytes, but only 48MiB (at
least for llvm-14):

    # dpkg -L llvm-14| xargs file | grep -v directory | cut -d: -f1 | xargs du -sch | grep total
    48M     total

Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>
---
 docker/test/base/Dockerfile     | 3 +++
 docker/test/fasttest/Dockerfile | 3 ---
 docker/test/util/Dockerfile     | 2 ++
 3 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/docker/test/base/Dockerfile b/docker/test/base/Dockerfile
index 55229e893de..2317f84e0cb 100644
--- a/docker/test/base/Dockerfile
+++ b/docker/test/base/Dockerfile
@@ -33,6 +33,9 @@ ENV TSAN_OPTIONS='halt_on_error=1 abort_on_error=1 history_size=7 memory_limit_m
 ENV UBSAN_OPTIONS='print_stacktrace=1'
 ENV MSAN_OPTIONS='abort_on_error=1 poison_in_dtor=1'
 
+# for external_symbolizer_path
+RUN ln -s /usr/bin/llvm-symbolizer-${LLVM_VERSION} /usr/bin/llvm-symbolizer
+
 RUN echo "en_US.UTF-8 UTF-8" > /etc/locale.gen && locale-gen en_US.UTF-8
 ENV LC_ALL en_US.UTF-8
 
diff --git a/docker/test/fasttest/Dockerfile b/docker/test/fasttest/Dockerfile
index 62cdcc3f830..912ff191e57 100644
--- a/docker/test/fasttest/Dockerfile
+++ b/docker/test/fasttest/Dockerfile
@@ -14,7 +14,6 @@ RUN apt-get update \
         libclang-${LLVM_VERSION}-dev \
         libclang-rt-${LLVM_VERSION}-dev \
         lld-${LLVM_VERSION} \
-        llvm-${LLVM_VERSION} \
         llvm-${LLVM_VERSION}-dev \
         lsof \
         ninja-build \
@@ -37,8 +36,6 @@ RUN pip3 install numpy==1.26.3 scipy==1.12.0 pandas==1.5.3 Jinja2==3.1.3
 
 # This symlink is required by gcc to find the lld linker
 RUN ln -s /usr/bin/lld-${LLVM_VERSION} /usr/bin/ld.lld
-# for external_symbolizer_path
-RUN ln -s /usr/bin/llvm-symbolizer-${LLVM_VERSION} /usr/bin/llvm-symbolizer
 # FIXME: workaround for "The imported target "merge-fdata" references the file" error
 # https://salsa.debian.org/pkg-llvm-team/llvm-toolchain/-/commit/992e52c0b156a5ba9c6a8a54f8c4857ddd3d371d
 RUN sed -i '/_IMPORT_CHECK_FILES_FOR_\(mlir-\|llvm-bolt\|merge-fdata\|MLIR\)/ {s|^|#|}' /usr/lib/llvm-${LLVM_VERSION}/lib/cmake/llvm/LLVMExports-*.cmake
diff --git a/docker/test/util/Dockerfile b/docker/test/util/Dockerfile
index 4f2dc9df849..5446adf3793 100644
--- a/docker/test/util/Dockerfile
+++ b/docker/test/util/Dockerfile
@@ -26,6 +26,8 @@ RUN apt-get update \
     && export CODENAME="$(lsb_release --codename --short | tr 'A-Z' 'a-z')" \
     && echo "deb https://apt.llvm.org/${CODENAME}/ llvm-toolchain-${CODENAME}-${LLVM_VERSION} main" >> \
         /etc/apt/sources.list \
+    && apt-get update \
+    && apt-get install --yes --no-install-recommends --verbose-versions llvm-${LLVM_VERSION} \
     && apt-get clean \
     && rm -rf /var/lib/apt/lists/* /var/cache/debconf /tmp/*
 

From 9b055c3a43039387b42e755efddd83b9a8934ca6 Mon Sep 17 00:00:00 2001
From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com>
Date: Mon, 11 Mar 2024 20:38:30 +0100
Subject: [PATCH 185/443] Use assert_cast to prevent nullptr dereference on bad
 column types in FunctionsConversion

---
 src/Functions/FunctionsConversion.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/src/Functions/FunctionsConversion.h b/src/Functions/FunctionsConversion.h
index 1522e76893e..f338af28240 100644
--- a/src/Functions/FunctionsConversion.h
+++ b/src/Functions/FunctionsConversion.h
@@ -4561,7 +4561,7 @@ arguments, result_type, input_rows_count); \
 
                 if (from_low_cardinality)
                 {
-                    const auto * col_low_cardinality = typeid_cast<const ColumnLowCardinality *>(arguments[0].column.get());
+                    const auto * col_low_cardinality = assert_cast<const ColumnLowCardinality *>(arguments[0].column.get());
 
                     if (skip_not_null_check && col_low_cardinality->containsNull())
                         throw Exception(ErrorCodes::CANNOT_INSERT_NULL_IN_ORDINARY_COLUMN, "Cannot convert NULL value to non-Nullable type");
@@ -4586,7 +4586,7 @@ arguments, result_type, input_rows_count); \
             if (to_low_cardinality)
             {
                 auto res_column = to_low_cardinality->createColumn();
-                auto * col_low_cardinality = typeid_cast<ColumnLowCardinality *>(res_column.get());
+                auto * col_low_cardinality = assert_cast<ColumnLowCardinality *>(res_column.get());
 
                 if (from_low_cardinality && !src_converted_to_full_column)
                 {

From 3a26b9c89ee3083884fde341c2af418bcde2f4cf Mon Sep 17 00:00:00 2001
From: Nikita Taranov <nikita.taranov@clickhouse.com>
Date: Mon, 11 Mar 2024 19:42:25 +0000
Subject: [PATCH 186/443] impl

---
 .../0_stateless/02887_mutations_subcolumns.reference        | 6 +++---
 tests/queries/0_stateless/02887_mutations_subcolumns.sql    | 4 ++--
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/tests/queries/0_stateless/02887_mutations_subcolumns.reference b/tests/queries/0_stateless/02887_mutations_subcolumns.reference
index c2d6cbbd225..1ccc83b48a3 100644
--- a/tests/queries/0_stateless/02887_mutations_subcolumns.reference
+++ b/tests/queries/0_stateless/02887_mutations_subcolumns.reference
@@ -5,6 +5,6 @@
 4	ttt
 5	ttt
 6	ttt
-{"a":"1","obj":{"k1":1,"k2":null,"k3":null}}
-{"a":"3","obj":{"k1":null,"k2":null,"k3":1}}
-{"a":"1","obj":{"k1":1,"k2":null,"k3":null}}
+1	[('k1',1)]
+3	[('k3',1)]
+1	[('k1',1)]
diff --git a/tests/queries/0_stateless/02887_mutations_subcolumns.sql b/tests/queries/0_stateless/02887_mutations_subcolumns.sql
index a01158e1b06..87b3009e929 100644
--- a/tests/queries/0_stateless/02887_mutations_subcolumns.sql
+++ b/tests/queries/0_stateless/02887_mutations_subcolumns.sql
@@ -40,9 +40,9 @@ INSERT INTO t_mutations_subcolumns VALUES (2, '{"k2": 1}');
 INSERT INTO t_mutations_subcolumns VALUES (3, '{"k3": 1}');
 
 ALTER TABLE t_mutations_subcolumns DELETE WHERE obj.k2 = 1;
-SELECT * FROM t_mutations_subcolumns ORDER BY a FORMAT JSONEachRow;
+SELECT a, arrayFilter(x -> not isNull(x.2), tupleToNameValuePairs(obj)) FROM t_mutations_subcolumns ORDER BY a;
 
 ALTER TABLE t_mutations_subcolumns DELETE WHERE isNull(obj.k1);
-SELECT * FROM t_mutations_subcolumns ORDER BY a FORMAT JSONEachRow;
+SELECT a, arrayFilter(x -> not isNull(x.2), tupleToNameValuePairs(obj)) FROM t_mutations_subcolumns ORDER BY a;
 
 DROP TABLE t_mutations_subcolumns;

From 2e74685ba6ea8a3cc32ff0e21d0ee657517ef5a4 Mon Sep 17 00:00:00 2001
From: avogar <avogar@clickhouse.com>
Date: Mon, 11 Mar 2024 19:58:43 +0000
Subject: [PATCH 187/443] Make variant tests a bit faster

---
 .../02941_variant_type_2.reference            | 80 +++++++++----------
 .../0_stateless/02941_variant_type_2.sh       | 12 +--
 ...different_local_and_global_order.reference | 30 +++----
 ...e_with_different_local_and_global_order.sh |  8 +-
 4 files changed, 65 insertions(+), 65 deletions(-)

diff --git a/tests/queries/0_stateless/02941_variant_type_2.reference b/tests/queries/0_stateless/02941_variant_type_2.reference
index 4b6d53c52ac..20a5176cb5e 100644
--- a/tests/queries/0_stateless/02941_variant_type_2.reference
+++ b/tests/queries/0_stateless/02941_variant_type_2.reference
@@ -1,51 +1,51 @@
 Memory
 test4 insert
 test4 select
-1000000
-200000
-200000
-200000
-200000
-200000
-200000
-200000
+500000
+100000
+100000
+100000
+100000
+100000
+100000
+100000
 MergeTree compact
 test4 insert
 test4 select
-1000000
-200000
-200000
-200000
-200000
-200000
-200000
-200000
+500000
+100000
+100000
+100000
+100000
+100000
+100000
+100000
 test4 select
-1000000
-200000
-200000
-200000
-200000
-200000
-200000
-200000
+500000
+100000
+100000
+100000
+100000
+100000
+100000
+100000
 MergeTree wide
 test4 insert
 test4 select
-1000000
-200000
-200000
-200000
-200000
-200000
-200000
-200000
+500000
+100000
+100000
+100000
+100000
+100000
+100000
+100000
 test4 select
-1000000
-200000
-200000
-200000
-200000
-200000
-200000
-200000
+500000
+100000
+100000
+100000
+100000
+100000
+100000
+100000
diff --git a/tests/queries/0_stateless/02941_variant_type_2.sh b/tests/queries/0_stateless/02941_variant_type_2.sh
index 509c537e7fc..d1fa0a777c9 100755
--- a/tests/queries/0_stateless/02941_variant_type_2.sh
+++ b/tests/queries/0_stateless/02941_variant_type_2.sh
@@ -12,12 +12,12 @@ CH_CLIENT="$CLICKHOUSE_CLIENT --allow_experimental_variant_type=1 --allow_suspic
 function test4_insert()
 {
     echo "test4 insert"
-    $CH_CLIENT -nmq "insert into test select number, NULL from numbers(200000);
-insert into test select number + 200000, number from numbers(200000);
-insert into test select number + 400000, 'str_' || toString(number) from numbers(200000);
-insert into test select number + 600000, ('lc_str_' || toString(number))::LowCardinality(String) from numbers(200000);
-insert into test select number + 800000, tuple(number, number + 1)::Tuple(a UInt32, b UInt32) from numbers(200000);
-insert into test select number + 1000000, range(number % 20 + 1)::Array(UInt64) from numbers(200000);"
+    $CH_CLIENT -nmq "insert into test select number, NULL from numbers(100000);
+insert into test select number + 100000, number from numbers(100000);
+insert into test select number + 200000, 'str_' || toString(number) from numbers(100000);
+insert into test select number + 300000, ('lc_str_' || toString(number))::LowCardinality(String) from numbers(100000);
+insert into test select number + 400000, tuple(number, number + 1)::Tuple(a UInt32, b UInt32) from numbers(100000);
+insert into test select number + 500000, range(number % 20 + 1)::Array(UInt64) from numbers(100000);"
 }
 
 function test4_select
diff --git a/tests/queries/0_stateless/02943_variant_type_with_different_local_and_global_order.reference b/tests/queries/0_stateless/02943_variant_type_with_different_local_and_global_order.reference
index 1736a307c42..4109a88997c 100644
--- a/tests/queries/0_stateless/02943_variant_type_with_different_local_and_global_order.reference
+++ b/tests/queries/0_stateless/02943_variant_type_with_different_local_and_global_order.reference
@@ -44,9 +44,9 @@ str_38	str_38	\N
 -----------------------------------------------------------------------------------------------------------
 test2 insert
 test2 select
-2500000
-750000
-1750000
+500000
+100000
+400000
 -----------------------------------------------------------------------------------------------------------
 MergeTree compact
 test1 insert
@@ -136,14 +136,14 @@ str_38	str_38	\N
 -----------------------------------------------------------------------------------------------------------
 test2 insert
 test2 select
-2500000
-750000
-1750000
+500000
+100000
+400000
 -----------------------------------------------------------------------------------------------------------
 test2 select
-2500000
-750000
-1750000
+500000
+100000
+400000
 -----------------------------------------------------------------------------------------------------------
 MergeTree wide
 test1 insert
@@ -233,12 +233,12 @@ str_38	str_38	\N
 -----------------------------------------------------------------------------------------------------------
 test2 insert
 test2 select
-2500000
-750000
-1750000
+500000
+100000
+400000
 -----------------------------------------------------------------------------------------------------------
 test2 select
-2500000
-750000
-1750000
+500000
+100000
+400000
 -----------------------------------------------------------------------------------------------------------
diff --git a/tests/queries/0_stateless/02943_variant_type_with_different_local_and_global_order.sh b/tests/queries/0_stateless/02943_variant_type_with_different_local_and_global_order.sh
index 3bb37719a3f..1d88757a5d6 100755
--- a/tests/queries/0_stateless/02943_variant_type_with_different_local_and_global_order.sh
+++ b/tests/queries/0_stateless/02943_variant_type_with_different_local_and_global_order.sh
@@ -29,10 +29,10 @@ function test1_select()
 function test2_insert()
 {
     echo "test2 insert"
-    $CH_CLIENT -q "insert into test select number, number::Variant(UInt64)::Variant(UInt64, Array(UInt64)) from numbers(1000000) settings max_insert_block_size = 100000, min_insert_block_size_rows=100000"
-    $CH_CLIENT -q "insert into test select number, if(number % 2, NULL, number)::Variant(UInt64)::Variant(UInt64, String, Array(UInt64)) as res from numbers(1000000, 1000000) settings max_insert_block_size = 100000, min_insert_block_size_rows=100000"
-    $CH_CLIENT -q "insert into test select number, if(number % 2, NULL, 'str_' || toString(number))::Variant(String)::Variant(UInt64, String, Array(UInt64)) as res from numbers(2000000, 1000000) settings max_insert_block_size = 100000, min_insert_block_size_rows=100000"
-    $CH_CLIENT -q "insert into test select number, if(number < 3500000, if(number % 2, NULL, number)::Variant(UInt64)::Variant(UInt64, String, Array(UInt64)), if(number % 2, NULL, 'str_' || toString(number))::Variant(String)::Variant(UInt64, String, Array(UInt64))) from numbers(3000000, 1000000) settings max_insert_block_size = 100000, min_insert_block_size_rows=100000"
+    $CH_CLIENT -q "insert into test select number, number::Variant(UInt64)::Variant(UInt64, Array(UInt64)) from numbers(200000) settings max_insert_block_size = 10000, min_insert_block_size_rows=10000"
+    $CH_CLIENT -q "insert into test select number, if(number % 2, NULL, number)::Variant(UInt64)::Variant(UInt64, String, Array(UInt64)) as res from numbers(200000, 200000) settings max_insert_block_size = 10000, min_insert_block_size_rows=10000"
+    $CH_CLIENT -q "insert into test select number, if(number % 2, NULL, 'str_' || toString(number))::Variant(String)::Variant(UInt64, String, Array(UInt64)) as res from numbers(400000, 200000) settings max_insert_block_size = 10000, min_insert_block_size_rows=10000"
+    $CH_CLIENT -q "insert into test select number, if(number < 3500000, if(number % 2, NULL, number)::Variant(UInt64)::Variant(UInt64, String, Array(UInt64)), if(number % 2, NULL, 'str_' || toString(number))::Variant(String)::Variant(UInt64, String, Array(UInt64))) from numbers(600000, 200000) settings max_insert_block_size = 10000, min_insert_block_size_rows=10000"
 }
 
 function test2_select()

From a90a6e9a271515dec58e4d4f716bcd591f245c00 Mon Sep 17 00:00:00 2001
From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com>
Date: Mon, 11 Mar 2024 21:05:44 +0100
Subject: [PATCH 188/443] Revert "Don't allow to set max_parallel_replicas to 0
 as it doesn't make sense"

---
 src/Client/ConnectionPoolWithFailover.cpp              | 10 ----------
 src/Client/HedgedConnectionsFactory.cpp                |  6 +-----
 src/Client/HedgedConnectionsFactory.h                  |  2 +-
 src/Interpreters/InterpreterSelectQuery.cpp            |  2 +-
 src/Planner/PlannerJoinTree.cpp                        |  4 ++--
 .../03001_max_parallel_replicas_zero_value.reference   |  0
 .../03001_max_parallel_replicas_zero_value.sql         |  5 -----
 7 files changed, 5 insertions(+), 24 deletions(-)
 delete mode 100644 tests/queries/0_stateless/03001_max_parallel_replicas_zero_value.reference
 delete mode 100644 tests/queries/0_stateless/03001_max_parallel_replicas_zero_value.sql

diff --git a/src/Client/ConnectionPoolWithFailover.cpp b/src/Client/ConnectionPoolWithFailover.cpp
index ad8ed0067d8..492fd4ae9e2 100644
--- a/src/Client/ConnectionPoolWithFailover.cpp
+++ b/src/Client/ConnectionPoolWithFailover.cpp
@@ -21,7 +21,6 @@ namespace ErrorCodes
 {
     extern const int LOGICAL_ERROR;
     extern const int ALL_CONNECTION_TRIES_FAILED;
-    extern const int BAD_ARGUMENTS;
 }
 
 
@@ -192,20 +191,11 @@ std::vector<ConnectionPoolWithFailover::TryResult> ConnectionPoolWithFailover::g
         max_entries = nested_pools.size();
     }
     else if (pool_mode == PoolMode::GET_ONE)
-    {
         max_entries = 1;
-    }
     else if (pool_mode == PoolMode::GET_MANY)
-    {
-        if (settings.max_parallel_replicas == 0)
-            throw Exception(ErrorCodes::BAD_ARGUMENTS, "The value of the setting max_parallel_replicas must be greater than 0");
-
         max_entries = settings.max_parallel_replicas;
-    }
     else
-    {
         throw DB::Exception(DB::ErrorCodes::LOGICAL_ERROR, "Unknown pool allocation mode");
-    }
 
     if (!priority_func)
         priority_func = makeGetPriorityFunc(settings);
diff --git a/src/Client/HedgedConnectionsFactory.cpp b/src/Client/HedgedConnectionsFactory.cpp
index 703cc1f8821..f5b074a0257 100644
--- a/src/Client/HedgedConnectionsFactory.cpp
+++ b/src/Client/HedgedConnectionsFactory.cpp
@@ -19,7 +19,6 @@ namespace ErrorCodes
     extern const int ALL_CONNECTION_TRIES_FAILED;
     extern const int ALL_REPLICAS_ARE_STALE;
     extern const int LOGICAL_ERROR;
-    extern const int BAD_ARGUMENTS;
 }
 
 HedgedConnectionsFactory::HedgedConnectionsFactory(
@@ -83,10 +82,7 @@ std::vector<Connection *> HedgedConnectionsFactory::getManyConnections(PoolMode
         }
         case PoolMode::GET_MANY:
         {
-            if (max_parallel_replicas == 0)
-                throw Exception(ErrorCodes::BAD_ARGUMENTS, "The value of the setting max_parallel_replicas must be greater than 0");
-
-            max_entries = std::min(max_parallel_replicas, shuffled_pools.size());
+            max_entries = max_parallel_replicas;
             break;
         }
     }
diff --git a/src/Client/HedgedConnectionsFactory.h b/src/Client/HedgedConnectionsFactory.h
index dd600d58e1e..ce7b553acdd 100644
--- a/src/Client/HedgedConnectionsFactory.h
+++ b/src/Client/HedgedConnectionsFactory.h
@@ -158,7 +158,7 @@ private:
     /// checking the number of requested replicas that are still in process).
     size_t requested_connections_count = 0;
 
-    const size_t max_parallel_replicas = 1;
+    const size_t max_parallel_replicas = 0;
     const bool skip_unavailable_shards = 0;
 };
 
diff --git a/src/Interpreters/InterpreterSelectQuery.cpp b/src/Interpreters/InterpreterSelectQuery.cpp
index e28d8366aa7..bcedba7346d 100644
--- a/src/Interpreters/InterpreterSelectQuery.cpp
+++ b/src/Interpreters/InterpreterSelectQuery.cpp
@@ -947,7 +947,7 @@ bool InterpreterSelectQuery::adjustParallelReplicasAfterAnalysis()
     if (number_of_replicas_to_use <= 1)
     {
         context->setSetting("allow_experimental_parallel_reading_from_replicas", Field(0));
-        context->setSetting("max_parallel_replicas", UInt64{1});
+        context->setSetting("max_parallel_replicas", UInt64{0});
         LOG_DEBUG(log, "Disabling parallel replicas because there aren't enough rows to read");
         return true;
     }
diff --git a/src/Planner/PlannerJoinTree.cpp b/src/Planner/PlannerJoinTree.cpp
index 0fe943e0bc7..7b3fb0c5c91 100644
--- a/src/Planner/PlannerJoinTree.cpp
+++ b/src/Planner/PlannerJoinTree.cpp
@@ -295,7 +295,7 @@ bool applyTrivialCountIfPossible(
 
         /// The query could use trivial count if it didn't use parallel replicas, so let's disable it
         query_context->setSetting("allow_experimental_parallel_reading_from_replicas", Field(0));
-        query_context->setSetting("max_parallel_replicas", UInt64{1});
+        query_context->setSetting("max_parallel_replicas", UInt64{0});
         LOG_TRACE(getLogger("Planner"), "Disabling parallel replicas to be able to use a trivial count optimization");
 
     }
@@ -756,7 +756,7 @@ JoinTreeQueryPlan buildQueryPlanForTableExpression(QueryTreeNodePtr table_expres
                     {
                         planner_context->getMutableQueryContext()->setSetting(
                             "allow_experimental_parallel_reading_from_replicas", Field(0));
-                        planner_context->getMutableQueryContext()->setSetting("max_parallel_replicas", UInt64{1});
+                        planner_context->getMutableQueryContext()->setSetting("max_parallel_replicas", UInt64{0});
                         LOG_DEBUG(getLogger("Planner"), "Disabling parallel replicas because there aren't enough rows to read");
                     }
                     else if (number_of_replicas_to_use < settings.max_parallel_replicas)
diff --git a/tests/queries/0_stateless/03001_max_parallel_replicas_zero_value.reference b/tests/queries/0_stateless/03001_max_parallel_replicas_zero_value.reference
deleted file mode 100644
index e69de29bb2d..00000000000
diff --git a/tests/queries/0_stateless/03001_max_parallel_replicas_zero_value.sql b/tests/queries/0_stateless/03001_max_parallel_replicas_zero_value.sql
deleted file mode 100644
index 611aa4777ba..00000000000
--- a/tests/queries/0_stateless/03001_max_parallel_replicas_zero_value.sql
+++ /dev/null
@@ -1,5 +0,0 @@
-drop table if exists test_d;
-create table test_d engine=Distributed(test_cluster_two_shard_three_replicas_localhost, system, numbers);
-select * from test_d limit 10 settings max_parallel_replicas = 0, prefer_localhost_replica = 0; --{serverError BAD_ARGUMENTS}
-drop table test_d;
-

From 120a1fdb5f817b442bf659da243407fb7003eaa1 Mon Sep 17 00:00:00 2001
From: johnnymatthews <9611008+johnnymatthews@users.noreply.github.com>
Date: Mon, 11 Mar 2024 17:24:33 -0300
Subject: [PATCH 189/443] Improves varPop docs. Adds varPopStable.

---
 .../aggregate-functions/reference/varpop.md   | 99 +++++++++++++++++--
 1 file changed, 91 insertions(+), 8 deletions(-)

diff --git a/docs/en/sql-reference/aggregate-functions/reference/varpop.md b/docs/en/sql-reference/aggregate-functions/reference/varpop.md
index 751688b0830..5f18bdc30f6 100644
--- a/docs/en/sql-reference/aggregate-functions/reference/varpop.md
+++ b/docs/en/sql-reference/aggregate-functions/reference/varpop.md
@@ -1,16 +1,99 @@
 ---
-slug: /en/sql-reference/aggregate-functions/reference/varpop
+title: "varPop"
+slug: "/en/sql-reference/aggregate-functions/reference/varpop"
 sidebar_position: 32
 ---
 
-# varPop(x)
+This page covers the `varPop` and `varPopStable` functions available in ClickHouse.
 
-Calculates the amount `Σ((x - x̅)^2) / n`, where `n` is the sample size and `x̅`is the average value of `x`.
+## varPop
 
-In other words, dispersion for a set of values. Returns `Float64`.
+Calculates the population covariance between two data columns. The population covariance measures the degree to which two variables vary together. Calculates the amount `Σ((x - x̅)^2) / n`, where `n` is the sample size and `x̅`is the average value of `x`.
 
-Alias: `VAR_POP`.
+**Syntax**
 
-:::note
-This function uses a numerically unstable algorithm. If you need [numerical stability](https://en.wikipedia.org/wiki/Numerical_stability) in calculations, use the `varPopStable` function. It works slower but provides a lower computational error.
-:::
\ No newline at end of file
+```sql
+covarPop(x, y)
+```
+
+**Parameters**
+
+- `x`: The first data column. [Numeric](../../../native-protocol/columns.md)
+- `y`: The second data column. [Numeric](../../../native-protocol/columns.md)
+
+**Returned value**
+
+Returns an integer of type `Float64`.
+
+**Implementation details**
+
+This function uses a numerically unstable algorithm. If you need numerical stability in calculations, use the slower but more stable [`varPopStable` function](#varPopStable). 
+
+**Example**
+
+```sql
+DROP TABLE IF EXISTS test_data;
+CREATE TABLE test_data
+(
+    x Int32,
+    y Int32
+)
+ENGINE = Memory;
+
+INSERT INTO test_data VALUES (1, 2), (2, 3), (3, 5), (4, 6), (5, 8);
+
+SELECT
+    covarPop(x, y) AS covar_pop
+FROM test_data;
+```
+
+```response
+3
+```
+
+## varPopStable
+
+Calculates population covariance between two data columns using a stable, numerically accurate method to calculate the variance. This function is designed to provide reliable results even with large datasets or values that might cause numerical instability in other implementations.
+
+**Syntax**
+
+```sql
+covarPopStable(x, y)
+```
+
+**Parameters**
+
+- `x`: The first data column. [String literal](../syntax#syntax-string-literal)
+- `y`: The second data column. [Expression](../syntax#syntax-expressions)
+
+**Returned value**
+
+Returns an integer of type `Float64`.
+
+**Implementation details**
+
+Unlike [`varPop()`](#varPop), this function uses a stable, numerically accurate algorithm to calculate the population variance to avoid issues like catastrophic cancellation or loss of precision. This function also handles `NaN` and `Inf` values correctly, excluding them from calculations.
+
+**Example**
+
+Query:
+
+```sql
+DROP TABLE IF EXISTS test_data;
+CREATE TABLE test_data
+(
+    x Int32,
+    y Int32
+)
+ENGINE = Memory;
+
+INSERT INTO test_data VALUES (1, 2), (2, 9), (9, 5), (4, 6), (5, 8);
+
+SELECT
+    covarPopStable(x, y) AS covar_pop_stable
+FROM test_data;
+```
+
+```response
+0.5999999999999999
+```

From 281dc8d29deba2980e6b191edefa3b62114d38a7 Mon Sep 17 00:00:00 2001
From: johnnymatthews <9611008+johnnymatthews@users.noreply.github.com>
Date: Mon, 11 Mar 2024 17:48:12 -0300
Subject: [PATCH 190/443] Improves varSamp docs. Adds varSampStable docs.

---
 .../aggregate-functions/reference/varsamp.md  | 126 ++++++++++++++++--
 1 file changed, 118 insertions(+), 8 deletions(-)

diff --git a/docs/en/sql-reference/aggregate-functions/reference/varsamp.md b/docs/en/sql-reference/aggregate-functions/reference/varsamp.md
index 9b2b94936ec..e75cb075ff8 100644
--- a/docs/en/sql-reference/aggregate-functions/reference/varsamp.md
+++ b/docs/en/sql-reference/aggregate-functions/reference/varsamp.md
@@ -1,18 +1,128 @@
 ---
+title: "varSamp"
 slug: /en/sql-reference/aggregate-functions/reference/varsamp
 sidebar_position: 33
 ---
 
-# varSamp
+This page contains information on the `varSamp` and `varSampStable` ClickHouse functions.
 
-Calculates the amount `Σ((x - x̅)^2) / (n - 1)`, where `n` is the sample size and `x̅`is the average value of `x`.
+## varSamp
 
-It represents an unbiased estimate of the variance of a random variable if passed values from its sample.
+Calculate the sample variance of a data set. 
 
-Returns `Float64`. When `n <= 1`, returns `+∞`.
+**Syntax**
 
-Alias: `VAR_SAMP`.
+```sql
+varSamp(expr)
+```
 
-:::note
-This function uses a numerically unstable algorithm. If you need [numerical stability](https://en.wikipedia.org/wiki/Numerical_stability) in calculations, use the `varSampStable` function. It works slower but provides a lower computational error.
-:::
+**Parameters**
+
+- `expr`: An expression representing the data set for which you want to calculate the sample variance. [Expression](../syntax#syntax-expressions)
+
+**Returned value**
+
+Returns a Float64 value representing the sample variance of the input data set.
+
+**Implementation details**
+
+The `varSamp()` function calculates the sample variance using the following formula:
+
+```plaintext
+∑(x - mean(x))^2 / (n - 1)
+```
+
+Where:
+
+- `x` is each individual data point in the data set.
+- `mean(x)` is the arithmetic mean of the data set.
+- `n` is the number of data points in the data set.
+
+The function assumes that the input data set represents a sample from a larger population. If you want to calculate the variance of the entire population (when you have the complete data set), you should use the [`varPop()` function](./varpop#varpop) instead.
+
+This function uses a numerically unstable algorithm. If you need numerical stability in calculations, use the slower but more stable [`varSampStable` function](#varSampStable).
+
+**Example**
+
+Query:
+
+```sql
+CREATE TABLE example_table
+(
+    id UInt64,
+    value Float64
+)
+ENGINE = MergeTree
+ORDER BY id;
+
+INSERT INTO example_table VALUES (1, 10.5), (2, 12.3), (3, 9.8), (4, 11.2), (5, 10.7);
+
+SELECT varSamp(value) FROM example_table;
+```
+
+Response:
+
+```response
+0.8650000000000091
+```
+
+## varSampStable
+
+Calculate the sample variance of a data set using a numerically stable algorithm.
+
+**Syntax**
+
+```sql
+varSampStable(expr)
+```
+
+**Parameters**
+
+- `expr`: An expression representing the data set for which you want to calculate the sample variance. [Expression](../syntax#syntax-expressions)
+
+**Returned value**
+
+The `varSampStable()` function returns a Float64 value representing the sample variance of the input data set.
+
+**Implementation details**
+
+The `varSampStable()` function calculates the sample variance using the same formula as the [`varSamp()`](#varSamp function):
+
+```plaintext
+∑(x - mean(x))^2 / (n - 1)
+```
+
+Where:
+- `x` is each individual data point in the data set.
+- `mean(x)` is the arithmetic mean of the data set.
+- `n` is the number of data points in the data set.
+
+The difference between `varSampStable()` and `varSamp()` is that `varSampStable()` is designed to provide a more deterministic and stable result when dealing with floating-point arithmetic. It uses an algorithm that minimizes the accumulation of rounding errors, which can be particularly important when dealing with large data sets or data with a wide range of values.
+
+Like `varSamp()`, the `varSampStable()` function assumes that the input data set represents a sample from a larger population. If you want to calculate the variance of the entire population (when you have the complete data set), you should use the [`varPopStable()` function](./varpop#varpopstable) instead.
+
+**Example**
+
+Query:
+
+```sql
+CREATE TABLE example_table
+(
+    id UInt64,
+    value Float64
+)
+ENGINE = MergeTree
+ORDER BY id;
+
+INSERT INTO example_table VALUES (1, 10.5), (2, 12.3), (3, 9.8), (4, 11.2), (5, 10.7);
+
+SELECT varSampStable(value) FROM example_table;
+```
+
+Response:
+
+```response
+0.865
+```
+
+This query calculates the sample variance of the `value` column in the `example_table` using the `varSampStable()` function. The result shows that the sample variance of the values `[10.5, 12.3, 9.8, 11.2, 10.7]` is approximately 0.865, which may differ slightly from the result of `varSamp()` due to the more precise handling of floating-point arithmetic.

From 563df9bdcb425810a0c2d3ecb11302e22039c048 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Mon, 11 Mar 2024 22:49:18 +0100
Subject: [PATCH 191/443] Fix multiple bugs in groupArraySorted

---
 .../AggregateFunctionGroupArraySorted.cpp                 | 8 +++++---
 .../0_stateless/03008_groupSortedArray_field.reference    | 3 +++
 .../queries/0_stateless/03008_groupSortedArray_field.sql  | 6 ++++++
 3 files changed, 14 insertions(+), 3 deletions(-)
 create mode 100644 tests/queries/0_stateless/03008_groupSortedArray_field.reference
 create mode 100644 tests/queries/0_stateless/03008_groupSortedArray_field.sql

diff --git a/src/AggregateFunctions/AggregateFunctionGroupArraySorted.cpp b/src/AggregateFunctions/AggregateFunctionGroupArraySorted.cpp
index 0e9856cfab9..0692ff28f18 100644
--- a/src/AggregateFunctions/AggregateFunctionGroupArraySorted.cpp
+++ b/src/AggregateFunctions/AggregateFunctionGroupArraySorted.cpp
@@ -310,10 +310,12 @@ public:
         {
             for (Field & element : values)
             {
-                UInt8 is_null = 0;
-                readBinary(is_null, buf);
-                if (!is_null)
+                bool has_value = 0;
+                readBinary(has_value, buf);
+                if (has_value)
                     serialization->deserializeBinary(element, buf, {});
+                else
+                    element = Field{};
             }
         }
         else
diff --git a/tests/queries/0_stateless/03008_groupSortedArray_field.reference b/tests/queries/0_stateless/03008_groupSortedArray_field.reference
new file mode 100644
index 00000000000..a7f89ebcf58
--- /dev/null
+++ b/tests/queries/0_stateless/03008_groupSortedArray_field.reference
@@ -0,0 +1,3 @@
+0A01003C79A557B3C43400C4865AA84C3B4B01000650BC18F7DE0B00FAAF43E708213401008ED706EA0A9F13007228F915F5602C0100C692CA8FB81405003A6D357047EB1A01008416B7C3239EE3FF7BE9483CDC61DC01003E133A7C081AF5FFC1ECC583F7E5EA01000000000000000000000000000000000100C4865AA84C3BCBFF3B79A557B3C4B4010024C46EF500F1ECFFDB3B910AFF0ED301005E2FC14EBAEAE5FFA1D03EB14515DA
+070109000000010600000001080000000103000000010500000001040000000107000000	AggregateFunction(groupArraySorted(10), Nullable(Decimal(3, 0)))
+[3,4,5,6,7,8,9]
diff --git a/tests/queries/0_stateless/03008_groupSortedArray_field.sql b/tests/queries/0_stateless/03008_groupSortedArray_field.sql
new file mode 100644
index 00000000000..6d2aea641a5
--- /dev/null
+++ b/tests/queries/0_stateless/03008_groupSortedArray_field.sql
@@ -0,0 +1,6 @@
+-- https://github.com/ClickHouse/ClickHouse/issues/61186
+SELECT hex(CAST(unhex('0A01003C79A557B3C43400C4865AA84C3B4B01000650BC18F7DE0B00FAAF43E708213401008ED706EA0A9F13007228F915F5602C0100C692CA8FB81405003A6D357047EB1A01008416B7C3239EE3FF7BE9483CDC61DC01003E133A7C081AF5FFC1ECC583F7E5EA01000000000000000000000000000000000100C4865AA84C3BCBFF3B79A557B3C4B4010024C46EF500F1ECFFDB3B910AFF0ED301005E2FC14EBAEAE5FFA1D03EB14515DA'),
+                'AggregateFunction(groupArraySorted(10), Decimal(38, 38))'));
+
+Select hex(groupArraySortedState(10)((number < 3 ? NULL : number)::Nullable(Decimal(3))) as t), toTypeName(t) from numbers(10);
+Select finalizeAggregation(unhex('070109000000010600000001080000000103000000010500000001040000000107000000')::AggregateFunction(groupArraySorted(10), Nullable(Decimal(3, 0))));

From 1b04cc0b4da6d32fd4741ea953dfed060f846d0b Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Tue, 12 Mar 2024 03:56:10 +0100
Subject: [PATCH 192/443] Fix strange log message

---
 src/Loggers/Loggers.cpp | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/src/Loggers/Loggers.cpp b/src/Loggers/Loggers.cpp
index 1d17585cc96..cc6e4691737 100644
--- a/src/Loggers/Loggers.cpp
+++ b/src/Loggers/Loggers.cpp
@@ -304,6 +304,9 @@ void Loggers::buildLoggers(Poco::Util::AbstractConfiguration & config, Poco::Log
 
         log_settings.turn_off_logger = DB::TextLog::shouldTurnOffLogger();
 
+        log_settings.database = config.getString("text_log.database", "system");
+        log_settings.table = config.getString("text_log.table", "text_log");
+
         split->addTextLog(DB::TextLog::getLogQueue(log_settings), text_log_level);
     }
 #endif

From c628eaca8ba19584fe36067dee8e6ec3e8f5cc4b Mon Sep 17 00:00:00 2001
From: Zhuo Qiu <jewelz.q.915@gmail.com>
Date: Tue, 26 Dec 2023 14:13:07 +0800
Subject: [PATCH 193/443] Consider deleted rows when selecting parts to merge

---
 src/Storages/MergeTree/IMergeTreeDataPart.cpp | 90 +++++++++++++++++++
 src/Storages/MergeTree/IMergeTreeDataPart.h   | 13 +++
 .../MergeTree/MergeFromLogEntryTask.cpp       |  2 +-
 src/Storages/MergeTree/MergeTreeData.cpp      |  1 +
 .../MergeTree/MergeTreeDataMergerMutator.cpp  | 11 ++-
 .../MergeTree/MergeTreeDataMergerMutator.h    |  2 +-
 .../MergeTree/MergeTreeDataWriter.cpp         |  1 +
 src/Storages/MergeTree/MergeTreeSettings.h    |  2 +
 .../MergeTree/MergedBlockOutputStream.cpp     |  5 ++
 .../MergeTree/MutateFromLogEntryTask.cpp      |  2 +-
 src/Storages/MergeTree/MutateTask.cpp         | 54 +++++++++++
 .../MergeTree/ReplicatedMergeTreeQueue.cpp    |  5 +-
 src/Storages/StorageMergeTree.cpp             |  4 +-
 .../03001_consider_lwd_when_merge.reference   |  3 +
 .../03001_consider_lwd_when_merge.sql         | 23 +++++
 15 files changed, 208 insertions(+), 10 deletions(-)
 create mode 100644 tests/queries/0_stateless/03001_consider_lwd_when_merge.reference
 create mode 100644 tests/queries/0_stateless/03001_consider_lwd_when_merge.sql

diff --git a/src/Storages/MergeTree/IMergeTreeDataPart.cpp b/src/Storages/MergeTree/IMergeTreeDataPart.cpp
index 3fea6d04944..c099512d636 100644
--- a/src/Storages/MergeTree/IMergeTreeDataPart.cpp
+++ b/src/Storages/MergeTree/IMergeTreeDataPart.cpp
@@ -609,6 +609,15 @@ UInt64 IMergeTreeDataPart::getMarksCount() const
     return index_granularity.getMarksCount();
 }
 
+UInt64 IMergeTreeDataPart::getExistingBytesOnDisk() const
+{
+    if (storage.getSettings()->exclude_deleted_rows_for_part_size_in_merge && supportLightweightDeleteMutate() && hasLightweightDelete()
+        && existing_rows_count.has_value() && existing_rows_count.value() < rows_count && rows_count > 0)
+        return bytes_on_disk * existing_rows_count.value() / rows_count;
+    else
+        return bytes_on_disk;
+}
+
 size_t IMergeTreeDataPart::getFileSizeOrZero(const String & file_name) const
 {
     auto checksum = checksums.files.find(file_name);
@@ -691,6 +700,7 @@ void IMergeTreeDataPart::loadColumnsChecksumsIndexes(bool require_columns_checks
 
         calculateColumnsAndSecondaryIndicesSizesOnDisk();
         loadRowsCount(); /// Must be called after loadIndexGranularity() as it uses the value of `index_granularity`.
+        loadExistingRowsCount(); /// Must be called after loadRowsCount() as it uses the value of `rows_count`.
         loadPartitionAndMinMaxIndex();
         if (!parent_part)
         {
@@ -1313,6 +1323,86 @@ void IMergeTreeDataPart::loadRowsCount()
     }
 }
 
+void IMergeTreeDataPart::loadExistingRowsCount()
+{
+    if (existing_rows_count.has_value())
+        return;
+
+    if (!rows_count || !storage.getSettings()->load_existing_rows_count_for_old_parts || !supportLightweightDeleteMutate()
+        || !hasLightweightDelete())
+        existing_rows_count = rows_count;
+    else
+        existing_rows_count = readExistingRowsCount();
+}
+
+UInt64 IMergeTreeDataPart::readExistingRowsCount()
+{
+    const size_t total_mark = getMarksCount();
+    if (!total_mark)
+        return rows_count;
+
+    NamesAndTypesList cols;
+    cols.push_back(LightweightDeleteDescription::FILTER_COLUMN);
+
+    StorageMetadataPtr metadata_ptr = storage.getInMemoryMetadataPtr();
+    StorageSnapshotPtr storage_snapshot_ptr = std::make_shared<StorageSnapshot>(storage, metadata_ptr);
+
+    MergeTreeReaderPtr reader = getReader(
+        cols,
+        storage_snapshot_ptr,
+        MarkRanges{MarkRange(0, total_mark)},
+        nullptr,
+        storage.getContext()->getMarkCache().get(),
+        std::make_shared<AlterConversions>(),
+        MergeTreeReaderSettings{},
+        ValueSizeMap{},
+        ReadBufferFromFileBase::ProfileCallback{});
+
+    if (!reader)
+    {
+        LOG_WARNING(storage.log, "Create reader failed while reading existing rows count");
+        return rows_count;
+    }
+
+    size_t current_mark = 0;
+    bool continue_reading = false;
+    size_t current_row = 0;
+    size_t existing_count = 0;
+
+    while (current_row < rows_count)
+    {
+        size_t rows_to_read = index_granularity.getMarkRows(current_mark);
+        continue_reading = (current_mark != 0);
+
+        Columns result;
+        result.resize(1);
+
+        size_t rows_read = reader->readRows(current_mark, total_mark, continue_reading, rows_to_read, result);
+        if (!rows_read)
+        {
+            LOG_WARNING(storage.log, "Part {} has lightweight delete, but _row_exists column not found", name);
+            return rows_count;
+        }
+
+        current_row += rows_read;
+        current_mark += (rows_to_read == rows_read);
+
+        const ColumnUInt8 * row_exists_col = typeid_cast<const ColumnUInt8 *>(result[0].get());
+        if (!row_exists_col)
+        {
+            LOG_WARNING(storage.log, "Part {} _row_exists column type is not UInt8", name);
+            return rows_count;
+        }
+
+        for (UInt8 row_exists : row_exists_col->getData())
+            if (row_exists)
+                existing_count++;
+    }
+
+    LOG_DEBUG(storage.log, "Part {} existing_rows_count = {}", name, existing_count);
+    return existing_count;
+}
+
 void IMergeTreeDataPart::appendFilesOfRowsCount(Strings & files)
 {
     files.push_back("count.txt");
diff --git a/src/Storages/MergeTree/IMergeTreeDataPart.h b/src/Storages/MergeTree/IMergeTreeDataPart.h
index aaae64a5970..8bd32e777bc 100644
--- a/src/Storages/MergeTree/IMergeTreeDataPart.h
+++ b/src/Storages/MergeTree/IMergeTreeDataPart.h
@@ -231,6 +231,9 @@ public:
 
     size_t rows_count = 0;
 
+    /// Existing rows count (excluding lightweight deleted rows)
+    std::optional<size_t> existing_rows_count;
+
     time_t modification_time = 0;
     /// When the part is removed from the working set. Changes once.
     mutable std::atomic<time_t> remove_time { std::numeric_limits<time_t>::max() };
@@ -373,6 +376,10 @@ public:
     void setBytesOnDisk(UInt64 bytes_on_disk_) { bytes_on_disk = bytes_on_disk_; }
     void setBytesUncompressedOnDisk(UInt64 bytes_uncompressed_on_disk_) { bytes_uncompressed_on_disk = bytes_uncompressed_on_disk_; }
 
+    /// Returns estimated size of existing rows if setting exclude_deleted_rows_for_part_size_in_merge is true
+    /// Otherwise returns bytes_on_disk
+    UInt64 getExistingBytesOnDisk() const;
+
     size_t getFileSizeOrZero(const String & file_name) const;
     auto getFilesChecksums() const { return checksums.files; }
 
@@ -499,6 +506,9 @@ public:
     /// True if here is lightweight deleted mask file in part.
     bool hasLightweightDelete() const;
 
+    /// Read existing rows count from _row_exists column
+    UInt64 readExistingRowsCount();
+
     void writeChecksums(const MergeTreeDataPartChecksums & checksums_, const WriteSettings & settings);
 
     /// Checks the consistency of this data part.
@@ -664,6 +674,9 @@ private:
     /// For the older format version calculates rows count from the size of a column with a fixed size.
     void loadRowsCount();
 
+    /// Load existing rows count from _row_exists column if load_existing_rows_count_for_old_parts is true.
+    void loadExistingRowsCount();
+
     static void appendFilesOfRowsCount(Strings & files);
 
     /// Loads ttl infos in json format from file ttl.txt. If file doesn't exists assigns ttl infos with all zeros
diff --git a/src/Storages/MergeTree/MergeFromLogEntryTask.cpp b/src/Storages/MergeTree/MergeFromLogEntryTask.cpp
index ae6e398026d..5ef004ec019 100644
--- a/src/Storages/MergeTree/MergeFromLogEntryTask.cpp
+++ b/src/Storages/MergeTree/MergeFromLogEntryTask.cpp
@@ -174,7 +174,7 @@ ReplicatedMergeMutateTaskBase::PrepareResult MergeFromLogEntryTask::prepare()
     }
 
     /// Start to make the main work
-    size_t estimated_space_for_merge = MergeTreeDataMergerMutator::estimateNeededDiskSpace(parts);
+    size_t estimated_space_for_merge = MergeTreeDataMergerMutator::estimateNeededDiskSpace(parts, true);
 
     /// Can throw an exception while reserving space.
     IMergeTreeDataPart::TTLInfos ttl_infos;
diff --git a/src/Storages/MergeTree/MergeTreeData.cpp b/src/Storages/MergeTree/MergeTreeData.cpp
index d56cf761cf4..5e05f75c1c5 100644
--- a/src/Storages/MergeTree/MergeTreeData.cpp
+++ b/src/Storages/MergeTree/MergeTreeData.cpp
@@ -8261,6 +8261,7 @@ std::pair<MergeTreeData::MutableDataPartPtr, scope_guard> MergeTreeData::createE
 
     new_data_part->setColumns(columns, {}, metadata_snapshot->getMetadataVersion());
     new_data_part->rows_count = block.rows();
+    new_data_part->existing_rows_count = block.rows();
 
     new_data_part->partition = partition;
 
diff --git a/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp b/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp
index 1bf1d4a3c29..90144a8cc8f 100644
--- a/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp
+++ b/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp
@@ -405,7 +405,7 @@ MergeTreeDataMergerMutator::MergeSelectingInfo MergeTreeDataMergerMutator::getPo
         }
 
         IMergeSelector::Part part_info;
-        part_info.size = part->getBytesOnDisk();
+        part_info.size = part->getExistingBytesOnDisk();
         part_info.age = res.current_time - part->modification_time;
         part_info.level = part->info.level;
         part_info.data = &part;
@@ -611,7 +611,7 @@ SelectPartsDecision MergeTreeDataMergerMutator::selectAllPartsToMergeWithinParti
             return SelectPartsDecision::CANNOT_SELECT;
         }
 
-        sum_bytes += (*it)->getBytesOnDisk();
+        sum_bytes += (*it)->getExistingBytesOnDisk();
 
         prev_it = it;
         ++it;
@@ -793,7 +793,7 @@ MergeTreeData::DataPartPtr MergeTreeDataMergerMutator::renameMergedTemporaryPart
 }
 
 
-size_t MergeTreeDataMergerMutator::estimateNeededDiskSpace(const MergeTreeData::DataPartsVector & source_parts)
+size_t MergeTreeDataMergerMutator::estimateNeededDiskSpace(const MergeTreeData::DataPartsVector & source_parts, const bool & is_merge)
 {
     size_t res = 0;
     time_t current_time = std::time(nullptr);
@@ -804,7 +804,10 @@ size_t MergeTreeDataMergerMutator::estimateNeededDiskSpace(const MergeTreeData::
         if (part_max_ttl && part_max_ttl <= current_time)
             continue;
 
-        res += part->getBytesOnDisk();
+        if (is_merge)
+            res += part->getExistingBytesOnDisk();
+        else
+            res += part->getBytesOnDisk();
     }
 
     return static_cast<size_t>(res * DISK_USAGE_COEFFICIENT_TO_RESERVE);
diff --git a/src/Storages/MergeTree/MergeTreeDataMergerMutator.h b/src/Storages/MergeTree/MergeTreeDataMergerMutator.h
index f3a3f51b6c3..731c5e1d176 100644
--- a/src/Storages/MergeTree/MergeTreeDataMergerMutator.h
+++ b/src/Storages/MergeTree/MergeTreeDataMergerMutator.h
@@ -193,7 +193,7 @@ public:
 
 
     /// The approximate amount of disk space needed for merge or mutation. With a surplus.
-    static size_t estimateNeededDiskSpace(const MergeTreeData::DataPartsVector & source_parts);
+    static size_t estimateNeededDiskSpace(const MergeTreeData::DataPartsVector & source_parts, const bool & is_merge);
 
 private:
     /** Select all parts belonging to the same partition.
diff --git a/src/Storages/MergeTree/MergeTreeDataWriter.cpp b/src/Storages/MergeTree/MergeTreeDataWriter.cpp
index fdac16ae19a..2ba74e44b40 100644
--- a/src/Storages/MergeTree/MergeTreeDataWriter.cpp
+++ b/src/Storages/MergeTree/MergeTreeDataWriter.cpp
@@ -537,6 +537,7 @@ MergeTreeDataWriter::TemporaryPart MergeTreeDataWriter::writeTempPartImpl(
 
     new_data_part->setColumns(columns, infos, metadata_snapshot->getMetadataVersion());
     new_data_part->rows_count = block.rows();
+    new_data_part->existing_rows_count = block.rows();
     new_data_part->partition = std::move(partition);
     new_data_part->minmax_idx = std::move(minmax_idx);
     new_data_part->is_temp = true;
diff --git a/src/Storages/MergeTree/MergeTreeSettings.h b/src/Storages/MergeTree/MergeTreeSettings.h
index 925dc973dc3..ea54f61b4b6 100644
--- a/src/Storages/MergeTree/MergeTreeSettings.h
+++ b/src/Storages/MergeTree/MergeTreeSettings.h
@@ -42,6 +42,7 @@ struct Settings;
     M(UInt64, compact_parts_max_bytes_to_buffer, 128 * 1024 * 1024, "Only available in ClickHouse Cloud", 0) \
     M(UInt64, compact_parts_max_granules_to_buffer, 128, "Only available in ClickHouse Cloud", 0) \
     M(UInt64, compact_parts_merge_max_bytes_to_prefetch_part, 16 * 1024 * 1024, "Only available in ClickHouse Cloud", 0) \
+    M(Bool, load_existing_rows_count_for_old_parts, false, "Whether to load existing_rows_count for existing parts. If false, existing_rows_count will be equal to rows_count for existing parts.", 0) \
     \
     /** Merge settings. */ \
     M(UInt64, merge_max_block_size, 8192, "How many rows in blocks should be formed for merge operations. By default has the same value as `index_granularity`.", 0) \
@@ -79,6 +80,7 @@ struct Settings;
     M(UInt64, number_of_mutations_to_throw, 1000, "If table has at least that many unfinished mutations, throw 'Too many mutations' exception. Disabled if set to 0", 0) \
     M(UInt64, min_delay_to_mutate_ms, 10, "Min delay of mutating MergeTree table in milliseconds, if there are a lot of unfinished mutations", 0) \
     M(UInt64, max_delay_to_mutate_ms, 1000, "Max delay of mutating MergeTree table in milliseconds, if there are a lot of unfinished mutations", 0) \
+    M(Bool, exclude_deleted_rows_for_part_size_in_merge, false, "Use an estimated source part size (excluding lightweight deleted rows) when selecting parts to merge", 0) \
     \
     /** Inserts settings. */ \
     M(UInt64, parts_to_delay_insert, 1000, "If table contains at least that many active parts in single partition, artificially slow down insert into table. Disabled if set to 0", 0) \
diff --git a/src/Storages/MergeTree/MergedBlockOutputStream.cpp b/src/Storages/MergeTree/MergedBlockOutputStream.cpp
index f2fe2e0f255..d8555d69788 100644
--- a/src/Storages/MergeTree/MergedBlockOutputStream.cpp
+++ b/src/Storages/MergeTree/MergedBlockOutputStream.cpp
@@ -188,6 +188,11 @@ MergedBlockOutputStream::Finalizer MergedBlockOutputStream::finalizePartAsync(
     new_part->index_granularity = writer->getIndexGranularity();
     new_part->calculateColumnsAndSecondaryIndicesSizesOnDisk();
 
+    /// In mutation, existing_rows_count is already calculated in PartMergerWriter
+    /// In merge situation, lightweight deleted rows was physically deleted, existing_rows_count equals rows_count
+    if (!new_part->existing_rows_count.has_value())
+        new_part->existing_rows_count = rows_count;
+
     if (default_codec != nullptr)
         new_part->default_codec = default_codec;
 
diff --git a/src/Storages/MergeTree/MutateFromLogEntryTask.cpp b/src/Storages/MergeTree/MutateFromLogEntryTask.cpp
index a9ff687fe4d..620b0e34c6a 100644
--- a/src/Storages/MergeTree/MutateFromLogEntryTask.cpp
+++ b/src/Storages/MergeTree/MutateFromLogEntryTask.cpp
@@ -49,7 +49,7 @@ ReplicatedMergeMutateTaskBase::PrepareResult MutateFromLogEntryTask::prepare()
     }
 
     /// TODO - some better heuristic?
-    size_t estimated_space_for_result = MergeTreeDataMergerMutator::estimateNeededDiskSpace({source_part});
+    size_t estimated_space_for_result = MergeTreeDataMergerMutator::estimateNeededDiskSpace({source_part}, false);
 
     if (entry.create_time + storage_settings_ptr->prefer_fetch_merged_part_time_threshold.totalSeconds() <= time(nullptr)
         && estimated_space_for_result >= storage_settings_ptr->prefer_fetch_merged_part_size_threshold)
diff --git a/src/Storages/MergeTree/MutateTask.cpp b/src/Storages/MergeTree/MutateTask.cpp
index 150cc27c369..3d31d2f05db 100644
--- a/src/Storages/MergeTree/MutateTask.cpp
+++ b/src/Storages/MergeTree/MutateTask.cpp
@@ -60,6 +60,26 @@ static bool checkOperationIsNotCanceled(ActionBlocker & merges_blocker, MergeLis
     return true;
 }
 
+static UInt64 getExistingRowsCount(const Block & block)
+{
+    auto column = block.getByName(LightweightDeleteDescription::FILTER_COLUMN.name).column;
+    const ColumnUInt8 * row_exists_col = typeid_cast<const ColumnUInt8 *>(column.get());
+
+    if (!row_exists_col)
+    {
+        LOG_WARNING(&Poco::Logger::get("MutationHelpers::getExistingRowsCount"), "_row_exists column type is not UInt8");
+        return block.rows();
+    }
+
+    UInt64 existing_count = 0;
+
+    for (UInt8 row_exists : row_exists_col->getData())
+        if (row_exists)
+            existing_count++;
+
+    return existing_count;
+}
+
 /** Split mutation commands into two parts:
 *   First part should be executed by mutations interpreter.
 *   Other is just simple drop/renames, so they can be executed without interpreter.
@@ -997,6 +1017,9 @@ struct MutationContext
     bool need_prefix = true;
 
     scope_guard temporary_directory_lock;
+
+    /// Whether this mutation contains lightweight delete
+    bool has_lightweight_delete;
 };
 
 using MutationContextPtr = std::shared_ptr<MutationContext>;
@@ -1191,6 +1214,7 @@ public:
             }
             case State::SUCCESS:
             {
+                finalize();
                 return false;
             }
         }
@@ -1226,6 +1250,11 @@ private:
     const ProjectionsDescription & projections;
 
     ExecutableTaskPtr merge_projection_parts_task_ptr;
+
+    /// Existing rows count calculated during part writing.
+    /// It is initialized in prepare(), calculated in mutateOriginalPartAndPrepareProjections()
+    /// and set to new_data_part in finalize()
+    size_t existing_rows_count;
 };
 
 
@@ -1238,6 +1267,8 @@ void PartMergerWriter::prepare()
         // We split the materialization into multiple stages similar to the process of INSERT SELECT query.
         projection_squashes.emplace_back(settings.min_insert_block_size_rows, settings.min_insert_block_size_bytes);
     }
+
+    existing_rows_count = 0;
 }
 
 
@@ -1251,6 +1282,9 @@ bool PartMergerWriter::mutateOriginalPartAndPrepareProjections()
 
         ctx->out->write(cur_block);
 
+        if (ctx->has_lightweight_delete)
+            existing_rows_count += MutationHelpers::getExistingRowsCount(cur_block);
+
         for (size_t i = 0, size = ctx->projections_to_build.size(); i < size; ++i)
         {
             const auto & projection = *ctx->projections_to_build[i];
@@ -1340,6 +1374,12 @@ bool PartMergerWriter::iterateThroughAllProjections()
     return true;
 }
 
+void PartMergerWriter::finalize()
+{
+    if (ctx->has_lightweight_delete)
+        ctx->new_data_part->existing_rows_count = existing_rows_count;
+}
+
 class MutateAllPartColumnsTask : public IExecutableTask
 {
 public:
@@ -2185,6 +2225,20 @@ bool MutateTask::prepare()
     if (ctx->mutating_pipeline_builder.initialized())
         ctx->execute_ttl_type = MutationHelpers::shouldExecuteTTL(ctx->metadata_snapshot, ctx->interpreter->getColumnDependencies());
 
+    if (ctx->updated_header.has(LightweightDeleteDescription::FILTER_COLUMN.name))
+    {
+        /// This mutation contains lightweight delete, reset existing_rows_count of new data part to 0
+        /// It will be updated while writing _row_exists column
+        ctx->has_lightweight_delete = true;
+    }
+    else
+    {
+        ctx->has_lightweight_delete = false;
+
+        /// This mutation does not contains lightweight delete, copy existing_rows_count from source part
+        ctx->new_data_part->existing_rows_count = ctx->source_part->existing_rows_count.value_or(ctx->source_part->rows_count);
+    }
+
     /// All columns from part are changed and may be some more that were missing before in part
     /// TODO We can materialize compact part without copying data
     if (!isWidePart(ctx->source_part) || !isFullPartStorage(ctx->source_part->getDataPartStorage())
diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeQueue.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeQueue.cpp
index 858eae4afd9..42f564f40da 100644
--- a/src/Storages/MergeTree/ReplicatedMergeTreeQueue.cpp
+++ b/src/Storages/MergeTree/ReplicatedMergeTreeQueue.cpp
@@ -1350,7 +1350,10 @@ bool ReplicatedMergeTreeQueue::shouldExecuteLogEntry(
             auto part = data.getPartIfExists(name, {MergeTreeDataPartState::PreActive, MergeTreeDataPartState::Active, MergeTreeDataPartState::Outdated});
             if (part)
             {
-                sum_parts_size_in_bytes += part->getBytesOnDisk();
+                if (entry.type == LogEntry::MERGE_PARTS)
+                    sum_parts_size_in_bytes += part->getExistingBytesOnDisk();
+                else
+                    sum_parts_size_in_bytes += part->getBytesOnDisk();
 
                 if (entry.type == LogEntry::MUTATE_PART && !storage.mutation_backoff_policy.partCanBeMutated(part->name))
                 {
diff --git a/src/Storages/StorageMergeTree.cpp b/src/Storages/StorageMergeTree.cpp
index 663e7f435b7..c816a6f0dce 100644
--- a/src/Storages/StorageMergeTree.cpp
+++ b/src/Storages/StorageMergeTree.cpp
@@ -1113,7 +1113,7 @@ MergeMutateSelectedEntryPtr StorageMergeTree::selectPartsToMerge(
     if (isTTLMergeType(future_part->merge_type))
         getContext()->getMergeList().bookMergeWithTTL();
 
-    merging_tagger = std::make_unique<CurrentlyMergingPartsTagger>(future_part, MergeTreeDataMergerMutator::estimateNeededDiskSpace(future_part->parts), *this, metadata_snapshot, false);
+    merging_tagger = std::make_unique<CurrentlyMergingPartsTagger>(future_part, MergeTreeDataMergerMutator::estimateNeededDiskSpace(future_part->parts, true), *this, metadata_snapshot, false);
     return std::make_shared<MergeMutateSelectedEntry>(future_part, std::move(merging_tagger), std::make_shared<MutationCommands>());
 }
 
@@ -1336,7 +1336,7 @@ MergeMutateSelectedEntryPtr StorageMergeTree::selectPartsToMutate(
             future_part->name = part->getNewName(new_part_info);
             future_part->part_format = part->getFormat();
 
-            tagger = std::make_unique<CurrentlyMergingPartsTagger>(future_part, MergeTreeDataMergerMutator::estimateNeededDiskSpace({part}), *this, metadata_snapshot, true);
+            tagger = std::make_unique<CurrentlyMergingPartsTagger>(future_part, MergeTreeDataMergerMutator::estimateNeededDiskSpace({part}, false), *this, metadata_snapshot, true);
             return std::make_shared<MergeMutateSelectedEntry>(future_part, std::move(tagger), commands, txn);
         }
     }
diff --git a/tests/queries/0_stateless/03001_consider_lwd_when_merge.reference b/tests/queries/0_stateless/03001_consider_lwd_when_merge.reference
new file mode 100644
index 00000000000..19920de3d3c
--- /dev/null
+++ b/tests/queries/0_stateless/03001_consider_lwd_when_merge.reference
@@ -0,0 +1,3 @@
+2
+2
+1
diff --git a/tests/queries/0_stateless/03001_consider_lwd_when_merge.sql b/tests/queries/0_stateless/03001_consider_lwd_when_merge.sql
new file mode 100644
index 00000000000..a65e8877020
--- /dev/null
+++ b/tests/queries/0_stateless/03001_consider_lwd_when_merge.sql
@@ -0,0 +1,23 @@
+DROP TABLE IF EXISTS lwd_merge;
+
+CREATE TABLE lwd_merge (id UInt64 CODEC(NONE))
+    ENGINE = MergeTree ORDER BY id
+SETTINGS max_bytes_to_merge_at_max_space_in_pool = 80000, exclude_deleted_rows_for_part_size_in_merge = 0;
+
+INSERT INTO lwd_merge SELECT number FROM numbers(10000);
+INSERT INTO lwd_merge SELECT number FROM numbers(10000, 10000);
+
+OPTIMIZE TABLE lwd_merge;
+SELECT count() FROM system.parts WHERE database = currentDatabase() AND table = 'lwd_merge' AND active = 1;
+
+DELETE FROM lwd_merge WHERE id % 10 > 0;
+
+OPTIMIZE TABLE lwd_merge;
+SELECT count() FROM system.parts WHERE database = currentDatabase() AND table = 'lwd_merge' AND active = 1;
+
+ALTER TABLE lwd_merge MODIFY SETTING exclude_deleted_rows_for_part_size_in_merge = 1;
+
+OPTIMIZE TABLE lwd_merge;
+SELECT count() FROM system.parts WHERE database = currentDatabase() AND table = 'lwd_merge' AND active = 1;
+
+DROP TABLE IF EXISTS lwd_merge;

From 4ad8141a162b3b7735e2f08c069e98b9c2ba2382 Mon Sep 17 00:00:00 2001
From: Zhuo Qiu <jewelz.q.915@gmail.com>
Date: Wed, 28 Feb 2024 19:54:21 -0600
Subject: [PATCH 194/443] Maintain compatibility of estimateNeededDiskSpace()

Co-authored-by: Yakov Olkhovskiy <99031427+yakov-olkhovskiy@users.noreply.github.com>
---
 src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp | 4 ++--
 src/Storages/MergeTree/MergeTreeDataMergerMutator.h   | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp b/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp
index 90144a8cc8f..53d49b51e8f 100644
--- a/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp
+++ b/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp
@@ -793,7 +793,7 @@ MergeTreeData::DataPartPtr MergeTreeDataMergerMutator::renameMergedTemporaryPart
 }
 
 
-size_t MergeTreeDataMergerMutator::estimateNeededDiskSpace(const MergeTreeData::DataPartsVector & source_parts, const bool & is_merge)
+size_t MergeTreeDataMergerMutator::estimateNeededDiskSpace(const MergeTreeData::DataPartsVector & source_parts, const bool & account_for_deleted)
 {
     size_t res = 0;
     time_t current_time = std::time(nullptr);
@@ -804,7 +804,7 @@ size_t MergeTreeDataMergerMutator::estimateNeededDiskSpace(const MergeTreeData::
         if (part_max_ttl && part_max_ttl <= current_time)
             continue;
 
-        if (is_merge)
+        if (account_for_deleted)
             res += part->getExistingBytesOnDisk();
         else
             res += part->getBytesOnDisk();
diff --git a/src/Storages/MergeTree/MergeTreeDataMergerMutator.h b/src/Storages/MergeTree/MergeTreeDataMergerMutator.h
index 731c5e1d176..669ee040af3 100644
--- a/src/Storages/MergeTree/MergeTreeDataMergerMutator.h
+++ b/src/Storages/MergeTree/MergeTreeDataMergerMutator.h
@@ -193,7 +193,7 @@ public:
 
 
     /// The approximate amount of disk space needed for merge or mutation. With a surplus.
-    static size_t estimateNeededDiskSpace(const MergeTreeData::DataPartsVector & source_parts, const bool & is_merge);
+    static size_t estimateNeededDiskSpace(const MergeTreeData::DataPartsVector & source_parts, const bool & account_for_deleted = false);
 
 private:
     /** Select all parts belonging to the same partition.

From 05969a39f390445c8d0df43b7077e0eb81db3538 Mon Sep 17 00:00:00 2001
From: Zhuo Qiu <jewelz.q.915@gmail.com>
Date: Tue, 12 Mar 2024 14:45:25 +0800
Subject: [PATCH 195/443] resolve conflicts

---
 src/Storages/MergeTree/IMergeTreeDataPart.cpp | 5 +++--
 src/Storages/MergeTree/MutateTask.cpp         | 4 ++--
 2 files changed, 5 insertions(+), 4 deletions(-)

diff --git a/src/Storages/MergeTree/IMergeTreeDataPart.cpp b/src/Storages/MergeTree/IMergeTreeDataPart.cpp
index c099512d636..5fede923252 100644
--- a/src/Storages/MergeTree/IMergeTreeDataPart.cpp
+++ b/src/Storages/MergeTree/IMergeTreeDataPart.cpp
@@ -1342,7 +1342,7 @@ UInt64 IMergeTreeDataPart::readExistingRowsCount()
         return rows_count;
 
     NamesAndTypesList cols;
-    cols.push_back(LightweightDeleteDescription::FILTER_COLUMN);
+    cols.emplace_back(RowExistsColumn::name, RowExistsColumn::type);
 
     StorageMetadataPtr metadata_ptr = storage.getInMemoryMetadataPtr();
     StorageSnapshotPtr storage_snapshot_ptr = std::make_shared<StorageSnapshot>(storage, metadata_ptr);
@@ -1351,7 +1351,8 @@ UInt64 IMergeTreeDataPart::readExistingRowsCount()
         cols,
         storage_snapshot_ptr,
         MarkRanges{MarkRange(0, total_mark)},
-        nullptr,
+        /*virtual_fields=*/ {},
+        /*uncompressed_cache=*/{},
         storage.getContext()->getMarkCache().get(),
         std::make_shared<AlterConversions>(),
         MergeTreeReaderSettings{},
diff --git a/src/Storages/MergeTree/MutateTask.cpp b/src/Storages/MergeTree/MutateTask.cpp
index 3d31d2f05db..4d1e60f450e 100644
--- a/src/Storages/MergeTree/MutateTask.cpp
+++ b/src/Storages/MergeTree/MutateTask.cpp
@@ -62,7 +62,7 @@ static bool checkOperationIsNotCanceled(ActionBlocker & merges_blocker, MergeLis
 
 static UInt64 getExistingRowsCount(const Block & block)
 {
-    auto column = block.getByName(LightweightDeleteDescription::FILTER_COLUMN.name).column;
+    auto column = block.getByName(RowExistsColumn::name).column;
     const ColumnUInt8 * row_exists_col = typeid_cast<const ColumnUInt8 *>(column.get());
 
     if (!row_exists_col)
@@ -2225,7 +2225,7 @@ bool MutateTask::prepare()
     if (ctx->mutating_pipeline_builder.initialized())
         ctx->execute_ttl_type = MutationHelpers::shouldExecuteTTL(ctx->metadata_snapshot, ctx->interpreter->getColumnDependencies());
 
-    if (ctx->updated_header.has(LightweightDeleteDescription::FILTER_COLUMN.name))
+    if (ctx->updated_header.has(RowExistsColumn::name))
     {
         /// This mutation contains lightweight delete, reset existing_rows_count of new data part to 0
         /// It will be updated while writing _row_exists column

From fb1e5923a333bad5336ccb2631c81c95f4eb57dc Mon Sep 17 00:00:00 2001
From: Max Kainov <max.kainov@clickhouse.com>
Date: Tue, 12 Mar 2024 09:12:25 +0000
Subject: [PATCH 196/443] CI: fix runner for arm perf tests

 #do_not_test
---
 tests/ci/ci_config.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/tests/ci/ci_config.py b/tests/ci/ci_config.py
index df8bfb1c2a8..7c213da27ec 100644
--- a/tests/ci/ci_config.py
+++ b/tests/ci/ci_config.py
@@ -629,7 +629,9 @@ class CIConfig:
 
         assert result, f"BUG, no runner for [{check_name}]"
 
-        if ("aarch" in check_name or "arm" in check_name) and "aarch" not in result:
+        if (
+            "aarch" in check_name.lower() or "arm64" in check_name.lower()
+        ) and "aarch" not in result:
             if result == Runners.STRESS_TESTER:
                 # FIXME: no arm stress tester group atm
                 result = Runners.FUNC_TESTER_ARM

From f6abe68b607dd1893e838b55f85245c716d41f8b Mon Sep 17 00:00:00 2001
From: Pablo Musa <pmusa@users.noreply.github.com>
Date: Tue, 12 Mar 2024 10:52:22 +0100
Subject: [PATCH 197/443] Clarify sentence about clusterAllReplicas

The explanation about `clusterAllReplicas` is not clear. This change tries to simplify the sentence and clarify the explanation.
---
 docs/en/sql-reference/table-functions/cluster.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/en/sql-reference/table-functions/cluster.md b/docs/en/sql-reference/table-functions/cluster.md
index ad92ab39183..136ff72e4a9 100644
--- a/docs/en/sql-reference/table-functions/cluster.md
+++ b/docs/en/sql-reference/table-functions/cluster.md
@@ -5,7 +5,7 @@ sidebar_label: cluster
 title: "cluster, clusterAllReplicas"
 ---
 
-Allows to access all shards in an existing cluster which configured in `remote_servers` section without creating a [Distributed](../../engines/table-engines/special/distributed.md) table. One replica of each shard is queried.
+Allows to access all shards (configured in the `remote_servers` section) of a cluster without creating a [Distributed](../../engines/table-engines/special/distributed.md) table. Only one replica of each shard is queried.
 
 `clusterAllReplicas` function — same as `cluster`, but all replicas are queried. Each replica in a cluster is used as a separate shard/connection.
 

From ccd47126fdcc82e57d0c648b82613b5672847702 Mon Sep 17 00:00:00 2001
From: Azat Khuzhin <a3at.mail@gmail.com>
Date: Tue, 12 Mar 2024 11:01:47 +0100
Subject: [PATCH 198/443] Fix 01603_insert_select_too_many_parts flakiness

CI: https://s3.amazonaws.com/clickhouse-test-reports/60695/048a042dc4963631a23358d3e454dcd8a9eaafa2/stateless_tests__aarch64_.html
Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>
---
 .../queries/0_stateless/01603_insert_select_too_many_parts.sql  | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tests/queries/0_stateless/01603_insert_select_too_many_parts.sql b/tests/queries/0_stateless/01603_insert_select_too_many_parts.sql
index a56b680e212..0c33c1d6b18 100644
--- a/tests/queries/0_stateless/01603_insert_select_too_many_parts.sql
+++ b/tests/queries/0_stateless/01603_insert_select_too_many_parts.sql
@@ -3,6 +3,8 @@ CREATE TABLE too_many_parts (x UInt64) ENGINE = MergeTree ORDER BY tuple() SETTI
 
 SYSTEM STOP MERGES too_many_parts;
 SET max_block_size = 1, min_insert_block_size_rows = 0, min_insert_block_size_bytes = 0;
+-- Avoid concurrent partse check to avoid flakiness
+SET max_threads=1, max_insert_threads=1;
 
 -- exception is not thrown if threshold is exceeded when multi-block INSERT is already started.
 INSERT INTO too_many_parts SELECT * FROM numbers(10);

From 47a4ce8a4e629f2d6321be1411d93c0bfabbbc95 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <benjamin.antal@clickhouse.com>
Date: Tue, 12 Mar 2024 10:02:33 +0000
Subject: [PATCH 199/443] Move `RangeHashedDictionary::getColumn` into source
 file

---
 src/Dictionaries/RangeHashedDictionary.cpp | 448 +++++++++++++++++++++
 src/Dictionaries/RangeHashedDictionary.h   | 421 -------------------
 2 files changed, 448 insertions(+), 421 deletions(-)
 create mode 100644 src/Dictionaries/RangeHashedDictionary.cpp

diff --git a/src/Dictionaries/RangeHashedDictionary.cpp b/src/Dictionaries/RangeHashedDictionary.cpp
new file mode 100644
index 00000000000..203561fc23d
--- /dev/null
+++ b/src/Dictionaries/RangeHashedDictionary.cpp
@@ -0,0 +1,448 @@
+#include <Dictionaries/RangeHashedDictionary.h>
+
+namespace DB
+{
+
+namespace ErrorCodes
+{
+    extern const int TYPE_MISMATCH;
+}
+
+template <DictionaryKeyType dictionary_key_type>
+ColumnPtr RangeHashedDictionary<dictionary_key_type>::getColumn(
+    const std::string & attribute_name,
+    const DataTypePtr & attribute_type,
+    const Columns & key_columns,
+    const DataTypes & key_types,
+    DefaultOrFilter default_or_filter) const
+{
+    bool is_short_circuit = std::holds_alternative<RefFilter>(default_or_filter);
+    assert(is_short_circuit || std::holds_alternative<RefDefault>(default_or_filter));
+
+    if (dictionary_key_type == DictionaryKeyType::Complex)
+    {
+        auto key_types_copy = key_types;
+        key_types_copy.pop_back();
+        dict_struct.validateKeyTypes(key_types_copy);
+    }
+
+    ColumnPtr result;
+
+    const auto & dictionary_attribute = dict_struct.getAttribute(attribute_name, attribute_type);
+    const size_t attribute_index = dict_struct.attribute_name_to_index.find(attribute_name)->second;
+    const auto & attribute = attributes[attribute_index];
+
+    /// Cast range column to storage type
+    Columns modified_key_columns = key_columns;
+    const ColumnPtr & range_storage_column = key_columns.back();
+    ColumnWithTypeAndName column_to_cast = {range_storage_column->convertToFullColumnIfConst(), key_types.back(), ""};
+    modified_key_columns.back() = castColumnAccurate(column_to_cast, dict_struct.range_min->type);
+
+    size_t keys_size = key_columns.front()->size();
+    bool is_attribute_nullable = attribute.is_value_nullable.has_value();
+
+    ColumnUInt8::MutablePtr col_null_map_to;
+    ColumnUInt8::Container * vec_null_map_to = nullptr;
+    if (is_attribute_nullable)
+    {
+        col_null_map_to = ColumnUInt8::create(keys_size, false);
+        vec_null_map_to = &col_null_map_to->getData();
+    }
+
+    auto type_call = [&](const auto & dictionary_attribute_type)
+    {
+        using Type = std::decay_t<decltype(dictionary_attribute_type)>;
+        using AttributeType = typename Type::AttributeType;
+        using ValueType = DictionaryValueType<AttributeType>;
+        using ColumnProvider = DictionaryAttributeColumnProvider<AttributeType>;
+
+        auto column = ColumnProvider::getColumn(dictionary_attribute, keys_size);
+
+        if (is_short_circuit)
+        {
+            IColumn::Filter & default_mask = std::get<RefFilter>(default_or_filter).get();
+            size_t keys_found = 0;
+
+            if constexpr (std::is_same_v<ValueType, Array>)
+            {
+                auto * out = column.get();
+
+                keys_found = getItemsShortCircuitImpl<ValueType, false>(
+                    attribute,
+                    modified_key_columns,
+                    [&](size_t, const Array & value, bool)
+                    {
+                        out->insert(value);
+                    },
+                    default_mask);
+            }
+            else if constexpr (std::is_same_v<ValueType, StringRef>)
+            {
+                auto * out = column.get();
+
+                if (is_attribute_nullable)
+                    keys_found = getItemsShortCircuitImpl<ValueType, true>(
+                        attribute,
+                        modified_key_columns,
+                        [&](size_t row, StringRef value, bool is_null)
+                        {
+                            (*vec_null_map_to)[row] = is_null;
+                            out->insertData(value.data, value.size);
+                        },
+                        default_mask);
+                else
+                    keys_found = getItemsShortCircuitImpl<ValueType, false>(
+                        attribute,
+                        modified_key_columns,
+                        [&](size_t, StringRef value, bool)
+                        {
+                            out->insertData(value.data, value.size);
+                        },
+                        default_mask);
+            }
+            else
+            {
+                auto & out = column->getData();
+
+                if (is_attribute_nullable)
+                    keys_found = getItemsShortCircuitImpl<ValueType, true>(
+                        attribute,
+                        modified_key_columns,
+                        [&](size_t row, const auto value, bool is_null)
+                        {
+                            (*vec_null_map_to)[row] = is_null;
+                            out[row] = value;
+                        },
+                        default_mask);
+                else
+                    keys_found = getItemsShortCircuitImpl<ValueType, false>(
+                        attribute,
+                        modified_key_columns,
+                        [&](size_t row, const auto value, bool)
+                        {
+                            out[row] = value;
+                        },
+                        default_mask);
+
+                out.resize(keys_found);
+            }
+
+            if (is_attribute_nullable)
+                vec_null_map_to->resize(keys_found);
+        }
+        else
+        {
+            const ColumnPtr & default_values_column = std::get<RefDefault>(default_or_filter).get();
+
+            DictionaryDefaultValueExtractor<AttributeType> default_value_extractor(
+            dictionary_attribute.null_value, default_values_column);
+
+            if constexpr (std::is_same_v<ValueType, Array>)
+            {
+                auto * out = column.get();
+
+                getItemsImpl<ValueType, false>(
+                    attribute,
+                    modified_key_columns,
+                    [&](size_t, const Array & value, bool)
+                    {
+                        out->insert(value);
+                    },
+                    default_value_extractor);
+            }
+            else if constexpr (std::is_same_v<ValueType, StringRef>)
+            {
+                auto * out = column.get();
+
+                if (is_attribute_nullable)
+                    getItemsImpl<ValueType, true>(
+                        attribute,
+                        modified_key_columns,
+                        [&](size_t row, StringRef value, bool is_null)
+                        {
+                            (*vec_null_map_to)[row] = is_null;
+                            out->insertData(value.data, value.size);
+                        },
+                        default_value_extractor);
+                else
+                    getItemsImpl<ValueType, false>(
+                        attribute,
+                        modified_key_columns,
+                        [&](size_t, StringRef value, bool)
+                        {
+                            out->insertData(value.data, value.size);
+                        },
+                        default_value_extractor);
+            }
+            else
+            {
+                auto & out = column->getData();
+
+                if (is_attribute_nullable)
+                    getItemsImpl<ValueType, true>(
+                        attribute,
+                        modified_key_columns,
+                        [&](size_t row, const auto value, bool is_null)
+                        {
+                            (*vec_null_map_to)[row] = is_null;
+                            out[row] = value;
+                        },
+                        default_value_extractor);
+                else
+                    getItemsImpl<ValueType, false>(
+                        attribute,
+                        modified_key_columns,
+                        [&](size_t row, const auto value, bool)
+                        {
+                            out[row] = value;
+                        },
+                        default_value_extractor);
+            }
+        }
+
+        result = std::move(column);
+    };
+
+    callOnDictionaryAttributeType(attribute.type, type_call);
+
+    if (is_attribute_nullable)
+        result = ColumnNullable::create(result, std::move(col_null_map_to));
+
+    return result;
+}
+
+template <DictionaryKeyType dictionary_key_type>
+template <typename AttributeType, bool is_nullable, typename ValueSetter, typename DefaultValueExtractor>
+void RangeHashedDictionary<dictionary_key_type>::getItemsImpl(
+    const Attribute & attribute,
+    const Columns & key_columns,
+    ValueSetter && set_value,
+    DefaultValueExtractor & default_value_extractor) const
+{
+    const auto & attribute_container = std::get<AttributeContainerType<AttributeType>>(attribute.container);
+
+    size_t keys_found = 0;
+
+    const ColumnPtr & range_column = key_columns.back();
+    auto key_columns_copy = key_columns;
+    key_columns_copy.pop_back();
+
+    DictionaryKeysArenaHolder<dictionary_key_type> arena_holder;
+    DictionaryKeysExtractor<dictionary_key_type> keys_extractor(key_columns_copy, arena_holder.getComplexKeyArena());
+    const size_t keys_size = keys_extractor.getKeysSize();
+
+    callOnRangeType(dict_struct.range_min->type, [&](const auto & types)
+    {
+        using Types = std::decay_t<decltype(types)>;
+        using RangeColumnType = typename Types::LeftType;
+        using RangeStorageType = typename RangeColumnType::ValueType;
+        using RangeInterval = Interval<RangeStorageType>;
+
+        const auto * range_column_typed = typeid_cast<const RangeColumnType *>(range_column.get());
+        if (!range_column_typed)
+            throw Exception(ErrorCodes::TYPE_MISMATCH,
+                "Dictionary {} range column type should be equal to {}",
+                getFullName(),
+                dict_struct.range_min->type->getName());
+
+        const auto & range_column_data = range_column_typed->getData();
+
+        const auto & key_attribute_container = std::get<KeyAttributeContainerType<RangeStorageType>>(key_attribute.container);
+
+        for (size_t key_index = 0; key_index < keys_size; ++key_index)
+        {
+            auto key = keys_extractor.extractCurrentKey();
+            const auto it = key_attribute_container.find(key);
+
+            if (it)
+            {
+                const auto date = range_column_data[key_index];
+                const auto & interval_tree = it->getMapped();
+
+                size_t value_index = 0;
+                std::optional<RangeInterval> range;
+
+                interval_tree.find(date, [&](auto & interval, auto & interval_value_index)
+                {
+                    if (range)
+                    {
+                        if (likely(configuration.lookup_strategy == RangeHashedDictionaryLookupStrategy::min) && interval < *range)
+                        {
+                            range = interval;
+                            value_index = interval_value_index;
+                        }
+                        else if (configuration.lookup_strategy == RangeHashedDictionaryLookupStrategy::max && interval > * range)
+                        {
+                            range = interval;
+                            value_index = interval_value_index;
+                        }
+                    }
+                    else
+                    {
+                        range = interval;
+                        value_index = interval_value_index;
+                    }
+
+                    return true;
+                });
+
+                if (range.has_value())
+                {
+                    ++keys_found;
+
+                    AttributeType value = attribute_container[value_index];
+
+                    if constexpr (is_nullable)
+                    {
+                        bool is_null = (*attribute.is_value_nullable)[value_index];
+                        set_value(key_index, value, is_null);
+                    }
+                    else
+                    {
+                        set_value(key_index, value, false);
+                    }
+
+                    keys_extractor.rollbackCurrentKey();
+                    continue;
+                }
+            }
+
+            if constexpr (is_nullable)
+                set_value(key_index, default_value_extractor[key_index], default_value_extractor.isNullAt(key_index));
+            else
+                set_value(key_index, default_value_extractor[key_index], false);
+
+            keys_extractor.rollbackCurrentKey();
+        }
+    });
+
+    query_count.fetch_add(keys_size, std::memory_order_relaxed);
+    found_count.fetch_add(keys_found, std::memory_order_relaxed);
+}
+
+template <DictionaryKeyType dictionary_key_type>
+template <typename AttributeType, bool is_nullable, typename ValueSetter>
+size_t RangeHashedDictionary<dictionary_key_type>::getItemsShortCircuitImpl(
+    const Attribute & attribute,
+    const Columns & key_columns,
+    ValueSetter && set_value,
+    IColumn::Filter & default_mask) const
+{
+    const auto & attribute_container = std::get<AttributeContainerType<AttributeType>>(attribute.container);
+
+    size_t keys_found = 0;
+
+    const ColumnPtr & range_column = key_columns.back();
+    auto key_columns_copy = key_columns;
+    key_columns_copy.pop_back();
+
+    DictionaryKeysArenaHolder<dictionary_key_type> arena_holder;
+    DictionaryKeysExtractor<dictionary_key_type> keys_extractor(key_columns_copy, arena_holder.getComplexKeyArena());
+    const size_t keys_size = keys_extractor.getKeysSize();
+    default_mask.resize(keys_size);
+
+    callOnRangeType(dict_struct.range_min->type, [&](const auto & types)
+    {
+        using Types = std::decay_t<decltype(types)>;
+        using RangeColumnType = typename Types::LeftType;
+        using RangeStorageType = typename RangeColumnType::ValueType;
+        using RangeInterval = Interval<RangeStorageType>;
+
+        const auto * range_column_typed = typeid_cast<const RangeColumnType *>(range_column.get());
+        if (!range_column_typed)
+            throw Exception(ErrorCodes::TYPE_MISMATCH,
+                "Dictionary {} range column type should be equal to {}",
+                getFullName(),
+                dict_struct.range_min->type->getName());
+
+        const auto & range_column_data = range_column_typed->getData();
+
+        const auto & key_attribute_container = std::get<KeyAttributeContainerType<RangeStorageType>>(key_attribute.container);
+
+        for (size_t key_index = 0; key_index < keys_size; ++key_index)
+        {
+            auto key = keys_extractor.extractCurrentKey();
+            const auto it = key_attribute_container.find(key);
+
+            if (it)
+            {
+                const auto date = range_column_data[key_index];
+                const auto & interval_tree = it->getMapped();
+
+                size_t value_index = 0;
+                std::optional<RangeInterval> range;
+
+                interval_tree.find(date, [&](auto & interval, auto & interval_value_index)
+                {
+                    if (range)
+                    {
+                        if (likely(configuration.lookup_strategy == RangeHashedDictionaryLookupStrategy::min) && interval < *range)
+                        {
+                            range = interval;
+                            value_index = interval_value_index;
+                        }
+                        else if (configuration.lookup_strategy == RangeHashedDictionaryLookupStrategy::max && interval > * range)
+                        {
+                            range = interval;
+                            value_index = interval_value_index;
+                        }
+                    }
+                    else
+                    {
+                        range = interval;
+                        value_index = interval_value_index;
+                    }
+
+                    return true;
+                });
+
+                if (range.has_value())
+                {
+                    default_mask[key_index] = 0;
+                    ++keys_found;
+
+                    AttributeType value = attribute_container[value_index];
+
+                    if constexpr (is_nullable)
+                    {
+                        bool is_null = (*attribute.is_value_nullable)[value_index];
+                        set_value(key_index, value, is_null);
+                    }
+                    else
+                    {
+                        set_value(key_index, value, false);
+                    }
+
+                    keys_extractor.rollbackCurrentKey();
+                    continue;
+                }
+            }
+
+            default_mask[key_index] = 1;
+
+            keys_extractor.rollbackCurrentKey();
+        }
+    });
+
+    query_count.fetch_add(keys_size, std::memory_order_relaxed);
+    found_count.fetch_add(keys_found, std::memory_order_relaxed);
+    return keys_found;
+}
+
+template
+ColumnPtr RangeHashedDictionary<DictionaryKeyType::Simple>::getColumn(
+    const std::string & attribute_name,
+    const DataTypePtr & attribute_type,
+    const Columns & key_columns,
+    const DataTypes & key_types,
+    DefaultOrFilter default_or_filter) const;
+
+template
+ColumnPtr RangeHashedDictionary<DictionaryKeyType::Complex>::getColumn(
+    const std::string & attribute_name,
+    const DataTypePtr & attribute_type,
+    const Columns & key_columns,
+    const DataTypes & key_types,
+    DefaultOrFilter default_or_filter) const;
+
+}
diff --git a/src/Dictionaries/RangeHashedDictionary.h b/src/Dictionaries/RangeHashedDictionary.h
index 509b991b30c..4a8008b9051 100644
--- a/src/Dictionaries/RangeHashedDictionary.h
+++ b/src/Dictionaries/RangeHashedDictionary.h
@@ -339,209 +339,6 @@ RangeHashedDictionary<dictionary_key_type>::RangeHashedDictionary(
     calculateBytesAllocated();
 }
 
-template <DictionaryKeyType dictionary_key_type>
-ColumnPtr RangeHashedDictionary<dictionary_key_type>::getColumn(
-    const std::string & attribute_name,
-    const DataTypePtr & attribute_type,
-    const Columns & key_columns,
-    const DataTypes & key_types,
-    DefaultOrFilter default_or_filter) const
-{
-    bool is_short_circuit = std::holds_alternative<RefFilter>(default_or_filter);
-    assert(is_short_circuit || std::holds_alternative<RefDefault>(default_or_filter));
-
-    if (dictionary_key_type == DictionaryKeyType::Complex)
-    {
-        auto key_types_copy = key_types;
-        key_types_copy.pop_back();
-        dict_struct.validateKeyTypes(key_types_copy);
-    }
-
-    ColumnPtr result;
-
-    const auto & dictionary_attribute = dict_struct.getAttribute(attribute_name, attribute_type);
-    const size_t attribute_index = dict_struct.attribute_name_to_index.find(attribute_name)->second;
-    const auto & attribute = attributes[attribute_index];
-
-    /// Cast range column to storage type
-    Columns modified_key_columns = key_columns;
-    const ColumnPtr & range_storage_column = key_columns.back();
-    ColumnWithTypeAndName column_to_cast = {range_storage_column->convertToFullColumnIfConst(), key_types.back(), ""};
-    modified_key_columns.back() = castColumnAccurate(column_to_cast, dict_struct.range_min->type);
-
-    size_t keys_size = key_columns.front()->size();
-    bool is_attribute_nullable = attribute.is_value_nullable.has_value();
-
-    ColumnUInt8::MutablePtr col_null_map_to;
-    ColumnUInt8::Container * vec_null_map_to = nullptr;
-    if (is_attribute_nullable)
-    {
-        col_null_map_to = ColumnUInt8::create(keys_size, false);
-        vec_null_map_to = &col_null_map_to->getData();
-    }
-
-    auto type_call = [&](const auto & dictionary_attribute_type)
-    {
-        using Type = std::decay_t<decltype(dictionary_attribute_type)>;
-        using AttributeType = typename Type::AttributeType;
-        using ValueType = DictionaryValueType<AttributeType>;
-        using ColumnProvider = DictionaryAttributeColumnProvider<AttributeType>;
-
-        auto column = ColumnProvider::getColumn(dictionary_attribute, keys_size);
-
-        if (is_short_circuit)
-        {
-            IColumn::Filter & default_mask = std::get<RefFilter>(default_or_filter).get();
-            size_t keys_found = 0;
-
-            if constexpr (std::is_same_v<ValueType, Array>)
-            {
-                auto * out = column.get();
-
-                keys_found = getItemsShortCircuitImpl<ValueType, false>(
-                    attribute,
-                    modified_key_columns,
-                    [&](size_t, const Array & value, bool)
-                    {
-                        out->insert(value);
-                    },
-                    default_mask);
-            }
-            else if constexpr (std::is_same_v<ValueType, StringRef>)
-            {
-                auto * out = column.get();
-
-                if (is_attribute_nullable)
-                    keys_found = getItemsShortCircuitImpl<ValueType, true>(
-                        attribute,
-                        modified_key_columns,
-                        [&](size_t row, StringRef value, bool is_null)
-                        {
-                            (*vec_null_map_to)[row] = is_null;
-                            out->insertData(value.data, value.size);
-                        },
-                        default_mask);
-                else
-                    keys_found = getItemsShortCircuitImpl<ValueType, false>(
-                        attribute,
-                        modified_key_columns,
-                        [&](size_t, StringRef value, bool)
-                        {
-                            out->insertData(value.data, value.size);
-                        },
-                        default_mask);
-            }
-            else
-            {
-                auto & out = column->getData();
-
-                if (is_attribute_nullable)
-                    keys_found = getItemsShortCircuitImpl<ValueType, true>(
-                        attribute,
-                        modified_key_columns,
-                        [&](size_t row, const auto value, bool is_null)
-                        {
-                            (*vec_null_map_to)[row] = is_null;
-                            out[row] = value;
-                        },
-                        default_mask);
-                else
-                    keys_found = getItemsShortCircuitImpl<ValueType, false>(
-                        attribute,
-                        modified_key_columns,
-                        [&](size_t row, const auto value, bool)
-                        {
-                            out[row] = value;
-                        },
-                        default_mask);
-
-                out.resize(keys_found);
-            }
-
-            if (is_attribute_nullable)
-                vec_null_map_to->resize(keys_found);
-        }
-        else
-        {
-            const ColumnPtr & default_values_column = std::get<RefDefault>(default_or_filter).get();
-
-            DictionaryDefaultValueExtractor<AttributeType> default_value_extractor(
-            dictionary_attribute.null_value, default_values_column);
-
-            if constexpr (std::is_same_v<ValueType, Array>)
-            {
-                auto * out = column.get();
-
-                getItemsImpl<ValueType, false>(
-                    attribute,
-                    modified_key_columns,
-                    [&](size_t, const Array & value, bool)
-                    {
-                        out->insert(value);
-                    },
-                    default_value_extractor);
-            }
-            else if constexpr (std::is_same_v<ValueType, StringRef>)
-            {
-                auto * out = column.get();
-
-                if (is_attribute_nullable)
-                    getItemsImpl<ValueType, true>(
-                        attribute,
-                        modified_key_columns,
-                        [&](size_t row, StringRef value, bool is_null)
-                        {
-                            (*vec_null_map_to)[row] = is_null;
-                            out->insertData(value.data, value.size);
-                        },
-                        default_value_extractor);
-                else
-                    getItemsImpl<ValueType, false>(
-                        attribute,
-                        modified_key_columns,
-                        [&](size_t, StringRef value, bool)
-                        {
-                            out->insertData(value.data, value.size);
-                        },
-                        default_value_extractor);
-            }
-            else
-            {
-                auto & out = column->getData();
-
-                if (is_attribute_nullable)
-                    getItemsImpl<ValueType, true>(
-                        attribute,
-                        modified_key_columns,
-                        [&](size_t row, const auto value, bool is_null)
-                        {
-                            (*vec_null_map_to)[row] = is_null;
-                            out[row] = value;
-                        },
-                        default_value_extractor);
-                else
-                    getItemsImpl<ValueType, false>(
-                        attribute,
-                        modified_key_columns,
-                        [&](size_t row, const auto value, bool)
-                        {
-                            out[row] = value;
-                        },
-                        default_value_extractor);
-            }
-        }
-
-        result = std::move(column);
-    };
-
-    callOnDictionaryAttributeType(attribute.type, type_call);
-
-    if (is_attribute_nullable)
-        result = ColumnNullable::create(result, std::move(col_null_map_to));
-
-    return result;
-}
-
 template <DictionaryKeyType dictionary_key_type>
 ColumnPtr RangeHashedDictionary<dictionary_key_type>::getColumnInternal(
     const std::string & attribute_name,
@@ -840,224 +637,6 @@ typename RangeHashedDictionary<dictionary_key_type>::Attribute RangeHashedDictio
     return attribute;
 }
 
-template <DictionaryKeyType dictionary_key_type>
-template <typename AttributeType, bool is_nullable, typename ValueSetter, typename DefaultValueExtractor>
-void RangeHashedDictionary<dictionary_key_type>::getItemsImpl(
-    const Attribute & attribute,
-    const Columns & key_columns,
-    ValueSetter && set_value,
-    DefaultValueExtractor & default_value_extractor) const
-{
-    const auto & attribute_container = std::get<AttributeContainerType<AttributeType>>(attribute.container);
-
-    size_t keys_found = 0;
-
-    const ColumnPtr & range_column = key_columns.back();
-    auto key_columns_copy = key_columns;
-    key_columns_copy.pop_back();
-
-    DictionaryKeysArenaHolder<dictionary_key_type> arena_holder;
-    DictionaryKeysExtractor<dictionary_key_type> keys_extractor(key_columns_copy, arena_holder.getComplexKeyArena());
-    const size_t keys_size = keys_extractor.getKeysSize();
-
-    callOnRangeType(dict_struct.range_min->type, [&](const auto & types)
-    {
-        using Types = std::decay_t<decltype(types)>;
-        using RangeColumnType = typename Types::LeftType;
-        using RangeStorageType = typename RangeColumnType::ValueType;
-        using RangeInterval = Interval<RangeStorageType>;
-
-        const auto * range_column_typed = typeid_cast<const RangeColumnType *>(range_column.get());
-        if (!range_column_typed)
-            throw Exception(ErrorCodes::TYPE_MISMATCH,
-                "Dictionary {} range column type should be equal to {}",
-                getFullName(),
-                dict_struct.range_min->type->getName());
-
-        const auto & range_column_data = range_column_typed->getData();
-
-        const auto & key_attribute_container = std::get<KeyAttributeContainerType<RangeStorageType>>(key_attribute.container);
-
-        for (size_t key_index = 0; key_index < keys_size; ++key_index)
-        {
-            auto key = keys_extractor.extractCurrentKey();
-            const auto it = key_attribute_container.find(key);
-
-            if (it)
-            {
-                const auto date = range_column_data[key_index];
-                const auto & interval_tree = it->getMapped();
-
-                size_t value_index = 0;
-                std::optional<RangeInterval> range;
-
-                interval_tree.find(date, [&](auto & interval, auto & interval_value_index)
-                {
-                    if (range)
-                    {
-                        if (likely(configuration.lookup_strategy == RangeHashedDictionaryLookupStrategy::min) && interval < *range)
-                        {
-                            range = interval;
-                            value_index = interval_value_index;
-                        }
-                        else if (configuration.lookup_strategy == RangeHashedDictionaryLookupStrategy::max && interval > * range)
-                        {
-                            range = interval;
-                            value_index = interval_value_index;
-                        }
-                    }
-                    else
-                    {
-                        range = interval;
-                        value_index = interval_value_index;
-                    }
-
-                    return true;
-                });
-
-                if (range.has_value())
-                {
-                    ++keys_found;
-
-                    AttributeType value = attribute_container[value_index];
-
-                    if constexpr (is_nullable)
-                    {
-                        bool is_null = (*attribute.is_value_nullable)[value_index];
-                        set_value(key_index, value, is_null);
-                    }
-                    else
-                    {
-                        set_value(key_index, value, false);
-                    }
-
-                    keys_extractor.rollbackCurrentKey();
-                    continue;
-                }
-            }
-
-            if constexpr (is_nullable)
-                set_value(key_index, default_value_extractor[key_index], default_value_extractor.isNullAt(key_index));
-            else
-                set_value(key_index, default_value_extractor[key_index], false);
-
-            keys_extractor.rollbackCurrentKey();
-        }
-    });
-
-    query_count.fetch_add(keys_size, std::memory_order_relaxed);
-    found_count.fetch_add(keys_found, std::memory_order_relaxed);
-}
-
-template <DictionaryKeyType dictionary_key_type>
-template <typename AttributeType, bool is_nullable, typename ValueSetter>
-size_t RangeHashedDictionary<dictionary_key_type>::getItemsShortCircuitImpl(
-    const Attribute & attribute,
-    const Columns & key_columns,
-    ValueSetter && set_value,
-    IColumn::Filter & default_mask) const
-{
-    const auto & attribute_container = std::get<AttributeContainerType<AttributeType>>(attribute.container);
-
-    size_t keys_found = 0;
-
-    const ColumnPtr & range_column = key_columns.back();
-    auto key_columns_copy = key_columns;
-    key_columns_copy.pop_back();
-
-    DictionaryKeysArenaHolder<dictionary_key_type> arena_holder;
-    DictionaryKeysExtractor<dictionary_key_type> keys_extractor(key_columns_copy, arena_holder.getComplexKeyArena());
-    const size_t keys_size = keys_extractor.getKeysSize();
-    default_mask.resize(keys_size);
-
-    callOnRangeType(dict_struct.range_min->type, [&](const auto & types)
-    {
-        using Types = std::decay_t<decltype(types)>;
-        using RangeColumnType = typename Types::LeftType;
-        using RangeStorageType = typename RangeColumnType::ValueType;
-        using RangeInterval = Interval<RangeStorageType>;
-
-        const auto * range_column_typed = typeid_cast<const RangeColumnType *>(range_column.get());
-        if (!range_column_typed)
-            throw Exception(ErrorCodes::TYPE_MISMATCH,
-                "Dictionary {} range column type should be equal to {}",
-                getFullName(),
-                dict_struct.range_min->type->getName());
-
-        const auto & range_column_data = range_column_typed->getData();
-
-        const auto & key_attribute_container = std::get<KeyAttributeContainerType<RangeStorageType>>(key_attribute.container);
-
-        for (size_t key_index = 0; key_index < keys_size; ++key_index)
-        {
-            auto key = keys_extractor.extractCurrentKey();
-            const auto it = key_attribute_container.find(key);
-
-            if (it)
-            {
-                const auto date = range_column_data[key_index];
-                const auto & interval_tree = it->getMapped();
-
-                size_t value_index = 0;
-                std::optional<RangeInterval> range;
-
-                interval_tree.find(date, [&](auto & interval, auto & interval_value_index)
-                {
-                    if (range)
-                    {
-                        if (likely(configuration.lookup_strategy == RangeHashedDictionaryLookupStrategy::min) && interval < *range)
-                        {
-                            range = interval;
-                            value_index = interval_value_index;
-                        }
-                        else if (configuration.lookup_strategy == RangeHashedDictionaryLookupStrategy::max && interval > * range)
-                        {
-                            range = interval;
-                            value_index = interval_value_index;
-                        }
-                    }
-                    else
-                    {
-                        range = interval;
-                        value_index = interval_value_index;
-                    }
-
-                    return true;
-                });
-
-                if (range.has_value())
-                {
-                    default_mask[key_index] = 0;
-                    ++keys_found;
-
-                    AttributeType value = attribute_container[value_index];
-
-                    if constexpr (is_nullable)
-                    {
-                        bool is_null = (*attribute.is_value_nullable)[value_index];
-                        set_value(key_index, value, is_null);
-                    }
-                    else
-                    {
-                        set_value(key_index, value, false);
-                    }
-
-                    keys_extractor.rollbackCurrentKey();
-                    continue;
-                }
-            }
-
-            default_mask[key_index] = 1;
-
-            keys_extractor.rollbackCurrentKey();
-        }
-    });
-
-    query_count.fetch_add(keys_size, std::memory_order_relaxed);
-    found_count.fetch_add(keys_found, std::memory_order_relaxed);
-    return keys_found;
-}
-
 template <DictionaryKeyType dictionary_key_type>
 template <typename AttributeType, bool is_nullable, typename ValueSetter>
 void RangeHashedDictionary<dictionary_key_type>::getItemsInternalImpl(

From f1f388d9d26dbaf0775e90fdc26f9085b83b1c26 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <benjamin.antal@clickhouse.com>
Date: Mon, 11 Mar 2024 16:33:12 +0000
Subject: [PATCH 200/443] Remove RangeHashedDictionary as exception from large
 objects check

---
 utils/check-style/check-large-objects.sh | 2 --
 1 file changed, 2 deletions(-)

diff --git a/utils/check-style/check-large-objects.sh b/utils/check-style/check-large-objects.sh
index 5c1276e5732..5b0e8e88df5 100755
--- a/utils/check-style/check-large-objects.sh
+++ b/utils/check-style/check-large-objects.sh
@@ -7,8 +7,6 @@ TU_EXCLUDES=(
     AggregateFunctionUniq
     FunctionsConversion
 
-    RangeHashedDictionary
-
     Aggregator
 )
 

From c78a029d15f3b8d4796b69ea370af0e3b8abf8e2 Mon Sep 17 00:00:00 2001
From: Azat Khuzhin <a3at.mail@gmail.com>
Date: Tue, 12 Mar 2024 11:12:27 +0100
Subject: [PATCH 201/443] tests: fix typo in
 01603_insert_select_too_many_parts.sql
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Co-authored-by: János Benjamin Antal <antaljanosbenjamin@users.noreply.github.com>
---
 .../queries/0_stateless/01603_insert_select_too_many_parts.sql  | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/queries/0_stateless/01603_insert_select_too_many_parts.sql b/tests/queries/0_stateless/01603_insert_select_too_many_parts.sql
index 0c33c1d6b18..9408ccf2bbf 100644
--- a/tests/queries/0_stateless/01603_insert_select_too_many_parts.sql
+++ b/tests/queries/0_stateless/01603_insert_select_too_many_parts.sql
@@ -3,7 +3,7 @@ CREATE TABLE too_many_parts (x UInt64) ENGINE = MergeTree ORDER BY tuple() SETTI
 
 SYSTEM STOP MERGES too_many_parts;
 SET max_block_size = 1, min_insert_block_size_rows = 0, min_insert_block_size_bytes = 0;
--- Avoid concurrent partse check to avoid flakiness
+-- Avoid concurrent parts check to avoid flakiness
 SET max_threads=1, max_insert_threads=1;
 
 -- exception is not thrown if threshold is exceeded when multi-block INSERT is already started.

From c9c73378eb169e3de7e7579c5ec7edd573e1b72e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Tue, 12 Mar 2024 11:52:17 +0100
Subject: [PATCH 202/443] Fix problem detected by ubsan

---
 .../AggregateFunctionGroupArraySorted.cpp                   | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/src/AggregateFunctions/AggregateFunctionGroupArraySorted.cpp b/src/AggregateFunctions/AggregateFunctionGroupArraySorted.cpp
index 0692ff28f18..1ee29c362cd 100644
--- a/src/AggregateFunctions/AggregateFunctionGroupArraySorted.cpp
+++ b/src/AggregateFunctions/AggregateFunctionGroupArraySorted.cpp
@@ -310,12 +310,12 @@ public:
         {
             for (Field & element : values)
             {
-                bool has_value = 0;
+                /// We must initialize the Field type since some internal functions (like operator=) use them
+                new (&element) Field;
+                bool has_value = false;
                 readBinary(has_value, buf);
                 if (has_value)
                     serialization->deserializeBinary(element, buf, {});
-                else
-                    element = Field{};
             }
         }
         else

From 4ce5245157418217b3f7e41724df3b8cf3dd3272 Mon Sep 17 00:00:00 2001
From: robot-clickhouse <robot-clickhouse@users.noreply.github.com>
Date: Tue, 12 Mar 2024 11:03:39 +0000
Subject: [PATCH 203/443] Automatic style fix

---
 tests/integration/test_disk_types/test.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tests/integration/test_disk_types/test.py b/tests/integration/test_disk_types/test.py
index 5047cdc605e..a53d073d30b 100644
--- a/tests/integration/test_disk_types/test.py
+++ b/tests/integration/test_disk_types/test.py
@@ -50,6 +50,8 @@ def test_different_types(cluster):
             assert (
                 fields[encrypted_col_ix] == "0"
             ), f"{fields[name_col_ix]} expected to be non-encrypted!"
+
+
 def test_different_types(cluster):
     node = cluster.instances["node"]
     response = TSV.toMat(node.query("SELECT * FROM system.disks FORMAT TSVWithNames"))

From c1cefe18744d06bebd67b6425ae5afc9cb922e78 Mon Sep 17 00:00:00 2001
From: Antonio Andelic <antonio@clickhouse.com>
Date: Tue, 12 Mar 2024 11:41:36 +0000
Subject: [PATCH 204/443] Fix keeper reconfig for standalone binary

---
 src/Coordination/Standalone/Context.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Coordination/Standalone/Context.cpp b/src/Coordination/Standalone/Context.cpp
index 264cf118501..75b81187973 100644
--- a/src/Coordination/Standalone/Context.cpp
+++ b/src/Coordination/Standalone/Context.cpp
@@ -374,7 +374,7 @@ void Context::updateKeeperConfiguration([[maybe_unused]] const Poco::Util::Abstr
     if (!shared->keeper_dispatcher)
         return;
 
-    shared->keeper_dispatcher->updateConfiguration(getConfigRef(), getMacros());
+    shared->keeper_dispatcher->updateConfiguration(config_, getMacros());
 }
 
 std::shared_ptr<zkutil::ZooKeeper> Context::getZooKeeper() const

From d431276045f6600af92562abf2b6387bd37d068c Mon Sep 17 00:00:00 2001
From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com>
Date: Tue, 12 Mar 2024 12:43:48 +0100
Subject: [PATCH 205/443] Fix usage of session_token in S3

---
 src/Storages/StorageS3.cpp | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/src/Storages/StorageS3.cpp b/src/Storages/StorageS3.cpp
index 11da394feec..ff055508aa6 100644
--- a/src/Storages/StorageS3.cpp
+++ b/src/Storages/StorageS3.cpp
@@ -1451,7 +1451,8 @@ void StorageS3::Configuration::connect(const ContextPtr & context)
             auth_settings.expiration_window_seconds.value_or(
                 context->getConfigRef().getUInt64("s3.expiration_window_seconds", S3::DEFAULT_EXPIRATION_WINDOW_SECONDS)),
             auth_settings.no_sign_request.value_or(context->getConfigRef().getBool("s3.no_sign_request", false)),
-        });
+        },
+        credentials.GetSessionToken());
 }
 
 void StorageS3::processNamedCollectionResult(StorageS3::Configuration & configuration, const NamedCollection & collection)

From 2e803f29f8f89eec0d9eb95b089ecad5dd6e18b7 Mon Sep 17 00:00:00 2001
From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com>
Date: Tue, 12 Mar 2024 12:45:02 +0100
Subject: [PATCH 206/443] Fix KeeperSnapshotManagerS3

---
 src/Coordination/KeeperSnapshotManagerS3.cpp | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/src/Coordination/KeeperSnapshotManagerS3.cpp b/src/Coordination/KeeperSnapshotManagerS3.cpp
index 80345db2524..796506a07db 100644
--- a/src/Coordination/KeeperSnapshotManagerS3.cpp
+++ b/src/Coordination/KeeperSnapshotManagerS3.cpp
@@ -121,7 +121,8 @@ void KeeperSnapshotManagerS3::updateS3Configuration(const Poco::Util::AbstractCo
                 auth_settings.use_insecure_imds_request.value_or(false),
                 auth_settings.expiration_window_seconds.value_or(S3::DEFAULT_EXPIRATION_WINDOW_SECONDS),
                 auth_settings.no_sign_request.value_or(false),
-            });
+            },
+            credentials.GetSessionToken());
 
         auto new_client = std::make_shared<KeeperSnapshotManagerS3::S3Configuration>(std::move(new_uri), std::move(auth_settings), std::move(client));
 

From 612ff3c5bc1f4878f869bd6e49cc1e60e58e3afd Mon Sep 17 00:00:00 2001
From: "Mikhail f. Shiryaev" <felixoid@clickhouse.com>
Date: Tue, 12 Mar 2024 12:35:33 +0100
Subject: [PATCH 207/443] Make every style-checker runner types scaling-out
 very quickly

---
 tests/ci/autoscale_runners_lambda/app.py            | 8 ++++++--
 tests/ci/autoscale_runners_lambda/test_autoscale.py | 2 +-
 2 files changed, 7 insertions(+), 3 deletions(-)

diff --git a/tests/ci/autoscale_runners_lambda/app.py b/tests/ci/autoscale_runners_lambda/app.py
index 1aa2e424320..a36c508482c 100644
--- a/tests/ci/autoscale_runners_lambda/app.py
+++ b/tests/ci/autoscale_runners_lambda/app.py
@@ -51,7 +51,7 @@ class Queue:
     label: str
 
 
-def get_scales() -> Tuple[int, int]:
+def get_scales(runner_type: str) -> Tuple[int, int]:
     "returns the multipliers for scaling down and up ASG by types"
     # Scaling down is quicker on the lack of running jobs than scaling up on
     # queue
@@ -63,8 +63,12 @@ def get_scales() -> Tuple[int, int]:
     # 10. I am trying 7 now.
     # 7 still looks a bit slow, so I try 6
     # Let's have it the same as the other ASG
+    #
+    # All type of style-checkers should be added very quickly to not block the workflows
     # UPDATE THE COMMENT ON CHANGES
     scale_up = 3
+    if "style" in runner_type:
+        scale_up = 1
     return scale_down, scale_up
 
 
@@ -95,7 +99,7 @@ def set_capacity(
             continue
         raise ValueError("Queue status is not in ['in_progress', 'queued']")
 
-    scale_down, scale_up = get_scales()
+    scale_down, scale_up = get_scales(runner_type)
     # With lyfecycle hooks some instances are actually free because some of
     # them are in 'Terminating:Wait' state
     effective_capacity = max(
diff --git a/tests/ci/autoscale_runners_lambda/test_autoscale.py b/tests/ci/autoscale_runners_lambda/test_autoscale.py
index 21a407276f9..75f178ac394 100644
--- a/tests/ci/autoscale_runners_lambda/test_autoscale.py
+++ b/tests/ci/autoscale_runners_lambda/test_autoscale.py
@@ -80,7 +80,7 @@ class TestSetCapacity(unittest.TestCase):
             ),
             TestCase("increase-1", 1, 13, 20, [Queue("queued", 23, "increase-1")], 17),
             TestCase(
-                "style-checker", 1, 13, 20, [Queue("queued", 33, "style-checker")], 20
+                "style-checker", 1, 13, 20, [Queue("queued", 19, "style-checker")], 19
             ),
             TestCase("increase-2", 1, 13, 20, [Queue("queued", 18, "increase-2")], 15),
             TestCase("increase-3", 1, 13, 20, [Queue("queued", 183, "increase-3")], 20),

From 50aab0b69f75c027b0b6dd53ff747492485eed53 Mon Sep 17 00:00:00 2001
From: "Mikhail f. Shiryaev" <felixoid@clickhouse.com>
Date: Tue, 12 Mar 2024 14:06:17 +0100
Subject: [PATCH 208/443] Fix annoying typo

---
 tests/ci/autoscale_runners_lambda/app.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/ci/autoscale_runners_lambda/app.py b/tests/ci/autoscale_runners_lambda/app.py
index a36c508482c..6c3d71708e9 100644
--- a/tests/ci/autoscale_runners_lambda/app.py
+++ b/tests/ci/autoscale_runners_lambda/app.py
@@ -142,7 +142,7 @@ def set_capacity(
 
         logging.info(
             "The ASG %s capacity will be increased to %s, current capacity=%s, "
-            "effective capacity=%sm maximum capacity=%s, running jobs=%s, queue size=%s",
+            "effective capacity=%s, maximum capacity=%s, running jobs=%s, queue size=%s",
             asg["AutoScalingGroupName"],
             desired_capacity,
             effective_capacity,

From c011901d01709dff91e8afe53121af7388842f33 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Tue, 12 Mar 2024 14:28:28 +0100
Subject: [PATCH 209/443] Fix coverage

---
 src/Functions/coverage.cpp | 1 +
 1 file changed, 1 insertion(+)

diff --git a/src/Functions/coverage.cpp b/src/Functions/coverage.cpp
index f4cac26df78..a1a43d0cf58 100644
--- a/src/Functions/coverage.cpp
+++ b/src/Functions/coverage.cpp
@@ -4,6 +4,7 @@
 #include <DataTypes/DataTypesNumber.h>
 #include <Columns/ColumnArray.h>
 #include <Columns/ColumnVector.h>
+#include <Columns/ColumnsNumber.h>
 #include <Columns/ColumnConst.h>
 #include <Functions/FunctionFactory.h>
 #include <Functions/IFunction.h>

From c27c1e1035772cff76f318a361153ca21f9adc55 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Tue, 12 Mar 2024 14:36:07 +0100
Subject: [PATCH 210/443] Fix build

---
 src/Functions/FunctionsConversion.cpp | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 94fd960a99e..e2d9c4a173e 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -3391,9 +3391,9 @@ arguments, result_type, input_rows_count); \
             case IntervalKind::Kind::INTERVAL_KIND: \
                 return createFunctionAdaptor(FunctionConvert<DataTypeInterval, NameToInterval##INTERVAL_KIND, PositiveMonotonicity>::create(), from_type);
 
-    static WrapperType createIntervalWrapper(const DataTypePtr & from_type, IntervalKind::Kind kind)
+    static WrapperType createIntervalWrapper(const DataTypePtr & from_type, IntervalKind kind)
     {
-        switch (kind)
+        switch (kind.kind)
         {
             GENERATE_INTERVAL_CASE(Nanosecond)
             GENERATE_INTERVAL_CASE(Microsecond)

From b5489ac9620311b160b8c5fa23c02afcc7271114 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <benjamin.antal@clickhouse.com>
Date: Tue, 12 Mar 2024 14:05:39 +0000
Subject: [PATCH 211/443] Print out all queries to have better insights

---
 ...materialized_views_ignore_errors.reference | 26 +++++++++++++++++--
 ..._logs_materialized_views_ignore_errors.sql |  8 +++---
 2 files changed, 29 insertions(+), 5 deletions(-)

diff --git a/tests/queries/0_stateless/02572_system_logs_materialized_views_ignore_errors.reference b/tests/queries/0_stateless/02572_system_logs_materialized_views_ignore_errors.reference
index 029f80b46b0..596a047c104 100644
--- a/tests/queries/0_stateless/02572_system_logs_materialized_views_ignore_errors.reference
+++ b/tests/queries/0_stateless/02572_system_logs_materialized_views_ignore_errors.reference
@@ -1,2 +1,24 @@
-11	queryfinish	OK
-11	querystart	OK
+"-- Attach MV to system.query_log and check that writing query_log will not fail\n\nset log_queries=1;","querystart","OK"
+"-- Attach MV to system.query_log and check that writing query_log will not fail\n\nset log_queries=1;","queryfinish","OK"
+"drop table if exists log_proxy_02572;","querystart","OK"
+"drop table if exists log_proxy_02572;","queryfinish","OK"
+"drop table if exists push_to_logs_proxy_mv_02572;","querystart","OK"
+"drop table if exists push_to_logs_proxy_mv_02572;","queryfinish","OK"
+"-- create log tables\nsystem flush logs;","querystart","OK"
+"-- create log tables\nsystem flush logs;","queryfinish","OK"
+"create table log_proxy_02572 as system.query_log engine=Distributed('test_shard_localhost', currentDatabase(), 'receiver_02572');","querystart","OK"
+"create table log_proxy_02572 as system.query_log engine=Distributed('test_shard_localhost', currentDatabase(), 'receiver_02572');","queryfinish","OK"
+"create materialized view push_to_logs_proxy_mv_02572 to log_proxy_02572 as select * from system.query_log;","querystart","OK"
+"create materialized view push_to_logs_proxy_mv_02572 to log_proxy_02572 as select * from system.query_log;","queryfinish","OK"
+"select 1 format Null;","querystart","OK"
+"select 1 format Null;","queryfinish","OK"
+"system flush logs;","querystart","OK"
+"system flush logs;","queryfinish","OK"
+"system flush logs;","querystart","OK"
+"system flush logs;","queryfinish","OK"
+"drop table log_proxy_02572;","querystart","OK"
+"drop table log_proxy_02572;","queryfinish","OK"
+"drop table push_to_logs_proxy_mv_02572;","querystart","OK"
+"drop table push_to_logs_proxy_mv_02572;","queryfinish","OK"
+"set log_queries=0;","querystart","OK"
+"set log_queries=0;","queryfinish","OK"
diff --git a/tests/queries/0_stateless/02572_system_logs_materialized_views_ignore_errors.sql b/tests/queries/0_stateless/02572_system_logs_materialized_views_ignore_errors.sql
index a7a74190821..2381639fba0 100644
--- a/tests/queries/0_stateless/02572_system_logs_materialized_views_ignore_errors.sql
+++ b/tests/queries/0_stateless/02572_system_logs_materialized_views_ignore_errors.sql
@@ -21,10 +21,12 @@ system flush logs;
 drop table log_proxy_02572;
 drop table push_to_logs_proxy_mv_02572;
 
+set log_queries=0;
+
 system flush logs;
 -- lower() to pass through clickhouse-test "exception" check
-select count(), lower(type::String), errorCodeToName(exception_code)
+select replaceAll(query, '\n', '\\n'), lower(type::String), errorCodeToName(exception_code)
     from system.query_log
     where current_database = currentDatabase()
-    group by 2, 3
-    order by 2;
+    order by event_time_microseconds
+    format CSV;

From 0b588480f5165af38675911ebba043fe410465db Mon Sep 17 00:00:00 2001
From: avogar <avogar@clickhouse.com>
Date: Tue, 12 Mar 2024 14:34:15 +0000
Subject: [PATCH 212/443] Fix lazy execution in dictGetOrDefault for
 RangeHashedDictionary

---
 src/Functions/FunctionsExternalDictionaries.h |  7 ++--
 .../03009_range_dict_get_or_default.reference |  1 +
 .../03009_range_dict_get_or_default.sql       | 34 +++++++++++++++++++
 3 files changed, 40 insertions(+), 2 deletions(-)
 create mode 100644 tests/queries/0_stateless/03009_range_dict_get_or_default.reference
 create mode 100644 tests/queries/0_stateless/03009_range_dict_get_or_default.sql

diff --git a/src/Functions/FunctionsExternalDictionaries.h b/src/Functions/FunctionsExternalDictionaries.h
index 261c728e9e1..011772baab9 100644
--- a/src/Functions/FunctionsExternalDictionaries.h
+++ b/src/Functions/FunctionsExternalDictionaries.h
@@ -324,12 +324,15 @@ public:
     String getName() const override { return name; }
 
     bool isVariadic() const override { return true; }
-    bool isShortCircuit(ShortCircuitSettings & settings, size_t /*number_of_arguments*/) const override
+    bool isShortCircuit(ShortCircuitSettings & settings, size_t number_of_arguments) const override
     {
         if constexpr (dictionary_get_function_type != DictionaryGetFunctionType::getOrDefault)
             return false;
 
-        settings.arguments_with_disabled_lazy_execution.insert({0, 1, 2});
+        /// We execute lazily only last argument with default expression.
+        for (size_t i = 0; i != number_of_arguments - 1; ++i)
+            settings.arguments_with_disabled_lazy_execution.insert(i);
+
         settings.enable_lazy_execution_for_common_descendants_of_arguments = false;
         settings.force_enable_lazy_execution = false;
         return true;
diff --git a/tests/queries/0_stateless/03009_range_dict_get_or_default.reference b/tests/queries/0_stateless/03009_range_dict_get_or_default.reference
new file mode 100644
index 00000000000..0cfbf08886f
--- /dev/null
+++ b/tests/queries/0_stateless/03009_range_dict_get_or_default.reference
@@ -0,0 +1 @@
+2
diff --git a/tests/queries/0_stateless/03009_range_dict_get_or_default.sql b/tests/queries/0_stateless/03009_range_dict_get_or_default.sql
new file mode 100644
index 00000000000..1f4b4073b9f
--- /dev/null
+++ b/tests/queries/0_stateless/03009_range_dict_get_or_default.sql
@@ -0,0 +1,34 @@
+DROP DICTIONARY IF EXISTS range_dictionary;
+DROP TABLE IF EXISTS range_dictionary_nullable_source_table;
+
+
+CREATE TABLE range_dictionary_nullable_source_table
+(
+  key UInt64,
+  start_date Date,
+  end_date Date,
+  value Nullable(UInt64)
+)
+ENGINE = TinyLog;
+
+INSERT INTO range_dictionary_nullable_source_table VALUES (0, toDate('2019-05-05'), toDate('2019-05-20'), 0), (1, toDate('2019-05-05'), toDate('2019-05-20'), NULL);
+
+CREATE DICTIONARY range_dictionary
+(
+  key UInt64,
+  start_date Date,
+  end_date Date,
+  value Nullable(UInt64) DEFAULT NULL
+)
+PRIMARY KEY key
+SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() TABLE 'range_dictionary_nullable_source_table'))
+LIFETIME(MIN 1 MAX 1000)
+LAYOUT(RANGE_HASHED())
+RANGE(MIN start_date MAX end_date);
+
+SELECT dictGetOrDefault('range_dictionary', 'value', toUInt64(2), toDate(toLowCardinality(materialize('2019-05-15'))), 2);
+
+
+DROP DICTIONARY IF EXISTS range_dictionary;
+DROP TABLE IF EXISTS range_dictionary_nullable_source_table;
+

From 94f15dd5e058fc9f68bf3644c895908bba849db6 Mon Sep 17 00:00:00 2001
From: Mikhail Koviazin <mikhail.koviazin@aiven.io>
Date: Tue, 12 Mar 2024 16:48:43 +0200
Subject: [PATCH 213/443] Fix `has()` function with `Nullable` column

Previous implementation didn't check for `null_map` when the given
argument was not `Null`. This commit adds the missing check.

Fixes #60214
---
 src/Functions/array/arrayIndex.h                       | 9 +++++++--
 tests/queries/0_stateless/00662_has_nullable.reference | 2 ++
 tests/queries/0_stateless/00662_has_nullable.sql       | 7 +++++++
 3 files changed, 16 insertions(+), 2 deletions(-)
 create mode 100644 tests/queries/0_stateless/00662_has_nullable.reference
 create mode 100644 tests/queries/0_stateless/00662_has_nullable.sql

diff --git a/src/Functions/array/arrayIndex.h b/src/Functions/array/arrayIndex.h
index 3b19f0b486a..cd537763b4a 100644
--- a/src/Functions/array/arrayIndex.h
+++ b/src/Functions/array/arrayIndex.h
@@ -1007,8 +1007,13 @@ private:
                         if (!(*null_map)[row])
                             continue;
                     }
-                    else if (!applyVisitor(FieldVisitorAccurateEquals(), arr[i], value))
-                        continue;
+                    else
+                    {
+                        if (null_map && (*null_map)[row])
+                            continue;
+                        if (!applyVisitor(FieldVisitorAccurateEquals(), arr[i], value))
+                            continue;
+                    }
 
                     ConcreteAction::apply(data[row], i);
 
diff --git a/tests/queries/0_stateless/00662_has_nullable.reference b/tests/queries/0_stateless/00662_has_nullable.reference
new file mode 100644
index 00000000000..97da99d655e
--- /dev/null
+++ b/tests/queries/0_stateless/00662_has_nullable.reference
@@ -0,0 +1,2 @@
+1	1
+\N	0
diff --git a/tests/queries/0_stateless/00662_has_nullable.sql b/tests/queries/0_stateless/00662_has_nullable.sql
new file mode 100644
index 00000000000..0d395871d9f
--- /dev/null
+++ b/tests/queries/0_stateless/00662_has_nullable.sql
@@ -0,0 +1,7 @@
+DROP TABLE IF EXISTS 00662_has_nullable;
+CREATE TABLE 00662_has_nullable(a Nullable(UInt64)) ENGINE = Memory;
+
+INSERT INTO 00662_has_nullable VALUES (1), (Null);
+SELECT a, has([0, 1], a) FROM 00662_has_nullable;
+
+DROP TABLE 00662_has_nullable;

From 01d0fba0853930de42d4c122c9f067bd91f9af4b Mon Sep 17 00:00:00 2001
From: yariks5s <yaroslav.briukhovetskyi@clickhouse.com>
Date: Tue, 12 Mar 2024 14:48:59 +0000
Subject: [PATCH 214/443] init

---
 src/Interpreters/InterpreterSelectQuery.cpp | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/src/Interpreters/InterpreterSelectQuery.cpp b/src/Interpreters/InterpreterSelectQuery.cpp
index bcedba7346d..3c84d086d85 100644
--- a/src/Interpreters/InterpreterSelectQuery.cpp
+++ b/src/Interpreters/InterpreterSelectQuery.cpp
@@ -2552,7 +2552,12 @@ void InterpreterSelectQuery::executeFetchColumns(QueryProcessingStage::Enum proc
 
         /// If necessary, we request more sources than the number of threads - to distribute the work evenly over the threads.
         if (max_streams > 1 && !is_sync_remote)
-            max_streams = static_cast<size_t>(max_streams * settings.max_streams_to_max_threads_ratio);
+        {
+            if (auto streams_with_ratio = max_streams * settings.max_streams_to_max_threads_ratio; streams_with_ratio > sizeof(size_t))
+                max_streams = static_cast<size_t>(streams_with_ratio);
+            else
+                throw Exception(ErrorCodes::INCORRECT_DATA, "Exceeded limit for `max_streams_to_max_threads_ratio`. Make sure that `max_streams * max_streams_to_max_threads_ratio` not exceeds {}, current value: {}", sizeof(size_t), streams_with_ratio);
+        }
 
         auto & prewhere_info = analysis_result.prewhere_info;
 

From 5576c170517c10757611868444b4f2aa7d9cf41c Mon Sep 17 00:00:00 2001
From: Peter <peter279k@gmail.com>
Date: Tue, 12 Mar 2024 22:57:06 +0800
Subject: [PATCH 215/443] Using straightforward approach to import GPG key

---
 docs/en/getting-started/install.md | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/docs/en/getting-started/install.md b/docs/en/getting-started/install.md
index 234420de374..f3bacf7d092 100644
--- a/docs/en/getting-started/install.md
+++ b/docs/en/getting-started/install.md
@@ -78,8 +78,8 @@ It is recommended to use official pre-compiled `deb` packages for Debian or Ubun
 
 #### Setup the Debian repository
 ``` bash
-sudo apt-get install -y apt-transport-https ca-certificates dirmngr
-sudo gpg --no-default-keyring --keyring /usr/share/keyrings/clickhouse-keyring.gpg --keyserver hkp://keyserver.ubuntu.com:80 --recv-keys 8919F6BD2B48D754
+sudo apt-get install -y apt-transport-https ca-certificates dirmngr curl gpg
+curl -fsSL 'https://packages.clickhouse.com/rpm/lts/repodata/repomd.xml.key' | sudo gpg --dearmor -o /usr/share/keyrings/clickhouse-keyring.gpg
 
 echo "deb [signed-by=/usr/share/keyrings/clickhouse-keyring.gpg] https://packages.clickhouse.com/deb stable main" | sudo tee \
     /etc/apt/sources.list.d/clickhouse.list

From 24f4cda4845698bc2e014087a46222d135541f36 Mon Sep 17 00:00:00 2001
From: Nikita Taranov <nikita.taranov@clickhouse.com>
Date: Tue, 12 Mar 2024 15:34:07 +0000
Subject: [PATCH 216/443] impl

---
 .../02908_many_requests_to_system_replicas.sh | 32 +++++++++++++------
 1 file changed, 22 insertions(+), 10 deletions(-)

diff --git a/tests/queries/0_stateless/02908_many_requests_to_system_replicas.sh b/tests/queries/0_stateless/02908_many_requests_to_system_replicas.sh
index d3eed891ab9..2c57545e603 100755
--- a/tests/queries/0_stateless/02908_many_requests_to_system_replicas.sh
+++ b/tests/queries/0_stateless/02908_many_requests_to_system_replicas.sh
@@ -12,22 +12,34 @@ CONCURRENCY=200
 
 echo "Creating $NUM_TABLES tables"
 
+function get_done_or_die_trying()
+{
+  # Sometimes curl produces errors like 'Recv failure: Connection reset by peer' and fails test, let's add a little bit of retries
+  for _ in $(seq 1 10)
+  do
+    curl "$CLICKHOUSE_URL" --silent --fail --show-error --data "$1" &>/dev/null && return
+  done
+
+  echo "Cannot successfully make request"
+  exit 1
+}
+
 function init_table()
 {
     set -e
     i=$1
-    curl $CLICKHOUSE_URL --silent --fail --show-error --data "CREATE TABLE test_02908_r1_$i (a UInt64) ENGINE=ReplicatedMergeTree('/02908/{database}/test_$i', 'r1') ORDER BY tuple()" 2>&1
-    curl $CLICKHOUSE_URL --silent --fail --show-error --data "CREATE TABLE test_02908_r2_$i (a UInt64) ENGINE=ReplicatedMergeTree('/02908/{database}/test_$i', 'r2') ORDER BY tuple()" 2>&1
-    curl $CLICKHOUSE_URL --silent --fail --show-error --data "CREATE TABLE test_02908_r3_$i (a UInt64) ENGINE=ReplicatedMergeTree('/02908/{database}/test_$i', 'r3') ORDER BY tuple()" 2>&1
+    get_done_or_die_trying "CREATE TABLE test_02908_r1_$i (a UInt64) ENGINE=ReplicatedMergeTree('/02908/{database}/test_$i', 'r1') ORDER BY tuple()"
+    get_done_or_die_trying "CREATE TABLE test_02908_r2_$i (a UInt64) ENGINE=ReplicatedMergeTree('/02908/{database}/test_$i', 'r2') ORDER BY tuple()"
+    get_done_or_die_trying "CREATE TABLE test_02908_r3_$i (a UInt64) ENGINE=ReplicatedMergeTree('/02908/{database}/test_$i', 'r3') ORDER BY tuple()"
 
-    curl $CLICKHOUSE_URL --silent --fail --show-error --data "INSERT INTO test_02908_r1_$i  SELECT rand64() FROM numbers(5);" 2>&1
+    get_done_or_die_trying "INSERT INTO test_02908_r1_$i  SELECT rand64() FROM numbers(5);"
 }
 
 export init_table;
 
-for i in `seq 1 $NUM_TABLES`;
+for i in $(seq 1 $NUM_TABLES)
 do
-    init_table $i &
+    init_table "$i" &
 done
 
 wait;
@@ -35,15 +47,15 @@ wait;
 
 echo "Making $CONCURRENCY requests to system.replicas"
 
-for i in `seq 1 $CONCURRENCY`;
+for i in $(seq 1 $CONCURRENCY)
 do
-    curl $CLICKHOUSE_URL --silent --fail --show-error --data "SELECT * FROM system.replicas WHERE database=currentDatabase() FORMAT Null;" 2>&1 || echo "query $i failed" &
+    curl "$CLICKHOUSE_URL" --silent --fail --show-error --data "SELECT * FROM system.replicas WHERE database=currentDatabase() FORMAT Null;" 2>&1 || echo "query $i failed" &
 done
 
 echo "Query system.replicas while waiting for other concurrent requests to finish"
 # lost_part_count column is read from ZooKeeper
-curl $CLICKHOUSE_URL --silent --fail --show-error --data "SELECT sum(lost_part_count) FROM system.replicas WHERE database=currentDatabase();" 2>&1;
+curl "$CLICKHOUSE_URL" --silent --fail --show-error --data "SELECT sum(lost_part_count) FROM system.replicas WHERE database=currentDatabase();" 2>&1;
 # is_leader column is filled without ZooKeeper
-curl $CLICKHOUSE_URL --silent --fail --show-error --data "SELECT sum(is_leader) FROM system.replicas WHERE database=currentDatabase();" 2>&1;
+curl "$CLICKHOUSE_URL" --silent --fail --show-error --data "SELECT sum(is_leader) FROM system.replicas WHERE database=currentDatabase();" 2>&1;
 
 wait;

From 07ee777c8d354526cf89c3647863916a0855de49 Mon Sep 17 00:00:00 2001
From: avogar <avogar@clickhouse.com>
Date: Tue, 12 Mar 2024 15:39:04 +0000
Subject: [PATCH 217/443] Restart CI


From 858ad2d68860d7993280b90c0f17ae3a0e84d712 Mon Sep 17 00:00:00 2001
From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com>
Date: Tue, 12 Mar 2024 16:39:10 +0100
Subject: [PATCH 218/443] beautify and change max_value

---
 src/Interpreters/InterpreterSelectQuery.cpp | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/src/Interpreters/InterpreterSelectQuery.cpp b/src/Interpreters/InterpreterSelectQuery.cpp
index 3c84d086d85..a314492c5b0 100644
--- a/src/Interpreters/InterpreterSelectQuery.cpp
+++ b/src/Interpreters/InterpreterSelectQuery.cpp
@@ -2553,10 +2553,10 @@ void InterpreterSelectQuery::executeFetchColumns(QueryProcessingStage::Enum proc
         /// If necessary, we request more sources than the number of threads - to distribute the work evenly over the threads.
         if (max_streams > 1 && !is_sync_remote)
         {
-            if (auto streams_with_ratio = max_streams * settings.max_streams_to_max_threads_ratio; streams_with_ratio > sizeof(size_t))
+            if (auto streams_with_ratio = max_streams * settings.max_streams_to_max_threads_ratio; streams_with_ratio < SIZE_MAX)
                 max_streams = static_cast<size_t>(streams_with_ratio);
             else
-                throw Exception(ErrorCodes::INCORRECT_DATA, "Exceeded limit for `max_streams_to_max_threads_ratio`. Make sure that `max_streams * max_streams_to_max_threads_ratio` not exceeds {}, current value: {}", sizeof(size_t), streams_with_ratio);
+                throw Exception(ErrorCodes::PARAMETER_OUT_OF_BOUND, "Exceeded limit for `max_streams` with `max_streams_to_max_threads_ratio`. Make sure that `max_streams * max_streams_to_max_threads_ratio` not exceeds {}, current value: {}", SIZE_MAX, streams_with_ratio);
         }
 
         auto & prewhere_info = analysis_result.prewhere_info;

From 2fbc11302148f3374e5ce69ed50adebbce9cbae4 Mon Sep 17 00:00:00 2001
From: "Mikhail f. Shiryaev" <mr.felixoid@gmail.com>
Date: Tue, 12 Mar 2024 17:00:44 +0100
Subject: [PATCH 219/443] Adjust the packages in installation docs

---
 docs/en/getting-started/install.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/en/getting-started/install.md b/docs/en/getting-started/install.md
index f3bacf7d092..ca689ef7995 100644
--- a/docs/en/getting-started/install.md
+++ b/docs/en/getting-started/install.md
@@ -78,7 +78,7 @@ It is recommended to use official pre-compiled `deb` packages for Debian or Ubun
 
 #### Setup the Debian repository
 ``` bash
-sudo apt-get install -y apt-transport-https ca-certificates dirmngr curl gpg
+sudo apt-get install -y apt-transport-https ca-certificates curl gnupg
 curl -fsSL 'https://packages.clickhouse.com/rpm/lts/repodata/repomd.xml.key' | sudo gpg --dearmor -o /usr/share/keyrings/clickhouse-keyring.gpg
 
 echo "deb [signed-by=/usr/share/keyrings/clickhouse-keyring.gpg] https://packages.clickhouse.com/deb stable main" | sudo tee \

From e1ea0671b132097182279831560f1bafc07c95ce Mon Sep 17 00:00:00 2001
From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com>
Date: Tue, 12 Mar 2024 17:27:25 +0100
Subject: [PATCH 220/443] Disable test 02998_primary_key_skip_columns.sql in
 sanitizer builds as it can be slow

---
 tests/queries/0_stateless/02998_primary_key_skip_columns.sql | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tests/queries/0_stateless/02998_primary_key_skip_columns.sql b/tests/queries/0_stateless/02998_primary_key_skip_columns.sql
index b2dadcc5e7c..b567155ab1f 100644
--- a/tests/queries/0_stateless/02998_primary_key_skip_columns.sql
+++ b/tests/queries/0_stateless/02998_primary_key_skip_columns.sql
@@ -1,3 +1,5 @@
+-- Tags: no-asan, no-tsan, no-msan, no-ubsan
+
 DROP TABLE IF EXISTS test;
 
 CREATE TABLE test (a UInt64, b UInt64, c UInt64) ENGINE = MergeTree ORDER BY (a, b, c) SETTINGS index_granularity = 1, primary_key_ratio_of_unique_prefix_values_to_skip_suffix_columns = 1;

From ccc6df0e432783e3d128fe1c7b6a22b399f2b510 Mon Sep 17 00:00:00 2001
From: Anton Popov <anton@clickhouse.com>
Date: Tue, 12 Mar 2024 16:43:55 +0000
Subject: [PATCH 221/443] fix aggregate function uniqExact

---
 .../AggregateFunctionUniq.h                   |  2 ++
 .../Combinators/AggregateFunctionArray.h      |  1 +
 .../Combinators/AggregateFunctionIf.h         |  1 +
 .../Combinators/AggregateFunctionMerge.h      |  1 +
 .../Combinators/AggregateFunctionNull.h       |  1 +
 .../Combinators/AggregateFunctionState.h      |  1 +
 src/AggregateFunctions/IAggregateFunction.h   |  4 +++
 src/Common/ColumnsHashingImpl.h               |  3 +-
 src/Common/ProfileEvents.cpp                  |  2 ++
 src/Interpreters/Aggregator.cpp               | 18 ++++++++--
 src/Interpreters/Aggregator.h                 |  1 +
 .../03008_optimize_equal_ranges.reference     |  8 +++++
 .../03008_optimize_equal_ranges.sql           | 25 +++++++++++++
 .../03008_uniq_exact_equal_ranges.reference   |  0
 .../03008_uniq_exact_equal_ranges.sql         | 36 +++++++++++++++++++
 15 files changed, 100 insertions(+), 4 deletions(-)
 create mode 100644 tests/queries/0_stateless/03008_optimize_equal_ranges.reference
 create mode 100644 tests/queries/0_stateless/03008_optimize_equal_ranges.sql
 create mode 100644 tests/queries/0_stateless/03008_uniq_exact_equal_ranges.reference
 create mode 100644 tests/queries/0_stateless/03008_uniq_exact_equal_ranges.sql

diff --git a/src/AggregateFunctions/AggregateFunctionUniq.h b/src/AggregateFunctions/AggregateFunctionUniq.h
index 8ac75e4451c..891f2ac4284 100644
--- a/src/AggregateFunctions/AggregateFunctionUniq.h
+++ b/src/AggregateFunctions/AggregateFunctionUniq.h
@@ -483,6 +483,7 @@ public:
     }
 
     bool isAbleToParallelizeMerge() const override { return is_able_to_parallelize_merge; }
+    bool canOptimizeEqualKeysRanges() const override { return !is_able_to_parallelize_merge; }
 
     void merge(AggregateDataPtr __restrict place, ConstAggregateDataPtr rhs, ThreadPool & thread_pool, Arena *) const override
     {
@@ -576,6 +577,7 @@ public:
     }
 
     bool isAbleToParallelizeMerge() const override { return is_able_to_parallelize_merge; }
+    bool canOptimizeEqualKeysRanges() const override { return !is_able_to_parallelize_merge; }
 
     void merge(AggregateDataPtr __restrict place, ConstAggregateDataPtr rhs, ThreadPool & thread_pool, Arena *) const override
     {
diff --git a/src/AggregateFunctions/Combinators/AggregateFunctionArray.h b/src/AggregateFunctions/Combinators/AggregateFunctionArray.h
index 7f38453f86b..6b918926d0d 100644
--- a/src/AggregateFunctions/Combinators/AggregateFunctionArray.h
+++ b/src/AggregateFunctions/Combinators/AggregateFunctionArray.h
@@ -142,6 +142,7 @@ public:
     }
 
     bool isAbleToParallelizeMerge() const override { return nested_func->isAbleToParallelizeMerge(); }
+    bool canOptimizeEqualKeysRanges() const override { return nested_func->canOptimizeEqualKeysRanges(); }
 
     void merge(AggregateDataPtr __restrict place, ConstAggregateDataPtr rhs, ThreadPool & thread_pool, Arena * arena) const override
     {
diff --git a/src/AggregateFunctions/Combinators/AggregateFunctionIf.h b/src/AggregateFunctions/Combinators/AggregateFunctionIf.h
index e81f2203e7b..df23398a10d 100644
--- a/src/AggregateFunctions/Combinators/AggregateFunctionIf.h
+++ b/src/AggregateFunctions/Combinators/AggregateFunctionIf.h
@@ -165,6 +165,7 @@ public:
     }
 
     bool isAbleToParallelizeMerge() const override { return nested_func->isAbleToParallelizeMerge(); }
+    bool canOptimizeEqualKeysRanges() const override { return nested_func->canOptimizeEqualKeysRanges(); }
 
     void merge(AggregateDataPtr __restrict place, ConstAggregateDataPtr rhs, ThreadPool & thread_pool, Arena * arena) const override
     {
diff --git a/src/AggregateFunctions/Combinators/AggregateFunctionMerge.h b/src/AggregateFunctions/Combinators/AggregateFunctionMerge.h
index 5b9e8e606af..53c24bd60c1 100644
--- a/src/AggregateFunctions/Combinators/AggregateFunctionMerge.h
+++ b/src/AggregateFunctions/Combinators/AggregateFunctionMerge.h
@@ -111,6 +111,7 @@ public:
     }
 
     bool isAbleToParallelizeMerge() const override { return nested_func->isAbleToParallelizeMerge(); }
+    bool canOptimizeEqualKeysRanges() const override { return nested_func->canOptimizeEqualKeysRanges(); }
 
     void merge(AggregateDataPtr __restrict place, ConstAggregateDataPtr rhs, ThreadPool & thread_pool, Arena * arena) const override
     {
diff --git a/src/AggregateFunctions/Combinators/AggregateFunctionNull.h b/src/AggregateFunctions/Combinators/AggregateFunctionNull.h
index 8b614f68540..ba72f960852 100644
--- a/src/AggregateFunctions/Combinators/AggregateFunctionNull.h
+++ b/src/AggregateFunctions/Combinators/AggregateFunctionNull.h
@@ -152,6 +152,7 @@ public:
     }
 
     bool isAbleToParallelizeMerge() const override { return nested_function->isAbleToParallelizeMerge(); }
+    bool canOptimizeEqualKeysRanges() const override { return nested_function->canOptimizeEqualKeysRanges(); }
 
     void merge(AggregateDataPtr __restrict place, ConstAggregateDataPtr rhs, ThreadPool & thread_pool, Arena * arena) const override
     {
diff --git a/src/AggregateFunctions/Combinators/AggregateFunctionState.h b/src/AggregateFunctions/Combinators/AggregateFunctionState.h
index 8335d21cb1e..b0ab6d49604 100644
--- a/src/AggregateFunctions/Combinators/AggregateFunctionState.h
+++ b/src/AggregateFunctions/Combinators/AggregateFunctionState.h
@@ -92,6 +92,7 @@ public:
     }
 
     bool isAbleToParallelizeMerge() const override { return nested_func->isAbleToParallelizeMerge(); }
+    bool canOptimizeEqualKeysRanges() const override { return nested_func->canOptimizeEqualKeysRanges(); }
 
     void merge(AggregateDataPtr __restrict place, ConstAggregateDataPtr rhs, ThreadPool & thread_pool, Arena * arena) const override
     {
diff --git a/src/AggregateFunctions/IAggregateFunction.h b/src/AggregateFunctions/IAggregateFunction.h
index 94bb121893d..499185320e6 100644
--- a/src/AggregateFunctions/IAggregateFunction.h
+++ b/src/AggregateFunctions/IAggregateFunction.h
@@ -162,6 +162,10 @@ public:
     /// Tells if merge() with thread pool parameter could be used.
     virtual bool isAbleToParallelizeMerge() const { return false; }
 
+    /// Return true if it is allowed to replace call of `addBatch`
+    /// to `addBatchSinglePlace` for ranges of consecutive equal keys.
+    virtual bool canOptimizeEqualKeysRanges() const { return true; }
+
     /// Should be used only if isAbleToParallelizeMerge() returned true.
     virtual void
     merge(AggregateDataPtr __restrict /*place*/, ConstAggregateDataPtr /*rhs*/, ThreadPool & /*thread_pool*/, Arena * /*arena*/) const
diff --git a/src/Common/ColumnsHashingImpl.h b/src/Common/ColumnsHashingImpl.h
index 7116160e94c..d68171a6566 100644
--- a/src/Common/ColumnsHashingImpl.h
+++ b/src/Common/ColumnsHashingImpl.h
@@ -62,7 +62,6 @@ struct LastElementCache
     bool check(const Key & key) const { return value.first == key; }
 
     bool hasOnlyOneValue() const { return found && misses == 1; }
-    UInt64 getMisses() const { return misses; }
 };
 
 template <typename Data>
@@ -232,7 +231,7 @@ public:
     ALWAYS_INLINE UInt64 getCacheMissesSinceLastReset() const
     {
         if constexpr (consecutive_keys_optimization)
-            return cache.getMisses();
+            return cache.misses;
         return 0;
     }
 
diff --git a/src/Common/ProfileEvents.cpp b/src/Common/ProfileEvents.cpp
index c1ac3d08245..8fd1e189977 100644
--- a/src/Common/ProfileEvents.cpp
+++ b/src/Common/ProfileEvents.cpp
@@ -533,6 +533,8 @@ The server successfully detected this situation and will download merged part fr
     \
     M(AggregationPreallocatedElementsInHashTables, "How many elements were preallocated in hash tables for aggregation.") \
     M(AggregationHashTablesInitializedAsTwoLevel, "How many hash tables were inited as two-level for aggregation.") \
+    M(AggregationProcessedBlocks, "How many blocks were processed by Aggregator") \
+    M(AggregationOptimizedEqualRangesOfKeys, "For how many blocks optimization of equal ranges of keys was applied") \
     \
     M(MetadataFromKeeperCacheHit, "Number of times an object storage metadata request was answered from cache without making request to Keeper") \
     M(MetadataFromKeeperCacheMiss, "Number of times an object storage metadata request had to be answered from Keeper") \
diff --git a/src/Interpreters/Aggregator.cpp b/src/Interpreters/Aggregator.cpp
index 80a98683867..7c9dac82eff 100644
--- a/src/Interpreters/Aggregator.cpp
+++ b/src/Interpreters/Aggregator.cpp
@@ -53,6 +53,8 @@ namespace ProfileEvents
     extern const Event OverflowThrow;
     extern const Event OverflowBreak;
     extern const Event OverflowAny;
+    extern const Event AggregationProcessedBlocks;
+    extern const Event AggregationOptimizedEqualRangesOfKeys;
 }
 
 namespace CurrentMetrics
@@ -985,6 +987,7 @@ void Aggregator::executeOnBlockSmall(
 {
     /// `result` will destroy the states of aggregate functions in the destructor
     result.aggregator = this;
+    ProfileEvents::increment(ProfileEvents::AggregationProcessedBlocks);
 
     /// How to perform the aggregation?
     if (result.empty())
@@ -1342,6 +1345,7 @@ void NO_INLINE Aggregator::executeImplBatch(
     if constexpr (use_compiled_functions)
     {
         std::vector<ColumnData> columns_data;
+        bool can_optimize_equal_keys_ranges = true;
 
         for (size_t i = 0; i < aggregate_functions.size(); ++i)
         {
@@ -1350,13 +1354,15 @@ void NO_INLINE Aggregator::executeImplBatch(
 
             AggregateFunctionInstruction * inst = aggregate_instructions + i;
             size_t arguments_size = inst->that->getArgumentTypes().size(); // NOLINT
+            can_optimize_equal_keys_ranges &= inst->can_optimize_equal_keys_ranges;
 
             for (size_t argument_index = 0; argument_index < arguments_size; ++argument_index)
                 columns_data.emplace_back(getColumnData(inst->batch_arguments[argument_index]));
         }
 
-        if (all_keys_are_const || (!no_more_keys && state.hasOnlyOneValueSinceLastReset()))
+        if (all_keys_are_const || (can_optimize_equal_keys_ranges && state.hasOnlyOneValueSinceLastReset()))
         {
+            ProfileEvents::increment(ProfileEvents::AggregationOptimizedEqualRangesOfKeys);
             auto add_into_aggregate_states_function_single_place = compiled_aggregate_functions_holder->compiled_aggregate_functions.add_into_aggregate_states_function_single_place;
             add_into_aggregate_states_function_single_place(row_begin, row_end, columns_data.data(), places[key_start]);
         }
@@ -1379,10 +1385,15 @@ void NO_INLINE Aggregator::executeImplBatch(
 
         AggregateFunctionInstruction * inst = aggregate_instructions + i;
 
-        if (all_keys_are_const || (!no_more_keys && state.hasOnlyOneValueSinceLastReset()))
+        if (all_keys_are_const || (inst->can_optimize_equal_keys_ranges && state.hasOnlyOneValueSinceLastReset()))
+        {
+            ProfileEvents::increment(ProfileEvents::AggregationOptimizedEqualRangesOfKeys);
             addBatchSinglePlace(row_begin, row_end, inst, places[key_start] + inst->state_offset, aggregates_pool);
+        }
         else
+        {
             addBatch(row_begin, row_end, inst, places.get(), aggregates_pool);
+        }
     }
 }
 
@@ -1510,6 +1521,7 @@ void NO_INLINE Aggregator::executeOnIntervalWithoutKey(
     /// `data_variants` will destroy the states of aggregate functions in the destructor
     data_variants.aggregator = this;
     data_variants.init(AggregatedDataVariants::Type::without_key);
+    ProfileEvents::increment(ProfileEvents::AggregationProcessedBlocks);
 
     AggregatedDataWithoutKey & res = data_variants.without_key;
 
@@ -1588,6 +1600,7 @@ void Aggregator::prepareAggregateInstructions(
         }
 
         aggregate_functions_instructions[i].has_sparse_arguments = has_sparse_arguments;
+        aggregate_functions_instructions[i].can_optimize_equal_keys_ranges = aggregate_functions[i]->canOptimizeEqualKeysRanges();
         aggregate_functions_instructions[i].arguments = aggregate_columns[i].data();
         aggregate_functions_instructions[i].state_offset = offsets_of_aggregate_states[i];
 
@@ -1640,6 +1653,7 @@ bool Aggregator::executeOnBlock(Columns columns,
 {
     /// `result` will destroy the states of aggregate functions in the destructor
     result.aggregator = this;
+    ProfileEvents::increment(ProfileEvents::AggregationProcessedBlocks);
 
     /// How to perform the aggregation?
     if (result.empty())
diff --git a/src/Interpreters/Aggregator.h b/src/Interpreters/Aggregator.h
index 375b8986101..2d3f497fec0 100644
--- a/src/Interpreters/Aggregator.h
+++ b/src/Interpreters/Aggregator.h
@@ -1221,6 +1221,7 @@ public:
         const IColumn ** batch_arguments{};
         const UInt64 * offsets{};
         bool has_sparse_arguments = false;
+        bool can_optimize_equal_keys_ranges = true;
     };
 
     /// Used for optimize_aggregation_in_order:
diff --git a/tests/queries/0_stateless/03008_optimize_equal_ranges.reference b/tests/queries/0_stateless/03008_optimize_equal_ranges.reference
new file mode 100644
index 00000000000..08f8008fca6
--- /dev/null
+++ b/tests/queries/0_stateless/03008_optimize_equal_ranges.reference
@@ -0,0 +1,8 @@
+0	30000
+1	30000
+2	30000
+0	449985000
+1	449985000
+2	449985000
+sum	1
+uniqExact	0
diff --git a/tests/queries/0_stateless/03008_optimize_equal_ranges.sql b/tests/queries/0_stateless/03008_optimize_equal_ranges.sql
new file mode 100644
index 00000000000..c6143fb7f51
--- /dev/null
+++ b/tests/queries/0_stateless/03008_optimize_equal_ranges.sql
@@ -0,0 +1,25 @@
+DROP TABLE IF EXISTS t_optimize_equal_ranges;
+
+CREATE TABLE t_optimize_equal_ranges (a UInt64, b String, c UInt64) ENGINE = MergeTree ORDER BY a;
+
+SET max_block_size = 1024;
+SET max_bytes_before_external_group_by = 0;
+SET optimize_aggregation_in_order = 0;
+
+INSERT INTO t_optimize_equal_ranges SELECT 0, toString(number), number FROM numbers(30000);
+INSERT INTO t_optimize_equal_ranges SELECT 1, toString(number), number FROM numbers(30000);
+INSERT INTO t_optimize_equal_ranges SELECT 2, toString(number), number FROM numbers(30000);
+
+SELECT a, uniqExact(b) FROM t_optimize_equal_ranges GROUP BY a ORDER BY a;
+SELECT a, sum(c) FROM t_optimize_equal_ranges GROUP BY a ORDER BY a;
+
+SYSTEM FLUSH LOGS;
+
+SELECT
+    used_aggregate_functions[1] AS func,
+    ProfileEvents['AggregationOptimizedEqualRangesOfKeys'] > 0
+FROM system.query_log
+WHERE type = 'QueryFinish' AND current_database = currentDatabase() AND query LIKE '%SELECT%FROM%t_optimize_equal_ranges%'
+ORDER BY func;
+
+DROP TABLE t_optimize_equal_ranges;
diff --git a/tests/queries/0_stateless/03008_uniq_exact_equal_ranges.reference b/tests/queries/0_stateless/03008_uniq_exact_equal_ranges.reference
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/queries/0_stateless/03008_uniq_exact_equal_ranges.sql b/tests/queries/0_stateless/03008_uniq_exact_equal_ranges.sql
new file mode 100644
index 00000000000..2e708f28cac
--- /dev/null
+++ b/tests/queries/0_stateless/03008_uniq_exact_equal_ranges.sql
@@ -0,0 +1,36 @@
+DROP TABLE IF EXISTS t_uniq_exact;
+
+CREATE TABLE t_uniq_exact (a UInt64, b String, c UInt64) ENGINE = MergeTree ORDER BY a;
+
+SET group_by_two_level_threshold_bytes = 1;
+SET group_by_two_level_threshold = 1;
+SET max_threads = 4;
+SET max_bytes_before_external_group_by = 0;
+SET optimize_aggregation_in_order = 0;
+
+INSERT INTO t_uniq_exact SELECT 0, randomPrintableASCII(5), rand() FROM numbers(300000);
+INSERT INTO t_uniq_exact SELECT 1, randomPrintableASCII(5), rand() FROM numbers(300000);
+INSERT INTO t_uniq_exact SELECT 2, randomPrintableASCII(5), rand() FROM numbers(300000);
+INSERT INTO t_uniq_exact SELECT 3, randomPrintableASCII(5), rand() FROM numbers(300000);
+INSERT INTO t_uniq_exact SELECT 4, randomPrintableASCII(5), rand() FROM numbers(300000);
+INSERT INTO t_uniq_exact SELECT 5, randomPrintableASCII(5), rand() FROM numbers(300000);
+INSERT INTO t_uniq_exact SELECT 6, randomPrintableASCII(5), rand() FROM numbers(300000);
+INSERT INTO t_uniq_exact SELECT 7, randomPrintableASCII(5), rand() FROM numbers(300000);
+INSERT INTO t_uniq_exact SELECT 8, randomPrintableASCII(5), rand() FROM numbers(300000);
+INSERT INTO t_uniq_exact SELECT 9, randomPrintableASCII(5), rand() FROM numbers(300000);
+
+OPTIMIZE TABLE t_uniq_exact FINAL;
+
+SELECT a, uniqExact(b) FROM t_uniq_exact GROUP BY a ORDER BY a
+SETTINGS min_hit_rate_to_use_consecutive_keys_optimization = 1.0
+EXCEPT
+SELECT a, uniqExact(b) FROM t_uniq_exact GROUP BY a ORDER BY a
+SETTINGS min_hit_rate_to_use_consecutive_keys_optimization = 0.5;
+
+SELECT a, sum(c) FROM t_uniq_exact GROUP BY a ORDER BY a
+SETTINGS min_hit_rate_to_use_consecutive_keys_optimization = 1.0
+EXCEPT
+SELECT a, sum(c) FROM t_uniq_exact GROUP BY a ORDER BY a
+SETTINGS min_hit_rate_to_use_consecutive_keys_optimization = 0.5;
+
+DROP TABLE t_uniq_exact;

From 53442f49140c9c8b391f5b3a76a9bf4ab45dbed0 Mon Sep 17 00:00:00 2001
From: johnnymatthews <9611008+johnnymatthews@users.noreply.github.com>
Date: Tue, 12 Mar 2024 13:48:47 -0300
Subject: [PATCH 222/443] Cleans up markdown.

---
 docs/en/sql-reference/aggregate-functions/reference/varpop.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/en/sql-reference/aggregate-functions/reference/varpop.md b/docs/en/sql-reference/aggregate-functions/reference/varpop.md
index 5f18bdc30f6..76472f62789 100644
--- a/docs/en/sql-reference/aggregate-functions/reference/varpop.md
+++ b/docs/en/sql-reference/aggregate-functions/reference/varpop.md
@@ -27,7 +27,7 @@ Returns an integer of type `Float64`.
 
 **Implementation details**
 
-This function uses a numerically unstable algorithm. If you need numerical stability in calculations, use the slower but more stable [`varPopStable` function](#varPopStable). 
+This function uses a numerically unstable algorithm. If you need numerical stability in calculations, use the slower but more stable [`varPopStable` function](#varPopStable).
 
 **Example**
 

From b3cfb8a2cba2f779e63cb35aec49fd263a24eb19 Mon Sep 17 00:00:00 2001
From: johnnymatthews <9611008+johnnymatthews@users.noreply.github.com>
Date: Tue, 12 Mar 2024 13:57:03 -0300
Subject: [PATCH 223/443] Adds sleep function docs.

---
 .../functions/other-functions.md              | 92 ++++++++++++++++++-
 1 file changed, 89 insertions(+), 3 deletions(-)

diff --git a/docs/en/sql-reference/functions/other-functions.md b/docs/en/sql-reference/functions/other-functions.md
index 739b688a0d2..de8ac67a936 100644
--- a/docs/en/sql-reference/functions/other-functions.md
+++ b/docs/en/sql-reference/functions/other-functions.md
@@ -300,11 +300,97 @@ The argument is internally still evaluated. Useful e.g. for benchmarks.
 
 ## sleep(seconds)
 
-Sleeps ‘seconds’ seconds for each data block. The sleep time can be specified as integer or as floating-point number.
+Used to introduce a delay or pause in the execution of a query. It is primarily used for testing and debugging purposes.
 
-## sleepEachRow(seconds)
+**Syntax**
 
-Sleeps ‘seconds’ seconds for each row. The sleep time can be specified as integer or as floating-point number.
+```sql
+sleep(seconds)
+```
+
+**Arguments**
+
+- `seconds`: [Int](../../sql-reference/data-types/int-uint.md) The number of seconds to pause the query execution to a maximum of 3 seconds. It can be a floating-point value to specify fractional seconds.
+
+**Returned value**
+
+This function does not return any value.
+
+**Example**
+
+```sql
+SELECT sleep(2);
+```
+
+This function does not return any value. However, if you run the function with `clickhouse client` you will see something similar to:
+
+```response
+SELECT sleep(2)
+
+Query id: 8aa9943e-a686-45e1-8317-6e8e3a5596ac
+
+┌─sleep(2)─┐
+│        0 │
+└──────────┘
+
+1 row in set. Elapsed: 2.012 sec.
+```
+
+This query will pause for 2 seconds before completing. During this time, no results will be returned, and the query will appear to be hanging or unresponsive.
+
+**Implementation details**
+
+The `sleep()` function is generally not used in production environments, as it can negatively impact query performance and system responsiveness. However, it can be useful in the following scenarios:
+
+1. **Testing**: When testing or benchmarking ClickHouse, you may want to simulate delays or introduce pauses to observe how the system behaves under certain conditions.
+2. **Debugging**: If you need to examine the state of the system or the execution of a query at a specific point in time, you can use `sleep()` to introduce a pause, allowing you to inspect or collect relevant information.
+3. **Simulation**: In some cases, you may want to simulate real-world scenarios where delays or pauses occur, such as network latency or external system dependencies.
+
+It's important to use the `sleep()` function judiciously and only when necessary, as it can potentially impact the overall performance and responsiveness of your ClickHouse system.
+
+## sleepEachRow
+
+Pauses the execution of a query for a specified number of seconds for each row in the result set.
+
+**Syntax**
+
+```sql
+sleepEachRow(seconds)
+```
+
+**Arguments**
+
+- `seconds`: [Int](../../sql-reference/data-types/int-uint.md) The number of seconds to pause the query execution for each row in the result set. It can be a floating-point value to specify fractional seconds.
+
+**Returned value**
+
+This function returns the same input values as it receives, without modifying them.
+
+**Example**
+
+```sql
+SELECT number, sleepEachRow(0.5) FROM system.numbers LIMIT 5;
+```
+
+```response
+┌─number─┬─sleepEachRow(0.5)─┐
+│      0 │                 0 │
+│      1 │                 0 │
+│      2 │                 0 │
+│      3 │                 0 │
+│      4 │                 0 │
+└────────┴───────────────────┘
+```
+
+But the output will be delayed, with a 0.5-second pause between each row.
+
+The `sleepEachRow()` function is primarily used for testing and debugging purposes, similar to the `sleep()` function. It allows you to simulate delays or introduce pauses in the processing of each row, which can be useful in scenarios such as:
+
+1. **Testing**: When testing or benchmarking ClickHouse's performance under specific conditions, you can use `sleepEachRow()` to simulate delays or introduce pauses for each row processed.
+2. **Debugging**: If you need to examine the state of the system or the execution of a query for each row processed, you can use `sleepEachRow()` to introduce pauses, allowing you to inspect or collect relevant information.
+3. **Simulation**: In some cases, you may want to simulate real-world scenarios where delays or pauses occur for each row processed, such as when dealing with external systems or network latencies.
+
+Like the [`sleep()` function](#sleep), it's important to use `sleepEachRow()` judiciously and only when necessary, as it can significantly impact the overall performance and responsiveness of your ClickHouse system, especially when dealing with large result sets.
 
 ## currentDatabase()
 

From 2847def5176509bdbefe0809c2abbe31baf540af Mon Sep 17 00:00:00 2001
From: johnnymatthews <9611008+johnnymatthews@users.noreply.github.com>
Date: Tue, 12 Mar 2024 13:57:34 -0300
Subject: [PATCH 224/443] Formats markdown with prettier.

---
 .../functions/other-functions.md              | 408 +++++++++---------
 1 file changed, 204 insertions(+), 204 deletions(-)

diff --git a/docs/en/sql-reference/functions/other-functions.md b/docs/en/sql-reference/functions/other-functions.md
index de8ac67a936..df443eec0de 100644
--- a/docs/en/sql-reference/functions/other-functions.md
+++ b/docs/en/sql-reference/functions/other-functions.md
@@ -17,7 +17,7 @@ Returns a named value from the [macros](../../operations/server-configuration-pa
 
 **Syntax**
 
-``` sql
+```sql
 getMacro(name);
 ```
 
@@ -35,7 +35,7 @@ Type: [String](../../sql-reference/data-types/string.md).
 
 Example `<macros>` section in the server configuration file:
 
-``` xml
+```xml
 <macros>
     <test>Value</test>
 </macros>
@@ -43,13 +43,13 @@ Example `<macros>` section in the server configuration file:
 
 Query:
 
-``` sql
+```sql
 SELECT getMacro('test');
 ```
 
 Result:
 
-``` text
+```text
 ┌─getMacro('test')─┐
 │ Value            │
 └──────────────────┘
@@ -57,12 +57,12 @@ Result:
 
 The same value can be retrieved as follows:
 
-``` sql
+```sql
 SELECT * FROM system.macros
 WHERE macro = 'test';
 ```
 
-``` text
+```text
 ┌─macro─┬─substitution─┐
 │ test  │ Value        │
 └───────┴──────────────┘
@@ -74,7 +74,7 @@ Returns the fully qualified domain name of the ClickHouse server.
 
 **Syntax**
 
-``` sql
+```sql
 fqdn();
 ```
 
@@ -88,13 +88,13 @@ Type: `String`.
 
 **Example**
 
-``` sql
+```sql
 SELECT FQDN();
 ```
 
 Result:
 
-``` text
+```text
 ┌─FQDN()──────────────────────────┐
 │ clickhouse.ru-central1.internal │
 └─────────────────────────────────┘
@@ -104,7 +104,7 @@ Result:
 
 Extracts the tail of a string following its last slash or backslash. This function if often used to extract the filename from a path.
 
-``` sql
+```sql
 basename(expr)
 ```
 
@@ -123,13 +123,13 @@ A string that contains:
 
 Query:
 
-``` sql
+```sql
 SELECT 'some/long/path/to/file' AS a, basename(a)
 ```
 
 Result:
 
-``` text
+```text
 ┌─a──────────────────────┬─basename('some\\long\\path\\to\\file')─┐
 │ some\long\path\to\file │ file                                   │
 └────────────────────────┴────────────────────────────────────────┘
@@ -137,13 +137,13 @@ Result:
 
 Query:
 
-``` sql
+```sql
 SELECT 'some\\long\\path\\to\\file' AS a, basename(a)
 ```
 
 Result:
 
-``` text
+```text
 ┌─a──────────────────────┬─basename('some\\long\\path\\to\\file')─┐
 │ some\long\path\to\file │ file                                   │
 └────────────────────────┴────────────────────────────────────────┘
@@ -151,13 +151,13 @@ Result:
 
 Query:
 
-``` sql
+```sql
 SELECT 'some-file-name' AS a, basename(a)
 ```
 
 Result:
 
-``` text
+```text
 ┌─a──────────────┬─basename('some-file-name')─┐
 │ some-file-name │ some-file-name             │
 └────────────────┴────────────────────────────┘
@@ -170,11 +170,11 @@ This function is used by the system to implement Pretty formats.
 
 `NULL` is represented as a string corresponding to `NULL` in `Pretty` formats.
 
-``` sql
+```sql
 SELECT visibleWidth(NULL)
 ```
 
-``` text
+```text
 ┌─visibleWidth(NULL)─┐
 │                  4 │
 └────────────────────┘
@@ -256,7 +256,7 @@ SELECT key, byteSize(u8) AS `byteSize(UInt8)`, byteSize(u16) AS `byteSize(UInt16
 
 Result:
 
-``` text
+```text
 Row 1:
 ──────
 key:               1
@@ -401,7 +401,7 @@ Useful in table engine parameters of `CREATE TABLE` queries where you need to sp
 
 Returns the name of the current user. In case of a distributed query, the name of the user who initiated the query is returned.
 
-``` sql
+```sql
 SELECT currentUser();
 ```
 
@@ -416,13 +416,13 @@ Type: `String`.
 
 **Example**
 
-``` sql
+```sql
 SELECT currentUser();
 ```
 
 Result:
 
-``` text
+```text
 ┌─currentUser()─┐
 │ default       │
 └───────────────┘
@@ -438,7 +438,7 @@ This function is mostly intended for development, debugging and demonstration.
 
 **Syntax**
 
-``` sql
+```sql
 isConstant(x)
 ```
 
@@ -457,13 +457,13 @@ Type: [UInt8](../../sql-reference/data-types/int-uint.md).
 
 Query:
 
-``` sql
+```sql
 SELECT isConstant(x + 1) FROM (SELECT 43 AS x)
 ```
 
 Result:
 
-``` text
+```text
 ┌─isConstant(plus(x, 1))─┐
 │                      1 │
 └────────────────────────┘
@@ -471,13 +471,13 @@ Result:
 
 Query:
 
-``` sql
+```sql
 WITH 3.14 AS pi SELECT isConstant(cos(pi))
 ```
 
 Result:
 
-``` text
+```text
 ┌─isConstant(cos(pi))─┐
 │                   1 │
 └─────────────────────┘
@@ -485,13 +485,13 @@ Result:
 
 Query:
 
-``` sql
+```sql
 SELECT isConstant(number) FROM numbers(1)
 ```
 
 Result:
 
-``` text
+```text
 ┌─isConstant(number)─┐
 │                  0 │
 └────────────────────┘
@@ -511,7 +511,7 @@ Checks whether a floating point value is finite.
 
 **Syntax**
 
-``` sql
+```sql
 ifNotFinite(x,y)
 ```
 
@@ -565,7 +565,7 @@ The band is drawn with accuracy to one eighth of a symbol.
 
 Example:
 
-``` sql
+```sql
 SELECT
     toHour(EventTime) AS h,
     count() AS c,
@@ -575,7 +575,7 @@ GROUP BY h
 ORDER BY h ASC
 ```
 
-``` text
+```text
 ┌──h─┬──────c─┬─bar────────────────┐
 │  0 │ 292907 │ █████████▋         │
 │  1 │ 180563 │ ██████             │
@@ -633,7 +633,7 @@ For example, the first argument could have type `Int64`, while the second argume
 
 Example:
 
-``` sql
+```sql
 SELECT
     transform(SearchEngineID, [2, 3], ['Yandex', 'Google'], 'Other') AS title,
     count() AS c
@@ -643,7 +643,7 @@ GROUP BY title
 ORDER BY c DESC
 ```
 
-``` text
+```text
 ┌─title─────┬──────c─┐
 │ Yandex    │ 498635 │
 │ Google    │ 229872 │
@@ -657,7 +657,7 @@ Similar to the other variation but has no ‘default’ argument. In case no mat
 
 Example:
 
-``` sql
+```sql
 SELECT
     transform(domain(Referer), ['yandex.ru', 'google.ru', 'vkontakte.ru'], ['www.yandex', 'example.com', 'vk.com']) AS s,
     count() AS c
@@ -667,7 +667,7 @@ ORDER BY count() DESC
 LIMIT 10
 ```
 
-``` text
+```text
 ┌─s──────────────┬───────c─┐
 │                │ 2906259 │
 │ www.yandex     │  867767 │
@@ -687,13 +687,13 @@ Given a size (number of bytes), this function returns a readable, rounded size w
 
 Example:
 
-``` sql
+```sql
 SELECT
     arrayJoin([1, 1024, 1024*1024, 192851925]) AS filesize_bytes,
     formatReadableDecimalSize(filesize_bytes) AS filesize
 ```
 
-``` text
+```text
 ┌─filesize_bytes─┬─filesize───┐
 │              1 │ 1.00 B     │
 │           1024 │ 1.02 KB   │
@@ -708,7 +708,7 @@ Given a size (number of bytes), this function returns a readable, rounded size w
 
 Example:
 
-``` sql
+```sql
 SELECT
     arrayJoin([1, 1024, 1024*1024, 192851925]) AS filesize_bytes,
     formatReadableSize(filesize_bytes) AS filesize
@@ -716,7 +716,7 @@ SELECT
 
 Alias: `FORMAT_BYTES`.
 
-``` text
+```text
 ┌─filesize_bytes─┬─filesize───┐
 │              1 │ 1.00 B     │
 │           1024 │ 1.00 KiB   │
@@ -731,13 +731,13 @@ Given a number, this function returns a rounded number with suffix (thousand, mi
 
 Example:
 
-``` sql
+```sql
 SELECT
     arrayJoin([1024, 1234 * 1000, (4567 * 1000) * 1000, 98765432101234]) AS number,
     formatReadableQuantity(number) AS number_for_humans
 ```
 
-``` text
+```text
 ┌─────────number─┬─number_for_humans─┐
 │           1024 │ 1.02 thousand     │
 │        1234000 │ 1.23 million      │
@@ -752,7 +752,7 @@ Given a time interval (delta) in seconds, this function returns a time delta wit
 
 **Syntax**
 
-``` sql
+```sql
 formatReadableTimeDelta(column[, maximum_unit, minimum_unit])
 ```
 
@@ -760,21 +760,22 @@ formatReadableTimeDelta(column[, maximum_unit, minimum_unit])
 
 - `column` — A column with a numeric time delta.
 - `maximum_unit` — Optional. Maximum unit to show.
-  * Acceptable values: `nanoseconds`, `microseconds`, `milliseconds`, `seconds`, `minutes`, `hours`, `days`, `months`, `years`.
-  * Default value: `years`.
+  - Acceptable values: `nanoseconds`, `microseconds`, `milliseconds`, `seconds`, `minutes`, `hours`, `days`, `months`, `years`.
+  - Default value: `years`.
 - `minimum_unit` — Optional. Minimum unit to show. All smaller units are truncated.
-  * Acceptable values: `nanoseconds`, `microseconds`, `milliseconds`, `seconds`, `minutes`, `hours`, `days`, `months`, `years`.
-  * If explicitly specified value is bigger than `maximum_unit`, an exception will be thrown.
-  * Default value: `seconds` if `maximum_unit` is `seconds` or bigger, `nanoseconds` otherwise.
+  - Acceptable values: `nanoseconds`, `microseconds`, `milliseconds`, `seconds`, `minutes`, `hours`, `days`, `months`, `years`.
+  - If explicitly specified value is bigger than `maximum_unit`, an exception will be thrown.
+  - Default value: `seconds` if `maximum_unit` is `seconds` or bigger, `nanoseconds` otherwise.
 
 **Example**
-``` sql
+
+```sql
 SELECT
     arrayJoin([100, 12345, 432546534]) AS elapsed,
     formatReadableTimeDelta(elapsed) AS time_delta
 ```
 
-``` text
+```text
 ┌────elapsed─┬─time_delta ─────────────────────────────────────────────────────┐
 │        100 │ 1 minute and 40 seconds                                         │
 │      12345 │ 3 hours, 25 minutes and 45 seconds                              │
@@ -782,13 +783,13 @@ SELECT
 └────────────┴─────────────────────────────────────────────────────────────────┘
 ```
 
-``` sql
+```sql
 SELECT
     arrayJoin([100, 12345, 432546534]) AS elapsed,
     formatReadableTimeDelta(elapsed, 'minutes') AS time_delta
 ```
 
-``` text
+```text
 ┌────elapsed─┬─time_delta ─────────────────────────────────────────────────────┐
 │        100 │ 1 minute and 40 seconds                                         │
 │      12345 │ 205 minutes and 45 seconds                                      │
@@ -824,7 +825,6 @@ parseTimeDelta(timestr)
 
 - `timestr` — A sequence of numbers followed by something resembling a time unit.
 
-
 **Returned value**
 
 - A floating-point number with the number of seconds.
@@ -936,7 +936,7 @@ The window function that provides access to a row at a specified offset before o
 
 **Syntax**
 
-``` sql
+```sql
 neighbor(column, offset[, default_value])
 ```
 
@@ -966,13 +966,13 @@ Type: type of data blocks affected or default value type.
 
 Query:
 
-``` sql
+```sql
 SELECT number, neighbor(number, 2) FROM system.numbers LIMIT 10;
 ```
 
 Result:
 
-``` text
+```text
 ┌─number─┬─neighbor(number, 2)─┐
 │      0 │                   2 │
 │      1 │                   3 │
@@ -989,13 +989,13 @@ Result:
 
 Query:
 
-``` sql
+```sql
 SELECT number, neighbor(number, 2, 999) FROM system.numbers LIMIT 10;
 ```
 
 Result:
 
-``` text
+```text
 ┌─number─┬─neighbor(number, 2, 999)─┐
 │      0 │                        2 │
 │      1 │                        3 │
@@ -1014,7 +1014,7 @@ This function can be used to compute year-over-year metric value:
 
 Query:
 
-``` sql
+```sql
 WITH toDate('2018-01-01') AS start_date
 SELECT
     toStartOfMonth(start_date + (number * 32)) AS month,
@@ -1026,7 +1026,7 @@ FROM numbers(16)
 
 Result:
 
-``` text
+```text
 ┌──────month─┬─money─┬─prev_year─┬─year_over_year─┐
 │ 2018-01-01 │    32 │         0 │              0 │
 │ 2018-02-01 │    63 │         0 │              0 │
@@ -1063,7 +1063,7 @@ To prevent that you can create a subquery with [ORDER BY](../../sql-reference/st
 
 Example:
 
-``` sql
+```sql
 SELECT
     EventID,
     EventTime,
@@ -1080,7 +1080,7 @@ FROM
 )
 ```
 
-``` text
+```text
 ┌─EventID─┬───────────EventTime─┬─delta─┐
 │    1106 │ 2016-11-24 00:00:04 │     0 │
 │    1107 │ 2016-11-24 00:00:05 │     1 │
@@ -1092,7 +1092,7 @@ FROM
 
 Please note that the block size affects the result. The internal state of `runningDifference` state is reset for each new block.
 
-``` sql
+```sql
 SELECT
     number,
     runningDifference(number + 1) AS diff
@@ -1100,7 +1100,7 @@ FROM numbers(100000)
 WHERE diff != 1
 ```
 
-``` text
+```text
 ┌─number─┬─diff─┐
 │      0 │    0 │
 └────────┴──────┘
@@ -1109,7 +1109,7 @@ WHERE diff != 1
 └────────┴──────┘
 ```
 
-``` sql
+```sql
 set max_block_size=100000 -- default value is 65536!
 
 SELECT
@@ -1119,7 +1119,7 @@ FROM numbers(100000)
 WHERE diff != 1
 ```
 
-``` text
+```text
 ┌─number─┬─diff─┐
 │      0 │    0 │
 └────────┴──────┘
@@ -1135,21 +1135,20 @@ Calculates the number of concurrent events.
 Each event has a start time and an end time. The start time is included in the event, while the end time is excluded. Columns with a start time and an end time must be of the same data type.
 The function calculates the total number of active (concurrent) events for each event start time.
 
-
 :::tip
 Events must be ordered by the start time in ascending order. If this requirement is violated the function raises an exception. Every data block is processed separately. If events from different data blocks overlap then they can not be processed correctly.
 :::
 
 **Syntax**
 
-``` sql
+```sql
 runningConcurrency(start, end)
 ```
 
 **Arguments**
 
 - `start` — A column with the start time of events. [Date](../../sql-reference/data-types/date.md), [DateTime](../../sql-reference/data-types/datetime.md), or [DateTime64](../../sql-reference/data-types/datetime64.md).
-- `end` — A column with the end time of events.  [Date](../../sql-reference/data-types/date.md), [DateTime](../../sql-reference/data-types/datetime.md), or [DateTime64](../../sql-reference/data-types/datetime64.md).
+- `end` — A column with the end time of events. [Date](../../sql-reference/data-types/date.md), [DateTime](../../sql-reference/data-types/datetime.md), or [DateTime64](../../sql-reference/data-types/datetime64.md).
 
 **Returned values**
 
@@ -1161,7 +1160,7 @@ Type: [UInt32](../../sql-reference/data-types/int-uint.md)
 
 Consider the table:
 
-``` text
+```text
 ┌──────start─┬────────end─┐
 │ 2021-03-03 │ 2021-03-11 │
 │ 2021-03-06 │ 2021-03-12 │
@@ -1172,13 +1171,13 @@ Consider the table:
 
 Query:
 
-``` sql
+```sql
 SELECT start, runningConcurrency(start, end) FROM example_table;
 ```
 
 Result:
 
-``` text
+```text
 ┌──────start─┬─runningConcurrency(start, end)─┐
 │ 2021-03-03 │                              1 │
 │ 2021-03-06 │                              2 │
@@ -1204,7 +1203,7 @@ Given a MAC address in format AA:BB:CC:DD:EE:FF (colon-separated numbers in hexa
 Returns the number of fields in [Enum](../../sql-reference/data-types/enum.md).
 An exception is thrown if the type is not `Enum`.
 
-``` sql
+```sql
 getSizeOfEnumType(value)
 ```
 
@@ -1218,11 +1217,11 @@ getSizeOfEnumType(value)
 
 **Example**
 
-``` sql
+```sql
 SELECT getSizeOfEnumType( CAST('a' AS Enum8('a' = 1, 'b' = 2) ) ) AS x
 ```
 
-``` text
+```text
 ┌─x─┐
 │ 2 │
 └───┘
@@ -1232,7 +1231,7 @@ SELECT getSizeOfEnumType( CAST('a' AS Enum8('a' = 1, 'b' = 2) ) ) AS x
 
 Returns the size on disk without considering compression.
 
-``` sql
+```sql
 blockSerializedSize(value[, value[, ...]])
 ```
 
@@ -1248,13 +1247,13 @@ blockSerializedSize(value[, value[, ...]])
 
 Query:
 
-``` sql
+```sql
 SELECT blockSerializedSize(maxState(1)) as x
 ```
 
 Result:
 
-``` text
+```text
 ┌─x─┐
 │ 2 │
 └───┘
@@ -1264,7 +1263,7 @@ Result:
 
 Returns the internal name of the data type that represents the value.
 
-``` sql
+```sql
 toColumnTypeName(value)
 ```
 
@@ -1280,13 +1279,13 @@ toColumnTypeName(value)
 
 Difference between `toTypeName ' and ' toColumnTypeName`:
 
-``` sql
+```sql
 SELECT toTypeName(CAST('2018-01-01 01:02:03' AS DateTime))
 ```
 
 Result:
 
-``` text
+```text
 ┌─toTypeName(CAST('2018-01-01 01:02:03', 'DateTime'))─┐
 │ DateTime                                            │
 └─────────────────────────────────────────────────────┘
@@ -1294,13 +1293,13 @@ Result:
 
 Query:
 
-``` sql
+```sql
 SELECT toColumnTypeName(CAST('2018-01-01 01:02:03' AS DateTime))
 ```
 
 Result:
 
-``` text
+```text
 ┌─toColumnTypeName(CAST('2018-01-01 01:02:03', 'DateTime'))─┐
 │ Const(UInt32)                                             │
 └───────────────────────────────────────────────────────────┘
@@ -1312,7 +1311,7 @@ The example shows that the `DateTime` data type is internally stored as `Const(U
 
 Outputs a detailed description of data structures in RAM
 
-``` sql
+```sql
 dumpColumnStructure(value)
 ```
 
@@ -1326,11 +1325,11 @@ dumpColumnStructure(value)
 
 **Example**
 
-``` sql
+```sql
 SELECT dumpColumnStructure(CAST('2018-01-01 01:02:03', 'DateTime'))
 ```
 
-``` text
+```text
 ┌─dumpColumnStructure(CAST('2018-01-01 01:02:03', 'DateTime'))─┐
 │ DateTime, Const(size = 1, UInt32(size = 1))                  │
 └──────────────────────────────────────────────────────────────┘
@@ -1342,7 +1341,7 @@ Returns the default value for the given data type.
 
 Does not include default values for custom columns set by the user.
 
-``` sql
+```sql
 defaultValueOfArgumentType(expression)
 ```
 
@@ -1360,13 +1359,13 @@ defaultValueOfArgumentType(expression)
 
 Query:
 
-``` sql
+```sql
 SELECT defaultValueOfArgumentType( CAST(1 AS Int8) )
 ```
 
 Result:
 
-``` text
+```text
 ┌─defaultValueOfArgumentType(CAST(1, 'Int8'))─┐
 │                                           0 │
 └─────────────────────────────────────────────┘
@@ -1374,13 +1373,13 @@ Result:
 
 Query:
 
-``` sql
+```sql
 SELECT defaultValueOfArgumentType( CAST(1 AS Nullable(Int8) ) )
 ```
 
 Result:
 
-``` text
+```text
 ┌─defaultValueOfArgumentType(CAST(1, 'Nullable(Int8)'))─┐
 │                                                  ᴺᵁᴸᴸ │
 └───────────────────────────────────────────────────────┘
@@ -1392,7 +1391,7 @@ Returns the default value for the given type name.
 
 Does not include default values for custom columns set by the user.
 
-``` sql
+```sql
 defaultValueOfTypeName(type)
 ```
 
@@ -1410,13 +1409,13 @@ defaultValueOfTypeName(type)
 
 Query:
 
-``` sql
+```sql
 SELECT defaultValueOfTypeName('Int8')
 ```
 
 Result:
 
-``` text
+```text
 ┌─defaultValueOfTypeName('Int8')─┐
 │                              0 │
 └────────────────────────────────┘
@@ -1424,13 +1423,13 @@ Result:
 
 Query:
 
-``` sql
+```sql
 SELECT defaultValueOfTypeName('Nullable(Int8)')
 ```
 
 Result:
 
-``` text
+```text
 ┌─defaultValueOfTypeName('Nullable(Int8)')─┐
 │                                     ᴺᵁᴸᴸ │
 └──────────────────────────────────────────┘
@@ -1542,7 +1541,7 @@ Creates an array with a single value.
 
 Used for the internal implementation of [arrayJoin](../../sql-reference/functions/array-join.md#functions_arrayjoin).
 
-``` sql
+```sql
 SELECT replicate(x, arr);
 ```
 
@@ -1561,13 +1560,13 @@ Type: `Array`.
 
 Query:
 
-``` sql
+```sql
 SELECT replicate(1, ['a', 'b', 'c'])
 ```
 
 Result:
 
-``` text
+```text
 ┌─replicate(1, ['a', 'b', 'c'])─┐
 │ [1,1,1]                       │
 └───────────────────────────────┘
@@ -1579,7 +1578,7 @@ Returns the amount of free space in the filesystem hosting the database persiste
 
 **Syntax**
 
-``` sql
+```sql
 filesystemAvailable()
 ```
 
@@ -1593,13 +1592,13 @@ Type: [UInt64](../../sql-reference/data-types/int-uint.md).
 
 Query:
 
-``` sql
+```sql
 SELECT formatReadableSize(filesystemAvailable()) AS "Available space";
 ```
 
 Result:
 
-``` text
+```text
 ┌─Available space─┐
 │ 30.75 GiB       │
 └─────────────────┘
@@ -1611,7 +1610,7 @@ Returns the total amount of the free space on the filesystem hosting the databas
 
 **Syntax**
 
-``` sql
+```sql
 filesystemFree()
 ```
 
@@ -1625,13 +1624,13 @@ Type: [UInt64](../../sql-reference/data-types/int-uint.md).
 
 Query:
 
-``` sql
+```sql
 SELECT formatReadableSize(filesystemFree()) AS "Free space";
 ```
 
 Result:
 
-``` text
+```text
 ┌─Free space─┐
 │ 32.39 GiB  │
 └────────────┘
@@ -1643,7 +1642,7 @@ Returns the capacity of the filesystem in bytes. Needs the [path](../../operatio
 
 **Syntax**
 
-``` sql
+```sql
 filesystemCapacity()
 ```
 
@@ -1657,13 +1656,13 @@ Type: [UInt64](../../sql-reference/data-types/int-uint.md).
 
 Query:
 
-``` sql
+```sql
 SELECT formatReadableSize(filesystemCapacity()) AS "Capacity";
 ```
 
 Result:
 
-``` text
+```text
 ┌─Capacity──┐
 │ 39.32 GiB │
 └───────────┘
@@ -1675,7 +1674,7 @@ Calculates the result of an aggregate function based on a single value. This fun
 
 **Syntax**
 
-``` sql
+```sql
 initializeAggregation (aggregate_function, arg1, arg2, ..., argN)
 ```
 
@@ -1697,6 +1696,7 @@ Query:
 ```sql
 SELECT uniqMerge(state) FROM (SELECT initializeAggregation('uniqState', number % 3) AS state FROM numbers(10000));
 ```
+
 Result:
 
 ```text
@@ -1749,7 +1749,7 @@ Given a state of aggregate function, this function returns the result of aggrega
 
 **Syntax**
 
-``` sql
+```sql
 finalizeAggregation(state)
 ```
 
@@ -1854,7 +1854,7 @@ The state is reset for each new block of data.
 
 **Syntax**
 
-``` sql
+```sql
 runningAccumulate(agg_state[, grouping]);
 ```
 
@@ -1875,13 +1875,13 @@ Consider how you can use `runningAccumulate` to find the cumulative sum of numbe
 
 Query:
 
-``` sql
+```sql
 SELECT k, runningAccumulate(sum_k) AS res FROM (SELECT number as k, sumState(k) AS sum_k FROM numbers(10) GROUP BY k ORDER BY k);
 ```
 
 Result:
 
-``` text
+```text
 ┌─k─┬─res─┐
 │ 0 │   0 │
 │ 1 │   1 │
@@ -1909,7 +1909,7 @@ The following example shows the `groupping` parameter usage:
 
 Query:
 
-``` sql
+```sql
 SELECT
     grouping,
     item,
@@ -1928,7 +1928,7 @@ FROM
 
 Result:
 
-``` text
+```text
 ┌─grouping─┬─item─┬─res─┐
 │        0 │    0 │   0 │
 │        0 │    1 │   1 │
@@ -1960,7 +1960,7 @@ Only supports tables created with the `ENGINE = Join(ANY, LEFT, <join_keys>)` st
 
 **Syntax**
 
-``` sql
+```sql
 joinGet(join_storage_table_name, `value_column`, join_keys)
 ```
 
@@ -1982,13 +1982,13 @@ More info about `join_use_nulls` in [Join operation](../../engines/table-engines
 
 Input table:
 
-``` sql
+```sql
 CREATE DATABASE db_test
 CREATE TABLE db_test.id_val(`id` UInt32, `val` UInt32) ENGINE = Join(ANY, LEFT, id) SETTINGS join_use_nulls = 1
 INSERT INTO db_test.id_val VALUES (1,11)(2,12)(4,13)
 ```
 
-``` text
+```text
 ┌─id─┬─val─┐
 │  4 │  13 │
 │  2 │  12 │
@@ -1998,13 +1998,13 @@ INSERT INTO db_test.id_val VALUES (1,11)(2,12)(4,13)
 
 Query:
 
-``` sql
+```sql
 SELECT joinGet(db_test.id_val, 'val', toUInt32(number)) from numbers(4) SETTINGS join_use_nulls = 1
 ```
 
 Result:
 
-``` text
+```text
 ┌─joinGet(db_test.id_val, 'val', toUInt32(number))─┐
 │                                                0 │
 │                                               11 │
@@ -2022,7 +2022,7 @@ This function is not available in ClickHouse Cloud.
 Evaluate an external catboost model. [CatBoost](https://catboost.ai) is an open-source gradient boosting library developed by Yandex for machine learning.
 Accepts a path to a catboost model and model arguments (features). Returns Float64.
 
-``` sql
+```sql
 SELECT feat1, ..., feat_n, catboostEvaluate('/path/to/model.bin', feat_1, ..., feat_n) AS prediction
 FROM data_table
 ```
@@ -2035,7 +2035,7 @@ Before evaluating catboost models, the `libcatboostmodel.<so|dylib>` library mus
 
 Next, specify the path to `libcatboostmodel.<so|dylib>` in the clickhouse configuration:
 
-``` xml
+```xml
 <clickhouse>
 ...
     <catboost_lib_path>/path/to/libcatboostmodel.so</catboost_lib_path>
@@ -2048,7 +2048,7 @@ At the first execution of `catboostEvaluate()`, the server starts the library br
 communicate using a HTTP interface. By default, port `9012` is used. A different port can be specified as follows - this is useful if port
 `9012` is already assigned to a different service.
 
-``` xml
+```xml
 <library_bridge>
     <port>9019</port>
 </library_bridge>
@@ -2072,13 +2072,13 @@ To use the `error_code` argument, configuration parameter `allow_custom_error_co
 
 **Example**
 
-``` sql
+```sql
 SELECT throwIf(number = 3, 'Too many') FROM numbers(10);
 ```
 
 Result:
 
-``` text
+```text
 ↙ Progress: 0.00 rows, 0.00 B (0.00 rows/s., 0.00 B/s.) Received exception from server (version 19.14.1):
 Code: 395. DB::Exception: Received from localhost:9000. DB::Exception: Too many.
 ```
@@ -2089,7 +2089,7 @@ Returns its argument. Intended for debugging and testing. Allows to cancel using
 
 **Syntax**
 
-``` sql
+```sql
 identity(x)
 ```
 
@@ -2097,13 +2097,13 @@ identity(x)
 
 Query:
 
-``` sql
+```sql
 SELECT identity(42);
 ```
 
 Result:
 
-``` text
+```text
 ┌─identity(42)─┐
 │           42 │
 └──────────────┘
@@ -2150,7 +2150,7 @@ Checks whether the [Decimal](../../sql-reference/data-types/decimal.md) value is
 
 **Syntax**
 
-``` sql
+```sql
 isDecimalOverflow(d, [p])
 ```
 
@@ -2168,7 +2168,7 @@ isDecimalOverflow(d, [p])
 
 Query:
 
-``` sql
+```sql
 SELECT isDecimalOverflow(toDecimal32(1000000000, 0), 9),
        isDecimalOverflow(toDecimal32(1000000000, 0)),
        isDecimalOverflow(toDecimal32(-1000000000, 0), 9),
@@ -2177,7 +2177,7 @@ SELECT isDecimalOverflow(toDecimal32(1000000000, 0), 9),
 
 Result:
 
-``` text
+```text
 1	1	1	1
 ```
 
@@ -2187,7 +2187,7 @@ Returns number of decimal digits need to represent a value.
 
 **Syntax**
 
-``` sql
+```sql
 countDigits(x)
 ```
 
@@ -2209,7 +2209,7 @@ For `Decimal` values takes into account their scales: calculates result over und
 
 Query:
 
-``` sql
+```sql
 SELECT countDigits(toDecimal32(1, 9)), countDigits(toDecimal32(-1, 9)),
        countDigits(toDecimal64(1, 18)), countDigits(toDecimal64(-1, 18)),
        countDigits(toDecimal128(1, 38)), countDigits(toDecimal128(-1, 38));
@@ -2217,7 +2217,7 @@ SELECT countDigits(toDecimal32(1, 9)), countDigits(toDecimal32(-1, 9)),
 
 Result:
 
-``` text
+```text
 10	10	19	19	39	39
 ```
 
@@ -2229,13 +2229,13 @@ Type: [LowCardinality(String)](../../sql-reference/data-types/lowcardinality.md)
 
 **Syntax**
 
-``` sql
+```sql
 errorCodeToName(1)
 ```
 
 Result:
 
-``` text
+```text
 UNSUPPORTED_METHOD
 ```
 
@@ -2246,7 +2246,7 @@ If executed in the context of a distributed table, this function generates a nor
 
 **Syntax**
 
-``` sql
+```sql
 tcpPort()
 ```
 
@@ -2264,13 +2264,13 @@ Type: [UInt16](../../sql-reference/data-types/int-uint.md).
 
 Query:
 
-``` sql
+```sql
 SELECT tcpPort();
 ```
 
 Result:
 
-``` text
+```text
 ┌─tcpPort()─┐
 │      9000 │
 └───────────┘
@@ -2288,7 +2288,7 @@ The command [SET PROFILE](../../sql-reference/statements/set.md#query-set) could
 
 **Syntax**
 
-``` sql
+```sql
 currentProfiles()
 ```
 
@@ -2300,11 +2300,11 @@ Type: [Array](../../sql-reference/data-types/array.md)([String](../../sql-refere
 
 ## enabledProfiles
 
- Returns settings profiles, assigned to the current user both explicitly and implicitly. Explicitly assigned profiles are the same as returned by the [currentProfiles](#current-profiles) function. Implicitly assigned profiles include parent profiles of other assigned profiles, profiles assigned via granted roles, profiles assigned via their own settings, and the main default profile (see the `default_profile` section in the main server configuration file).
+Returns settings profiles, assigned to the current user both explicitly and implicitly. Explicitly assigned profiles are the same as returned by the [currentProfiles](#current-profiles) function. Implicitly assigned profiles include parent profiles of other assigned profiles, profiles assigned via granted roles, profiles assigned via their own settings, and the main default profile (see the `default_profile` section in the main server configuration file).
 
 **Syntax**
 
-``` sql
+```sql
 enabledProfiles()
 ```
 
@@ -2320,7 +2320,7 @@ Returns all the profiles specified at the current user's definition (see [CREATE
 
 **Syntax**
 
-``` sql
+```sql
 defaultProfiles()
 ```
 
@@ -2336,7 +2336,7 @@ Returns the roles assigned to the current user. The roles can be changed by the
 
 **Syntax**
 
-``` sql
+```sql
 currentRoles()
 ```
 
@@ -2352,7 +2352,7 @@ Returns the names of the current roles and the roles, granted to some of the cur
 
 **Syntax**
 
-``` sql
+```sql
 enabledRoles()
 ```
 
@@ -2368,7 +2368,7 @@ Returns the roles which are enabled by default for the current user when he logs
 
 **Syntax**
 
-``` sql
+```sql
 defaultRoles()
 ```
 
@@ -2384,7 +2384,7 @@ Returns the server port number. When the port is not used by the server, throws
 
 **Syntax**
 
-``` sql
+```sql
 getServerPort(port_name)
 ```
 
@@ -2392,16 +2392,16 @@ getServerPort(port_name)
 
 - `port_name` — The name of the server port. [String](../../sql-reference/data-types/string.md#string). Possible values:
 
-    - 'tcp_port'
-    - 'tcp_port_secure'
-    - 'http_port'
-    - 'https_port'
-    - 'interserver_http_port'
-    - 'interserver_https_port'
-    - 'mysql_port'
-    - 'postgresql_port'
-    - 'grpc_port'
-    - 'prometheus.port'
+  - 'tcp_port'
+  - 'tcp_port_secure'
+  - 'http_port'
+  - 'https_port'
+  - 'interserver_http_port'
+  - 'interserver_https_port'
+  - 'mysql_port'
+  - 'postgresql_port'
+  - 'grpc_port'
+  - 'prometheus.port'
 
 **Returned value**
 
@@ -2413,13 +2413,13 @@ Type: [UInt16](../../sql-reference/data-types/int-uint.md).
 
 Query:
 
-``` sql
+```sql
 SELECT getServerPort('tcp_port');
 ```
 
 Result:
 
-``` text
+```text
 ┌─getServerPort('tcp_port')─┐
 │ 9000                      │
 └───────────────────────────┘
@@ -2433,7 +2433,7 @@ In contrast to [initialQueryID](#initial-query-id) function, `queryID` can retur
 
 **Syntax**
 
-``` sql
+```sql
 queryID()
 ```
 
@@ -2447,7 +2447,7 @@ Type: [String](../../sql-reference/data-types/string.md)
 
 Query:
 
-``` sql
+```sql
 CREATE TABLE tmp (str String) ENGINE = Log;
 INSERT INTO tmp (*) VALUES ('a');
 SELECT count(DISTINCT t) FROM (SELECT queryID() AS t FROM remote('127.0.0.{1..3}', currentDatabase(), 'tmp') GROUP BY queryID());
@@ -2455,7 +2455,7 @@ SELECT count(DISTINCT t) FROM (SELECT queryID() AS t FROM remote('127.0.0.{1..3}
 
 Result:
 
-``` text
+```text
 ┌─count()─┐
 │ 3       │
 └─────────┘
@@ -2469,7 +2469,7 @@ In contrast to [queryID](#query-id) function, `initialQueryID` returns the same
 
 **Syntax**
 
-``` sql
+```sql
 initialQueryID()
 ```
 
@@ -2483,7 +2483,7 @@ Type: [String](../../sql-reference/data-types/string.md)
 
 Query:
 
-``` sql
+```sql
 CREATE TABLE tmp (str String) ENGINE = Log;
 INSERT INTO tmp (*) VALUES ('a');
 SELECT count(DISTINCT t) FROM (SELECT initialQueryID() AS t FROM remote('127.0.0.{1..3}', currentDatabase(), 'tmp') GROUP BY queryID());
@@ -2491,7 +2491,7 @@ SELECT count(DISTINCT t) FROM (SELECT initialQueryID() AS t FROM remote('127.0.0
 
 Result:
 
-``` text
+```text
 ┌─count()─┐
 │ 1       │
 └─────────┘
@@ -2504,7 +2504,7 @@ If a query is not distributed then constant value `0` is returned.
 
 **Syntax**
 
-``` sql
+```sql
 shardNum()
 ```
 
@@ -2520,7 +2520,7 @@ In the following example a configuration with two shards is used. The query is e
 
 Query:
 
-``` sql
+```sql
 CREATE TABLE shard_num_example (dummy UInt8)
     ENGINE=Distributed(test_cluster_two_shards_localhost, system, one, dummy);
 SELECT dummy, shardNum(), shardCount() FROM shard_num_example;
@@ -2528,7 +2528,7 @@ SELECT dummy, shardNum(), shardCount() FROM shard_num_example;
 
 Result:
 
-``` text
+```text
 ┌─dummy─┬─shardNum()─┬─shardCount()─┐
 │     0 │          2 │            2 │
 │     0 │          1 │            2 │
@@ -2546,7 +2546,7 @@ If a query is not distributed then constant value `0` is returned.
 
 **Syntax**
 
-``` sql
+```sql
 shardCount()
 ```
 
@@ -2566,7 +2566,7 @@ Returns a string with the current OS kernel version.
 
 **Syntax**
 
-``` sql
+```sql
 getOSKernelVersion()
 ```
 
@@ -2584,13 +2584,13 @@ Type: [String](../../sql-reference/data-types/string.md).
 
 Query:
 
-``` sql
+```sql
 SELECT getOSKernelVersion();
 ```
 
 Result:
 
-``` text
+```text
 ┌─getOSKernelVersion()────┐
 │ Linux 4.15.0-55-generic │
 └─────────────────────────┘
@@ -2602,7 +2602,7 @@ Returns the uptime of the current ZooKeeper session in seconds.
 
 **Syntax**
 
-``` sql
+```sql
 zookeeperSessionUptime()
 ```
 
@@ -2620,13 +2620,13 @@ Type: [UInt32](../../sql-reference/data-types/int-uint.md).
 
 Query:
 
-``` sql
+```sql
 SELECT zookeeperSessionUptime();
 ```
 
 Result:
 
-``` text
+```text
 ┌─zookeeperSessionUptime()─┐
 │                      286 │
 └──────────────────────────┘
@@ -2638,7 +2638,7 @@ Generates random table structure in a format `column1_name column1_type, column2
 
 **Syntax**
 
-``` sql
+```sql
 generateRandomStructure([number_of_columns, seed])
 ```
 
@@ -2659,13 +2659,13 @@ Type: [String](../../sql-reference/data-types/string.md).
 
 Query:
 
-``` sql
+```sql
 SELECT generateRandomStructure()
 ```
 
 Result:
 
-``` text
+```text
 ┌─generateRandomStructure()─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────┐
 │ c1 Decimal32(5), c2 Date, c3 Tuple(LowCardinality(String), Int128, UInt64, UInt16, UInt8, IPv6), c4 Array(UInt128), c5 UInt32, c6 IPv4, c7 Decimal256(64), c8 Decimal128(3), c9 UInt256, c10 UInt64, c11 DateTime │
 └───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────┘
@@ -2673,13 +2673,13 @@ Result:
 
 Query:
 
-``` sql
+```sql
 SELECT generateRandomStructure(1)
 ```
 
 Result:
 
-``` text
+```text
 ┌─generateRandomStructure(1)─┐
 │ c1 Map(UInt256, UInt16)    │
 └────────────────────────────┘
@@ -2687,13 +2687,13 @@ Result:
 
 Query:
 
-``` sql
+```sql
 SELECT generateRandomStructure(NULL, 33)
 ```
 
 Result:
 
-``` text
+```text
 ┌─generateRandomStructure(NULL, 33)─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────┐
 │ c1 DateTime, c2 Enum8('c2V0' = 0, 'c2V1' = 1, 'c2V2' = 2, 'c2V3' = 3), c3 LowCardinality(Nullable(FixedString(30))), c4 Int16, c5 Enum8('c5V0' = 0, 'c5V1' = 1, 'c5V2' = 2, 'c5V3' = 3), c6 Nullable(UInt8), c7 String, c8 Nested(e1 IPv4, e2 UInt8, e3 UInt16, e4 UInt16, e5 Int32, e6 Map(Date, Decimal256(70))) │
 └────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────┘
@@ -2709,7 +2709,7 @@ Converts ClickHouse table structure to CapnProto schema.
 
 **Syntax**
 
-``` sql
+```sql
 structureToCapnProtoSchema(structure)
 ```
 
@@ -2720,7 +2720,7 @@ structureToCapnProtoSchema(structure)
 
 **Returned value**
 
-- CapnProto schema 
+- CapnProto schema
 
 Type: [String](../../sql-reference/data-types/string.md).
 
@@ -2728,13 +2728,13 @@ Type: [String](../../sql-reference/data-types/string.md).
 
 Query:
 
-``` sql
+```sql
 SELECT structureToCapnProtoSchema('column1 String, column2 UInt32, column3 Array(String)') FORMAT RawBLOB
 ```
 
 Result:
 
-``` text
+```text
 @0xf96402dd754d0eb7;
 
 struct Message
@@ -2747,13 +2747,13 @@ struct Message
 
 Query:
 
-``` sql
+```sql
 SELECT structureToCapnProtoSchema('column1 Nullable(String), column2 Tuple(element1 UInt32, element2 Array(String)), column3 Map(String, String)') FORMAT RawBLOB
 ```
 
 Result:
 
-``` text
+```text
 @0xd1c8320fecad2b7f;
 
 struct Message
@@ -2788,13 +2788,13 @@ struct Message
 
 Query:
 
-``` sql
+```sql
 SELECT structureToCapnProtoSchema('column1 String, column2 UInt32', 'Root') FORMAT RawBLOB
 ```
 
 Result:
 
-``` text
+```text
 @0x96ab2d4ab133c6e1;
 
 struct Root
@@ -2810,7 +2810,7 @@ Converts ClickHouse table structure to Protobuf schema.
 
 **Syntax**
 
-``` sql
+```sql
 structureToProtobufSchema(structure)
 ```
 
@@ -2829,13 +2829,13 @@ Type: [String](../../sql-reference/data-types/string.md).
 
 Query:
 
-``` sql
+```sql
 SELECT structureToProtobufSchema('column1 String, column2 UInt32, column3 Array(String)') FORMAT RawBLOB
 ```
 
 Result:
 
-``` text
+```text
 syntax = "proto3";
 
 message Message
@@ -2848,13 +2848,13 @@ message Message
 
 Query:
 
-``` sql
+```sql
 SELECT structureToProtobufSchema('column1 Nullable(String), column2 Tuple(element1 UInt32, element2 Array(String)), column3 Map(String, String)') FORMAT RawBLOB
 ```
 
 Result:
 
-``` text
+```text
 syntax = "proto3";
 
 message Message
@@ -2872,13 +2872,13 @@ message Message
 
 Query:
 
-``` sql
+```sql
 SELECT structureToProtobufSchema('column1 String, column2 UInt32', 'Root') FORMAT RawBLOB
 ```
 
 Result:
 
-``` text
+```text
 syntax = "proto3";
 
 message Root
@@ -3050,7 +3050,7 @@ Calculates minimum required sample size for an A/B test comparing conversions (p
 
 **Syntax**
 
-``` sql
+```sql
 minSampleSizeConversion(baseline, mde, power, alpha)
 ```
 
@@ -3075,13 +3075,13 @@ A named [Tuple](../data-types/tuple.md) with 3 elements:
 
 The following query calculates the required sample size for an A/B test with baseline conversion of 25%, MDE of 3%, significance level of 5%, and the desired statistical power of 80%:
 
-``` sql
+```sql
 SELECT minSampleSizeConversion(0.25, 0.03, 0.80, 0.05) AS sample_size;
 ```
 
 Result:
 
-``` text
+```text
 ┌─sample_size───────────────────┐
 │ (3396.077603219163,0.22,0.28) │
 └───────────────────────────────┘
@@ -3093,7 +3093,7 @@ Calculates minimum required sample size for an A/B test comparing means of a con
 
 **Syntax**
 
-``` sql
+```sql
 minSampleSizeContinous(baseline, sigma, mde, power, alpha)
 ```
 
@@ -3105,7 +3105,7 @@ Uses the formula described in [this article](https://towardsdatascience.com/requ
 
 - `baseline` — Baseline value of a metric. [Integer](../data-types/int-uint.md) or [Float](../data-types/float.md).
 - `sigma` — Baseline standard deviation of a metric. [Integer](../data-types/int-uint.md) or [Float](../data-types/float.md).
-- `mde` — Minimum detectable effect (MDE) as percentage of the baseline value (e.g. for a baseline value 112.25 the MDE 0.03 means an expected change to 112.25 ± 112.25*0.03). [Integer](../data-types/int-uint.md) or [Float](../data-types/float.md).
+- `mde` — Minimum detectable effect (MDE) as percentage of the baseline value (e.g. for a baseline value 112.25 the MDE 0.03 means an expected change to 112.25 ± 112.25\*0.03). [Integer](../data-types/int-uint.md) or [Float](../data-types/float.md).
 - `power` — Required statistical power of a test (1 - probability of Type II error). [Integer](../data-types/int-uint.md) or [Float](../data-types/float.md).
 - `alpha` — Required significance level of a test (probability of Type I error). [Integer](../data-types/int-uint.md) or [Float](../data-types/float.md).
 
@@ -3121,13 +3121,13 @@ A named [Tuple](../data-types/tuple.md) with 3 elements:
 
 The following query calculates the required sample size for an A/B test on a metric with baseline value of 112.25, standard deviation of 21.1, MDE of 3%, significance level of 5%, and the desired statistical power of 80%:
 
-``` sql
+```sql
 SELECT minSampleSizeContinous(112.25, 21.1, 0.03, 0.80, 0.05) AS sample_size;
 ```
 
 Result:
 
-``` text
+```text
 ┌─sample_size───────────────────────────┐
 │ (616.2931945826209,108.8825,115.6175) │
 └───────────────────────────────────────┘

From 8abb85948c4b401a61cdf6dc6cf33b7ac9df2279 Mon Sep 17 00:00:00 2001
From: Yatsishin Ilya <2159081+qoega@users.noreply.github.com>
Date: Tue, 12 Mar 2024 17:03:11 +0000
Subject: [PATCH 225/443] Fix flaky test and add separate to show unexpected
 behaviour

---
 .../01603_insert_select_too_many_parts.sql       |  3 ++-
 ...t_select_too_many_parts_multithread.reference |  1 +
 ..._insert_select_too_many_parts_multithread.sql | 16 ++++++++++++++++
 3 files changed, 19 insertions(+), 1 deletion(-)
 create mode 100644 tests/queries/0_stateless/02997_insert_select_too_many_parts_multithread.reference
 create mode 100644 tests/queries/0_stateless/02997_insert_select_too_many_parts_multithread.sql

diff --git a/tests/queries/0_stateless/01603_insert_select_too_many_parts.sql b/tests/queries/0_stateless/01603_insert_select_too_many_parts.sql
index a56b680e212..eea52282cf4 100644
--- a/tests/queries/0_stateless/01603_insert_select_too_many_parts.sql
+++ b/tests/queries/0_stateless/01603_insert_select_too_many_parts.sql
@@ -5,7 +5,8 @@ SYSTEM STOP MERGES too_many_parts;
 SET max_block_size = 1, min_insert_block_size_rows = 0, min_insert_block_size_bytes = 0;
 
 -- exception is not thrown if threshold is exceeded when multi-block INSERT is already started.
-INSERT INTO too_many_parts SELECT * FROM numbers(10);
+-- Single thread is used as different threads check it separately https://github.com/ClickHouse/ClickHouse/issues/61158
+INSERT INTO too_many_parts SELECT * FROM numbers(10) SETTINGS max_insert_threads=1;
 SELECT count() FROM too_many_parts;
 
 -- exception is thrown if threshold is exceeded on new INSERT.
diff --git a/tests/queries/0_stateless/02997_insert_select_too_many_parts_multithread.reference b/tests/queries/0_stateless/02997_insert_select_too_many_parts_multithread.reference
new file mode 100644
index 00000000000..29d6383b52c
--- /dev/null
+++ b/tests/queries/0_stateless/02997_insert_select_too_many_parts_multithread.reference
@@ -0,0 +1 @@
+100
diff --git a/tests/queries/0_stateless/02997_insert_select_too_many_parts_multithread.sql b/tests/queries/0_stateless/02997_insert_select_too_many_parts_multithread.sql
new file mode 100644
index 00000000000..00cf262add5
--- /dev/null
+++ b/tests/queries/0_stateless/02997_insert_select_too_many_parts_multithread.sql
@@ -0,0 +1,16 @@
+# Tags: disabled
+# TODO: Fix parts_to_throw_insert logic for parallel  MergeTreeSink onStart calls
+DROP TABLE IF EXISTS too_many_parts;
+
+CREATE TABLE too_many_parts (x UInt64) ENGINE = MergeTree ORDER BY tuple() SETTINGS parts_to_delay_insert = 5, parts_to_throw_insert = 5;
+
+SYSTEM STOP MERGES too_many_parts;
+SET max_block_size = 1, min_insert_block_size_rows = 0, min_insert_block_size_bytes = 0, max_threads=100, max_insert_threads=100;
+
+-- exception is not thrown if threshold is exceeded when multi-block INSERT is already started.
+INSERT INTO too_many_parts SELECT * FROM numbers_mt(100);
+SELECT count() FROM too_many_parts;
+
+INSERT INTO too_many_parts SELECT * FROM numbers_mt(10); --  { serverError 252 }
+
+DROP TABLE too_many_parts;

From f3f1f611a45500f82c581e5ec34810e77a02cfdc Mon Sep 17 00:00:00 2001
From: Nikita Mikhaylov <nikitamikhaylov@clickhouse.com>
Date: Tue, 12 Mar 2024 17:26:42 +0000
Subject: [PATCH 226/443] Better

---
 base/base/defines.h                      | 10 ++++++++--
 src/Parsers/ASTQueryWithTableAndOutput.h |  1 +
 2 files changed, 9 insertions(+), 2 deletions(-)

diff --git a/base/base/defines.h b/base/base/defines.h
index 1f02748633d..3308315b27f 100644
--- a/base/base/defines.h
+++ b/base/base/defines.h
@@ -108,16 +108,22 @@
         {
             [[noreturn]] void abortOnFailedAssertion(const String & description);
         }
-        #define chassert(x) do { static_cast<bool>(x) ? void(0) : ::DB::abortOnFailedAssertion(#x); } while (0)
+        #define chassert_1(x, ...) do { static_cast<bool>(x) ? void(0) : ::DB::abortOnFailedAssertion(#x); } while (0)
+        #define chassert_2(x, comment, ...) do { static_cast<bool>(x) ? void(0) : ::DB::abortOnFailedAssertion(comment); } while (0)
         #define UNREACHABLE() abort()
         // clang-format off
     #else
         /// Here sizeof() trick is used to suppress unused warning for result,
         /// since simple "(void)x" will evaluate the expression, while
         /// "sizeof(!(x))" will not.
-        #define chassert(x) (void)sizeof(!(x))
+        #define chassert_1(x, ...) (void)sizeof(!(x))
+        #define chassert_2(x, comment, ...) (void)sizeof(!(x))
         #define UNREACHABLE() __builtin_unreachable()
     #endif
+        #define CHASSERT_IMPL(_1,_2, N,...) N(_1, _2)
+        #define CHASSERT_IMPL_(tuple) CHASSERT_IMPL tuple
+        #define chassert(...) CHASSERT_IMPL_((__VA_ARGS__, chassert_2, chassert_1))
+
 #endif
 
 /// Macros for Clang Thread Safety Analysis (TSA). They can be safely ignored by other compilers.
diff --git a/src/Parsers/ASTQueryWithTableAndOutput.h b/src/Parsers/ASTQueryWithTableAndOutput.h
index 1b8621fb63b..358291d9aa8 100644
--- a/src/Parsers/ASTQueryWithTableAndOutput.h
+++ b/src/Parsers/ASTQueryWithTableAndOutput.h
@@ -61,6 +61,7 @@ protected:
             settings.ostr << '.';
         }
 
+        chassert(table != nullptr, "Table is empty for the ASTQueryWithTableAndOutputImpl.");
         table->formatImpl(settings, state, frame);
     }
 };

From ca282ab4510c3f06f6433de8349b28ce3527f436 Mon Sep 17 00:00:00 2001
From: Nikita Mikhaylov <nikitamikhaylov@clickhouse.com>
Date: Tue, 12 Mar 2024 17:33:09 +0000
Subject: [PATCH 227/443] Better

---
 base/base/defines.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/base/base/defines.h b/base/base/defines.h
index 3308315b27f..627c50c27d2 100644
--- a/base/base/defines.h
+++ b/base/base/defines.h
@@ -120,9 +120,9 @@
         #define chassert_2(x, comment, ...) (void)sizeof(!(x))
         #define UNREACHABLE() __builtin_unreachable()
     #endif
-        #define CHASSERT_IMPL(_1,_2, N,...) N(_1, _2)
-        #define CHASSERT_IMPL_(tuple) CHASSERT_IMPL tuple
-        #define chassert(...) CHASSERT_IMPL_((__VA_ARGS__, chassert_2, chassert_1))
+        #define CHASSERT_DISPATCH(_1,_2, N,...) N(_1, _2)
+        #define CHASSERT_INVOKE(tuple) CHASSERT_DISPATCH tuple
+        #define chassert(...) CHASSERT_INVOKE((__VA_ARGS__, chassert_2, chassert_1))
 
 #endif
 

From ddd2d92d0fbd31dad7e091ce1d828eb65fc34338 Mon Sep 17 00:00:00 2001
From: Smita Kulkarni <Smita.Kulkarni@clickhouse.com>
Date: Tue, 12 Mar 2024 18:40:20 +0100
Subject: [PATCH 228/443] Update curl to curl with data race fix

---
 contrib/curl | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/contrib/curl b/contrib/curl
index 5ce164e0e92..1a05e833f8f 160000
--- a/contrib/curl
+++ b/contrib/curl
@@ -1 +1 @@
-Subproject commit 5ce164e0e9290c96eb7d502173426c0a135ec008
+Subproject commit 1a05e833f8f7140628b27882b10525fd9ec4b873

From 296f7a1da2e5f5297fd26eb878071e4e60ac21fa Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Tue, 12 Mar 2024 19:02:58 +0100
Subject: [PATCH 229/443] Fix

---
 src/Disks/ObjectStorages/Web/WebObjectStorage.cpp | 4 ++++
 src/IO/ReadWriteBufferFromHTTP.cpp                | 2 +-
 src/IO/ReadWriteBufferFromHTTP.h                  | 3 +++
 3 files changed, 8 insertions(+), 1 deletion(-)

diff --git a/src/Disks/ObjectStorages/Web/WebObjectStorage.cpp b/src/Disks/ObjectStorages/Web/WebObjectStorage.cpp
index 4adb92cf5c8..7e942a6cf6f 100644
--- a/src/Disks/ObjectStorages/Web/WebObjectStorage.cpp
+++ b/src/Disks/ObjectStorages/Web/WebObjectStorage.cpp
@@ -86,6 +86,10 @@ WebObjectStorage::loadFiles(const String & path, const std::unique_lock<std::sha
             loaded_files.emplace_back(file_path);
         }
 
+        /// Check for not found url after read attempt, because of delayed initialization.
+        if (metadata_buf->hasNotFoundURL())
+            return {};
+
         auto [it, inserted] = files.add(path, FileData::createDirectoryInfo(true));
         if (!inserted)
         {
diff --git a/src/IO/ReadWriteBufferFromHTTP.cpp b/src/IO/ReadWriteBufferFromHTTP.cpp
index bcbec97537a..fdc8ef04d2e 100644
--- a/src/IO/ReadWriteBufferFromHTTP.cpp
+++ b/src/IO/ReadWriteBufferFromHTTP.cpp
@@ -449,6 +449,7 @@ bool ReadWriteBufferFromHTTP::nextImpl()
                     if (http_skip_not_found_url && e.getHTTPStatus() == Poco::Net::HTTPResponse::HTTPStatus::HTTP_NOT_FOUND)
                     {
                         next_result = false;
+                        has_not_found_url = true;
                         return;
                     }
 
@@ -740,4 +741,3 @@ ReadWriteBufferFromHTTP::HTTPFileInfo ReadWriteBufferFromHTTP::parseFileInfo(con
 }
 
 }
-
diff --git a/src/IO/ReadWriteBufferFromHTTP.h b/src/IO/ReadWriteBufferFromHTTP.h
index d0d792430c0..1b7437ea0c6 100644
--- a/src/IO/ReadWriteBufferFromHTTP.h
+++ b/src/IO/ReadWriteBufferFromHTTP.h
@@ -79,6 +79,7 @@ private:
 
     const bool use_external_buffer;
     const bool http_skip_not_found_url;
+    bool has_not_found_url = false;
 
     std::function<void(std::ostream &)> out_stream_callback;
 
@@ -183,6 +184,8 @@ public:
 
     std::optional<time_t> tryGetLastModificationTime();
 
+    bool hasNotFoundURL() const { return has_not_found_url; }
+
     HTTPFileInfo getFileInfo();
     static HTTPFileInfo parseFileInfo(const Poco::Net::HTTPResponse & response, size_t requested_range_begin);
 };

From f35987928bc41b37e3579d0a881562357efa32c7 Mon Sep 17 00:00:00 2001
From: Anton Popov <anton@clickhouse.com>
Date: Tue, 12 Mar 2024 18:05:14 +0000
Subject: [PATCH 230/443] free memory earlier in inserts

---
 src/Interpreters/AsynchronousInsertQueue.cpp  | 45 ++++++++++++-------
 src/Interpreters/AsynchronousInsertQueue.h    | 13 ++++++
 .../MergeTree/MergeTreeDataWriter.cpp         | 14 +++---
 src/Storages/MergeTree/MergeTreeDataWriter.h  |  4 +-
 src/Storages/MergeTree/MergeTreeSink.cpp      |  6 ++-
 .../MergeTree/ReplicatedMergeTreeSink.cpp     |  8 +++-
 6 files changed, 64 insertions(+), 26 deletions(-)

diff --git a/src/Interpreters/AsynchronousInsertQueue.cpp b/src/Interpreters/AsynchronousInsertQueue.cpp
index 7d56dbabe3c..12d1a5cc861 100644
--- a/src/Interpreters/AsynchronousInsertQueue.cpp
+++ b/src/Interpreters/AsynchronousInsertQueue.cpp
@@ -148,19 +148,25 @@ AsynchronousInsertQueue::InsertData::Entry::Entry(
 {
 }
 
+void AsynchronousInsertQueue::InsertData::Entry::resetChunk()
+{
+    if (chunk.empty())
+        return;
+
+    // To avoid races on counter of user's MemoryTracker we should free memory at this moment.
+    // Entries data must be destroyed in context of user who runs async insert.
+    // Each entry in the list may correspond to a different user,
+    // so we need to switch current thread's MemoryTracker.
+    MemoryTrackerSwitcher switcher(user_memory_tracker);
+    chunk = {};
+}
+
 void AsynchronousInsertQueue::InsertData::Entry::finish(std::exception_ptr exception_)
 {
     if (finished.exchange(true))
         return;
 
-    {
-        // To avoid races on counter of user's MemoryTracker we should free memory at this moment.
-        // Entries data must be destroyed in context of user who runs async insert.
-        // Each entry in the list may correspond to a different user,
-        // so we need to switch current thread's MemoryTracker.
-        MemoryTrackerSwitcher switcher(user_memory_tracker);
-        chunk = {};
-    }
+    resetChunk();
 
     if (exception_)
     {
@@ -224,7 +230,7 @@ AsynchronousInsertQueue::~AsynchronousInsertQueue()
             auto & shard = queue_shards[i];
 
             shard.are_tasks_available.notify_one();
-            assert(dump_by_first_update_threads[i].joinable());
+            chassert(dump_by_first_update_threads[i].joinable());
             dump_by_first_update_threads[i].join();
 
             if (flush_on_shutdown)
@@ -510,14 +516,13 @@ void AsynchronousInsertQueue::validateSettings(const Settings & settings, Logger
     /// Adaptive timeout settings.
     const auto min_ms = std::chrono::milliseconds(settings.async_insert_busy_timeout_min_ms);
 
-    if (min_ms > max_ms)
-        if (log)
-            LOG_WARNING(
-                log,
-                "Setting 'async_insert_busy_timeout_min_ms'={} is greater than 'async_insert_busy_timeout_max_ms'={}. Ignoring "
-                "'async_insert_busy_timeout_min_ms'",
-                min_ms.count(),
-                max_ms.count());
+    if (min_ms > max_ms && log)
+        LOG_WARNING(
+            log,
+            "Setting 'async_insert_busy_timeout_min_ms'={} is greater than 'async_insert_busy_timeout_max_ms'={}. Ignoring "
+            "'async_insert_busy_timeout_min_ms'",
+            min_ms.count(),
+            max_ms.count());
 
     if (settings.async_insert_busy_timeout_increase_rate <= 0)
         throw Exception(ErrorCodes::INVALID_SETTING_VALUE, "Setting 'async_insert_busy_timeout_increase_rate' must be greater than zero");
@@ -953,14 +958,18 @@ Chunk AsynchronousInsertQueue::processEntriesWithParsing(
                 "Expected entry with data kind Parsed. Got: {}", entry->chunk.getDataKind());
 
         auto buffer = std::make_unique<ReadBufferFromString>(*bytes);
+
         size_t num_bytes = bytes->size();
         size_t num_rows = executor.execute(*buffer);
+
         total_rows += num_rows;
         chunk_info->offsets.push_back(total_rows);
         chunk_info->tokens.push_back(entry->async_dedup_token);
 
         add_to_async_insert_log(entry, query_for_logging, current_exception, num_rows, num_bytes, data->timeout_ms);
+
         current_exception.clear();
+        entry->resetChunk();
     }
 
     Chunk chunk(executor.getResultColumns(), total_rows);
@@ -1011,6 +1020,8 @@ Chunk AsynchronousInsertQueue::processPreprocessedEntries(
 
         const auto & query_for_logging = get_query_by_format(entry->format);
         add_to_async_insert_log(entry, query_for_logging, "", block->rows(), block->bytes(), data->timeout_ms);
+
+        entry->resetChunk();
     }
 
     Chunk chunk(std::move(result_columns), total_rows);
diff --git a/src/Interpreters/AsynchronousInsertQueue.h b/src/Interpreters/AsynchronousInsertQueue.h
index f60b3d343fb..f4044563f34 100644
--- a/src/Interpreters/AsynchronousInsertQueue.h
+++ b/src/Interpreters/AsynchronousInsertQueue.h
@@ -117,6 +117,17 @@ private:
                 return DataKind::Parsed;
         }
 
+        bool empty() const
+        {
+            return std::visit([]<typename T>(const T & arg)
+            {
+                if constexpr (std::is_same_v<T, Block>)
+                    return arg.rows() == 0;
+                else
+                    return arg.empty();
+            }, *this);
+        }
+
         const String * asString() const { return std::get_if<String>(this); }
         const Block * asBlock() const { return std::get_if<Block>(this); }
     };
@@ -140,7 +151,9 @@ private:
                 const String & format_,
                 MemoryTracker * user_memory_tracker_);
 
+            void resetChunk();
             void finish(std::exception_ptr exception_ = nullptr);
+
             std::future<void> getFuture() { return promise.get_future(); }
             bool isFinished() const { return finished; }
 
diff --git a/src/Storages/MergeTree/MergeTreeDataWriter.cpp b/src/Storages/MergeTree/MergeTreeDataWriter.cpp
index fdac16ae19a..30466e1b388 100644
--- a/src/Storages/MergeTree/MergeTreeDataWriter.cpp
+++ b/src/Storages/MergeTree/MergeTreeDataWriter.cpp
@@ -241,7 +241,7 @@ std::vector<AsyncInsertInfoPtr> scatterAsyncInsertInfoBySelector(AsyncInsertInfo
 }
 
 BlocksWithPartition MergeTreeDataWriter::splitBlockIntoParts(
-    const Block & block, size_t max_parts, const StorageMetadataPtr & metadata_snapshot, ContextPtr context, AsyncInsertInfoPtr async_insert_info)
+    Block && block, size_t max_parts, const StorageMetadataPtr & metadata_snapshot, ContextPtr context, AsyncInsertInfoPtr async_insert_info)
 {
     BlocksWithPartition result;
     if (!block || !block.rows())
@@ -320,7 +320,7 @@ BlocksWithPartition MergeTreeDataWriter::splitBlockIntoParts(
 }
 
 Block MergeTreeDataWriter::mergeBlock(
-    const Block & block,
+    Block && block,
     SortDescription sort_description,
     const Names & partition_key_columns,
     IColumn::Permutation *& permutation,
@@ -410,7 +410,11 @@ MergeTreeDataWriter::TemporaryPart MergeTreeDataWriter::writeTempPartWithoutPref
 }
 
 MergeTreeDataWriter::TemporaryPart MergeTreeDataWriter::writeTempPartImpl(
-    BlockWithPartition & block_with_partition, const StorageMetadataPtr & metadata_snapshot, ContextPtr context, int64_t block_number, bool need_tmp_prefix)
+    BlockWithPartition & block_with_partition,
+    const StorageMetadataPtr & metadata_snapshot,
+    ContextPtr context,
+    int64_t block_number,
+    bool need_tmp_prefix)
 {
     TemporaryPart temp_part;
     Block & block = block_with_partition.block;
@@ -498,7 +502,7 @@ MergeTreeDataWriter::TemporaryPart MergeTreeDataWriter::writeTempPartImpl(
     if (context->getSettingsRef().optimize_on_insert)
     {
         ProfileEventTimeIncrement<Microseconds> watch(ProfileEvents::MergeTreeDataWriterMergingBlocksMicroseconds);
-        block = mergeBlock(block, sort_description, partition_key_columns, perm_ptr, data.merging_params);
+        block = mergeBlock(std::move(block), sort_description, partition_key_columns, perm_ptr, data.merging_params);
     }
 
     /// Size of part would not be greater than block.bytes() + epsilon
@@ -717,7 +721,7 @@ MergeTreeDataWriter::TemporaryPart MergeTreeDataWriter::writeProjectionPartImpl(
 
         MergeTreeData::MergingParams projection_merging_params;
         projection_merging_params.mode = MergeTreeData::MergingParams::Aggregating;
-        block = mergeBlock(block, sort_description, {}, perm_ptr, projection_merging_params);
+        block = mergeBlock(std::move(block), sort_description, {}, perm_ptr, projection_merging_params);
     }
 
     /// This effectively chooses minimal compression method:
diff --git a/src/Storages/MergeTree/MergeTreeDataWriter.h b/src/Storages/MergeTree/MergeTreeDataWriter.h
index 8fb8b82dbe6..9dffea0a471 100644
--- a/src/Storages/MergeTree/MergeTreeDataWriter.h
+++ b/src/Storages/MergeTree/MergeTreeDataWriter.h
@@ -53,7 +53,7 @@ public:
       *  (split rows by partition)
       * Works deterministically: if same block was passed, function will return same result in same order.
       */
-    static BlocksWithPartition splitBlockIntoParts(const Block & block, size_t max_parts, const StorageMetadataPtr & metadata_snapshot, ContextPtr context, AsyncInsertInfoPtr async_insert_info = nullptr);
+    static BlocksWithPartition splitBlockIntoParts(Block && block, size_t max_parts, const StorageMetadataPtr & metadata_snapshot, ContextPtr context, AsyncInsertInfoPtr async_insert_info = nullptr);
 
     /// This structure contains not completely written temporary part.
     /// Some writes may happen asynchronously, e.g. for blob storages.
@@ -107,7 +107,7 @@ public:
         size_t block_num);
 
     static Block mergeBlock(
-        const Block & block,
+        Block && block,
         SortDescription sort_description,
         const Names & partition_key_columns,
         IColumn::Permutation *& permutation,
diff --git a/src/Storages/MergeTree/MergeTreeSink.cpp b/src/Storages/MergeTree/MergeTreeSink.cpp
index 87b0a04d244..2da7565daad 100644
--- a/src/Storages/MergeTree/MergeTreeSink.cpp
+++ b/src/Storages/MergeTree/MergeTreeSink.cpp
@@ -63,7 +63,7 @@ void MergeTreeSink::consume(Chunk chunk)
     if (!storage_snapshot->object_columns.empty())
         convertDynamicColumnsToTuples(block, storage_snapshot);
 
-    auto part_blocks = storage.writer.splitBlockIntoParts(block, max_parts_per_block, metadata_snapshot, context);
+    auto part_blocks = storage.writer.splitBlockIntoParts(std::move(block), max_parts_per_block, metadata_snapshot, context);
 
     using DelayedPartitions = std::vector<MergeTreeSink::DelayedChunk::Partition>;
     DelayedPartitions partitions;
@@ -87,6 +87,10 @@ void MergeTreeSink::consume(Chunk chunk)
             elapsed_ns = watch.elapsed();
         }
 
+        /// Reset earlier to free memory
+        current_block.block.clear();
+        current_block.partition.clear();
+
         /// If optimize_on_insert setting is true, current_block could become empty after merge
         /// and we didn't create part.
         if (!temp_part.part)
diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp
index e50b4007d64..768a680d9bd 100644
--- a/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp
+++ b/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp
@@ -288,7 +288,7 @@ void ReplicatedMergeTreeSinkImpl<async_insert>::consume(Chunk chunk)
             throw Exception(ErrorCodes::LOGICAL_ERROR, "No chunk info for async inserts");
     }
 
-    auto part_blocks = storage.writer.splitBlockIntoParts(block, max_parts_per_block, metadata_snapshot, context, async_insert_info);
+    auto part_blocks = storage.writer.splitBlockIntoParts(std::move(block), max_parts_per_block, metadata_snapshot, context, async_insert_info);
 
     using DelayedPartition = typename ReplicatedMergeTreeSinkImpl<async_insert>::DelayedChunk::Partition;
     using DelayedPartitions = std::vector<DelayedPartition>;
@@ -383,6 +383,12 @@ void ReplicatedMergeTreeSinkImpl<async_insert>::consume(Chunk chunk)
             partitions = DelayedPartitions{};
         }
 
+        if constexpr (!async_insert)
+        {
+            /// Reset earlier to free memory.
+            current_block.block.clear();
+            current_block.partition.clear();
+        }
 
         partitions.emplace_back(DelayedPartition(
             log,

From c947484fe0d788dd46384b90987b59694f0a0b77 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Tue, 12 Mar 2024 19:36:10 +0100
Subject: [PATCH 231/443] Fxi again

---
 tests/integration/test_disk_types/test.py       | 15 ++++++++++++---
 .../test_endpoint_macro_substitution/test.py    | 17 ++++++++++++-----
 2 files changed, 24 insertions(+), 8 deletions(-)

diff --git a/tests/integration/test_disk_types/test.py b/tests/integration/test_disk_types/test.py
index a53d073d30b..b9b8ef2010d 100644
--- a/tests/integration/test_disk_types/test.py
+++ b/tests/integration/test_disk_types/test.py
@@ -87,19 +87,28 @@ def test_different_types(cluster):
 def test_select_by_type(cluster):
     node = cluster.instances["node"]
     for name, disk_type in list(disk_types.items()):
-        if disk_type != "S3":
+        if disk_type == "Local":
             assert (
                 node.query(
                     "SELECT name FROM system.disks WHERE type='" + disk_type + "'"
                 )
                 == name + "\n"
             )
-        else:
+        elif disk_type == "S3":
             assert (
                 node.query(
-                    "SELECT name FROM system.disks WHERE type='"
+                    "SELECT name FROM system.disks WHERE object_storage_type='"
                     + disk_type
                     + "' ORDER BY name"
                 )
                 == "disk_encrypted\ndisk_s3\n"
             )
+        else:
+            assert (
+                node.query(
+                    "SELECT name FROM system.disks WHERE object_storage_type='"
+                    + disk_type
+                    + "'"
+                )
+                == name + "\n"
+            )
diff --git a/tests/integration/test_endpoint_macro_substitution/test.py b/tests/integration/test_endpoint_macro_substitution/test.py
index e161d8e82ff..ee72fb9b492 100644
--- a/tests/integration/test_endpoint_macro_substitution/test.py
+++ b/tests/integration/test_endpoint_macro_substitution/test.py
@@ -67,22 +67,29 @@ def test_different_types(cluster):
 
 def test_select_by_type(cluster):
     node = cluster.instances["node"]
-    fs = HdfsClient(hosts=cluster.hdfs_ip)
-
     for name, disk_type in list(disk_types.items()):
-        if disk_type != "S3":
+        if disk_type == "Local":
             assert (
                 node.query(
                     "SELECT name FROM system.disks WHERE type='" + disk_type + "'"
                 )
                 == name + "\n"
             )
-        else:
+        elif disk_type == "S3":
             assert (
                 node.query(
-                    "SELECT name FROM system.disks WHERE type='"
+                    "SELECT name FROM system.disks WHERE object_storage_type='"
                     + disk_type
                     + "' ORDER BY name"
                 )
                 == "disk_encrypted\ndisk_s3\n"
             )
+        else:
+            assert (
+                node.query(
+                    "SELECT name FROM system.disks WHERE object_storage_type='"
+                    + disk_type
+                    + "'"
+                )
+                == name + "\n"
+            )

From ec80cc475ed8c0c3126aceb5db35ff20b9295451 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Tue, 12 Mar 2024 19:58:23 +0100
Subject: [PATCH 232/443] Fix formatting of overlay database

---
 src/Databases/DatabasesOverlay.cpp                        | 4 +++-
 .../0_stateless/03009_format_show_database.reference      | 2 ++
 tests/queries/0_stateless/03009_format_show_database.sh   | 8 ++++++++
 3 files changed, 13 insertions(+), 1 deletion(-)
 create mode 100644 tests/queries/0_stateless/03009_format_show_database.reference
 create mode 100755 tests/queries/0_stateless/03009_format_show_database.sh

diff --git a/src/Databases/DatabasesOverlay.cpp b/src/Databases/DatabasesOverlay.cpp
index 8cea3441698..c8705254e73 100644
--- a/src/Databases/DatabasesOverlay.cpp
+++ b/src/Databases/DatabasesOverlay.cpp
@@ -149,7 +149,9 @@ ASTPtr DatabasesOverlay::getCreateTableQueryImpl(const String & name, ContextPtr
  */
 ASTPtr DatabasesOverlay::getCreateDatabaseQuery() const
 {
-    return std::make_shared<ASTCreateQuery>();
+    auto query = std::make_shared<ASTCreateQuery>();
+    query->setDatabase(getDatabaseName());
+    return query;
 }
 
 String DatabasesOverlay::getTableDataPath(const String & table_name) const
diff --git a/tests/queries/0_stateless/03009_format_show_database.reference b/tests/queries/0_stateless/03009_format_show_database.reference
new file mode 100644
index 00000000000..83cfd4c1a68
--- /dev/null
+++ b/tests/queries/0_stateless/03009_format_show_database.reference
@@ -0,0 +1,2 @@
+CREATE DATABASE default
+UNKNOWN_DATABASE
diff --git a/tests/queries/0_stateless/03009_format_show_database.sh b/tests/queries/0_stateless/03009_format_show_database.sh
new file mode 100755
index 00000000000..7f33ad7b1e1
--- /dev/null
+++ b/tests/queries/0_stateless/03009_format_show_database.sh
@@ -0,0 +1,8 @@
+#!/usr/bin/env bash
+
+CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
+# shellcheck source=../shell_config.sh
+. "$CUR_DIR"/../shell_config.sh
+
+$CLICKHOUSE_LOCAL -q "show database default"
+$CLICKHOUSE_LOCAL -q "show database default2" 2>&1 | grep -o 'UNKNOWN_DATABASE'

From 15484bedadf62d05fdcaea842ce9d86bdafe57f3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Tue, 12 Mar 2024 21:03:26 +0100
Subject: [PATCH 233/443] Fix clone for ASTQueryWithTableAndOutputImpl derived
 classes

---
 src/Parsers/TablePropertiesQueriesASTs.h | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/src/Parsers/TablePropertiesQueriesASTs.h b/src/Parsers/TablePropertiesQueriesASTs.h
index e8e4bd8adb3..81ad975aa37 100644
--- a/src/Parsers/TablePropertiesQueriesASTs.h
+++ b/src/Parsers/TablePropertiesQueriesASTs.h
@@ -85,6 +85,15 @@ using ASTShowCreateDictionaryQuery = ASTQueryWithTableAndOutputImpl<ASTShowCreat
 
 class ASTExistsDatabaseQuery : public ASTQueryWithTableAndOutputImpl<ASTExistsDatabaseQueryIDAndQueryNames>
 {
+public:
+    ASTPtr clone() const override
+    {
+        auto res = std::make_shared<ASTExistsDatabaseQuery>(*this);
+        res->children.clear();
+        cloneTableOptions(*res);
+        return res;
+    }
+
 protected:
     void formatQueryImpl(const FormatSettings & settings, FormatState & state, FormatStateStacked frame) const override
     {
@@ -98,6 +107,15 @@ protected:
 
 class ASTShowCreateDatabaseQuery : public ASTQueryWithTableAndOutputImpl<ASTShowCreateDatabaseQueryIDAndQueryNames>
 {
+public:
+    ASTPtr clone() const override
+    {
+        auto res = std::make_shared<ASTShowCreateDatabaseQuery>(*this);
+        res->children.clear();
+        cloneTableOptions(*res);
+        return res;
+    }
+
 protected:
     void formatQueryImpl(const FormatSettings & settings, FormatState & state, FormatStateStacked frame) const override
     {

From c7c64636d018f1d07c7ea84c2fe9a61285bfe863 Mon Sep 17 00:00:00 2001
From: Han Fei <hanfei19910905@gmail.com>
Date: Tue, 12 Mar 2024 23:31:33 +0100
Subject: [PATCH 234/443] add a test

---
 programs/keeper/Keeper.cpp                    | 22 +++++---
 programs/server/Server.cpp                    |  2 +-
 src/Common/CgroupsMemoryUsageObserver.cpp     | 18 +++++--
 src/Common/CgroupsMemoryUsageObserver.h       | 10 ++--
 tests/integration/helpers/cluster.py          |  9 ++++
 .../test_memory_limit_observer/__init__.py    |  0
 .../config/text_log.xml                       |  7 +++
 .../test_memory_limit_observer/test.py        | 51 +++++++++++++++++++
 8 files changed, 101 insertions(+), 18 deletions(-)
 create mode 100644 tests/integration/test_memory_limit_observer/__init__.py
 create mode 100644 tests/integration/test_memory_limit_observer/config/text_log.xml
 create mode 100644 tests/integration/test_memory_limit_observer/test.py

diff --git a/programs/keeper/Keeper.cpp b/programs/keeper/Keeper.cpp
index 31e0b7dc576..5ec1d2a6568 100644
--- a/programs/keeper/Keeper.cpp
+++ b/programs/keeper/Keeper.cpp
@@ -625,15 +625,23 @@ try
     main_config_reloader->start();
 
     std::optional<CgroupsMemoryUsageObserver> observer;
-    auto cgroups_memory_observer_wait_time = config().getUInt64("keeper_server.cgroups_memory_observer_wait_time", 1);
-    if (cgroups_memory_observer_wait_time > 0)
+    auto cgroups_memory_observer_wait_time = config().getUInt64("keeper_server.cgroups_memory_observer_wait_time", 15);
+    if (cgroups_memory_observer_wait_time != 0)
     {
-        observer.emplace(std::chrono::seconds(cgroups_memory_observer_wait_time));
-        observer->startThread();
-        observer->setOnMemoryLimitUpdate([&]()
+        try
         {
-            main_config_reloader->reload();
-        });
+            observer.emplace(std::chrono::seconds(cgroups_memory_observer_wait_time));
+            observer->setOnMemoryLimitUpdate([&]()
+            {
+                main_config_reloader->reload();
+            });
+            observer->startThread();
+        }
+        catch (Exception &)
+        {
+            tryLogCurrentException(log, "Disabling cgroup memory observer because of an error during initialization");
+        }
+
     }
 
 
diff --git a/programs/server/Server.cpp b/programs/server/Server.cpp
index 33a30a1de2c..8e672f3f73e 100644
--- a/programs/server/Server.cpp
+++ b/programs/server/Server.cpp
@@ -1362,7 +1362,6 @@ try
                 cgroups_memory_usage_observer->setLimits(
                     static_cast<uint64_t>(max_server_memory_usage * hard_limit_ratio),
                     static_cast<uint64_t>(max_server_memory_usage * soft_limit_ratio));
-                cgroups_memory_usage_observer->startThread();
             }
 
             size_t merges_mutations_memory_usage_soft_limit = new_server_settings.merges_mutations_memory_usage_soft_limit;
@@ -1707,6 +1706,7 @@ try
         {
             main_config_reloader->reload();
         });
+        cgroups_memory_usage_observer->startThread();
     }
 
     /// Reload config in SYSTEM RELOAD CONFIG query.
diff --git a/src/Common/CgroupsMemoryUsageObserver.cpp b/src/Common/CgroupsMemoryUsageObserver.cpp
index 3fda51a119f..e9121754230 100644
--- a/src/Common/CgroupsMemoryUsageObserver.cpp
+++ b/src/Common/CgroupsMemoryUsageObserver.cpp
@@ -97,6 +97,12 @@ void CgroupsMemoryUsageObserver::setLimits(uint64_t hard_limit_, uint64_t soft_l
     LOG_INFO(log, "Set new limits, soft limit: {}, hard limit: {}", ReadableSize(soft_limit_), ReadableSize(hard_limit_));
 }
 
+void CgroupsMemoryUsageObserver::setOnMemoryLimitUpdate(UpdateMemLimitCallbackFn on_memory_limit_update_)
+{
+    std::lock_guard<std::mutex> set_limit_lock(set_limit_mutex);
+    on_memory_limit_update = on_memory_limit_update_;
+}
+
 uint64_t CgroupsMemoryUsageObserver::readMemoryUsage() const
 {
     return file.readMemoryUsage();
@@ -275,7 +281,10 @@ void CgroupsMemoryUsageObserver::stopThread()
 void CgroupsMemoryUsageObserver::runThread()
 {
     setThreadName("CgrpMemUsgObsr");
-    last_memory_amount = getMemoryAmount();
+
+    last_process_memory_amount = getMemoryAmount();
+    LOG_INFO(log, "Init memory amount is {} bytes", last_process_memory_amount);
+
     std::unique_lock lock(thread_mutex);
     while (true)
     {
@@ -284,10 +293,11 @@ void CgroupsMemoryUsageObserver::runThread()
 
         try
         {
-            uint64_t memory_limit = getMemoryAmount();
-            if (memory_limit != last_memory_amount)
+            uint64_t process_memory_limit = getMemoryAmount();
+            if (process_memory_limit != last_process_memory_amount)
             {
-                last_memory_amount = memory_limit;
+                LOG_INFO(log, "Find memory amount change, old limit is {} bytes, new limit is {} bytes", last_process_memory_amount, process_memory_limit);
+                last_process_memory_amount = process_memory_limit;
                 /// if we find memory amount changes, we just reload config.
                 /// Reloading config will check the memory amount again and calculate soft/hard limit again.
                 on_memory_limit_update();
diff --git a/src/Common/CgroupsMemoryUsageObserver.h b/src/Common/CgroupsMemoryUsageObserver.h
index fc0d847af4f..24acc0d8f40 100644
--- a/src/Common/CgroupsMemoryUsageObserver.h
+++ b/src/Common/CgroupsMemoryUsageObserver.h
@@ -27,10 +27,8 @@ public:
     ~CgroupsMemoryUsageObserver();
 
     void setLimits(uint64_t hard_limit_, uint64_t soft_limit_);
-    void setOnMemoryLimitUpdate(std::function<void()> on_memory_limit_update_)
-    {
-        on_memory_limit_update = on_memory_limit_update_;
-    }
+    using UpdateMemLimitCallbackFn = std::function<void()>;
+    void setOnMemoryLimitUpdate(UpdateMemLimitCallbackFn on_memory_limit_update_);
     void startThread();
 
     size_t getHardLimit() const { return hard_limit; }
@@ -49,7 +47,7 @@ private:
     using CallbackFn = std::function<void(bool)>;
     CallbackFn on_hard_limit;
     CallbackFn on_soft_limit;
-    std::function<void()> on_memory_limit_update;
+    UpdateMemLimitCallbackFn on_memory_limit_update;
 
     uint64_t last_usage = 0;
 
@@ -80,7 +78,7 @@ private:
     std::condition_variable cond;
     ThreadFromGlobalPool thread;
     bool quit = false;
-    uint64_t last_memory_amount;
+    uint64_t last_process_memory_amount;
 };
 
 #else
diff --git a/tests/integration/helpers/cluster.py b/tests/integration/helpers/cluster.py
index 52c0d8a8ee5..b695b493db7 100644
--- a/tests/integration/helpers/cluster.py
+++ b/tests/integration/helpers/cluster.py
@@ -1618,6 +1618,7 @@ class ClickHouseCluster:
         with_installed_binary=False,
         external_dirs=None,
         tmpfs=None,
+        mem_limit=None,
         zookeeper_docker_compose_path=None,
         minio_certs_dir=None,
         minio_data_dir=None,
@@ -1728,6 +1729,7 @@ class ClickHouseCluster:
             with_installed_binary=with_installed_binary,
             external_dirs=external_dirs,
             tmpfs=tmpfs or [],
+            mem_limit=mem_limit,
             config_root_name=config_root_name,
             extra_configs=extra_configs,
         )
@@ -3203,6 +3205,7 @@ services:
             {krb5_conf}
         entrypoint: {entrypoint_cmd}
         tmpfs: {tmpfs}
+        {mem_limit}
         cap_add:
             - SYS_PTRACE
             - NET_ADMIN
@@ -3288,6 +3291,7 @@ class ClickHouseInstance:
         with_installed_binary=False,
         external_dirs=None,
         tmpfs=None,
+        mem_limit=None,
         config_root_name="clickhouse",
         extra_configs=[],
     ):
@@ -3299,6 +3303,10 @@ class ClickHouseInstance:
 
         self.external_dirs = external_dirs
         self.tmpfs = tmpfs or []
+        if mem_limit is not None:
+            self.mem_limit = "mem_limit : " + mem_limit
+        else:
+            self.mem_limit = ""
         self.base_config_dir = (
             p.abspath(p.join(base_path, base_config_dir)) if base_config_dir else None
         )
@@ -4644,6 +4652,7 @@ class ClickHouseInstance:
                     db_dir=db_dir,
                     external_dirs_volumes=external_dirs_volumes,
                     tmpfs=str(self.tmpfs),
+                    mem_limit=self.mem_limit,
                     logs_dir=logs_dir,
                     depends_on=str(depends_on),
                     user=os.getuid(),
diff --git a/tests/integration/test_memory_limit_observer/__init__.py b/tests/integration/test_memory_limit_observer/__init__.py
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/integration/test_memory_limit_observer/config/text_log.xml b/tests/integration/test_memory_limit_observer/config/text_log.xml
new file mode 100644
index 00000000000..5c1cc663d0f
--- /dev/null
+++ b/tests/integration/test_memory_limit_observer/config/text_log.xml
@@ -0,0 +1,7 @@
+<clickhouse>
+    <text_log>
+        <database>system</database>
+        <table>text_log</table>
+        <flush_interval_milliseconds>500</flush_interval_milliseconds>
+    </text_log>
+</clickhouse>
diff --git a/tests/integration/test_memory_limit_observer/test.py b/tests/integration/test_memory_limit_observer/test.py
new file mode 100644
index 00000000000..cf10f46be6a
--- /dev/null
+++ b/tests/integration/test_memory_limit_observer/test.py
@@ -0,0 +1,51 @@
+import pytest
+import logging
+import time
+
+from helpers.cluster import ClickHouseCluster, run_and_check
+
+cluster = ClickHouseCluster(__file__)
+
+node1 = cluster.add_instance(
+    "node1", main_configs=["config/text_log.xml"], mem_limit = "5g"
+)
+
+@pytest.fixture(scope="module")
+def started_cluster():
+    try:
+        cluster.start()
+
+        yield cluster
+
+    finally:
+        cluster.shutdown()
+
+def get_latest_mem_limit():
+    for _ in range(10):
+        try:
+            mem_limit = float(node1.query(
+            '''
+            select extract(message, '\\d+\\.\\d+') from system.text_log
+            where message like '%Setting max_server_memory_usage was set to%' and
+            message not like '%like%' order by event_time desc limit 1
+            '''
+            ).strip())
+            return mem_limit
+        except Exception as e:
+            time.sleep(1)
+    raise Exception("Cannot get memory limit")
+
+def test_observe_memory_limit(started_cluster):
+    original_max_mem = get_latest_mem_limit()
+    #container_id = cluster.get_container_id('node1')
+    #cmd = "docker update --memory=10g {}".format(node1.docker_id)
+    logging.debug(f"get original memory limit {original_max_mem}")
+    run_and_check([
+        'docker', 'update', '--memory=10g', node1.docker_id
+    ])
+    for _ in range(30):
+        time.sleep(10)
+        new_max_mem =  get_latest_mem_limit()
+        logging.debug(f"get new memory limit {new_max_mem}")
+        if new_max_mem > original_max_mem:
+            return

From ebd934e28696c1d01f6ee225c62b4e76a026c337 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 13 Mar 2024 05:05:23 +0100
Subject: [PATCH 235/443] Continue

---
 src/Functions/FunctionsConversion.cpp | 390 +++++++++++++-------------
 1 file changed, 192 insertions(+), 198 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index e2d9c4a173e..846b6c31062 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -179,6 +179,131 @@ struct ToDateTimeImpl
 };
 
 
+/// Implementation of toDate function.
+
+template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ToDateTransform32Or64
+{
+    static constexpr auto name = "toDate";
+
+    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
+    {
+        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+        {
+            if (from > MAX_DATETIME_TIMESTAMP) [[unlikely]]
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type Date", from);
+        }
+        /// if value is smaller (or equal) than maximum day value for Date, than treat it as day num,
+        /// otherwise treat it as unix timestamp. This is a bit weird, but we leave this behavior.
+        if (from <= DATE_LUT_MAX_DAY_NUM)
+            return from;
+        else
+            return time_zone.toDayNum(std::min(time_t(from), time_t(MAX_DATETIME_TIMESTAMP)));
+    }
+};
+
+
+template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ToDateTransform32Or64Signed
+{
+    static constexpr auto name = "toDate";
+
+    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
+    {
+        // TODO: decide narrow or extended range based on FromType
+        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+        {
+            if (from < 0 || from > MAX_DATE_TIMESTAMP) [[unlikely]]
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type Date", from);
+        }
+        else
+        {
+            if (from < 0)
+                return 0;
+        }
+        return (from <= DATE_LUT_MAX_DAY_NUM)
+            ? static_cast<ToType>(from)
+            : time_zone.toDayNum(std::min(time_t(from), time_t(MAX_DATE_TIMESTAMP)));
+    }
+};
+
+template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ToDateTransform8Or16Signed
+{
+    static constexpr auto name = "toDate";
+
+    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
+    {
+        if (from < 0)
+        {
+            if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type Date", from);
+            else
+                return 0;
+        }
+        return from;
+    }
+};
+
+/// Implementation of toDate32 function.
+
+template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ToDate32Transform32Or64
+{
+    static constexpr auto name = "toDate32";
+
+    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
+    {
+        if (from < DATE_LUT_MAX_EXTEND_DAY_NUM)
+            return static_cast<ToType>(from);
+        else
+        {
+            if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+            {
+                if (from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
+                    throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type Date32", from);
+            }
+            return time_zone.toDayNum(std::min(time_t(from), time_t(MAX_DATETIME64_TIMESTAMP)));
+        }
+    }
+};
+
+template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ToDate32Transform32Or64Signed
+{
+    static constexpr auto name = "toDate32";
+
+    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
+    {
+        static const Int32 daynum_min_offset = -static_cast<Int32>(time_zone.getDayNumOffsetEpoch());
+
+        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+        {
+            if (from < daynum_min_offset || from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type Date32", from);
+        }
+
+        if (from < daynum_min_offset)
+            return daynum_min_offset;
+
+        return (from < DATE_LUT_MAX_EXTEND_DAY_NUM)
+            ? static_cast<ToType>(from)
+            : time_zone.toDayNum(std::min(time_t(Int64(from)), time_t(MAX_DATETIME64_TIMESTAMP)));
+    }
+};
+
+template <typename FromType, typename ToType>
+struct ToDate32Transform8Or16Signed
+{
+    static constexpr auto name = "toDate32";
+
+    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
+    {
+        return from;
+    }
+};
+
+
 /// Function toUnixTimestamp has exactly the same implementation as toDateTime of String type.
 struct NameToUnixTimestamp { static constexpr auto name = "toUnixTimestamp"; };
 
@@ -241,6 +366,73 @@ struct ConvertImpl
             return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeImpl<date_time_overflow_behavior>, false>::execute(
                 arguments, result_type, input_rows_count);
         }
+        /** Special case of converting Int8, Int16, (U)Int32 or (U)Int64 (and also, for convenience,
+          * Float32, Float64) to Date. If the
+          * number is less than 65536, then it is treated as DayNum, and if it's greater or equals to 65536,
+          * then treated as unix timestamp. If the number exceeds UInt32, saturate to MAX_UINT32 then as DayNum.
+          * It's a bit illogical, as we actually have two functions in one.
+          * But allows to support frequent case,
+          *  when user write toDate(UInt32), expecting conversion of unix timestamp to Date.
+          *  (otherwise such usage would be frequent mistake).
+          */
+        else if constexpr ((
+                std::is_same_v<FromDataType, DataTypeUInt32>
+                || std::is_same_v<FromDataType, DataTypeUInt64>)
+            && std::is_same_v<ToDataType, DataTypeDate>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform32Or64<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr ((
+                std::is_same_v<FromDataType, DataTypeInt8>
+                || std::is_same_v<FromDataType, DataTypeInt16>)
+            && std::is_same_v<ToDataType, DataTypeDate>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform8Or16Signed<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr ((
+                std::is_same_v<FromDataType, DataTypeInt32>
+                || std::is_same_v<FromDataType, DataTypeInt64>
+                || std::is_same_v<FromDataType, DataTypeFloat32>
+                || std::is_same_v<FromDataType, DataTypeFloat64>)
+            && std::is_same_v<ToDataType, DataTypeDate>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform32Or64Signed<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr ((
+                std::is_same_v<FromDataType, DataTypeUInt32>
+                || std::is_same_v<FromDataType, DataTypeUInt64>)
+            && std::is_same_v<ToDataType, DataTypeDate32>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform32Or64<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr ((
+                std::is_same_v<FromDataType, DataTypeInt8>
+                || std::is_same_v<FromDataType, DataTypeInt16>)
+            && std::is_same_v<ToDataType, DataTypeDate32>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform8Or16Signed<typename FromDataType::FieldType, UInt16>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr ((
+                std::is_same_v<FromDataType, DataTypeInt32>
+                || std::is_same_v<FromDataType, DataTypeInt64>
+                || std::is_same_v<FromDataType, DataTypeFloat32>
+                || std::is_same_v<FromDataType, DataTypeFloat64>)
+            && std::is_same_v<ToDataType, DataTypeDate32>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform32Or64Signed<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
         else
         {
             using ColVecFrom = typename FromDataType::ColumnType;
@@ -490,208 +682,10 @@ struct ConvertImpl
 };
 
 
-/// Implementation of toDate function.
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTransform32Or64
-{
-    static constexpr auto name = "toDate";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from > MAX_DATETIME_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type Date", from);
-        }
-        /// if value is smaller (or equal) than maximum day value for Date, than treat it as day num,
-        /// otherwise treat it as unix timestamp. This is a bit weird, but we leave this behavior.
-        if (from <= DATE_LUT_MAX_DAY_NUM)
-            return from;
-        else
-            return time_zone.toDayNum(std::min(time_t(from), time_t(MAX_DATETIME_TIMESTAMP)));
-    }
-};
-
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTransform32Or64Signed
-{
-    static constexpr auto name = "toDate";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
-    {
-        // TODO: decide narrow or extended range based on FromType
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from < 0 || from > MAX_DATE_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type Date", from);
-        }
-        else
-        {
-            if (from < 0)
-                return 0;
-        }
-        return (from <= DATE_LUT_MAX_DAY_NUM)
-            ? static_cast<ToType>(from)
-            : time_zone.toDayNum(std::min(time_t(from), time_t(MAX_DATE_TIMESTAMP)));
-    }
-};
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTransform8Or16Signed
-{
-    static constexpr auto name = "toDate";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
-    {
-        if (from < 0)
-        {
-            if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type Date", from);
-            else
-                return 0;
-        }
-        return from;
-    }
-};
-
 template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
 struct ConvertImpl<DataTypeDateTime64, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
         : DateTimeTransformImpl<DataTypeDateTime64, DataTypeDate32, TransformDateTime64<ToDate32Impl>, false> {};
 
-/// Implementation of toDate32 function.
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDate32Transform32Or64
-{
-    static constexpr auto name = "toDate32";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
-    {
-        if (from < DATE_LUT_MAX_EXTEND_DAY_NUM)
-            return static_cast<ToType>(from);
-        else
-        {
-            if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-            {
-                if (from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
-                    throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type Date32", from);
-            }
-            return time_zone.toDayNum(std::min(time_t(from), time_t(MAX_DATETIME64_TIMESTAMP)));
-        }
-    }
-};
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDate32Transform32Or64Signed
-{
-    static constexpr auto name = "toDate32";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
-    {
-        static const Int32 daynum_min_offset = -static_cast<Int32>(time_zone.getDayNumOffsetEpoch());
-
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from < daynum_min_offset || from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type Date32", from);
-        }
-
-        if (from < daynum_min_offset)
-            return daynum_min_offset;
-
-        return (from < DATE_LUT_MAX_EXTEND_DAY_NUM)
-            ? static_cast<ToType>(from)
-            : time_zone.toDayNum(std::min(time_t(Int64(from)), time_t(MAX_DATETIME64_TIMESTAMP)));
-    }
-};
-
-template <typename FromType, typename ToType>
-struct ToDate32Transform8Or16Signed
-{
-    static constexpr auto name = "toDate32";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
-    {
-        return from;
-    }
-};
-
-/** Special case of converting Int8, Int16, (U)Int32 or (U)Int64 (and also, for convenience,
-  * Float32, Float64) to Date. If the
-  * number is less than 65536, then it is treated as DayNum, and if it's greater or equals to 65536,
-  * then treated as unix timestamp. If the number exceeds UInt32, saturate to MAX_UINT32 then as DayNum.
-  * It's a bit illogical, as we actually have two functions in one.
-  * But allows to support frequent case,
-  *  when user write toDate(UInt32), expecting conversion of unix timestamp to Date.
-  *  (otherwise such usage would be frequent mistake).
-  */
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeUInt32, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeUInt32, DataTypeDate, ToDateTransform32Or64<UInt32, UInt16, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeUInt64, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeUInt64, DataTypeDate, ToDateTransform32Or64<UInt64, UInt16, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt8, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt8, DataTypeDate, ToDateTransform8Or16Signed<Int8, UInt16, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt16, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt16, DataTypeDate, ToDateTransform8Or16Signed<Int16, UInt16, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt32, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt32, DataTypeDate, ToDateTransform32Or64Signed<Int32, UInt16, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt64, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt64, DataTypeDate, ToDateTransform32Or64Signed<Int64, UInt16, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat32, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat32, DataTypeDate, ToDateTransform32Or64Signed<Float32, UInt16, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat64, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat64, DataTypeDate, ToDateTransform32Or64Signed<Float64, UInt16, default_date_time_overflow_behavior>, false> {};
-
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeUInt32, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeUInt32, DataTypeDate32, ToDate32Transform32Or64<UInt32, Int32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeUInt64, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeUInt64, DataTypeDate32, ToDate32Transform32Or64<UInt64, Int32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt8, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt8, DataTypeDate32, ToDate32Transform8Or16Signed<Int8, Int32>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt16, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt16, DataTypeDate32, ToDate32Transform8Or16Signed<Int16, Int32>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt32, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt32, DataTypeDate32, ToDate32Transform32Or64Signed<Int32, Int32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt64, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt64, DataTypeDate32, ToDate32Transform32Or64Signed<Int64, Int32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat32, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat32, DataTypeDate32, ToDate32Transform32Or64Signed<Float32, Int32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat64, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat64, DataTypeDate32, ToDate32Transform32Or64Signed<Float64, Int32, default_date_time_overflow_behavior>, false> {};
-
 
 template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
 struct ToDateTimeTransform64

From dbfda047014ab1b3e4cd81e46c5ce623387d61fe Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 13 Mar 2024 05:18:51 +0100
Subject: [PATCH 236/443] Continue

---
 src/Functions/FunctionsConversion.cpp | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 846b6c31062..33ef12b9d23 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -366,6 +366,12 @@ struct ConvertImpl
             return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeImpl<date_time_overflow_behavior>, false>::execute(
                 arguments, result_type, input_rows_count);
         }
+        else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime64> && std::is_same_v<ToDataType, DataTypeDate32>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<DataTypeDateTime64, DataTypeDate32, TransformDateTime64<ToDate32Impl>, false>::execute(
+                arguments, result_type, input_rows_count, TransformDateTime64<ToDate32Impl>(assert_cast<const DataTypeDateTime64 &>(*named_from.type).getScale()));
+        }
         /** Special case of converting Int8, Int16, (U)Int32 or (U)Int64 (and also, for convenience,
           * Float32, Float64) to Date. If the
           * number is less than 65536, then it is treated as DayNum, and if it's greater or equals to 65536,
@@ -682,11 +688,6 @@ struct ConvertImpl
 };
 
 
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDateTime64, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-        : DateTimeTransformImpl<DataTypeDateTime64, DataTypeDate32, TransformDateTime64<ToDate32Impl>, false> {};
-
-
 template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
 struct ToDateTimeTransform64
 {

From a5383a4619a18f0979594e25701f8f7893d62d56 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 13 Mar 2024 05:27:05 +0100
Subject: [PATCH 237/443] Continue

---
 src/Functions/FunctionsConversion.cpp | 410 +++++++++++++-------------
 1 file changed, 205 insertions(+), 205 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 33ef12b9d23..ffcb0086e27 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -303,6 +303,183 @@ struct ToDate32Transform8Or16Signed
     }
 };
 
+template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ToDateTimeTransform64
+{
+    static constexpr auto name = "toDateTime";
+
+    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
+    {
+        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+        {
+            if (from > MAX_DATETIME_TIMESTAMP) [[unlikely]]
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime", from);
+        }
+        return static_cast<ToType>(std::min(time_t(from), time_t(MAX_DATETIME_TIMESTAMP)));
+    }
+};
+
+template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ToDateTimeTransformSigned
+{
+    static constexpr auto name = "toDateTime";
+
+    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
+    {
+        if (from < 0)
+        {
+            if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime", from);
+            else
+                return 0;
+        }
+        return from;
+    }
+};
+
+template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ToDateTimeTransform64Signed
+{
+    static constexpr auto name = "toDateTime";
+
+    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
+    {
+        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+        {
+            if (from < 0 || from > MAX_DATETIME_TIMESTAMP) [[unlikely]]
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime", from);
+        }
+
+        if (from < 0)
+            return 0;
+        return static_cast<ToType>(std::min(time_t(from), time_t(MAX_DATETIME_TIMESTAMP)));
+    }
+};
+
+/** Conversion of numeric to DateTime64
+  */
+
+template <typename FromType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ToDateTime64TransformUnsigned
+{
+    static constexpr auto name = "toDateTime64";
+
+    const DateTime64::NativeType scale_multiplier = 1;
+
+    ToDateTime64TransformUnsigned(UInt32 scale = 0) /// NOLINT
+        : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
+    {}
+
+    NO_SANITIZE_UNDEFINED DateTime64::NativeType execute(FromType from, const DateLUTImpl &) const
+    {
+        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+        {
+            if (from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime64", from);
+            else
+                return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(from, 0, scale_multiplier);
+        }
+        else
+            return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(std::min<time_t>(from, MAX_DATETIME64_TIMESTAMP), 0, scale_multiplier);
+    }
+};
+template <typename FromType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ToDateTime64TransformSigned
+{
+    static constexpr auto name = "toDateTime64";
+
+    const DateTime64::NativeType scale_multiplier = 1;
+
+    ToDateTime64TransformSigned(UInt32 scale = 0) /// NOLINT
+        : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
+    {}
+
+    NO_SANITIZE_UNDEFINED DateTime64::NativeType execute(FromType from, const DateLUTImpl &) const
+    {
+        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+        {
+            if (from < MIN_DATETIME64_TIMESTAMP || from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime64", from);
+        }
+        from = static_cast<FromType>(std::max<time_t>(from, MIN_DATETIME64_TIMESTAMP));
+        from = static_cast<FromType>(std::min<time_t>(from, MAX_DATETIME64_TIMESTAMP));
+
+        return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(from, 0, scale_multiplier);
+    }
+};
+template <typename FromDataType, typename FromType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+struct ToDateTime64TransformFloat
+{
+    static constexpr auto name = "toDateTime64";
+
+    const UInt32 scale = 1;
+
+    ToDateTime64TransformFloat(UInt32 scale_ = 0) /// NOLINT
+        : scale(scale_)
+    {}
+
+    NO_SANITIZE_UNDEFINED DateTime64::NativeType execute(FromType from, const DateLUTImpl &) const
+    {
+        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
+        {
+            if (from < MIN_DATETIME64_TIMESTAMP || from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
+                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime64", from);
+        }
+
+        from = std::max(from, static_cast<FromType>(MIN_DATETIME64_TIMESTAMP));
+        from = std::min(from, static_cast<FromType>(MAX_DATETIME64_TIMESTAMP));
+        return convertToDecimal<FromDataType, DataTypeDateTime64>(from, scale);
+    }
+};
+
+/** Conversion of DateTime64 to Date or DateTime: discards fractional part.
+ */
+template <typename Transform>
+struct FromDateTime64Transform
+{
+    static constexpr auto name = Transform::name;
+
+    const DateTime64::NativeType scale_multiplier = 1;
+
+    FromDateTime64Transform(UInt32 scale) /// NOLINT
+        : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
+    {}
+
+    auto execute(DateTime64::NativeType dt, const DateLUTImpl & time_zone) const
+    {
+        const auto c = DecimalUtils::splitWithScaleMultiplier(DateTime64(dt), scale_multiplier);
+        return Transform::execute(static_cast<UInt32>(c.whole), time_zone);
+    }
+};
+
+struct ToDateTime64Transform
+{
+    static constexpr auto name = "toDateTime64";
+
+    const DateTime64::NativeType scale_multiplier = 1;
+
+    ToDateTime64Transform(UInt32 scale = 0) /// NOLINT
+        : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
+    {}
+
+    DateTime64::NativeType execute(UInt16 d, const DateLUTImpl & time_zone) const
+    {
+        const auto dt = ToDateTimeImpl<>::execute(d, time_zone);
+        return execute(dt, time_zone);
+    }
+
+    DateTime64::NativeType execute(Int32 d, const DateLUTImpl & time_zone) const
+    {
+        Int64 dt = static_cast<Int64>(time_zone.fromDayNum(ExtendedDayNum(d)));
+        return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(dt, 0, scale_multiplier);
+    }
+
+    DateTime64::NativeType execute(UInt32 dt, const DateLUTImpl & /*time_zone*/) const
+    {
+        return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(dt, 0, scale_multiplier);
+    }
+};
+
 
 /// Function toUnixTimestamp has exactly the same implementation as toDateTime of String type.
 struct NameToUnixTimestamp { static constexpr auto name = "toUnixTimestamp"; };
@@ -439,6 +616,34 @@ struct ConvertImpl
             return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform32Or64Signed<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::execute(
                 arguments, result_type, input_rows_count);
         }
+        /// Special case of converting Int8, Int16, Int32 or (U)Int64 (and also, for convenience, Float32, Float64) to DateTime.
+        else if constexpr ((
+                std::is_same_v<FromDataType, DataTypeInt8>
+                || std::is_same_v<FromDataType, DataTypeInt16>
+                || std::is_same_v<FromDataType, DataTypeInt32>)
+            && std::is_same_v<ToDataType, DataTypeDateTime>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeTransformSigned<typename FromDataType::FieldType, UInt32, default_date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr (std::is_same_v<FromDataType, DataTypeUInt64>
+            && std::is_same_v<ToDataType, DataTypeDateTime>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeTransform64<typename FromDataType::FieldType, UInt32, default_date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr ((
+                std::is_same_v<FromDataType, DataTypeInt64>
+                || std::is_same_v<FromDataType, DataTypeFloat32>
+                || std::is_same_v<FromDataType, DataTypeFloat64>)
+            && std::is_same_v<ToDataType, DataTypeDateTime>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeTransform64Signed<typename FromDataType::FieldType, UInt32, default_date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
         else
         {
             using ColVecFrom = typename FromDataType::ColumnType;
@@ -688,164 +893,6 @@ struct ConvertImpl
 };
 
 
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTimeTransform64
-{
-    static constexpr auto name = "toDateTime";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from > MAX_DATETIME_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime", from);
-        }
-        return static_cast<ToType>(std::min(time_t(from), time_t(MAX_DATETIME_TIMESTAMP)));
-    }
-};
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTimeTransformSigned
-{
-    static constexpr auto name = "toDateTime";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
-    {
-        if (from < 0)
-        {
-            if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime", from);
-            else
-                return 0;
-        }
-        return from;
-    }
-};
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTimeTransform64Signed
-{
-    static constexpr auto name = "toDateTime";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from < 0 || from > MAX_DATETIME_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime", from);
-        }
-
-        if (from < 0)
-            return 0;
-        return static_cast<ToType>(std::min(time_t(from), time_t(MAX_DATETIME_TIMESTAMP)));
-    }
-};
-
-/// Special case of converting Int8, Int16, Int32 or (U)Int64 (and also, for convenience, Float32, Float64) to DateTime.
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt8, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt8, DataTypeDateTime, ToDateTimeTransformSigned<Int8, UInt32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt16, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt16, DataTypeDateTime, ToDateTimeTransformSigned<Int16, UInt32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt32, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt32, DataTypeDateTime, ToDateTimeTransformSigned<Int32, UInt32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt64, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt64, DataTypeDateTime, ToDateTimeTransform64Signed<Int64, UInt32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeUInt64, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeUInt64, DataTypeDateTime, ToDateTimeTransform64<UInt64, UInt32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat32, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat32, DataTypeDateTime, ToDateTimeTransform64Signed<Float32, UInt32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat64, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat64, DataTypeDateTime, ToDateTimeTransform64Signed<Float64, UInt32, default_date_time_overflow_behavior>, false> {};
-
-/** Conversion of numeric to DateTime64
-  */
-
-template <typename FromType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTime64TransformUnsigned
-{
-    static constexpr auto name = "toDateTime64";
-
-    const DateTime64::NativeType scale_multiplier = 1;
-
-    ToDateTime64TransformUnsigned(UInt32 scale = 0) /// NOLINT
-        : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
-    {}
-
-    NO_SANITIZE_UNDEFINED DateTime64::NativeType execute(FromType from, const DateLUTImpl &) const
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime64", from);
-            else
-                return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(from, 0, scale_multiplier);
-        }
-        else
-            return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(std::min<time_t>(from, MAX_DATETIME64_TIMESTAMP), 0, scale_multiplier);
-    }
-};
-template <typename FromType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTime64TransformSigned
-{
-    static constexpr auto name = "toDateTime64";
-
-    const DateTime64::NativeType scale_multiplier = 1;
-
-    ToDateTime64TransformSigned(UInt32 scale = 0) /// NOLINT
-        : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
-    {}
-
-    NO_SANITIZE_UNDEFINED DateTime64::NativeType execute(FromType from, const DateLUTImpl &) const
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from < MIN_DATETIME64_TIMESTAMP || from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime64", from);
-        }
-        from = static_cast<FromType>(std::max<time_t>(from, MIN_DATETIME64_TIMESTAMP));
-        from = static_cast<FromType>(std::min<time_t>(from, MAX_DATETIME64_TIMESTAMP));
-
-        return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(from, 0, scale_multiplier);
-    }
-};
-template <typename FromDataType, typename FromType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTime64TransformFloat
-{
-    static constexpr auto name = "toDateTime64";
-
-    const UInt32 scale = 1;
-
-    ToDateTime64TransformFloat(UInt32 scale_ = 0) /// NOLINT
-        : scale(scale_)
-    {}
-
-    NO_SANITIZE_UNDEFINED DateTime64::NativeType execute(FromType from, const DateLUTImpl &) const
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from < MIN_DATETIME64_TIMESTAMP || from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime64", from);
-        }
-
-        from = std::max(from, static_cast<FromType>(MIN_DATETIME64_TIMESTAMP));
-        from = std::min(from, static_cast<FromType>(MAX_DATETIME64_TIMESTAMP));
-        return convertToDecimal<FromDataType, DataTypeDateTime64>(from, scale);
-    }
-};
-
 template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
 struct ConvertImpl<DataTypeInt8, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
     : DateTimeTransformImpl<DataTypeInt8, DataTypeDateTime64, ToDateTime64TransformSigned<Int8, date_time_overflow_behavior>, false> {};
@@ -875,26 +922,6 @@ struct ConvertImpl<DataTypeFloat64, DataTypeDateTime64, Name, ConvertDefaultBeha
     : DateTimeTransformImpl<DataTypeFloat64, DataTypeDateTime64, ToDateTime64TransformFloat<DataTypeFloat64, Float64, date_time_overflow_behavior>, false> {};
 
 
-/** Conversion of DateTime64 to Date or DateTime: discards fractional part.
- */
-template <typename Transform>
-struct FromDateTime64Transform
-{
-    static constexpr auto name = Transform::name;
-
-    const DateTime64::NativeType scale_multiplier = 1;
-
-    FromDateTime64Transform(UInt32 scale) /// NOLINT
-        : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
-    {}
-
-    auto execute(DateTime64::NativeType dt, const DateLUTImpl & time_zone) const
-    {
-        const auto c = DecimalUtils::splitWithScaleMultiplier(DateTime64(dt), scale_multiplier);
-        return Transform::execute(static_cast<UInt32>(c.whole), time_zone);
-    }
-};
-
 /** Conversion of DateTime64 to Date or DateTime: discards fractional part.
  */
 template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
@@ -905,33 +932,6 @@ template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_over
 struct ConvertImpl<DataTypeDateTime64, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
     : DateTimeTransformImpl<DataTypeDateTime64, DataTypeDateTime, TransformDateTime64<ToDateTimeImpl<date_time_overflow_behavior>>, false> {};
 
-struct ToDateTime64Transform
-{
-    static constexpr auto name = "toDateTime64";
-
-    const DateTime64::NativeType scale_multiplier = 1;
-
-    ToDateTime64Transform(UInt32 scale = 0) /// NOLINT
-        : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
-    {}
-
-    DateTime64::NativeType execute(UInt16 d, const DateLUTImpl & time_zone) const
-    {
-        const auto dt = ToDateTimeImpl<>::execute(d, time_zone);
-        return execute(dt, time_zone);
-    }
-
-    DateTime64::NativeType execute(Int32 d, const DateLUTImpl & time_zone) const
-    {
-        Int64 dt = static_cast<Int64>(time_zone.fromDayNum(ExtendedDayNum(d)));
-        return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(dt, 0, scale_multiplier);
-    }
-
-    DateTime64::NativeType execute(UInt32 dt, const DateLUTImpl & /*time_zone*/) const
-    {
-        return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(dt, 0, scale_multiplier);
-    }
-};
 
 /** Conversion of Date or DateTime to DateTime64: add zero sub-second part.
   */

From 5336a8911794662d6a207c5e4adda14c93871562 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 13 Mar 2024 05:36:09 +0100
Subject: [PATCH 238/443] Continue

---
 src/Functions/FunctionsConversion.cpp | 56 +++++++++++++--------------
 1 file changed, 27 insertions(+), 29 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index ffcb0086e27..af240f88560 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -644,6 +644,33 @@ struct ConvertImpl
             return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeTransform64Signed<typename FromDataType::FieldType, UInt32, default_date_time_overflow_behavior>, false>::execute(
                 arguments, result_type, input_rows_count);
         }
+        else if constexpr ((
+                std::is_same_v<FromDataType, DataTypeInt8>
+                || std::is_same_v<FromDataType, DataTypeInt16>
+                || std::is_same_v<FromDataType, DataTypeInt32>
+                || std::is_same_v<FromDataType, DataTypeInt64>)
+            && std::is_same_v<ToDataType, DataTypeDateTime64>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64TransformSigned<typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr (std::is_same_v<FromDataType, DataTypeUInt64>
+            && std::is_same_v<ToDataType, DataTypeDateTime64>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64TransformUnsigned<UInt64, default_date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr ((
+                std::is_same_v<FromDataType, DataTypeFloat32>
+                || std::is_same_v<FromDataType, DataTypeFloat64>)
+            && std::is_same_v<ToDataType, DataTypeDateTime64>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64TransformFloat<FromDataType, typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
         else
         {
             using ColVecFrom = typename FromDataType::ColumnType;
@@ -893,35 +920,6 @@ struct ConvertImpl
 };
 
 
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt8, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt8, DataTypeDateTime64, ToDateTime64TransformSigned<Int8, date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt16, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt16, DataTypeDateTime64, ToDateTime64TransformSigned<Int16, date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt32, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt32, DataTypeDateTime64, ToDateTime64TransformSigned<Int32, date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt64, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt64, DataTypeDateTime64, ToDateTime64TransformSigned<Int64, date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeUInt64, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeUInt64, DataTypeDateTime64, ToDateTime64TransformUnsigned<UInt64, date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat32, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat32, DataTypeDateTime64, ToDateTime64TransformFloat<DataTypeFloat32, Float32, date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat64, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat64, DataTypeDateTime64, ToDateTime64TransformFloat<DataTypeFloat64, Float64, date_time_overflow_behavior>, false> {};
-
-
 /** Conversion of DateTime64 to Date or DateTime: discards fractional part.
  */
 template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>

From 460692ac74d68a38bad1b4786b3bbabd377189e6 Mon Sep 17 00:00:00 2001
From: Yakov Olkhovskiy <99031427+yakov-olkhovskiy@users.noreply.github.com>
Date: Wed, 13 Mar 2024 00:36:23 -0400
Subject: [PATCH 239/443] Update
 docs/en/sql-reference/functions/other-functions.md

---
 docs/en/sql-reference/functions/other-functions.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/en/sql-reference/functions/other-functions.md b/docs/en/sql-reference/functions/other-functions.md
index df443eec0de..288905c83da 100644
--- a/docs/en/sql-reference/functions/other-functions.md
+++ b/docs/en/sql-reference/functions/other-functions.md
@@ -360,7 +360,7 @@ sleepEachRow(seconds)
 
 **Arguments**
 
-- `seconds`: [Int](../../sql-reference/data-types/int-uint.md) The number of seconds to pause the query execution for each row in the result set. It can be a floating-point value to specify fractional seconds.
+- `seconds`: [Int](../../sql-reference/data-types/int-uint.md) The number of seconds to pause the query execution for each row in the result set to a maximum of 3 seconds. It can be a floating-point value to specify fractional seconds.
 
 **Returned value**
 

From b870d9d1cd767eddcbe98cb31abe0167bbdab488 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 13 Mar 2024 05:45:33 +0100
Subject: [PATCH 240/443] Continue

---
 src/Functions/FunctionsConversion.cpp | 239 +++++++++++++-------------
 1 file changed, 119 insertions(+), 120 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index af240f88560..574e99bd6a6 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -433,7 +433,7 @@ struct ToDateTime64TransformFloat
 };
 
 /** Conversion of DateTime64 to Date or DateTime: discards fractional part.
- */
+  */
 template <typename Transform>
 struct FromDateTime64Transform
 {
@@ -480,6 +480,98 @@ struct ToDateTime64Transform
     }
 };
 
+/** Transformation of numbers, dates, datetimes to strings: through formatting.
+  */
+template <typename DataType>
+struct FormatImpl
+{
+    template <typename ReturnType = void>
+    static ReturnType execute(const typename DataType::FieldType x, WriteBuffer & wb, const DataType *, const DateLUTImpl *)
+    {
+        writeText(x, wb);
+        return ReturnType(true);
+    }
+};
+
+template <>
+struct FormatImpl<DataTypeDate>
+{
+    template <typename ReturnType = void>
+    static ReturnType execute(const DataTypeDate::FieldType x, WriteBuffer & wb, const DataTypeDate *, const DateLUTImpl * time_zone)
+    {
+        writeDateText(DayNum(x), wb, *time_zone);
+        return ReturnType(true);
+    }
+};
+
+template <>
+struct FormatImpl<DataTypeDate32>
+{
+    template <typename ReturnType = void>
+    static ReturnType execute(const DataTypeDate32::FieldType x, WriteBuffer & wb, const DataTypeDate32 *, const DateLUTImpl * time_zone)
+    {
+        writeDateText(ExtendedDayNum(x), wb, *time_zone);
+        return ReturnType(true);
+    }
+};
+
+template <>
+struct FormatImpl<DataTypeDateTime>
+{
+    template <typename ReturnType = void>
+    static ReturnType execute(const DataTypeDateTime::FieldType x, WriteBuffer & wb, const DataTypeDateTime *, const DateLUTImpl * time_zone)
+    {
+        writeDateTimeText(x, wb, *time_zone);
+        return ReturnType(true);
+    }
+};
+
+template <>
+struct FormatImpl<DataTypeDateTime64>
+{
+    template <typename ReturnType = void>
+    static ReturnType execute(const DataTypeDateTime64::FieldType x, WriteBuffer & wb, const DataTypeDateTime64 * type, const DateLUTImpl * time_zone)
+    {
+        writeDateTimeText(DateTime64(x), type->getScale(), wb, *time_zone);
+        return ReturnType(true);
+    }
+};
+
+
+template <typename FieldType>
+struct FormatImpl<DataTypeEnum<FieldType>>
+{
+    template <typename ReturnType = void>
+    static ReturnType execute(const FieldType x, WriteBuffer & wb, const DataTypeEnum<FieldType> * type, const DateLUTImpl *)
+    {
+        static constexpr bool throw_exception = std::is_same_v<ReturnType, void>;
+
+        if constexpr (throw_exception)
+        {
+            writeString(type->getNameForValue(x), wb);
+        }
+        else
+        {
+            StringRef res;
+            bool is_ok = type->getNameForValue(x, res);
+            if (is_ok)
+                writeString(res, wb);
+            return ReturnType(is_ok);
+        }
+    }
+};
+
+template <typename FieldType>
+struct FormatImpl<DataTypeDecimal<FieldType>>
+{
+    template <typename ReturnType = void>
+    static ReturnType execute(const FieldType x, WriteBuffer & wb, const DataTypeDecimal<FieldType> * type, const DateLUTImpl *)
+    {
+        writeText(x, type->getScale(), wb, false);
+        return ReturnType(true);
+    }
+};
+
 
 /// Function toUnixTimestamp has exactly the same implementation as toDateTime of String type.
 struct NameToUnixTimestamp { static constexpr auto name = "toUnixTimestamp"; };
@@ -671,6 +763,32 @@ struct ConvertImpl
             return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64TransformFloat<FromDataType, typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::execute(
                 arguments, result_type, input_rows_count);
         }
+        /// Conversion of DateTime64 to Date or DateTime: discards fractional part.
+        else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime64>
+            && std::is_same_v<ToDataType, DataTypeDate>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, TransformDateTime64<ToDateImpl<date_time_overflow_behavior>>, false>::execute(
+                arguments, result_type, input_rows_count, TransformDateTime64<ToDateImpl<date_time_overflow_behavior>>(assert_cast<const DataTypeDateTime64 &>(*named_from.type).getScale()));
+        }
+        else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime64>
+            && std::is_same_v<ToDataType, DataTypeDateTime>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, TransformDateTime64<ToDateTimeImpl<date_time_overflow_behavior>>, false>::execute(
+                arguments, result_type, input_rows_count, TransformDateTime64<ToDateTimeImpl<date_time_overflow_behavior>>(assert_cast<const DataTypeDateTime64 &>(*named_from.type).getScale()));
+        }
+        /// Conversion of Date or DateTime to DateTime64: add zero sub-second part.
+        else if constexpr ((
+                std::is_same_v<FromDataType, DataTypeDate>
+                || std::is_same_v<FromDataType, DataTypeDate32>
+                || std::is_same_v<FromDataType, DataTypeDateTime>)
+            && std::is_same_v<ToDataType, DataTypeDateTime64>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64Transform, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
         else
         {
             using ColVecFrom = typename FromDataType::ColumnType;
@@ -920,125 +1038,6 @@ struct ConvertImpl
 };
 
 
-/** Conversion of DateTime64 to Date or DateTime: discards fractional part.
- */
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDateTime64, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDateTime64, DataTypeDate, TransformDateTime64<ToDateImpl<date_time_overflow_behavior>>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDateTime64, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDateTime64, DataTypeDateTime, TransformDateTime64<ToDateTimeImpl<date_time_overflow_behavior>>, false> {};
-
-
-/** Conversion of Date or DateTime to DateTime64: add zero sub-second part.
-  */
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDate, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDate, DataTypeDateTime64, ToDateTime64Transform> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDate32, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDate32, DataTypeDateTime64, ToDateTime64Transform> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDateTime, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDateTime, DataTypeDateTime64, ToDateTime64Transform> {};
-
-
-/** Transformation of numbers, dates, datetimes to strings: through formatting.
-  */
-template <typename DataType>
-struct FormatImpl
-{
-    template <typename ReturnType = void>
-    static ReturnType execute(const typename DataType::FieldType x, WriteBuffer & wb, const DataType *, const DateLUTImpl *)
-    {
-        writeText(x, wb);
-        return ReturnType(true);
-    }
-};
-
-template <>
-struct FormatImpl<DataTypeDate>
-{
-    template <typename ReturnType = void>
-    static ReturnType execute(const DataTypeDate::FieldType x, WriteBuffer & wb, const DataTypeDate *, const DateLUTImpl * time_zone)
-    {
-        writeDateText(DayNum(x), wb, *time_zone);
-        return ReturnType(true);
-    }
-};
-
-template <>
-struct FormatImpl<DataTypeDate32>
-{
-    template <typename ReturnType = void>
-    static ReturnType execute(const DataTypeDate32::FieldType x, WriteBuffer & wb, const DataTypeDate32 *, const DateLUTImpl * time_zone)
-    {
-        writeDateText(ExtendedDayNum(x), wb, *time_zone);
-        return ReturnType(true);
-    }
-};
-
-template <>
-struct FormatImpl<DataTypeDateTime>
-{
-    template <typename ReturnType = void>
-    static ReturnType execute(const DataTypeDateTime::FieldType x, WriteBuffer & wb, const DataTypeDateTime *, const DateLUTImpl * time_zone)
-    {
-        writeDateTimeText(x, wb, *time_zone);
-        return ReturnType(true);
-    }
-};
-
-template <>
-struct FormatImpl<DataTypeDateTime64>
-{
-    template <typename ReturnType = void>
-    static ReturnType execute(const DataTypeDateTime64::FieldType x, WriteBuffer & wb, const DataTypeDateTime64 * type, const DateLUTImpl * time_zone)
-    {
-        writeDateTimeText(DateTime64(x), type->getScale(), wb, *time_zone);
-        return ReturnType(true);
-    }
-};
-
-
-template <typename FieldType>
-struct FormatImpl<DataTypeEnum<FieldType>>
-{
-    template <typename ReturnType = void>
-    static ReturnType execute(const FieldType x, WriteBuffer & wb, const DataTypeEnum<FieldType> * type, const DateLUTImpl *)
-    {
-        static constexpr bool throw_exception = std::is_same_v<ReturnType, void>;
-
-        if constexpr (throw_exception)
-        {
-            writeString(type->getNameForValue(x), wb);
-        }
-        else
-        {
-            StringRef res;
-            bool is_ok = type->getNameForValue(x, res);
-            if (is_ok)
-                writeString(res, wb);
-            return ReturnType(is_ok);
-        }
-    }
-};
-
-template <typename FieldType>
-struct FormatImpl<DataTypeDecimal<FieldType>>
-{
-    template <typename ReturnType = void>
-    static ReturnType execute(const FieldType x, WriteBuffer & wb, const DataTypeDecimal<FieldType> * type, const DateLUTImpl *)
-    {
-        writeText(x, type->getScale(), wb, false);
-        return ReturnType(true);
-    }
-};
-
-
 /// DataTypeEnum<T> to DataType<T> free conversion
 template <typename FieldType, typename Name>
 struct ConvertImpl<DataTypeEnum<FieldType>, DataTypeNumber<FieldType>, Name, ConvertDefaultBehaviorTag>

From 8d9a8fdfc1d91723513bd66184662a810e01b2fe Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 13 Mar 2024 05:52:01 +0100
Subject: [PATCH 241/443] Continue

---
 src/Functions/FunctionsConversion.cpp | 25 +++++++------------------
 1 file changed, 7 insertions(+), 18 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 574e99bd6a6..2ce508c4924 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -609,9 +609,12 @@ struct ConvertImpl
     {
         const ColumnWithTypeAndName & named_from = arguments[0];
 
-        if constexpr (std::is_same_v<FromDataType, ToDataType> && !FromDataType::is_parametric)
+        if constexpr ((std::is_same_v<FromDataType, ToDataType> && !FromDataType::is_parametric)
+            || (std::is_same_v<FromDataType, DataTypeEnum8> && std::is_same_v<ToDataType, DataTypeInt8>)
+            || (std::is_same_v<FromDataType, DataTypeEnum16> && std::is_same_v<ToDataType, DataTypeInt16>))
         {
             /// If types are the same, reuse the columns.
+            /// Conversions between Enum and the underlying type are also free.
             return named_from.column;
         }
         else if constexpr ((std::is_same_v<FromDataType, DataTypeDateTime> || std::is_same_v<FromDataType, DataTypeDate32>)
@@ -1038,25 +1041,11 @@ struct ConvertImpl
 };
 
 
-/// DataTypeEnum<T> to DataType<T> free conversion
-template <typename FieldType, typename Name>
-struct ConvertImpl<DataTypeEnum<FieldType>, DataTypeNumber<FieldType>, Name, ConvertDefaultBehaviorTag>
-{
-    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/)
-    {
-        return arguments[0].column;
-    }
-};
-
 inline ColumnUInt8::MutablePtr copyNullMap(ColumnPtr col)
 {
-    ColumnUInt8::MutablePtr null_map = nullptr;
-    if (const auto * col_null = checkAndGetColumn<ColumnNullable>(col.get()))
-    {
-        null_map = ColumnUInt8::create();
-        null_map->insertRangeFrom(col_null->getNullMapColumn(), 0, col_null->size());
-    }
-    return null_map;
+    if (const auto * col_nullable = checkAndGetColumn<ColumnNullable>(col.get()))
+        return col_nullable->getNullMapColumn().mutate();
+    return nullptr;
 }
 
 template <typename FromDataType, typename Name>

From 17a7696a378c5cfaad18214284e340f1d2306170 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 13 Mar 2024 06:06:07 +0100
Subject: [PATCH 242/443] Continue

---
 src/Functions/FunctionsConversion.cpp | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 2ce508c4924..56a0d74c60c 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -1041,11 +1041,15 @@ struct ConvertImpl
 };
 
 
-inline ColumnUInt8::MutablePtr copyNullMap(ColumnPtr col)
+ColumnUInt8::MutablePtr copyNullMap(ColumnPtr col)
 {
+    ColumnUInt8::MutablePtr null_map = nullptr;
     if (const auto * col_nullable = checkAndGetColumn<ColumnNullable>(col.get()))
-        return col_nullable->getNullMapColumn().mutate();
-    return nullptr;
+    {
+        null_map = ColumnUInt8::create();
+        null_map->insertRangeFrom(col_nullable->getNullMapColumn(), 0, col_nullable->size());
+    }
+    return null_map;
 }
 
 template <typename FromDataType, typename Name>

From ecd6b88831a4251acf3f62d52a43302fa1ef534c Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 13 Mar 2024 06:15:17 +0100
Subject: [PATCH 243/443] Continue

---
 src/Functions/FunctionsConversion.cpp | 1272 ++++++++++++-------------
 1 file changed, 635 insertions(+), 637 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 56a0d74c60c..16f547939fa 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -572,475 +572,6 @@ struct FormatImpl<DataTypeDecimal<FieldType>>
     }
 };
 
-
-/// Function toUnixTimestamp has exactly the same implementation as toDateTime of String type.
-struct NameToUnixTimestamp { static constexpr auto name = "toUnixTimestamp"; };
-
-struct AccurateConvertStrategyAdditions
-{
-    UInt32 scale { 0 };
-};
-
-struct AccurateOrNullConvertStrategyAdditions
-{
-    UInt32 scale { 0 };
-};
-
-
-struct ConvertDefaultBehaviorTag {};
-struct ConvertReturnNullOnErrorTag {};
-struct ConvertReturnZeroOnErrorTag {};
-
-/** Conversion of number types to each other, enums to numbers, dates and datetimes to numbers and back: done by straight assignment.
-  *  (Date is represented internally as number of days from some day; DateTime - as unix timestamp)
-  */
-template <typename FromDataType, typename ToDataType, typename Name,
-    typename SpecialTag = ConvertDefaultBehaviorTag,
-    FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior = default_date_time_overflow_behavior>
-struct ConvertImpl
-{
-    using FromFieldType = typename FromDataType::FieldType;
-    using ToFieldType = typename ToDataType::FieldType;
-
-    template <typename Additions = void *>
-    static ColumnPtr NO_SANITIZE_UNDEFINED execute(
-        const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type [[maybe_unused]], size_t input_rows_count,
-        Additions additions [[maybe_unused]] = Additions())
-    {
-        const ColumnWithTypeAndName & named_from = arguments[0];
-
-        if constexpr ((std::is_same_v<FromDataType, ToDataType> && !FromDataType::is_parametric)
-            || (std::is_same_v<FromDataType, DataTypeEnum8> && std::is_same_v<ToDataType, DataTypeInt8>)
-            || (std::is_same_v<FromDataType, DataTypeEnum16> && std::is_same_v<ToDataType, DataTypeInt16>))
-        {
-            /// If types are the same, reuse the columns.
-            /// Conversions between Enum and the underlying type are also free.
-            return named_from.column;
-        }
-        else if constexpr ((std::is_same_v<FromDataType, DataTypeDateTime> || std::is_same_v<FromDataType, DataTypeDate32>)
-            && std::is_same_v<ToDataType, DataTypeDate>)
-        {
-            /// Conversion of DateTime to Date: throw off time component.
-            /// Conversion of Date32 to Date.
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateImpl<date_time_overflow_behavior>, false>::execute(
-                arguments, result_type, input_rows_count);
-        }
-        else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime> && std::is_same_v<ToDataType, DataTypeDate32>)
-        {
-            /// Conversion of DateTime to Date: throw off time component.
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Impl, false>::execute(
-                arguments, result_type, input_rows_count);
-        }
-        else if constexpr ((std::is_same_v<FromDataType, DataTypeDate> || std::is_same_v<FromDataType, DataTypeDate32>)
-            && std::is_same_v<ToDataType, DataTypeDateTime>)
-        {
-            /// Conversion from Date/Date32 to DateTime.
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeImpl<date_time_overflow_behavior>, false>::execute(
-                arguments, result_type, input_rows_count);
-        }
-        else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime64> && std::is_same_v<ToDataType, DataTypeDate32>
-            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
-        {
-            return DateTimeTransformImpl<DataTypeDateTime64, DataTypeDate32, TransformDateTime64<ToDate32Impl>, false>::execute(
-                arguments, result_type, input_rows_count, TransformDateTime64<ToDate32Impl>(assert_cast<const DataTypeDateTime64 &>(*named_from.type).getScale()));
-        }
-        /** Special case of converting Int8, Int16, (U)Int32 or (U)Int64 (and also, for convenience,
-          * Float32, Float64) to Date. If the
-          * number is less than 65536, then it is treated as DayNum, and if it's greater or equals to 65536,
-          * then treated as unix timestamp. If the number exceeds UInt32, saturate to MAX_UINT32 then as DayNum.
-          * It's a bit illogical, as we actually have two functions in one.
-          * But allows to support frequent case,
-          *  when user write toDate(UInt32), expecting conversion of unix timestamp to Date.
-          *  (otherwise such usage would be frequent mistake).
-          */
-        else if constexpr ((
-                std::is_same_v<FromDataType, DataTypeUInt32>
-                || std::is_same_v<FromDataType, DataTypeUInt64>)
-            && std::is_same_v<ToDataType, DataTypeDate>
-            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
-        {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform32Or64<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::execute(
-                arguments, result_type, input_rows_count);
-        }
-        else if constexpr ((
-                std::is_same_v<FromDataType, DataTypeInt8>
-                || std::is_same_v<FromDataType, DataTypeInt16>)
-            && std::is_same_v<ToDataType, DataTypeDate>
-            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
-        {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform8Or16Signed<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::execute(
-                arguments, result_type, input_rows_count);
-        }
-        else if constexpr ((
-                std::is_same_v<FromDataType, DataTypeInt32>
-                || std::is_same_v<FromDataType, DataTypeInt64>
-                || std::is_same_v<FromDataType, DataTypeFloat32>
-                || std::is_same_v<FromDataType, DataTypeFloat64>)
-            && std::is_same_v<ToDataType, DataTypeDate>
-            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
-        {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform32Or64Signed<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::execute(
-                arguments, result_type, input_rows_count);
-        }
-        else if constexpr ((
-                std::is_same_v<FromDataType, DataTypeUInt32>
-                || std::is_same_v<FromDataType, DataTypeUInt64>)
-            && std::is_same_v<ToDataType, DataTypeDate32>
-            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
-        {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform32Or64<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::execute(
-                arguments, result_type, input_rows_count);
-        }
-        else if constexpr ((
-                std::is_same_v<FromDataType, DataTypeInt8>
-                || std::is_same_v<FromDataType, DataTypeInt16>)
-            && std::is_same_v<ToDataType, DataTypeDate32>
-            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
-        {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform8Or16Signed<typename FromDataType::FieldType, UInt16>, false>::execute(
-                arguments, result_type, input_rows_count);
-        }
-        else if constexpr ((
-                std::is_same_v<FromDataType, DataTypeInt32>
-                || std::is_same_v<FromDataType, DataTypeInt64>
-                || std::is_same_v<FromDataType, DataTypeFloat32>
-                || std::is_same_v<FromDataType, DataTypeFloat64>)
-            && std::is_same_v<ToDataType, DataTypeDate32>
-            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
-        {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform32Or64Signed<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::execute(
-                arguments, result_type, input_rows_count);
-        }
-        /// Special case of converting Int8, Int16, Int32 or (U)Int64 (and also, for convenience, Float32, Float64) to DateTime.
-        else if constexpr ((
-                std::is_same_v<FromDataType, DataTypeInt8>
-                || std::is_same_v<FromDataType, DataTypeInt16>
-                || std::is_same_v<FromDataType, DataTypeInt32>)
-            && std::is_same_v<ToDataType, DataTypeDateTime>
-            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
-        {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeTransformSigned<typename FromDataType::FieldType, UInt32, default_date_time_overflow_behavior>, false>::execute(
-                arguments, result_type, input_rows_count);
-        }
-        else if constexpr (std::is_same_v<FromDataType, DataTypeUInt64>
-            && std::is_same_v<ToDataType, DataTypeDateTime>
-            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
-        {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeTransform64<typename FromDataType::FieldType, UInt32, default_date_time_overflow_behavior>, false>::execute(
-                arguments, result_type, input_rows_count);
-        }
-        else if constexpr ((
-                std::is_same_v<FromDataType, DataTypeInt64>
-                || std::is_same_v<FromDataType, DataTypeFloat32>
-                || std::is_same_v<FromDataType, DataTypeFloat64>)
-            && std::is_same_v<ToDataType, DataTypeDateTime>
-            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
-        {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeTransform64Signed<typename FromDataType::FieldType, UInt32, default_date_time_overflow_behavior>, false>::execute(
-                arguments, result_type, input_rows_count);
-        }
-        else if constexpr ((
-                std::is_same_v<FromDataType, DataTypeInt8>
-                || std::is_same_v<FromDataType, DataTypeInt16>
-                || std::is_same_v<FromDataType, DataTypeInt32>
-                || std::is_same_v<FromDataType, DataTypeInt64>)
-            && std::is_same_v<ToDataType, DataTypeDateTime64>
-            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
-        {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64TransformSigned<typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::execute(
-                arguments, result_type, input_rows_count);
-        }
-        else if constexpr (std::is_same_v<FromDataType, DataTypeUInt64>
-            && std::is_same_v<ToDataType, DataTypeDateTime64>
-            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
-        {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64TransformUnsigned<UInt64, default_date_time_overflow_behavior>, false>::execute(
-                arguments, result_type, input_rows_count);
-        }
-        else if constexpr ((
-                std::is_same_v<FromDataType, DataTypeFloat32>
-                || std::is_same_v<FromDataType, DataTypeFloat64>)
-            && std::is_same_v<ToDataType, DataTypeDateTime64>
-            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
-        {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64TransformFloat<FromDataType, typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::execute(
-                arguments, result_type, input_rows_count);
-        }
-        /// Conversion of DateTime64 to Date or DateTime: discards fractional part.
-        else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime64>
-            && std::is_same_v<ToDataType, DataTypeDate>
-            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
-        {
-            return DateTimeTransformImpl<FromDataType, ToDataType, TransformDateTime64<ToDateImpl<date_time_overflow_behavior>>, false>::execute(
-                arguments, result_type, input_rows_count, TransformDateTime64<ToDateImpl<date_time_overflow_behavior>>(assert_cast<const DataTypeDateTime64 &>(*named_from.type).getScale()));
-        }
-        else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime64>
-            && std::is_same_v<ToDataType, DataTypeDateTime>
-            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
-        {
-            return DateTimeTransformImpl<FromDataType, ToDataType, TransformDateTime64<ToDateTimeImpl<date_time_overflow_behavior>>, false>::execute(
-                arguments, result_type, input_rows_count, TransformDateTime64<ToDateTimeImpl<date_time_overflow_behavior>>(assert_cast<const DataTypeDateTime64 &>(*named_from.type).getScale()));
-        }
-        /// Conversion of Date or DateTime to DateTime64: add zero sub-second part.
-        else if constexpr ((
-                std::is_same_v<FromDataType, DataTypeDate>
-                || std::is_same_v<FromDataType, DataTypeDate32>
-                || std::is_same_v<FromDataType, DataTypeDateTime>)
-            && std::is_same_v<ToDataType, DataTypeDateTime64>
-            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
-        {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64Transform, false>::execute(
-                arguments, result_type, input_rows_count);
-        }
-        else
-        {
-            using ColVecFrom = typename FromDataType::ColumnType;
-            using ColVecTo = typename ToDataType::ColumnType;
-
-            if constexpr ((IsDataTypeDecimal<FromDataType> || IsDataTypeDecimal<ToDataType>)
-                && !(std::is_same_v<DataTypeDateTime64, FromDataType> || std::is_same_v<DataTypeDateTime64, ToDataType>)
-                && (!IsDataTypeDecimalOrNumber<FromDataType> || !IsDataTypeDecimalOrNumber<ToDataType>))
-            {
-                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                    named_from.column->getName(), Name::name);
-            }
-
-            const ColVecFrom * col_from = checkAndGetColumn<ColVecFrom>(named_from.column.get());
-            if (!col_from)
-                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                    named_from.column->getName(), Name::name);
-
-            typename ColVecTo::MutablePtr col_to = nullptr;
-
-            if constexpr (IsDataTypeDecimal<ToDataType>)
-            {
-                UInt32 scale;
-
-                if constexpr (std::is_same_v<Additions, AccurateConvertStrategyAdditions>
-                    || std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-                {
-                    scale = additions.scale;
-                }
-                else
-                {
-                    scale = additions;
-                }
-
-                col_to = ColVecTo::create(0, scale);
-            }
-            else
-                col_to = ColVecTo::create();
-
-            const auto & vec_from = col_from->getData();
-            auto & vec_to = col_to->getData();
-            vec_to.resize(input_rows_count);
-
-            ColumnUInt8::MutablePtr col_null_map_to;
-            ColumnUInt8::Container * vec_null_map_to [[maybe_unused]] = nullptr;
-            if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-            {
-                col_null_map_to = ColumnUInt8::create(input_rows_count, false);
-                vec_null_map_to = &col_null_map_to->getData();
-            }
-
-            bool result_is_bool = isBool(result_type);
-            for (size_t i = 0; i < input_rows_count; ++i)
-            {
-                if constexpr (std::is_same_v<ToDataType, DataTypeUInt8>)
-                {
-                    if (result_is_bool)
-                    {
-                        vec_to[i] = vec_from[i] != FromFieldType(0);
-                        continue;
-                    }
-                }
-
-                if constexpr (std::is_same_v<FromDataType, DataTypeUUID> && std::is_same_v<ToDataType, DataTypeUInt128>)
-                {
-                    static_assert(
-                        std::is_same_v<DataTypeUInt128::FieldType, DataTypeUUID::FieldType::UnderlyingType>,
-                        "UInt128 and UUID types must be same");
-
-                    vec_to[i].items[1] = vec_from[i].toUnderType().items[0];
-                    vec_to[i].items[0] = vec_from[i].toUnderType().items[1];
-                }
-                else if constexpr (std::is_same_v<FromDataType, DataTypeIPv6> && std::is_same_v<ToDataType, DataTypeUInt128>)
-                {
-                    static_assert(
-                        std::is_same_v<DataTypeUInt128::FieldType, DataTypeUUID::FieldType::UnderlyingType>,
-                        "UInt128 and IPv6 types must be same");
-
-                    vec_to[i].items[1] = std::byteswap(vec_from[i].toUnderType().items[0]);
-                    vec_to[i].items[0] = std::byteswap(vec_from[i].toUnderType().items[1]);
-                }
-                else if constexpr (std::is_same_v<FromDataType, DataTypeUUID> != std::is_same_v<ToDataType, DataTypeUUID>)
-                {
-                    throw Exception(ErrorCodes::NOT_IMPLEMENTED,
-                                    "Conversion between numeric types and UUID is not supported. "
-                                    "Probably the passed UUID is unquoted");
-                }
-                else if constexpr (
-                    (std::is_same_v<FromDataType, DataTypeIPv4> != std::is_same_v<ToDataType, DataTypeIPv4>)
-                    && !(is_any_of<FromDataType, DataTypeUInt8, DataTypeUInt16, DataTypeUInt32, DataTypeUInt64, DataTypeIPv6>
-                        || is_any_of<ToDataType, DataTypeUInt32, DataTypeUInt64, DataTypeUInt128, DataTypeUInt256, DataTypeIPv6>))
-                {
-                    throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Conversion from {} to {} is not supported",
-                                    TypeName<typename FromDataType::FieldType>, TypeName<typename ToDataType::FieldType>);
-                }
-                else if constexpr (std::is_same_v<FromDataType, DataTypeIPv6> != std::is_same_v<ToDataType, DataTypeIPv6>
-                    && !(std::is_same_v<ToDataType, DataTypeIPv4> || std::is_same_v<FromDataType, DataTypeIPv4>))
-                {
-                    throw Exception(ErrorCodes::NOT_IMPLEMENTED,
-                                    "Conversion between numeric types and IPv6 is not supported. "
-                                    "Probably the passed IPv6 is unquoted");
-                }
-                else if constexpr (IsDataTypeDecimal<FromDataType> || IsDataTypeDecimal<ToDataType>)
-                {
-                    if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-                    {
-                        ToFieldType result;
-                        bool convert_result = false;
-
-                        if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeDecimal<ToDataType>)
-                            convert_result = tryConvertDecimals<FromDataType, ToDataType>(vec_from[i], col_from->getScale(), col_to->getScale(), result);
-                        else if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeNumber<ToDataType>)
-                            convert_result = tryConvertFromDecimal<FromDataType, ToDataType>(vec_from[i], col_from->getScale(), result);
-                        else if constexpr (IsDataTypeNumber<FromDataType> && IsDataTypeDecimal<ToDataType>)
-                            convert_result = tryConvertToDecimal<FromDataType, ToDataType>(vec_from[i], col_to->getScale(), result);
-
-                        if (convert_result)
-                            vec_to[i] = result;
-                        else
-                        {
-                            vec_to[i] = static_cast<ToFieldType>(0);
-                            (*vec_null_map_to)[i] = true;
-                        }
-                    }
-                    else
-                    {
-                        if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeDecimal<ToDataType>)
-                            vec_to[i] = convertDecimals<FromDataType, ToDataType>(vec_from[i], col_from->getScale(), col_to->getScale());
-                        else if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeNumber<ToDataType>)
-                            vec_to[i] = convertFromDecimal<FromDataType, ToDataType>(vec_from[i], col_from->getScale());
-                        else if constexpr (IsDataTypeNumber<FromDataType> && IsDataTypeDecimal<ToDataType>)
-                            vec_to[i] = convertToDecimal<FromDataType, ToDataType>(vec_from[i], col_to->getScale());
-                        else
-                            throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Unsupported data type in conversion function");
-                    }
-                }
-                else if constexpr (std::is_same_v<ToDataType, DataTypeIPv4> && std::is_same_v<FromDataType, DataTypeIPv6>)
-                {
-                    const uint8_t ip4_cidr[] {0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x00, 0x00, 0x00, 0x00};
-                    const uint8_t * src = reinterpret_cast<const uint8_t *>(&vec_from[i].toUnderType());
-                    if (!matchIPv6Subnet(src, ip4_cidr, 96))
-                    {
-                        char addr[IPV6_MAX_TEXT_LENGTH + 1] {};
-                        char * paddr = addr;
-                        formatIPv6(src, paddr);
-
-                        throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "IPv6 {} in column {} is not in IPv4 mapping block", addr, named_from.column->getName());
-                    }
-
-                    uint8_t * dst = reinterpret_cast<uint8_t *>(&vec_to[i].toUnderType());
-                    if constexpr (std::endian::native == std::endian::little)
-                    {
-                        dst[0] = src[15];
-                        dst[1] = src[14];
-                        dst[2] = src[13];
-                        dst[3] = src[12];
-                    }
-                    else
-                    {
-                        dst[0] = src[12];
-                        dst[1] = src[13];
-                        dst[2] = src[14];
-                        dst[3] = src[15];
-                    }
-                }
-                else if constexpr (std::is_same_v<ToDataType, DataTypeIPv6> && std::is_same_v<FromDataType, DataTypeIPv4>)
-                {
-                    const uint8_t * src = reinterpret_cast<const uint8_t *>(&vec_from[i].toUnderType());
-                    uint8_t * dst = reinterpret_cast<uint8_t *>(&vec_to[i].toUnderType());
-                    std::memset(dst, '\0', IPV6_BINARY_LENGTH);
-                    dst[10] = dst[11] = 0xff;
-
-                    if constexpr (std::endian::native == std::endian::little)
-                    {
-                        dst[12] = src[3];
-                        dst[13] = src[2];
-                        dst[14] = src[1];
-                        dst[15] = src[0];
-                    }
-                    else
-                    {
-                        dst[12] = src[0];
-                        dst[13] = src[1];
-                        dst[14] = src[2];
-                        dst[15] = src[3];
-                    }
-                }
-                else if constexpr (std::is_same_v<ToDataType, DataTypeIPv4> && std::is_same_v<FromDataType, DataTypeUInt64>)
-                {
-                    vec_to[i] = static_cast<ToFieldType>(static_cast<IPv4::UnderlyingType>(vec_from[i]));
-                }
-                else if constexpr (std::is_same_v<Name, NameToUnixTimestamp>
-                    && (std::is_same_v<FromDataType, DataTypeDate> || std::is_same_v<FromDataType, DataTypeDate32>))
-                {
-                    vec_to[i] = static_cast<ToFieldType>(vec_from[i] * DATE_SECONDS_PER_DAY);
-                }
-                else
-                {
-                    /// If From Data is Nan or Inf and we convert to integer type, throw exception
-                    if constexpr (std::is_floating_point_v<FromFieldType> && !std::is_floating_point_v<ToFieldType>)
-                    {
-                        if (!isFinite(vec_from[i]))
-                        {
-                            if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-                            {
-                                vec_to[i] = 0;
-                                (*vec_null_map_to)[i] = true;
-                                continue;
-                            }
-                            else
-                                throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Unexpected inf or nan to integer conversion");
-                        }
-                    }
-
-                    if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>
-                            || std::is_same_v<Additions, AccurateConvertStrategyAdditions>)
-                    {
-                        bool convert_result = accurate::convertNumeric(vec_from[i], vec_to[i]);
-
-                        if (!convert_result)
-                        {
-                            if (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-                            {
-                                vec_to[i] = 0;
-                                (*vec_null_map_to)[i] = true;
-                            }
-                            else
-                            {
-                                throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Value in column {} cannot be safely converted into type {}",
-                                    named_from.column->getName(), result_type->getName());
-                            }
-                        }
-                    }
-                    else
-                    {
-                        vec_to[i] = static_cast<ToFieldType>(vec_from[i]);
-                    }
-                }
-            }
-
-            if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-                return ColumnNullable::create(std::move(col_to), std::move(col_null_map_to));
-            else
-                return col_to;
-        }
-    }
-};
-
-
 ColumnUInt8::MutablePtr copyNullMap(ColumnPtr col)
 {
     ColumnUInt8::MutablePtr null_map = nullptr;
@@ -1052,174 +583,6 @@ ColumnUInt8::MutablePtr copyNullMap(ColumnPtr col)
     return null_map;
 }
 
-template <typename FromDataType, typename Name>
-requires (!std::is_same_v<FromDataType, DataTypeString>)
-struct ConvertImpl<FromDataType, DataTypeString, Name, ConvertDefaultBehaviorTag>
-{
-    using FromFieldType = typename FromDataType::FieldType;
-    using ColVecType = ColumnVectorOrDecimal<FromFieldType>;
-
-    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/)
-    {
-        if constexpr (IsDataTypeDateOrDateTime<FromDataType>)
-        {
-            auto datetime_arg = arguments[0];
-
-            const DateLUTImpl * time_zone = nullptr;
-            const ColumnConst * time_zone_column = nullptr;
-
-            if (arguments.size() == 1)
-            {
-                auto non_null_args = createBlockWithNestedColumns(arguments);
-                time_zone = &extractTimeZoneFromFunctionArguments(non_null_args, 1, 0);
-            }
-            else /// When we have a column for timezone
-            {
-                datetime_arg.column = datetime_arg.column->convertToFullColumnIfConst();
-
-                if constexpr (std::is_same_v<FromDataType, DataTypeDate> || std::is_same_v<FromDataType, DataTypeDate32>)
-                    time_zone = &DateLUT::instance();
-                /// For argument of Date or DateTime type, second argument with time zone could be specified.
-                if constexpr (std::is_same_v<FromDataType, DataTypeDateTime> || std::is_same_v<FromDataType, DataTypeDateTime64>)
-                {
-                    if ((time_zone_column = checkAndGetColumnConst<ColumnString>(arguments[1].column.get())))
-                    {
-                        auto non_null_args = createBlockWithNestedColumns(arguments);
-                        time_zone = &extractTimeZoneFromFunctionArguments(non_null_args, 1, 0);
-                    }
-                }
-            }
-            const auto & col_with_type_and_name = columnGetNested(datetime_arg);
-
-            if (const auto col_from = checkAndGetColumn<ColVecType>(col_with_type_and_name.column.get()))
-            {
-                auto col_to = ColumnString::create();
-
-                const typename ColVecType::Container & vec_from = col_from->getData();
-                ColumnString::Chars & data_to = col_to->getChars();
-                ColumnString::Offsets & offsets_to = col_to->getOffsets();
-                size_t size = vec_from.size();
-
-                if constexpr (std::is_same_v<FromDataType, DataTypeDate>)
-                    data_to.resize(size * (strlen("YYYY-MM-DD") + 1));
-                else if constexpr (std::is_same_v<FromDataType, DataTypeDate32>)
-                    data_to.resize(size * (strlen("YYYY-MM-DD") + 1));
-                else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime>)
-                    data_to.resize(size * (strlen("YYYY-MM-DD hh:mm:ss") + 1));
-                else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime64>)
-                    data_to.resize(size * (strlen("YYYY-MM-DD hh:mm:ss.") + col_from->getScale() + 1));
-                else
-                    data_to.resize(size * 3);   /// Arbitrary
-
-                offsets_to.resize(size);
-
-                WriteBufferFromVector<ColumnString::Chars> write_buffer(data_to);
-                const auto & type = static_cast<const FromDataType &>(*col_with_type_and_name.type);
-
-                ColumnUInt8::MutablePtr null_map = copyNullMap(datetime_arg.column);
-
-                if (!null_map && arguments.size() > 1)
-                    null_map = copyNullMap(arguments[1].column->convertToFullColumnIfConst());
-
-                if (null_map)
-                {
-                    for (size_t i = 0; i < size; ++i)
-                    {
-                        if (!time_zone_column && arguments.size() > 1)
-                        {
-                            if (!arguments[1].column.get()->getDataAt(i).toString().empty())
-                                time_zone = &DateLUT::instance(arguments[1].column.get()->getDataAt(i).toString());
-                            else
-                                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Provided time zone must be non-empty");
-                        }
-                        bool is_ok = FormatImpl<FromDataType>::template execute<bool>(vec_from[i], write_buffer, &type, time_zone);
-                        null_map->getData()[i] |= !is_ok;
-                        writeChar(0, write_buffer);
-                        offsets_to[i] = write_buffer.count();
-                    }
-                }
-                else
-                {
-                    for (size_t i = 0; i < size; ++i)
-                    {
-                        if (!time_zone_column && arguments.size() > 1)
-                        {
-                            if (!arguments[1].column.get()->getDataAt(i).toString().empty())
-                                time_zone = &DateLUT::instance(arguments[1].column.get()->getDataAt(i).toString());
-                            else
-                                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Provided time zone must be non-empty");
-                        }
-                        FormatImpl<FromDataType>::template execute<void>(vec_from[i], write_buffer, &type, time_zone);
-                        writeChar(0, write_buffer);
-                        offsets_to[i] = write_buffer.count();
-                    }
-                }
-
-                write_buffer.finalize();
-
-                if (null_map)
-                    return ColumnNullable::create(std::move(col_to), std::move(null_map));
-                return col_to;
-            }
-            else
-                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                        arguments[0].column->getName(), Name::name);
-        }
-        else
-        {
-            ColumnUInt8::MutablePtr null_map = copyNullMap(arguments[0].column);
-
-            const auto & col_with_type_and_name = columnGetNested(arguments[0]);
-            const auto & type = static_cast<const FromDataType &>(*col_with_type_and_name.type);
-
-            if (const auto col_from = checkAndGetColumn<ColVecType>(col_with_type_and_name.column.get()))
-            {
-                auto col_to = ColumnString::create();
-
-                const typename ColVecType::Container & vec_from = col_from->getData();
-                ColumnString::Chars & data_to = col_to->getChars();
-                ColumnString::Offsets & offsets_to = col_to->getOffsets();
-                size_t size = vec_from.size();
-
-                data_to.resize(size * 3);
-                offsets_to.resize(size);
-
-                WriteBufferFromVector<ColumnString::Chars> write_buffer(data_to);
-
-                if (null_map)
-                {
-                    for (size_t i = 0; i < size; ++i)
-                    {
-                        bool is_ok = FormatImpl<FromDataType>::template execute<bool>(vec_from[i], write_buffer, &type, nullptr);
-                        /// We don't use timezones in this branch
-                        null_map->getData()[i] |= !is_ok;
-                        writeChar(0, write_buffer);
-                        offsets_to[i] = write_buffer.count();
-                    }
-                }
-                else
-                {
-                    for (size_t i = 0; i < size; ++i)
-                    {
-                        FormatImpl<FromDataType>::template execute<void>(vec_from[i], write_buffer, &type, nullptr);
-                        writeChar(0, write_buffer);
-                        offsets_to[i] = write_buffer.count();
-                    }
-                }
-
-                write_buffer.finalize();
-
-                if (null_map)
-                    return ColumnNullable::create(std::move(col_to), std::move(null_map));
-                return col_to;
-            }
-            else
-                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                        arguments[0].column->getName(), Name::name);
-        }
-    }
-};
-
 
 /// Generic conversion of any type to String or FixedString via serialization to text.
 template <typename StringColumnType>
@@ -1774,6 +1137,641 @@ struct ConvertThroughParsing
 };
 
 
+/// Function toUnixTimestamp has exactly the same implementation as toDateTime of String type.
+struct NameToUnixTimestamp { static constexpr auto name = "toUnixTimestamp"; };
+
+struct AccurateConvertStrategyAdditions
+{
+    UInt32 scale { 0 };
+};
+
+struct AccurateOrNullConvertStrategyAdditions
+{
+    UInt32 scale { 0 };
+};
+
+
+struct ConvertDefaultBehaviorTag {};
+struct ConvertReturnNullOnErrorTag {};
+struct ConvertReturnZeroOnErrorTag {};
+
+/** Conversion of number types to each other, enums to numbers, dates and datetimes to numbers and back: done by straight assignment.
+  *  (Date is represented internally as number of days from some day; DateTime - as unix timestamp)
+  */
+template <typename FromDataType, typename ToDataType, typename Name,
+    typename SpecialTag = ConvertDefaultBehaviorTag,
+    FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior = default_date_time_overflow_behavior>
+struct ConvertImpl
+{
+    using FromFieldType = typename FromDataType::FieldType;
+    using ToFieldType = typename ToDataType::FieldType;
+
+    template <typename Additions = void *>
+    static ColumnPtr NO_SANITIZE_UNDEFINED execute(
+        const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type [[maybe_unused]], size_t input_rows_count,
+        Additions additions [[maybe_unused]] = Additions())
+    {
+        const ColumnWithTypeAndName & named_from = arguments[0];
+
+        if constexpr ((std::is_same_v<FromDataType, ToDataType> && !FromDataType::is_parametric)
+            || (std::is_same_v<FromDataType, DataTypeEnum8> && std::is_same_v<ToDataType, DataTypeInt8>)
+            || (std::is_same_v<FromDataType, DataTypeEnum16> && std::is_same_v<ToDataType, DataTypeInt16>))
+        {
+            /// If types are the same, reuse the columns.
+            /// Conversions between Enum and the underlying type are also free.
+            return named_from.column;
+        }
+        else if constexpr ((std::is_same_v<FromDataType, DataTypeDateTime> || std::is_same_v<FromDataType, DataTypeDate32>)
+            && std::is_same_v<ToDataType, DataTypeDate>)
+        {
+            /// Conversion of DateTime to Date: throw off time component.
+            /// Conversion of Date32 to Date.
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateImpl<date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime> && std::is_same_v<ToDataType, DataTypeDate32>)
+        {
+            /// Conversion of DateTime to Date: throw off time component.
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Impl, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr ((std::is_same_v<FromDataType, DataTypeDate> || std::is_same_v<FromDataType, DataTypeDate32>)
+            && std::is_same_v<ToDataType, DataTypeDateTime>)
+        {
+            /// Conversion from Date/Date32 to DateTime.
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeImpl<date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime64> && std::is_same_v<ToDataType, DataTypeDate32>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<DataTypeDateTime64, DataTypeDate32, TransformDateTime64<ToDate32Impl>, false>::execute(
+                arguments, result_type, input_rows_count, TransformDateTime64<ToDate32Impl>(assert_cast<const DataTypeDateTime64 &>(*named_from.type).getScale()));
+        }
+        /** Special case of converting Int8, Int16, (U)Int32 or (U)Int64 (and also, for convenience,
+          * Float32, Float64) to Date. If the
+          * number is less than 65536, then it is treated as DayNum, and if it's greater or equals to 65536,
+          * then treated as unix timestamp. If the number exceeds UInt32, saturate to MAX_UINT32 then as DayNum.
+          * It's a bit illogical, as we actually have two functions in one.
+          * But allows to support frequent case,
+          *  when user write toDate(UInt32), expecting conversion of unix timestamp to Date.
+          *  (otherwise such usage would be frequent mistake).
+          */
+        else if constexpr ((
+                std::is_same_v<FromDataType, DataTypeUInt32>
+                || std::is_same_v<FromDataType, DataTypeUInt64>)
+            && std::is_same_v<ToDataType, DataTypeDate>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform32Or64<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr ((
+                std::is_same_v<FromDataType, DataTypeInt8>
+                || std::is_same_v<FromDataType, DataTypeInt16>)
+            && std::is_same_v<ToDataType, DataTypeDate>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform8Or16Signed<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr ((
+                std::is_same_v<FromDataType, DataTypeInt32>
+                || std::is_same_v<FromDataType, DataTypeInt64>
+                || std::is_same_v<FromDataType, DataTypeFloat32>
+                || std::is_same_v<FromDataType, DataTypeFloat64>)
+            && std::is_same_v<ToDataType, DataTypeDate>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform32Or64Signed<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr ((
+                std::is_same_v<FromDataType, DataTypeUInt32>
+                || std::is_same_v<FromDataType, DataTypeUInt64>)
+            && std::is_same_v<ToDataType, DataTypeDate32>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform32Or64<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr ((
+                std::is_same_v<FromDataType, DataTypeInt8>
+                || std::is_same_v<FromDataType, DataTypeInt16>)
+            && std::is_same_v<ToDataType, DataTypeDate32>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform8Or16Signed<typename FromDataType::FieldType, UInt16>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr ((
+                std::is_same_v<FromDataType, DataTypeInt32>
+                || std::is_same_v<FromDataType, DataTypeInt64>
+                || std::is_same_v<FromDataType, DataTypeFloat32>
+                || std::is_same_v<FromDataType, DataTypeFloat64>)
+            && std::is_same_v<ToDataType, DataTypeDate32>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform32Or64Signed<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        /// Special case of converting Int8, Int16, Int32 or (U)Int64 (and also, for convenience, Float32, Float64) to DateTime.
+        else if constexpr ((
+                std::is_same_v<FromDataType, DataTypeInt8>
+                || std::is_same_v<FromDataType, DataTypeInt16>
+                || std::is_same_v<FromDataType, DataTypeInt32>)
+            && std::is_same_v<ToDataType, DataTypeDateTime>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeTransformSigned<typename FromDataType::FieldType, UInt32, default_date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr (std::is_same_v<FromDataType, DataTypeUInt64>
+            && std::is_same_v<ToDataType, DataTypeDateTime>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeTransform64<typename FromDataType::FieldType, UInt32, default_date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr ((
+                std::is_same_v<FromDataType, DataTypeInt64>
+                || std::is_same_v<FromDataType, DataTypeFloat32>
+                || std::is_same_v<FromDataType, DataTypeFloat64>)
+            && std::is_same_v<ToDataType, DataTypeDateTime>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeTransform64Signed<typename FromDataType::FieldType, UInt32, default_date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr ((
+                std::is_same_v<FromDataType, DataTypeInt8>
+                || std::is_same_v<FromDataType, DataTypeInt16>
+                || std::is_same_v<FromDataType, DataTypeInt32>
+                || std::is_same_v<FromDataType, DataTypeInt64>)
+            && std::is_same_v<ToDataType, DataTypeDateTime64>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64TransformSigned<typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr (std::is_same_v<FromDataType, DataTypeUInt64>
+            && std::is_same_v<ToDataType, DataTypeDateTime64>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64TransformUnsigned<UInt64, default_date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr ((
+                std::is_same_v<FromDataType, DataTypeFloat32>
+                || std::is_same_v<FromDataType, DataTypeFloat64>)
+            && std::is_same_v<ToDataType, DataTypeDateTime64>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64TransformFloat<FromDataType, typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        /// Conversion of DateTime64 to Date or DateTime: discards fractional part.
+        else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime64>
+            && std::is_same_v<ToDataType, DataTypeDate>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, TransformDateTime64<ToDateImpl<date_time_overflow_behavior>>, false>::execute(
+                arguments, result_type, input_rows_count, TransformDateTime64<ToDateImpl<date_time_overflow_behavior>>(assert_cast<const DataTypeDateTime64 &>(*named_from.type).getScale()));
+        }
+        else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime64>
+            && std::is_same_v<ToDataType, DataTypeDateTime>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, TransformDateTime64<ToDateTimeImpl<date_time_overflow_behavior>>, false>::execute(
+                arguments, result_type, input_rows_count, TransformDateTime64<ToDateTimeImpl<date_time_overflow_behavior>>(assert_cast<const DataTypeDateTime64 &>(*named_from.type).getScale()));
+        }
+        /// Conversion of Date or DateTime to DateTime64: add zero sub-second part.
+        else if constexpr ((
+                std::is_same_v<FromDataType, DataTypeDate>
+                || std::is_same_v<FromDataType, DataTypeDate32>
+                || std::is_same_v<FromDataType, DataTypeDateTime>)
+            && std::is_same_v<ToDataType, DataTypeDateTime64>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64Transform, false>::execute(
+                arguments, result_type, input_rows_count);
+        }
+        else if constexpr (IsDataTypeDateOrDateTime<FromDataType>
+            && std::is_same_v<ToDataType, DataTypeString>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            /// Date or DateTime to String
+
+            using ColVecType = ColumnVectorOrDecimal<FromFieldType>;
+
+            auto datetime_arg = arguments[0];
+
+            const DateLUTImpl * time_zone = nullptr;
+            const ColumnConst * time_zone_column = nullptr;
+
+            if (arguments.size() == 1)
+            {
+                auto non_null_args = createBlockWithNestedColumns(arguments);
+                time_zone = &extractTimeZoneFromFunctionArguments(non_null_args, 1, 0);
+            }
+            else /// When we have a column for timezone
+            {
+                datetime_arg.column = datetime_arg.column->convertToFullColumnIfConst();
+
+                if constexpr (std::is_same_v<FromDataType, DataTypeDate> || std::is_same_v<FromDataType, DataTypeDate32>)
+                    time_zone = &DateLUT::instance();
+                /// For argument of Date or DateTime type, second argument with time zone could be specified.
+                if constexpr (std::is_same_v<FromDataType, DataTypeDateTime> || std::is_same_v<FromDataType, DataTypeDateTime64>)
+                {
+                    if ((time_zone_column = checkAndGetColumnConst<ColumnString>(arguments[1].column.get())))
+                    {
+                        auto non_null_args = createBlockWithNestedColumns(arguments);
+                        time_zone = &extractTimeZoneFromFunctionArguments(non_null_args, 1, 0);
+                    }
+                }
+            }
+            const auto & col_with_type_and_name = columnGetNested(datetime_arg);
+
+            if (const auto col_from = checkAndGetColumn<ColVecType>(col_with_type_and_name.column.get()))
+            {
+                auto col_to = ColumnString::create();
+
+                const typename ColVecType::Container & vec_from = col_from->getData();
+                ColumnString::Chars & data_to = col_to->getChars();
+                ColumnString::Offsets & offsets_to = col_to->getOffsets();
+                size_t size = vec_from.size();
+
+                if constexpr (std::is_same_v<FromDataType, DataTypeDate>)
+                    data_to.resize(size * (strlen("YYYY-MM-DD") + 1));
+                else if constexpr (std::is_same_v<FromDataType, DataTypeDate32>)
+                    data_to.resize(size * (strlen("YYYY-MM-DD") + 1));
+                else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime>)
+                    data_to.resize(size * (strlen("YYYY-MM-DD hh:mm:ss") + 1));
+                else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime64>)
+                    data_to.resize(size * (strlen("YYYY-MM-DD hh:mm:ss.") + col_from->getScale() + 1));
+                else
+                    data_to.resize(size * 3);   /// Arbitrary
+
+                offsets_to.resize(size);
+
+                WriteBufferFromVector<ColumnString::Chars> write_buffer(data_to);
+                const auto & type = static_cast<const FromDataType &>(*col_with_type_and_name.type);
+
+                ColumnUInt8::MutablePtr null_map = copyNullMap(datetime_arg.column);
+
+                if (!null_map && arguments.size() > 1)
+                    null_map = copyNullMap(arguments[1].column->convertToFullColumnIfConst());
+
+                if (null_map)
+                {
+                    for (size_t i = 0; i < size; ++i)
+                    {
+                        if (!time_zone_column && arguments.size() > 1)
+                        {
+                            if (!arguments[1].column.get()->getDataAt(i).toString().empty())
+                                time_zone = &DateLUT::instance(arguments[1].column.get()->getDataAt(i).toString());
+                            else
+                                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Provided time zone must be non-empty");
+                        }
+                        bool is_ok = FormatImpl<FromDataType>::template execute<bool>(vec_from[i], write_buffer, &type, time_zone);
+                        null_map->getData()[i] |= !is_ok;
+                        writeChar(0, write_buffer);
+                        offsets_to[i] = write_buffer.count();
+                    }
+                }
+                else
+                {
+                    for (size_t i = 0; i < size; ++i)
+                    {
+                        if (!time_zone_column && arguments.size() > 1)
+                        {
+                            if (!arguments[1].column.get()->getDataAt(i).toString().empty())
+                                time_zone = &DateLUT::instance(arguments[1].column.get()->getDataAt(i).toString());
+                            else
+                                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Provided time zone must be non-empty");
+                        }
+                        FormatImpl<FromDataType>::template execute<void>(vec_from[i], write_buffer, &type, time_zone);
+                        writeChar(0, write_buffer);
+                        offsets_to[i] = write_buffer.count();
+                    }
+                }
+
+                write_buffer.finalize();
+
+                if (null_map)
+                    return ColumnNullable::create(std::move(col_to), std::move(null_map));
+                return col_to;
+            }
+            else
+                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
+                        arguments[0].column->getName(), Name::name);
+        }
+        else if constexpr (std::is_same_v<ToDataType, DataTypeString>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            /// Anything else to String.
+
+            using ColVecType = ColumnVectorOrDecimal<FromFieldType>;
+
+            ColumnUInt8::MutablePtr null_map = copyNullMap(arguments[0].column);
+
+            const auto & col_with_type_and_name = columnGetNested(arguments[0]);
+            const auto & type = static_cast<const FromDataType &>(*col_with_type_and_name.type);
+
+            if (const auto col_from = checkAndGetColumn<ColVecType>(col_with_type_and_name.column.get()))
+            {
+                auto col_to = ColumnString::create();
+
+                const typename ColVecType::Container & vec_from = col_from->getData();
+                ColumnString::Chars & data_to = col_to->getChars();
+                ColumnString::Offsets & offsets_to = col_to->getOffsets();
+                size_t size = vec_from.size();
+
+                data_to.resize(size * 3);
+                offsets_to.resize(size);
+
+                WriteBufferFromVector<ColumnString::Chars> write_buffer(data_to);
+
+                if (null_map)
+                {
+                    for (size_t i = 0; i < size; ++i)
+                    {
+                        bool is_ok = FormatImpl<FromDataType>::template execute<bool>(vec_from[i], write_buffer, &type, nullptr);
+                        /// We don't use timezones in this branch
+                        null_map->getData()[i] |= !is_ok;
+                        writeChar(0, write_buffer);
+                        offsets_to[i] = write_buffer.count();
+                    }
+                }
+                else
+                {
+                    for (size_t i = 0; i < size; ++i)
+                    {
+                        FormatImpl<FromDataType>::template execute<void>(vec_from[i], write_buffer, &type, nullptr);
+                        writeChar(0, write_buffer);
+                        offsets_to[i] = write_buffer.count();
+                    }
+                }
+
+                write_buffer.finalize();
+
+                if (null_map)
+                    return ColumnNullable::create(std::move(col_to), std::move(null_map));
+                return col_to;
+            }
+            else
+                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
+                        arguments[0].column->getName(), Name::name);
+        }
+        else
+        {
+            using ColVecFrom = typename FromDataType::ColumnType;
+            using ColVecTo = typename ToDataType::ColumnType;
+
+            if constexpr ((IsDataTypeDecimal<FromDataType> || IsDataTypeDecimal<ToDataType>)
+                && !(std::is_same_v<DataTypeDateTime64, FromDataType> || std::is_same_v<DataTypeDateTime64, ToDataType>)
+                && (!IsDataTypeDecimalOrNumber<FromDataType> || !IsDataTypeDecimalOrNumber<ToDataType>))
+            {
+                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
+                    named_from.column->getName(), Name::name);
+            }
+
+            const ColVecFrom * col_from = checkAndGetColumn<ColVecFrom>(named_from.column.get());
+            if (!col_from)
+                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
+                    named_from.column->getName(), Name::name);
+
+            typename ColVecTo::MutablePtr col_to = nullptr;
+
+            if constexpr (IsDataTypeDecimal<ToDataType>)
+            {
+                UInt32 scale;
+
+                if constexpr (std::is_same_v<Additions, AccurateConvertStrategyAdditions>
+                    || std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
+                {
+                    scale = additions.scale;
+                }
+                else
+                {
+                    scale = additions;
+                }
+
+                col_to = ColVecTo::create(0, scale);
+            }
+            else
+                col_to = ColVecTo::create();
+
+            const auto & vec_from = col_from->getData();
+            auto & vec_to = col_to->getData();
+            vec_to.resize(input_rows_count);
+
+            ColumnUInt8::MutablePtr col_null_map_to;
+            ColumnUInt8::Container * vec_null_map_to [[maybe_unused]] = nullptr;
+            if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
+            {
+                col_null_map_to = ColumnUInt8::create(input_rows_count, false);
+                vec_null_map_to = &col_null_map_to->getData();
+            }
+
+            bool result_is_bool = isBool(result_type);
+            for (size_t i = 0; i < input_rows_count; ++i)
+            {
+                if constexpr (std::is_same_v<ToDataType, DataTypeUInt8>)
+                {
+                    if (result_is_bool)
+                    {
+                        vec_to[i] = vec_from[i] != FromFieldType(0);
+                        continue;
+                    }
+                }
+
+                if constexpr (std::is_same_v<FromDataType, DataTypeUUID> && std::is_same_v<ToDataType, DataTypeUInt128>)
+                {
+                    static_assert(
+                        std::is_same_v<DataTypeUInt128::FieldType, DataTypeUUID::FieldType::UnderlyingType>,
+                        "UInt128 and UUID types must be same");
+
+                    vec_to[i].items[1] = vec_from[i].toUnderType().items[0];
+                    vec_to[i].items[0] = vec_from[i].toUnderType().items[1];
+                }
+                else if constexpr (std::is_same_v<FromDataType, DataTypeIPv6> && std::is_same_v<ToDataType, DataTypeUInt128>)
+                {
+                    static_assert(
+                        std::is_same_v<DataTypeUInt128::FieldType, DataTypeUUID::FieldType::UnderlyingType>,
+                        "UInt128 and IPv6 types must be same");
+
+                    vec_to[i].items[1] = std::byteswap(vec_from[i].toUnderType().items[0]);
+                    vec_to[i].items[0] = std::byteswap(vec_from[i].toUnderType().items[1]);
+                }
+                else if constexpr (std::is_same_v<FromDataType, DataTypeUUID> != std::is_same_v<ToDataType, DataTypeUUID>)
+                {
+                    throw Exception(ErrorCodes::NOT_IMPLEMENTED,
+                                    "Conversion between numeric types and UUID is not supported. "
+                                    "Probably the passed UUID is unquoted");
+                }
+                else if constexpr (
+                    (std::is_same_v<FromDataType, DataTypeIPv4> != std::is_same_v<ToDataType, DataTypeIPv4>)
+                    && !(is_any_of<FromDataType, DataTypeUInt8, DataTypeUInt16, DataTypeUInt32, DataTypeUInt64, DataTypeIPv6>
+                        || is_any_of<ToDataType, DataTypeUInt32, DataTypeUInt64, DataTypeUInt128, DataTypeUInt256, DataTypeIPv6>))
+                {
+                    throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Conversion from {} to {} is not supported",
+                                    TypeName<typename FromDataType::FieldType>, TypeName<typename ToDataType::FieldType>);
+                }
+                else if constexpr (std::is_same_v<FromDataType, DataTypeIPv6> != std::is_same_v<ToDataType, DataTypeIPv6>
+                    && !(std::is_same_v<ToDataType, DataTypeIPv4> || std::is_same_v<FromDataType, DataTypeIPv4>))
+                {
+                    throw Exception(ErrorCodes::NOT_IMPLEMENTED,
+                                    "Conversion between numeric types and IPv6 is not supported. "
+                                    "Probably the passed IPv6 is unquoted");
+                }
+                else if constexpr (IsDataTypeDecimal<FromDataType> || IsDataTypeDecimal<ToDataType>)
+                {
+                    if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
+                    {
+                        ToFieldType result;
+                        bool convert_result = false;
+
+                        if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeDecimal<ToDataType>)
+                            convert_result = tryConvertDecimals<FromDataType, ToDataType>(vec_from[i], col_from->getScale(), col_to->getScale(), result);
+                        else if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeNumber<ToDataType>)
+                            convert_result = tryConvertFromDecimal<FromDataType, ToDataType>(vec_from[i], col_from->getScale(), result);
+                        else if constexpr (IsDataTypeNumber<FromDataType> && IsDataTypeDecimal<ToDataType>)
+                            convert_result = tryConvertToDecimal<FromDataType, ToDataType>(vec_from[i], col_to->getScale(), result);
+
+                        if (convert_result)
+                            vec_to[i] = result;
+                        else
+                        {
+                            vec_to[i] = static_cast<ToFieldType>(0);
+                            (*vec_null_map_to)[i] = true;
+                        }
+                    }
+                    else
+                    {
+                        if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeDecimal<ToDataType>)
+                            vec_to[i] = convertDecimals<FromDataType, ToDataType>(vec_from[i], col_from->getScale(), col_to->getScale());
+                        else if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeNumber<ToDataType>)
+                            vec_to[i] = convertFromDecimal<FromDataType, ToDataType>(vec_from[i], col_from->getScale());
+                        else if constexpr (IsDataTypeNumber<FromDataType> && IsDataTypeDecimal<ToDataType>)
+                            vec_to[i] = convertToDecimal<FromDataType, ToDataType>(vec_from[i], col_to->getScale());
+                        else
+                            throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Unsupported data type in conversion function");
+                    }
+                }
+                else if constexpr (std::is_same_v<ToDataType, DataTypeIPv4> && std::is_same_v<FromDataType, DataTypeIPv6>)
+                {
+                    const uint8_t ip4_cidr[] {0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x00, 0x00, 0x00, 0x00};
+                    const uint8_t * src = reinterpret_cast<const uint8_t *>(&vec_from[i].toUnderType());
+                    if (!matchIPv6Subnet(src, ip4_cidr, 96))
+                    {
+                        char addr[IPV6_MAX_TEXT_LENGTH + 1] {};
+                        char * paddr = addr;
+                        formatIPv6(src, paddr);
+
+                        throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "IPv6 {} in column {} is not in IPv4 mapping block", addr, named_from.column->getName());
+                    }
+
+                    uint8_t * dst = reinterpret_cast<uint8_t *>(&vec_to[i].toUnderType());
+                    if constexpr (std::endian::native == std::endian::little)
+                    {
+                        dst[0] = src[15];
+                        dst[1] = src[14];
+                        dst[2] = src[13];
+                        dst[3] = src[12];
+                    }
+                    else
+                    {
+                        dst[0] = src[12];
+                        dst[1] = src[13];
+                        dst[2] = src[14];
+                        dst[3] = src[15];
+                    }
+                }
+                else if constexpr (std::is_same_v<ToDataType, DataTypeIPv6> && std::is_same_v<FromDataType, DataTypeIPv4>)
+                {
+                    const uint8_t * src = reinterpret_cast<const uint8_t *>(&vec_from[i].toUnderType());
+                    uint8_t * dst = reinterpret_cast<uint8_t *>(&vec_to[i].toUnderType());
+                    std::memset(dst, '\0', IPV6_BINARY_LENGTH);
+                    dst[10] = dst[11] = 0xff;
+
+                    if constexpr (std::endian::native == std::endian::little)
+                    {
+                        dst[12] = src[3];
+                        dst[13] = src[2];
+                        dst[14] = src[1];
+                        dst[15] = src[0];
+                    }
+                    else
+                    {
+                        dst[12] = src[0];
+                        dst[13] = src[1];
+                        dst[14] = src[2];
+                        dst[15] = src[3];
+                    }
+                }
+                else if constexpr (std::is_same_v<ToDataType, DataTypeIPv4> && std::is_same_v<FromDataType, DataTypeUInt64>)
+                {
+                    vec_to[i] = static_cast<ToFieldType>(static_cast<IPv4::UnderlyingType>(vec_from[i]));
+                }
+                else if constexpr (std::is_same_v<Name, NameToUnixTimestamp>
+                    && (std::is_same_v<FromDataType, DataTypeDate> || std::is_same_v<FromDataType, DataTypeDate32>))
+                {
+                    vec_to[i] = static_cast<ToFieldType>(vec_from[i] * DATE_SECONDS_PER_DAY);
+                }
+                else
+                {
+                    /// If From Data is Nan or Inf and we convert to integer type, throw exception
+                    if constexpr (std::is_floating_point_v<FromFieldType> && !std::is_floating_point_v<ToFieldType>)
+                    {
+                        if (!isFinite(vec_from[i]))
+                        {
+                            if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
+                            {
+                                vec_to[i] = 0;
+                                (*vec_null_map_to)[i] = true;
+                                continue;
+                            }
+                            else
+                                throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Unexpected inf or nan to integer conversion");
+                        }
+                    }
+
+                    if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>
+                            || std::is_same_v<Additions, AccurateConvertStrategyAdditions>)
+                    {
+                        bool convert_result = accurate::convertNumeric(vec_from[i], vec_to[i]);
+
+                        if (!convert_result)
+                        {
+                            if (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
+                            {
+                                vec_to[i] = 0;
+                                (*vec_null_map_to)[i] = true;
+                            }
+                            else
+                            {
+                                throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Value in column {} cannot be safely converted into type {}",
+                                    named_from.column->getName(), result_type->getName());
+                            }
+                        }
+                    }
+                    else
+                    {
+                        vec_to[i] = static_cast<ToFieldType>(vec_from[i]);
+                    }
+                }
+            }
+
+            if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
+                return ColumnNullable::create(std::move(col_to), std::move(col_null_map_to));
+            else
+                return col_to;
+        }
+    }
+};
+
+
 template <typename ToDataType, typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
 requires (!std::is_same_v<ToDataType, DataTypeString>)
 struct ConvertImpl<DataTypeString, ToDataType, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>

From f339c88079e2b0173b29e343872a576767b56e7b Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 13 Mar 2024 06:32:31 +0100
Subject: [PATCH 244/443] Continue

---
 src/Functions/FunctionsConversion.cpp | 61 +++++++++++++--------------
 1 file changed, 29 insertions(+), 32 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 16f547939fa..5001abcb5d2 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -1163,13 +1163,10 @@ template <typename FromDataType, typename ToDataType, typename Name,
     FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior = default_date_time_overflow_behavior>
 struct ConvertImpl
 {
-    using FromFieldType = typename FromDataType::FieldType;
-    using ToFieldType = typename ToDataType::FieldType;
-
     template <typename Additions = void *>
     static ColumnPtr NO_SANITIZE_UNDEFINED execute(
         const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type [[maybe_unused]], size_t input_rows_count,
-        Additions additions [[maybe_unused]] = Additions())
+        Additions additions = Additions())
     {
         const ColumnWithTypeAndName & named_from = arguments[0];
 
@@ -1206,7 +1203,7 @@ struct ConvertImpl
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
             return DateTimeTransformImpl<DataTypeDateTime64, DataTypeDate32, TransformDateTime64<ToDate32Impl>, false>::execute(
-                arguments, result_type, input_rows_count, TransformDateTime64<ToDate32Impl>(assert_cast<const DataTypeDateTime64 &>(*named_from.type).getScale()));
+                arguments, result_type, input_rows_count, additions);
         }
         /** Special case of converting Int8, Int16, (U)Int32 or (U)Int64 (and also, for convenience,
           * Float32, Float64) to Date. If the
@@ -1336,14 +1333,14 @@ struct ConvertImpl
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
             return DateTimeTransformImpl<FromDataType, ToDataType, TransformDateTime64<ToDateImpl<date_time_overflow_behavior>>, false>::execute(
-                arguments, result_type, input_rows_count, TransformDateTime64<ToDateImpl<date_time_overflow_behavior>>(assert_cast<const DataTypeDateTime64 &>(*named_from.type).getScale()));
+                arguments, result_type, input_rows_count, additions);
         }
         else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime64>
             && std::is_same_v<ToDataType, DataTypeDateTime>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
             return DateTimeTransformImpl<FromDataType, ToDataType, TransformDateTime64<ToDateTimeImpl<date_time_overflow_behavior>>, false>::execute(
-                arguments, result_type, input_rows_count, TransformDateTime64<ToDateTimeImpl<date_time_overflow_behavior>>(assert_cast<const DataTypeDateTime64 &>(*named_from.type).getScale()));
+                arguments, result_type, input_rows_count, additions);
         }
         /// Conversion of Date or DateTime to DateTime64: add zero sub-second part.
         else if constexpr ((
@@ -1362,6 +1359,7 @@ struct ConvertImpl
         {
             /// Date or DateTime to String
 
+            using FromFieldType = typename FromDataType::FieldType;
             using ColVecType = ColumnVectorOrDecimal<FromFieldType>;
 
             auto datetime_arg = arguments[0];
@@ -1471,6 +1469,7 @@ struct ConvertImpl
         {
             /// Anything else to String.
 
+            using FromFieldType = typename FromDataType::FieldType;
             using ColVecType = ColumnVectorOrDecimal<FromFieldType>;
 
             ColumnUInt8::MutablePtr null_map = copyNullMap(arguments[0].column);
@@ -1523,8 +1522,31 @@ struct ConvertImpl
                 throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
                         arguments[0].column->getName(), Name::name);
         }
+        else if constexpr ((std::is_same_v<FromDataType, DataTypeString> || std::is_same_v<FromDataType, DataTypeFixedString>)
+            && !std::is_same_v<ToDataType, DataTypeString>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            return ConvertThroughParsing<FromDataType, ToDataType, Name, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::Normal>::execute(
+                arguments, result_type, input_rows_count, additions);
+        }
+        else if constexpr ((std::is_same_v<FromDataType, DataTypeString> || std::is_same_v<FromDataType, DataTypeFixedString>)
+            && !std::is_same_v<ToDataType, DataTypeString>
+            && std::is_same_v<SpecialTag, ConvertReturnNullOnErrorTag>)
+        {
+            return ConvertThroughParsing<FromDataType, ToDataType, Name, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::Normal>::execute(
+                arguments, result_type, input_rows_count, additions);
+        }
+        else if constexpr ((std::is_same_v<FromDataType, DataTypeString> || std::is_same_v<FromDataType, DataTypeFixedString>)
+            && is_any_of<ToDataType, DataTypeIPv4, DataTypeIPv6>
+            && std::is_same_v<SpecialTag, ConvertReturnZeroOnErrorTag>)
+        {
+            return ConvertThroughParsing<FromDataType, ToDataType, Name, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::Normal>::execute(
+                arguments, result_type, input_rows_count, additions);
+        }
         else
         {
+            using FromFieldType = typename FromDataType::FieldType;
+            using ToFieldType = typename ToDataType::FieldType;
             using ColVecFrom = typename FromDataType::ColumnType;
             using ColVecTo = typename ToDataType::ColumnType;
 
@@ -1772,31 +1794,6 @@ struct ConvertImpl
 };
 
 
-template <typename ToDataType, typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-requires (!std::is_same_v<ToDataType, DataTypeString>)
-struct ConvertImpl<DataTypeString, ToDataType, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : ConvertThroughParsing<DataTypeString, ToDataType, Name, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::Normal> {};
-
-template <typename ToDataType, typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-requires (!std::is_same_v<ToDataType, DataTypeFixedString>)
-struct ConvertImpl<DataTypeFixedString, ToDataType, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : ConvertThroughParsing<DataTypeFixedString, ToDataType, Name, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::Normal> {};
-
-template <typename ToDataType, typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-requires (!std::is_same_v<ToDataType, DataTypeString>)
-struct ConvertImpl<DataTypeString, ToDataType, Name, ConvertReturnNullOnErrorTag, date_time_overflow_behavior>
-    : ConvertThroughParsing<DataTypeString, ToDataType, Name, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::Normal> {};
-
-template <typename ToDataType, typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-requires (!std::is_same_v<ToDataType, DataTypeFixedString>)
-struct ConvertImpl<DataTypeFixedString, ToDataType, Name, ConvertReturnNullOnErrorTag, date_time_overflow_behavior>
-    : ConvertThroughParsing<DataTypeFixedString, ToDataType, Name, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::Normal> {};
-
-template <typename FromDataType, typename ToDataType, typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-requires (is_any_of<FromDataType, DataTypeString, DataTypeFixedString> && is_any_of<ToDataType, DataTypeIPv4, DataTypeIPv6>)
-struct ConvertImpl<FromDataType, ToDataType, Name, ConvertReturnZeroOnErrorTag, date_time_overflow_behavior>
-    : ConvertThroughParsing<FromDataType, ToDataType, Name, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::Normal> {};
-
 /// Generic conversion of any type from String. Used for complex types: Array and Tuple or types with custom serialization.
 struct ConvertImplGenericFromString
 {

From 044eed8d8231cb7f22f46fd0b394a15698e7257f Mon Sep 17 00:00:00 2001
From: Han Fei <hanfei19910905@gmail.com>
Date: Wed, 13 Mar 2024 06:40:47 +0100
Subject: [PATCH 245/443] fix style

---
 .../test_memory_limit_observer/test.py        | 28 +++++++++----------
 1 file changed, 14 insertions(+), 14 deletions(-)

diff --git a/tests/integration/test_memory_limit_observer/test.py b/tests/integration/test_memory_limit_observer/test.py
index cf10f46be6a..54e8db89eac 100644
--- a/tests/integration/test_memory_limit_observer/test.py
+++ b/tests/integration/test_memory_limit_observer/test.py
@@ -7,9 +7,10 @@ from helpers.cluster import ClickHouseCluster, run_and_check
 cluster = ClickHouseCluster(__file__)
 
 node1 = cluster.add_instance(
-    "node1", main_configs=["config/text_log.xml"], mem_limit = "5g"
+    "node1", main_configs=["config/text_log.xml"], mem_limit="5g"
 )
 
+
 @pytest.fixture(scope="module")
 def started_cluster():
     try:
@@ -20,16 +21,19 @@ def started_cluster():
     finally:
         cluster.shutdown()
 
+
 def get_latest_mem_limit():
     for _ in range(10):
         try:
-            mem_limit = float(node1.query(
-            '''
-            select extract(message, '\\d+\\.\\d+') from system.text_log
-            where message like '%Setting max_server_memory_usage was set to%' and
-            message not like '%like%' order by event_time desc limit 1
-            '''
-            ).strip())
+            mem_limit = float(
+                node1.query(
+                    """
+                        select extract(message, '\\d+\\.\\d+') from system.text_log
+                        where message like '%Setting max_server_memory_usage was set to%' and
+                        message not like '%like%' order by event_time desc limit 1
+                    """
+                ).strip()
+            )
             return mem_limit
         except Exception as e:
             time.sleep(1)
@@ -37,15 +41,11 @@ def get_latest_mem_limit():
 
 def test_observe_memory_limit(started_cluster):
     original_max_mem = get_latest_mem_limit()
-    #container_id = cluster.get_container_id('node1')
-    #cmd = "docker update --memory=10g {}".format(node1.docker_id)
     logging.debug(f"get original memory limit {original_max_mem}")
-    run_and_check([
-        'docker', 'update', '--memory=10g', node1.docker_id
-    ])
+    run_and_check(['docker', 'update', '--memory=10g', node1.docker_id])
     for _ in range(30):
         time.sleep(10)
-        new_max_mem =  get_latest_mem_limit()
+        new_max_mem = get_latest_mem_limit()
         logging.debug(f"get new memory limit {new_max_mem}")
         if new_max_mem > original_max_mem:
             return

From 87db039d8952ca0e2ed32c7eda0f49600f9078c4 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 13 Mar 2024 07:06:06 +0100
Subject: [PATCH 246/443] Continue

---
 src/Functions/FunctionsConversion.cpp | 119 ++++++++++++--------------
 1 file changed, 55 insertions(+), 64 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 5001abcb5d2..fd328e8af42 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -1354,8 +1354,7 @@ struct ConvertImpl
                 arguments, result_type, input_rows_count);
         }
         else if constexpr (IsDataTypeDateOrDateTime<FromDataType>
-            && std::is_same_v<ToDataType, DataTypeString>
-            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+            && std::is_same_v<ToDataType, DataTypeString>)
         {
             /// Date or DateTime to String
 
@@ -1464,6 +1463,54 @@ struct ConvertImpl
                 throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
                         arguments[0].column->getName(), Name::name);
         }
+        /// Conversion from FixedString to String.
+        /// Cutting sequences of zero bytes from end of strings.
+        else if constexpr (std::is_same_v<ToDataType, DataTypeString>
+            && std::is_same_v<FromDataType, DataTypeFixedString>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+            ColumnUInt8::MutablePtr null_map = copyNullMap(arguments[0].column);
+            const auto & nested =  columnGetNested(arguments[0]);
+            if (const ColumnFixedString * col_from = checkAndGetColumn<ColumnFixedString>(nested.column.get()))
+            {
+                auto col_to = ColumnString::create();
+
+                const ColumnFixedString::Chars & data_from = col_from->getChars();
+                ColumnString::Chars & data_to = col_to->getChars();
+                ColumnString::Offsets & offsets_to = col_to->getOffsets();
+                size_t size = col_from->size();
+                size_t n = col_from->getN();
+                data_to.resize(size * (n + 1)); /// + 1 - zero terminator
+                offsets_to.resize(size);
+
+                size_t offset_from = 0;
+                size_t offset_to = 0;
+                for (size_t i = 0; i < size; ++i)
+                {
+                    if (!null_map || !null_map->getData()[i])
+                    {
+                        size_t bytes_to_copy = n;
+                        while (bytes_to_copy > 0 && data_from[offset_from + bytes_to_copy - 1] == 0)
+                            --bytes_to_copy;
+
+                        memcpy(&data_to[offset_to], &data_from[offset_from], bytes_to_copy);
+                        offset_to += bytes_to_copy;
+                    }
+                    data_to[offset_to] = 0;
+                    ++offset_to;
+                    offsets_to[i] = offset_to;
+                    offset_from += n;
+                }
+
+                data_to.resize(offset_to);
+                if (result_type->isNullable() && null_map)
+                    return ColumnNullable::create(std::move(col_to), std::move(null_map));
+                return col_to;
+            }
+            else
+                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
+                        arguments[0].column->getName(), Name::name);
+        }
         else if constexpr (std::is_same_v<ToDataType, DataTypeString>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
@@ -1543,6 +1590,12 @@ struct ConvertImpl
             return ConvertThroughParsing<FromDataType, ToDataType, Name, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::Normal>::execute(
                 arguments, result_type, input_rows_count, additions);
         }
+        else if constexpr (std::is_same_v<Name, NameToUnixTimestamp>
+            && std::is_same_v<FromDataType, DataTypeString>
+            && std::is_same_v<ToDataType, DataTypeUInt32>
+            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
+        {
+        }
         else
         {
             using FromFieldType = typename FromDataType::FieldType;
@@ -1863,68 +1916,6 @@ struct ConvertImplGenericFromString
 };
 
 
-template <>
-struct ConvertImpl<DataTypeString, DataTypeUInt32, NameToUnixTimestamp, ConvertDefaultBehaviorTag>
-    : ConvertImpl<DataTypeString, DataTypeDateTime, NameToUnixTimestamp, ConvertDefaultBehaviorTag> {};
-
-template <>
-struct ConvertImpl<DataTypeString, DataTypeUInt32, NameToUnixTimestamp, ConvertReturnNullOnErrorTag>
-    : ConvertImpl<DataTypeString, DataTypeDateTime, NameToUnixTimestamp, ConvertReturnNullOnErrorTag> {};
-
-
-/** Conversion from FixedString to String.
-  * Cutting sequences of zero bytes from end of strings.
-  */
-template <typename Name>
-struct ConvertImpl<DataTypeFixedString, DataTypeString, Name, ConvertDefaultBehaviorTag>
-{
-    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr & return_type, size_t /*input_rows_count*/)
-    {
-        ColumnUInt8::MutablePtr null_map = copyNullMap(arguments[0].column);
-        const auto & nested =  columnGetNested(arguments[0]);
-        if (const ColumnFixedString * col_from = checkAndGetColumn<ColumnFixedString>(nested.column.get()))
-        {
-            auto col_to = ColumnString::create();
-
-            const ColumnFixedString::Chars & data_from = col_from->getChars();
-            ColumnString::Chars & data_to = col_to->getChars();
-            ColumnString::Offsets & offsets_to = col_to->getOffsets();
-            size_t size = col_from->size();
-            size_t n = col_from->getN();
-            data_to.resize(size * (n + 1)); /// + 1 - zero terminator
-            offsets_to.resize(size);
-
-            size_t offset_from = 0;
-            size_t offset_to = 0;
-            for (size_t i = 0; i < size; ++i)
-            {
-                if (!null_map || !null_map->getData()[i])
-                {
-                    size_t bytes_to_copy = n;
-                    while (bytes_to_copy > 0 && data_from[offset_from + bytes_to_copy - 1] == 0)
-                        --bytes_to_copy;
-
-                    memcpy(&data_to[offset_to], &data_from[offset_from], bytes_to_copy);
-                    offset_to += bytes_to_copy;
-                }
-                data_to[offset_to] = 0;
-                ++offset_to;
-                offsets_to[i] = offset_to;
-                offset_from += n;
-            }
-
-            data_to.resize(offset_to);
-            if (return_type->isNullable() && null_map)
-                return ColumnNullable::create(std::move(col_to), std::move(null_map));
-            return col_to;
-        }
-        else
-            throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                    arguments[0].column->getName(), Name::name);
-    }
-};
-
-
 /// Declared early because used below.
 struct NameToDate { static constexpr auto name = "toDate"; };
 struct NameToDate32 { static constexpr auto name = "toDate32"; };

From f8dfd8c03a05f45729f0dce588da4c9b0200739e Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 13 Mar 2024 07:15:07 +0100
Subject: [PATCH 247/443] Less garbage

---
 src/Functions/FunctionsConversion.cpp | 30 +++++++++++++--------------
 1 file changed, 15 insertions(+), 15 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index fd328e8af42..5461661386e 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -2389,7 +2389,6 @@ public:
     static constexpr bool to_datetime64 = std::is_same_v<ToDataType, DataTypeDateTime64>;
 
     static FunctionPtr create(ContextPtr) { return std::make_shared<FunctionConvertFromString>(); }
-    static FunctionPtr create() { return std::make_shared<FunctionConvertFromString>(); }
 
     String getName() const override
     {
@@ -2516,28 +2515,29 @@ public:
         ColumnPtr result_column;
 
         if constexpr (to_decimal)
+        {
             result_column = executeInternal<ToDataType>(arguments, result_type, input_rows_count,
                 assert_cast<const ToDataType &>(*removeNullable(result_type)).getScale());
-        else
+        }
+        else if (isDateTime64<Name, ToDataType>(arguments))
         {
-            if (isDateTime64<Name, ToDataType>(arguments))
-            {
-                UInt64 scale = to_datetime64 ? DataTypeDateTime64::default_scale : 0;
-                if (arguments.size() > 1)
-                    scale = extractToDecimalScale(arguments[1]);
+            UInt64 scale = to_datetime64 ? DataTypeDateTime64::default_scale : 0;
+            if (arguments.size() > 1)
+                scale = extractToDecimalScale(arguments[1]);
 
-                if (scale == 0)
-                    result_column = executeInternal<DataTypeDateTime>(arguments, result_type, input_rows_count);
-                else
-                {
-                    result_column = executeInternal<DataTypeDateTime64>(arguments, result_type, input_rows_count, static_cast<UInt32>(scale));
-                }
+            if (scale == 0)
+            {
+                result_column = executeInternal<DataTypeDateTime>(arguments, result_type, input_rows_count);
             }
             else
             {
-                result_column = executeInternal<ToDataType>(arguments, result_type, input_rows_count);
+                result_column = executeInternal<DataTypeDateTime64>(arguments, result_type, input_rows_count, static_cast<UInt32>(scale));
             }
         }
+        else
+        {
+            result_column = executeInternal<ToDataType>(arguments, result_type, input_rows_count);
+        }
 
         if (!result_column)
             throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of argument of function {}. "
@@ -3204,7 +3204,7 @@ private:
         {
             /// In case when converting to Nullable type, we apply different parsing rule,
             /// that will not throw an exception but return NULL in case of malformed input.
-            FunctionPtr function = FunctionConvertFromString<ToDataType, FunctionCastName, ConvertFromStringExceptionMode::Null>::create();
+            FunctionPtr function = FunctionConvertFromString<ToDataType, FunctionCastName, ConvertFromStringExceptionMode::Null>::create(context);
             return createFunctionAdaptor(function, from_type);
         }
         else if (!can_apply_accurate_cast)

From cb3ffc6f1a6da589739d94b82a965c3bd5835e43 Mon Sep 17 00:00:00 2001
From: Han Fei <hanfei19910905@gmail.com>
Date: Wed, 13 Mar 2024 07:28:41 +0100
Subject: [PATCH 248/443] fix style

---
 tests/integration/test_memory_limit_observer/test.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tests/integration/test_memory_limit_observer/test.py b/tests/integration/test_memory_limit_observer/test.py
index 54e8db89eac..408aa5e4018 100644
--- a/tests/integration/test_memory_limit_observer/test.py
+++ b/tests/integration/test_memory_limit_observer/test.py
@@ -39,10 +39,11 @@ def get_latest_mem_limit():
             time.sleep(1)
     raise Exception("Cannot get memory limit")
 
+
 def test_observe_memory_limit(started_cluster):
     original_max_mem = get_latest_mem_limit()
     logging.debug(f"get original memory limit {original_max_mem}")
-    run_and_check(['docker', 'update', '--memory=10g', node1.docker_id])
+    run_and_check(["docker", "update", "--memory=10g", node1.docker_id])
     for _ in range(30):
         time.sleep(10)
         new_max_mem = get_latest_mem_limit()

From 65a541fbdbac4ec3a00a1f0271867e68e6cdef87 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 13 Mar 2024 07:48:14 +0100
Subject: [PATCH 249/443] Tighten

---
 utils/check-style/check-large-objects.sh | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/utils/check-style/check-large-objects.sh b/utils/check-style/check-large-objects.sh
index 3e2a385bdd0..9b78d6196e3 100755
--- a/utils/check-style/check-large-objects.sh
+++ b/utils/check-style/check-large-objects.sh
@@ -6,10 +6,7 @@ export LC_ALL=C # The "total" should be printed without localization
 
 TU_EXCLUDES=(
     AggregateFunctionUniq
-    FunctionsConversion
-
     RangeHashedDictionary
-
     Aggregator
 )
 

From 486e8537a80abe1b238839f207f35a9f9030011e Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 13 Mar 2024 08:02:27 +0100
Subject: [PATCH 250/443] Fix error

---
 src/Functions/FunctionsConversion.cpp | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 5461661386e..67d6e202255 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -1309,14 +1309,14 @@ struct ConvertImpl
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
             return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64TransformSigned<typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::execute(
-                arguments, result_type, input_rows_count);
+                arguments, result_type, input_rows_count, additions);
         }
         else if constexpr (std::is_same_v<FromDataType, DataTypeUInt64>
             && std::is_same_v<ToDataType, DataTypeDateTime64>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
             return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64TransformUnsigned<UInt64, default_date_time_overflow_behavior>, false>::execute(
-                arguments, result_type, input_rows_count);
+                arguments, result_type, input_rows_count, additions);
         }
         else if constexpr ((
                 std::is_same_v<FromDataType, DataTypeFloat32>
@@ -1325,7 +1325,7 @@ struct ConvertImpl
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
             return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64TransformFloat<FromDataType, typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::execute(
-                arguments, result_type, input_rows_count);
+                arguments, result_type, input_rows_count, additions);
         }
         /// Conversion of DateTime64 to Date or DateTime: discards fractional part.
         else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime64>
@@ -1351,7 +1351,7 @@ struct ConvertImpl
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
             return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64Transform, false>::execute(
-                arguments, result_type, input_rows_count);
+                arguments, result_type, input_rows_count, additions);
         }
         else if constexpr (IsDataTypeDateOrDateTime<FromDataType>
             && std::is_same_v<ToDataType, DataTypeString>)

From 87846b2c98205f95db69d99ede102e600615c54a Mon Sep 17 00:00:00 2001
From: Mikhail Koviazin <mikhail.koviazin@aiven.io>
Date: Wed, 13 Mar 2024 09:14:55 +0200
Subject: [PATCH 251/443] Added more tests

---
 .../0_stateless/00662_has_nullable.reference  | 12 +++++++
 .../0_stateless/00662_has_nullable.sql        | 32 +++++++++++++++++++
 2 files changed, 44 insertions(+)

diff --git a/tests/queries/0_stateless/00662_has_nullable.reference b/tests/queries/0_stateless/00662_has_nullable.reference
index 97da99d655e..1ac93f25a39 100644
--- a/tests/queries/0_stateless/00662_has_nullable.reference
+++ b/tests/queries/0_stateless/00662_has_nullable.reference
@@ -1,2 +1,14 @@
+Nullable(UInt64), non-null array
 1	1
 \N	0
+Non-nullable UInt64, nullable array
+0	0
+1	1
+2	1
+Nullable(UInt64), nullable array
+0	0
+\N	1
+1	1
+All NULLs
+0	0
+\N	1
diff --git a/tests/queries/0_stateless/00662_has_nullable.sql b/tests/queries/0_stateless/00662_has_nullable.sql
index 0d395871d9f..3fd3bd46baa 100644
--- a/tests/queries/0_stateless/00662_has_nullable.sql
+++ b/tests/queries/0_stateless/00662_has_nullable.sql
@@ -1,7 +1,39 @@
 DROP TABLE IF EXISTS 00662_has_nullable;
+
+SELECT 'Nullable(UInt64), non-null array';
 CREATE TABLE 00662_has_nullable(a Nullable(UInt64)) ENGINE = Memory;
 
 INSERT INTO 00662_has_nullable VALUES (1), (Null);
 SELECT a, has([0, 1], a) FROM 00662_has_nullable;
 
 DROP TABLE 00662_has_nullable;
+
+--------------------------------------------------------------------------------
+
+SELECT 'Non-nullable UInt64, nullable array';
+CREATE TABLE 00662_has_nullable(a UInt64) ENGINE = Memory;
+
+INSERT INTO 00662_has_nullable VALUES (0), (1), (2);
+SELECT a, has([NULL, 1, 2], a) FROM 00662_has_nullable;
+
+DROP TABLE 00662_has_nullable;
+
+--------------------------------------------------------------------------------
+
+SELECT 'Nullable(UInt64), nullable array';
+CREATE TABLE 00662_has_nullable(a Nullable(UInt64)) ENGINE = Memory;
+
+INSERT INTO 00662_has_nullable VALUES (0), (Null), (1);
+SELECT a, has([NULL, 1, 2], a) FROM 00662_has_nullable;
+
+DROP TABLE 00662_has_nullable;
+
+--------------------------------------------------------------------------------
+
+SELECT 'All NULLs';
+CREATE TABLE 00662_has_nullable(a Nullable(UInt64)) ENGINE = Memory;
+
+INSERT INTO 00662_has_nullable VALUES (0), (Null);
+SELECT a, has([NULL, NULL], a) FROM 00662_has_nullable;
+
+DROP TABLE 00662_has_nullable;

From 514f8392f93b9488ab4b52011102891a851b51db Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 13 Mar 2024 10:25:37 +0100
Subject: [PATCH 252/443] Fix error

---
 src/Functions/FunctionsConversion.cpp | 15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 67d6e202255..70cbf31bcb3 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -1569,15 +1569,20 @@ struct ConvertImpl
                 throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
                         arguments[0].column->getName(), Name::name);
         }
+        else if constexpr (std::is_same_v<Name, NameToUnixTimestamp>
+            && std::is_same_v<FromDataType, DataTypeString>
+            && std::is_same_v<ToDataType, DataTypeUInt32>)
+        {
+            return ConvertImpl<FromDataType, DataTypeDateTime, Name, SpecialTag, date_time_overflow_behavior>::execute(
+                arguments, result_type, input_rows_count, additions);
+        }
         else if constexpr ((std::is_same_v<FromDataType, DataTypeString> || std::is_same_v<FromDataType, DataTypeFixedString>)
-            && !std::is_same_v<ToDataType, DataTypeString>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
             return ConvertThroughParsing<FromDataType, ToDataType, Name, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::Normal>::execute(
                 arguments, result_type, input_rows_count, additions);
         }
         else if constexpr ((std::is_same_v<FromDataType, DataTypeString> || std::is_same_v<FromDataType, DataTypeFixedString>)
-            && !std::is_same_v<ToDataType, DataTypeString>
             && std::is_same_v<SpecialTag, ConvertReturnNullOnErrorTag>)
         {
             return ConvertThroughParsing<FromDataType, ToDataType, Name, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::Normal>::execute(
@@ -1590,12 +1595,6 @@ struct ConvertImpl
             return ConvertThroughParsing<FromDataType, ToDataType, Name, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::Normal>::execute(
                 arguments, result_type, input_rows_count, additions);
         }
-        else if constexpr (std::is_same_v<Name, NameToUnixTimestamp>
-            && std::is_same_v<FromDataType, DataTypeString>
-            && std::is_same_v<ToDataType, DataTypeUInt32>
-            && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
-        {
-        }
         else
         {
             using FromFieldType = typename FromDataType::FieldType;

From 47dfefd6b3983bad2c5598c19b14df679f605d93 Mon Sep 17 00:00:00 2001
From: Alexander Sapin <alesapin@gmail.com>
Date: Wed, 13 Mar 2024 10:37:51 +0100
Subject: [PATCH 253/443] config substitutions

---
 src/Common/Config/ConfigProcessor.cpp         | 21 +++++++-
 src/Common/Config/ConfigProcessor.h           | 11 +++--
 src/Common/tests/gtest_config_processor.cpp   | 49 +++++++++++++++++++
 .../configs/config_zk_include_test.xml        |  8 +++
 .../test_config_substitutions/test.py         | 11 ++++-
 5 files changed, 93 insertions(+), 7 deletions(-)
 create mode 100644 tests/integration/test_config_substitutions/configs/config_zk_include_test.xml

diff --git a/src/Common/Config/ConfigProcessor.cpp b/src/Common/Config/ConfigProcessor.cpp
index 641e7ddcdaa..57e9f9efc10 100644
--- a/src/Common/Config/ConfigProcessor.cpp
+++ b/src/Common/Config/ConfigProcessor.cpp
@@ -173,10 +173,12 @@ static void deleteAttributesRecursive(Node * root)
 static void mergeAttributes(Element & config_element, Element & with_element)
 {
     auto * with_element_attributes = with_element.attributes();
+    //std::cerr << "MERGE ATTRIBUTES:" << with_element_attributes->length() << "\n";
 
     for (size_t i = 0; i < with_element_attributes->length(); ++i)
     {
         auto * attr = with_element_attributes->item(i);
+        //std::cerr << "ATTR NAME:" << attr->nodeName() << std::endl;
         config_element.setAttribute(attr->nodeName(), attr->getNodeValue());
     }
 
@@ -276,6 +278,8 @@ void ConfigProcessor::mergeRecursive(XMLDocumentPtr config, Node * config_root,
     for (Node * node = config_root->firstChild(); node;)
     {
         Node * next_node = node->nextSibling();
+        //if (next_node)
+        //    std::cerr << "NEXT NODE:"  << next_node->nodeName() << std::endl;
         /// Remove text from the original config node.
         if (node->nodeType() == Node::TEXT_NODE && !allWhitespace(node->getNodeValue()))
         {
@@ -283,6 +287,7 @@ void ConfigProcessor::mergeRecursive(XMLDocumentPtr config, Node * config_root,
         }
         else if (node->nodeType() == Node::ELEMENT_NODE)
         {
+            std::cerr << "NODES IN SOURCE: " << node->nodeName() << std::endl;
             config_element_by_id.insert(ElementsByIdentifier::value_type(getElementIdentifier(node), node));
         }
         node = next_node;
@@ -296,6 +301,7 @@ void ConfigProcessor::mergeRecursive(XMLDocumentPtr config, Node * config_root,
         bool remove = false;
         if (with_node->nodeType() == Node::ELEMENT_NODE)
         {
+            //std::cerr << "WITH NODE: " << with_node->nodeName() << std::endl;
             Element & with_element = dynamic_cast<Element &>(*with_node);
             remove = with_element.hasAttribute("remove");
             bool replace = with_element.hasAttribute("replace");
@@ -303,11 +309,13 @@ void ConfigProcessor::mergeRecursive(XMLDocumentPtr config, Node * config_root,
             if (remove && replace)
                 throw Poco::Exception("both remove and replace attributes set for element <" + with_node->nodeName() + ">");
 
+
             ElementsByIdentifier::iterator it = config_element_by_id.find(getElementIdentifier(with_node));
 
             if (it != config_element_by_id.end())
             {
                 Node * config_node = it->second;
+                //std::cerr << "SUBNODE NODE: " << config_node->nodeName() << std::endl;
                 config_element_by_id.erase(it);
 
                 if (remove)
@@ -316,12 +324,14 @@ void ConfigProcessor::mergeRecursive(XMLDocumentPtr config, Node * config_root,
                 }
                 else if (replace)
                 {
+                    //std::cerr << "REPLACE!!!" << std::endl;
                     with_element.removeAttribute("replace");
                     NodePtr new_node = config->importNode(with_node, true);
                     config_root->replaceChild(new_node, config_node);
                 }
                 else
                 {
+                    //std::cerr << "SUBNODE NODE HERE: " << config_node->nodeName() << std::endl;
                     Element & config_element = dynamic_cast<Element &>(*config_node);
 
                     /// Remove substitution attributes from the merge target node if source node already has a value
@@ -333,11 +343,17 @@ void ConfigProcessor::mergeRecursive(XMLDocumentPtr config, Node * config_root,
                     mergeAttributes(config_element, with_element);
                     mergeRecursive(config, config_node, with_node);
                 }
+                //std::cerr << "DONE\n";
                 merged = true;
             }
         }
+        else
+        {
+            //std::cerr << "ELEMENT NOT FOUND\n";
+        }
         if (!merged && !remove)
         {
+            //std::cerr << "NOTHING hAPPENED\n";
             /// Since we didn't find a pair to this node in default config, we will paste it as is.
             /// But it may have some child nodes which have attributes like "replace" or "remove".
             /// They are useless in preprocessed configuration.
@@ -433,6 +449,7 @@ void ConfigProcessor::doIncludesRecursive(
     auto process_include = [&](const Node * include_attr, const std::function<const Node * (const std::string &)> & get_node, const char * error_msg)
     {
         const std::string & name = include_attr->getNodeValue();
+        LOG_DEBUG(log, "PROCESS INCLUDE {}", name);
         const Node * node_to_include = get_node(name);
         if (!node_to_include)
         {
@@ -453,13 +470,15 @@ void ConfigProcessor::doIncludesRecursive(
             /// Replace the whole node not just contents.
             if (node->nodeName() == "include")
             {
+                LOG_DEBUG(log, "Include here for node {}", name);
                 const NodeListPtr children = node_to_include->childNodes();
                 Node * next_child = nullptr;
                 for (Node * child = children->item(0); child; child = next_child)
                 {
                     next_child = child->nextSibling();
                     NodePtr new_node = config->importNode(child, true);
-                    node->parentNode()->insertBefore(new_node, node);
+                    //node->parentNode()->insertBefore(new_node, node);
+                    mergeRecursive(config, node->parentNode(), new_node);
                 }
 
                 node->parentNode()->removeChild(node);
diff --git a/src/Common/Config/ConfigProcessor.h b/src/Common/Config/ConfigProcessor.h
index 5712c36d737..9201ec7b77d 100644
--- a/src/Common/Config/ConfigProcessor.h
+++ b/src/Common/Config/ConfigProcessor.h
@@ -119,6 +119,11 @@ public:
 
     static inline const auto SUBSTITUTION_ATTRS = {"incl", "from_zk", "from_env"};
 
+    /// If config root node name is not 'clickhouse' and merging config's root node names doesn't match, bypasses merging and returns false.
+    /// For compatibility root node 'yandex' considered equal to 'clickhouse'.
+    bool merge(XMLDocumentPtr config, XMLDocumentPtr with);
+    void mergeRecursive(XMLDocumentPtr config, Poco::XML::Node * config_root, const Poco::XML::Node * with_root);
+
 private:
     const std::string path;
     std::string preprocessed_path;
@@ -145,11 +150,7 @@ private:
     void hideRecursive(Poco::XML::Node * config_root);
     XMLDocumentPtr hideElements(XMLDocumentPtr xml_tree);
 
-    void mergeRecursive(XMLDocumentPtr config, Poco::XML::Node * config_root, const Poco::XML::Node * with_root);
-
-    /// If config root node name is not 'clickhouse' and merging config's root node names doesn't match, bypasses merging and returns false.
-    /// For compatibility root node 'yandex' considered equal to 'clickhouse'.
-    bool merge(XMLDocumentPtr config, XMLDocumentPtr with);
+    //void mergeRecursive(XMLDocumentPtr config, Poco::XML::Node * config_root, const Poco::XML::Node * with_root);
 
     void doIncludesRecursive(
             XMLDocumentPtr config,
diff --git a/src/Common/tests/gtest_config_processor.cpp b/src/Common/tests/gtest_config_processor.cpp
index f01460d515b..d5397dd54c2 100644
--- a/src/Common/tests/gtest_config_processor.cpp
+++ b/src/Common/tests/gtest_config_processor.cpp
@@ -2,6 +2,7 @@
 #include <IO/WriteBufferFromFile.h>
 #include <IO/WriteHelpers.h>
 #include <Poco/Timestamp.h>
+#include <Common/XMLUtils.h>
 #include <Poco/Util/XMLConfiguration.h>
 #include <base/scope_guard.h>
 #include <gtest/gtest.h>
@@ -63,3 +64,51 @@ TEST(Common, ConfigProcessorManyElements)
     /// More that 5 min is way too slow
     ASSERT_LE(enumerate_elapsed_ms, 300*1000);
 }
+
+TEST(Common, ConfigProcessorMerge)
+{
+    namespace fs = std::filesystem;
+
+    auto path = fs::path("/tmp/test_config_processor/");
+
+    fs::remove_all(path);
+    fs::create_directories(path);
+    fs::create_directories(path / "config.d");
+    //SCOPE_EXIT({ fs::remove_all(path); });
+
+    auto config_file = std::make_unique<Poco::File>(path / "config.xml");
+    {
+        DB::WriteBufferFromFile out(config_file->path());
+        writeString("<clickhouse>\n", out);
+        writeString("<merge_tree><min_bytes_for_wide_part>1</min_bytes_for_wide_part></merge_tree>\n", out);
+        writeString("</clickhouse>\n", out);
+    }
+    DB::XMLDocumentPtr config_xml;
+    DB::ConfigProcessor processor(config_file->path(), /* throw_on_bad_incl = */ false, /* log_to_console = */ false);
+    {
+        bool has_zk_includes;
+        config_xml = processor.processConfig(&has_zk_includes);
+    }
+
+    auto small_part_file = std::make_unique<Poco::File>(path / "config.d" / "part.xml");
+    {
+        DB::WriteBufferFromFile out(small_part_file->path());
+        writeString("<merge_tree><min_bytes_for_wide_part>33</min_bytes_for_wide_part></merge_tree>\n", out);
+    }
+    DB::XMLDocumentPtr part_xml;
+
+    {
+        DB::ConfigProcessor tiny(small_part_file->path(), /* throw_on_bad_incl = */ false, /* log_to_console = */ false);
+        bool has_zk_includes;
+        part_xml = tiny.processConfig(&has_zk_includes);
+    }
+
+    auto * root_node = DB::XMLUtils::getRootNode(config_xml);
+    auto * part_node = DB::XMLUtils::getRootNode(part_xml);
+    auto * new_node = config_xml->importNode(part_node, true);
+    auto * deep_root = root_node->getNodeByPath("merge_tree");
+    processor.mergeRecursive(config_xml, deep_root, new_node);
+    DB::ConfigurationPtr configuration(new Poco::Util::XMLConfiguration(config_xml));
+
+    ASSERT_EQ(configuration->getUInt64("merge_tree.min_bytes_for_wide_part"), 33);
+}
diff --git a/tests/integration/test_config_substitutions/configs/config_zk_include_test.xml b/tests/integration/test_config_substitutions/configs/config_zk_include_test.xml
new file mode 100644
index 00000000000..4a0ed623c4e
--- /dev/null
+++ b/tests/integration/test_config_substitutions/configs/config_zk_include_test.xml
@@ -0,0 +1,8 @@
+<?xml version="1.0" encoding="utf-8"?>
+<clickhouse>
+  <merge_tree>
+    <min_bytes_for_wide_part>1</min_bytes_for_wide_part>
+  </merge_tree>
+
+  <include from_zk="/min_bytes_for_wide_part"/>
+ </clickhouse>
diff --git a/tests/integration/test_config_substitutions/test.py b/tests/integration/test_config_substitutions/test.py
index 564985b2f50..08834e20423 100644
--- a/tests/integration/test_config_substitutions/test.py
+++ b/tests/integration/test_config_substitutions/test.py
@@ -13,7 +13,7 @@ node2 = cluster.add_instance(
     env_variables={"MAX_QUERY_SIZE": "55555"},
 )
 node3 = cluster.add_instance(
-    "node3", user_configs=["configs/config_zk.xml"], with_zookeeper=True
+    "node3", user_configs=["configs/config_zk.xml",], main_configs=["configs/config_zk_include_test.xml"], with_zookeeper=True
 )
 node4 = cluster.add_instance(
     "node4",
@@ -62,6 +62,11 @@ def start_cluster():
                 value=b"<user_2><password></password><profile>default</profile></user_2>",
                 makepath=True,
             )
+            zk.create(
+                path="/min_bytes_for_wide_part",
+                value=b"<merge_tree><min_bytes_for_wide_part>33</min_bytes_for_wide_part></merge_tree>",
+                makepath=True,
+            )
 
         cluster.add_zookeeper_startup_command(create_zk_roots)
 
@@ -237,3 +242,7 @@ def test_allow_databases(start_cluster):
         ).strip()
         == ""
     )
+
+def test_config_multiple_zk_substitutions(start_cluster):
+    #print(node3.query("SELECT * FROM system.merge_tree_settings"))
+    print(node3.query("SELECT * FROM system.merge_tree_settings WHERE changed=1"))

From 5ef241cc850fe95240145f86329e259f5609de31 Mon Sep 17 00:00:00 2001
From: Dmitry Novik <n0vik@clickhouse.com>
Date: Wed, 13 Mar 2024 11:30:11 +0100
Subject: [PATCH 254/443] WIP on virtual columns in StorageMerge

---
 src/Storages/StorageMerge.cpp | 42 ++++++++++++++++++++++++++++-------
 1 file changed, 34 insertions(+), 8 deletions(-)

diff --git a/src/Storages/StorageMerge.cpp b/src/Storages/StorageMerge.cpp
index 8410f0a8df8..ec03545d767 100644
--- a/src/Storages/StorageMerge.cpp
+++ b/src/Storages/StorageMerge.cpp
@@ -56,12 +56,17 @@
 #include <Storages/checkAndGetLiteralArgument.h>
 #include <base/defines.h>
 #include <base/range.h>
+#include "Common/logger_useful.h"
 #include <Common/Exception.h>
 #include <Common/assert_cast.h>
 #include <Common/checkStackSize.h>
 #include <Common/typeid_cast.h>
+#include "Analyzer/QueryNode.h"
+#include "Core/QueryProcessingStage.h"
+#include "IO/WriteHelpers.h"
 #include <Core/NamesAndTypes.h>
 #include <Functions/FunctionFactory.h>
+#include <Poco/Logger.h>
 
 namespace DB
 {
@@ -798,10 +803,13 @@ QueryTreeNodePtr replaceTableExpressionAndRemoveJoin(
     const ContextPtr & context,
     const Names & required_column_names)
 {
+    LOG_DEBUG(&Poco::Logger::get("replaceTableExpressionAndRemoveJoin"), "BEFORE:\n{}", query->dumpTree());
     auto * query_node = query->as<QueryNode>();
     auto join_tree_type = query_node->getJoinTree()->getNodeType();
     auto modified_query = query_node->cloneAndReplace(original_table_expression, replacement_table_expression);
 
+    LOG_DEBUG(&Poco::Logger::get("replaceTableExpressionAndRemoveJoin"), "AFTER:\n{}", modified_query->dumpTree());
+
     // For the case when join tree is just a table or a table function we don't need to do anything more.
     if (join_tree_type == QueryTreeNodeType::TABLE || join_tree_type == QueryTreeNodeType::TABLE_FUNCTION)
         return modified_query;
@@ -880,6 +888,7 @@ SelectQueryInfo ReadFromMerge::getModifiedQueryInfo(const ContextPtr & modified_
     if (modified_query_info.table_expression)
     {
         auto replacement_table_expression = std::make_shared<TableNode>(storage, storage_lock, storage_snapshot_);
+        replacement_table_expression->setAlias(modified_query_info.table_expression->getAlias());
         if (query_info.table_expression_modifiers)
             replacement_table_expression->setTableExpressionModifiers(*query_info.table_expression_modifiers);
 
@@ -960,6 +969,8 @@ SelectQueryInfo ReadFromMerge::getModifiedQueryInfo(const ContextPtr & modified_
                 column_name_to_node);
         }
 
+        LOG_DEBUG(&Poco::Logger::get("getModifiedQueryInfo"), "{}", modified_query_info.query_tree->dumpTree());
+
         modified_query_info.query = queryNodeToSelectQuery(modified_query_info.query_tree);
     }
     else
@@ -1020,7 +1031,7 @@ QueryPipelineBuilderPtr ReadFromMerge::createSources(
     const auto & [database_name, storage, _, table_name] = storage_with_lock;
     bool allow_experimental_analyzer = context->getSettingsRef().allow_experimental_analyzer;
     auto storage_stage
-        = storage->getQueryProcessingStage(context, QueryProcessingStage::Complete, storage_snapshot_, modified_query_info);
+        = storage->getQueryProcessingStage(context, processed_stage, storage_snapshot_, modified_query_info);
 
     builder = plan.buildQueryPipeline(
         QueryPlanOptimizationSettings::fromContext(context), BuildQueryPipelineSettings::fromContext(context));
@@ -1047,10 +1058,23 @@ QueryPipelineBuilderPtr ReadFromMerge::createSources(
 
         Block pipe_header = builder->getHeader();
 
-        if (has_database_virtual_column && common_header.has("_database") && !pipe_header.has("_database"))
+        auto get_column_options = GetColumnsOptions(GetColumnsOptions::All).withExtendedObjects().withVirtuals();
+        if (storage_snapshot_->storage.supportsSubcolumns())
+            get_column_options.withSubcolumns();
+
+        LOG_DEBUG(&Poco::Logger::get("createSources"), "Processed:{}\nStorage:{}", toString(processed_stage), toString(storage_stage));
+        
+        String table_alias;
+        if (allow_experimental_analyzer)
+            table_alias = modified_query_info.query_tree->as<QueryNode>()->getJoinTree()->as<TableNode>()->getAlias();
+
+        String database_column = table_alias.empty() || processed_stage == QueryProcessingStage::FetchColumns ? "_database" : table_alias + "._database";
+        String table_column = table_alias.empty() || processed_stage == QueryProcessingStage::FetchColumns ? "_table" : table_alias + "._table";
+
+        if (has_database_virtual_column && common_header.has(database_column) && (storage_stage == QueryProcessingStage::FetchColumns || dynamic_cast<const StorageDistributed *>(&storage_snapshot_->storage) != nullptr))
         {
             ColumnWithTypeAndName column;
-            column.name = "_database";
+            column.name = database_column;
             column.type = std::make_shared<DataTypeLowCardinality>(std::make_shared<DataTypeString>());
             column.column = column.type->createColumnConst(0, Field(database_name));
 
@@ -1062,10 +1086,10 @@ QueryPipelineBuilderPtr ReadFromMerge::createSources(
                                         { return std::make_shared<ExpressionTransform>(stream_header, adding_column_actions); });
         }
 
-        if (has_table_virtual_column && common_header.has("_table") && !pipe_header.has("_table"))
+        if (has_table_virtual_column && common_header.has(table_column) && (storage_stage == QueryProcessingStage::FetchColumns || dynamic_cast<const StorageDistributed *>(&storage_snapshot_->storage) != nullptr))
         {
             ColumnWithTypeAndName column;
-            column.name = "_table";
+            column.name = table_column;
             column.type = std::make_shared<DataTypeLowCardinality>(std::make_shared<DataTypeString>());
             column.column = column.type->createColumnConst(0, Field(table_name));
 
@@ -1080,7 +1104,7 @@ QueryPipelineBuilderPtr ReadFromMerge::createSources(
         /// Subordinary tables could have different but convertible types, like numeric types of different width.
         /// We must return streams with structure equals to structure of Merge table.
         convertAndFilterSourceStream(
-            header, modified_query_info, storage_snapshot_, aliases, row_policy_data_opt, context, *builder, processed_stage);
+            header, modified_query_info, storage_snapshot_, aliases, row_policy_data_opt, context, *builder, storage_stage);
     }
 
     return builder;
@@ -1433,7 +1457,7 @@ void ReadFromMerge::convertAndFilterSourceStream(
     const RowPolicyDataOpt & row_policy_data_opt,
     ContextPtr local_context,
     QueryPipelineBuilder & builder,
-    QueryProcessingStage::Enum processed_stage)
+    QueryProcessingStage::Enum processed_stage [[maybe_unused]])
 {
     Block before_block_header = builder.getHeader();
 
@@ -1493,7 +1517,7 @@ void ReadFromMerge::convertAndFilterSourceStream(
     ActionsDAG::MatchColumnsMode convert_actions_match_columns_mode = ActionsDAG::MatchColumnsMode::Name;
 
     if (local_context->getSettingsRef().allow_experimental_analyzer
-        && (processed_stage != QueryProcessingStage::FetchColumns || dynamic_cast<const StorageDistributed *>(&snapshot->storage) != nullptr))
+        && (processed_stage != QueryProcessingStage::FetchColumns))
         convert_actions_match_columns_mode = ActionsDAG::MatchColumnsMode::Position;
 
     if (row_policy_data_opt)
@@ -1501,6 +1525,8 @@ void ReadFromMerge::convertAndFilterSourceStream(
         row_policy_data_opt->addFilterTransform(builder);
     }
 
+    LOG_DEBUG(&Poco::Logger::get("convertAndFilterSourceStream"), "SOURCE:\n{}\nRESULT:\n{}", builder.getHeader().dumpStructure(), header.dumpStructure());
+
     auto convert_actions_dag = ActionsDAG::makeConvertingActions(builder.getHeader().getColumnsWithTypeAndName(),
                                                                 header.getColumnsWithTypeAndName(),
                                                                 convert_actions_match_columns_mode);

From 2e49140de7695095831f52b32fc965a6b07ffd3e Mon Sep 17 00:00:00 2001
From: Dmitry Novik <n0vik@clickhouse.com>
Date: Wed, 13 Mar 2024 11:37:42 +0100
Subject: [PATCH 255/443] Small progress

---
 src/Storages/StorageMerge.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Storages/StorageMerge.cpp b/src/Storages/StorageMerge.cpp
index ec03545d767..d5fd0f51a62 100644
--- a/src/Storages/StorageMerge.cpp
+++ b/src/Storages/StorageMerge.cpp
@@ -1517,7 +1517,7 @@ void ReadFromMerge::convertAndFilterSourceStream(
     ActionsDAG::MatchColumnsMode convert_actions_match_columns_mode = ActionsDAG::MatchColumnsMode::Name;
 
     if (local_context->getSettingsRef().allow_experimental_analyzer
-        && (processed_stage != QueryProcessingStage::FetchColumns))
+        && (processed_stage == QueryProcessingStage::FetchColumns && dynamic_cast<const StorageDistributed *>(&snapshot->storage) != nullptr))
         convert_actions_match_columns_mode = ActionsDAG::MatchColumnsMode::Position;
 
     if (row_policy_data_opt)

From c0bfafa203091abbe2f5af6d1253249fc432ad64 Mon Sep 17 00:00:00 2001
From: Yatsishin Ilya <2159081+qoega@users.noreply.github.com>
Date: Wed, 13 Mar 2024 10:48:02 +0000
Subject: [PATCH 256/443] fix tags

---
 .../02997_insert_select_too_many_parts_multithread.sql        | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/queries/0_stateless/02997_insert_select_too_many_parts_multithread.sql b/tests/queries/0_stateless/02997_insert_select_too_many_parts_multithread.sql
index 00cf262add5..2dfc8094115 100644
--- a/tests/queries/0_stateless/02997_insert_select_too_many_parts_multithread.sql
+++ b/tests/queries/0_stateless/02997_insert_select_too_many_parts_multithread.sql
@@ -1,5 +1,5 @@
-# Tags: disabled
-# TODO: Fix parts_to_throw_insert logic for parallel  MergeTreeSink onStart calls
+--  Tags: disabled
+-- TODO: Fix parts_to_throw_insert logic for parallel  MergeTreeSink onStart calls
 DROP TABLE IF EXISTS too_many_parts;
 
 CREATE TABLE too_many_parts (x UInt64) ENGINE = MergeTree ORDER BY tuple() SETTINGS parts_to_delay_insert = 5, parts_to_throw_insert = 5;

From a704ea510a2c180d46ba89031915624619d3b74c Mon Sep 17 00:00:00 2001
From: Konstantin Bogdanov <thevar1able@users.noreply.github.com>
Date: Wed, 13 Mar 2024 12:03:42 +0100
Subject: [PATCH 257/443] Fix `test_placement_info` (#61057)

* Fix test

* Fix last minute changes

* Automatic style fix

---------

Co-authored-by: robot-clickhouse <robot-clickhouse@users.noreply.github.com>
---
 tests/integration/helpers/mock_servers.py      |  2 +-
 .../configs/imds_bootstrap.xml                 |  9 +++++++++
 tests/integration/test_placement_info/test.py  | 18 ++++++++++--------
 3 files changed, 20 insertions(+), 9 deletions(-)
 create mode 100644 tests/integration/test_placement_info/configs/imds_bootstrap.xml

diff --git a/tests/integration/helpers/mock_servers.py b/tests/integration/helpers/mock_servers.py
index e4655ffeeaf..f2181d85e12 100644
--- a/tests/integration/helpers/mock_servers.py
+++ b/tests/integration/helpers/mock_servers.py
@@ -33,7 +33,7 @@ def start_mock_servers(cluster, script_dir, mocks, timeout=100):
 
         cluster.exec_in_container(
             container_id,
-            ["python", server_name, str(port)],
+            ["python3", server_name, str(port)],
             detach=True,
         )
 
diff --git a/tests/integration/test_placement_info/configs/imds_bootstrap.xml b/tests/integration/test_placement_info/configs/imds_bootstrap.xml
new file mode 100644
index 00000000000..5b2a77e0663
--- /dev/null
+++ b/tests/integration/test_placement_info/configs/imds_bootstrap.xml
@@ -0,0 +1,9 @@
+<clickhouse>
+    <s3>
+        <use_environment_credentials>1</use_environment_credentials>
+    </s3>
+    <placement>
+        <use_imds>0</use_imds>
+        <availability_zone>ci-placeholder</availability_zone>
+    </placement>
+</clickhouse>
diff --git a/tests/integration/test_placement_info/test.py b/tests/integration/test_placement_info/test.py
index 1b93a3eae0b..32fd2fa57d7 100644
--- a/tests/integration/test_placement_info/test.py
+++ b/tests/integration/test_placement_info/test.py
@@ -2,16 +2,14 @@ import pytest
 from helpers.cluster import ClickHouseCluster
 from helpers.mock_servers import start_mock_servers
 import os
-import time
 
-METADATA_SERVER_HOSTNAME = "resolver"
+METADATA_SERVER_HOSTNAME = "node_imds"
 METADATA_SERVER_PORT = 8080
 
 cluster = ClickHouseCluster(__file__)
 node_imds = cluster.add_instance(
     "node_imds",
-    with_minio=True,
-    main_configs=["configs/imds.xml"],
+    main_configs=["configs/imds_bootstrap.xml"],
     env_variables={
         "AWS_EC2_METADATA_SERVICE_ENDPOINT": f"http://{METADATA_SERVER_HOSTNAME}:{METADATA_SERVER_PORT}",
     },
@@ -32,10 +30,10 @@ node_missing_value = cluster.add_instance(
 )
 
 
-def start_metadata_server():
+def start_metadata_server(started_cluster):
     script_dir = os.path.join(os.path.dirname(__file__), "metadata_servers")
     start_mock_servers(
-        cluster,
+        started_cluster,
         script_dir,
         [
             (
@@ -51,13 +49,17 @@ def start_metadata_server():
 def start_cluster():
     try:
         cluster.start()
-        start_metadata_server()
-        yield
+        start_metadata_server(cluster)
+        yield cluster
     finally:
         cluster.shutdown()
 
 
 def test_placement_info_from_imds():
+    with open(os.path.join(os.path.dirname(__file__), "configs/imds.xml"), "r") as f:
+        node_imds.replace_config(
+            "/etc/clickhouse-server/config.d/imds_bootstrap.xml", f.read()
+        )
     node_imds.stop_clickhouse(kill=True)
     node_imds.start_clickhouse()
 

From 0353121dccb87403ff08334137fd9cecbb8953f1 Mon Sep 17 00:00:00 2001
From: peter279k <peter279k@gmail.com>
Date: Wed, 13 Mar 2024 19:17:27 +0800
Subject: [PATCH 258/443] Improve related NULL functions usage

---
 docs/en/sql-reference/functions/functions-for-nulls.md | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/docs/en/sql-reference/functions/functions-for-nulls.md b/docs/en/sql-reference/functions/functions-for-nulls.md
index 91c04cfded3..e73d6c899e7 100644
--- a/docs/en/sql-reference/functions/functions-for-nulls.md
+++ b/docs/en/sql-reference/functions/functions-for-nulls.md
@@ -14,7 +14,7 @@ Returns whether the argument is [NULL](../../sql-reference/syntax.md#null).
 isNull(x)
 ```
 
-Alias: `ISNULL`.
+Alias: `IS NULL`.
 
 **Arguments**
 
@@ -58,6 +58,8 @@ Returns whether the argument is not [NULL](../../sql-reference/syntax.md#null-li
 isNotNull(x)
 ```
 
+Alias: `IS NOT NULL`.
+
 **Arguments:**
 
 - `x` — A value of non-compound data type.
@@ -100,6 +102,8 @@ Returns whether the argument is 0 (zero) or [NULL](../../sql-reference/syntax.md
 isZeroOrNull(x)
 ```
 
+Alias: `x = 0 OR x IS NULL`.
+
 **Arguments:**
 
 - `x` — A value of non-compound data type.

From 3931351ec4806769048d2638f54323f1ae89e056 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Wed, 13 Mar 2024 12:36:47 +0100
Subject: [PATCH 259/443] Pass timeout through setting

---
 src/Core/Settings.h                               |  1 +
 src/Disks/IO/CachedOnDiskReadBufferFromFile.cpp   |  5 +++--
 src/Disks/IO/CachedOnDiskWriteBufferFromFile.cpp  | 10 +++++++---
 src/Disks/IO/CachedOnDiskWriteBufferFromFile.h    | 15 +++++++++------
 src/IO/ReadSettings.h                             |  1 +
 src/IO/WriteSettings.h                            |  1 +
 src/Interpreters/Cache/FileCache.cpp              |  5 +++--
 src/Interpreters/Cache/FileCache.h                |  3 ++-
 src/Interpreters/Cache/FileCache_fwd.h            |  1 -
 src/Interpreters/Cache/FileSegment.cpp            |  4 ++--
 src/Interpreters/Cache/FileSegment.h              |  2 +-
 src/Interpreters/Cache/Metadata.cpp               |  6 +++++-
 .../Cache/WriteBufferToFileSegment.cpp            |  8 +++++++-
 src/Interpreters/Cache/WriteBufferToFileSegment.h |  2 ++
 src/Interpreters/Context.cpp                      |  2 ++
 15 files changed, 46 insertions(+), 20 deletions(-)

diff --git a/src/Core/Settings.h b/src/Core/Settings.h
index d70a6cf51c5..7ba335099e6 100644
--- a/src/Core/Settings.h
+++ b/src/Core/Settings.h
@@ -778,6 +778,7 @@ class IColumn;
     M(UInt64, filesystem_cache_max_download_size, (128UL * 1024 * 1024 * 1024), "Max remote filesystem cache size that can be downloaded by a single query", 0) \
     M(Bool, throw_on_error_from_cache_on_write_operations, false, "Ignore error from cache when caching on write operations (INSERT, merges)", 0) \
     M(UInt64, filesystem_cache_segments_batch_size, 20, "Limit on size of a single batch of file segments that a read buffer can request from cache. Too low value will lead to excessive requests to cache, too large may slow down eviction from cache", 0) \
+    M(UInt64, filesystem_cache_reserve_space_wait_lock_timeout_milliseconds, 1000, "Wait time to lock cache for sapce reservation in filesystem cache", 0) \
     \
     M(Bool, use_page_cache_for_disks_without_file_cache, false, "Use userspace page cache for remote disks that don't have filesystem cache enabled.", 0) \
     M(Bool, read_from_page_cache_if_exists_otherwise_bypass_cache, false, "Use userspace page cache in passive mode, similar to read_from_filesystem_cache_if_exists_otherwise_bypass_cache.", 0) \
diff --git a/src/Disks/IO/CachedOnDiskReadBufferFromFile.cpp b/src/Disks/IO/CachedOnDiskReadBufferFromFile.cpp
index 47ee5858562..1e108b481ee 100644
--- a/src/Disks/IO/CachedOnDiskReadBufferFromFile.cpp
+++ b/src/Disks/IO/CachedOnDiskReadBufferFromFile.cpp
@@ -637,7 +637,8 @@ void CachedOnDiskReadBufferFromFile::predownload(FileSegment & file_segment)
 
             ProfileEvents::increment(ProfileEvents::CachedReadBufferReadFromSourceBytes, current_impl_buffer_size);
 
-            bool continue_predownload = file_segment.reserve(current_predownload_size);
+            bool continue_predownload = file_segment.reserve(
+                current_predownload_size, settings.filesystem_cache_reserve_space_wait_lock_timeout_milliseconds);
             if (continue_predownload)
             {
                 LOG_TEST(log, "Left to predownload: {}, buffer size: {}", bytes_to_predownload, current_impl_buffer_size);
@@ -992,7 +993,7 @@ bool CachedOnDiskReadBufferFromFile::nextImplStep()
         {
             chassert(file_offset_of_buffer_end + size - 1 <= file_segment.range().right);
 
-            bool success = file_segment.reserve(size);
+            bool success = file_segment.reserve(size, settings.filesystem_cache_reserve_space_wait_lock_timeout_milliseconds);
             if (success)
             {
                 chassert(file_segment.getCurrentWriteOffset() == static_cast<size_t>(implementation_buffer->getPosition()));
diff --git a/src/Disks/IO/CachedOnDiskWriteBufferFromFile.cpp b/src/Disks/IO/CachedOnDiskWriteBufferFromFile.cpp
index faed55de713..f4e309f461e 100644
--- a/src/Disks/IO/CachedOnDiskWriteBufferFromFile.cpp
+++ b/src/Disks/IO/CachedOnDiskWriteBufferFromFile.cpp
@@ -26,16 +26,18 @@ FileSegmentRangeWriter::FileSegmentRangeWriter(
     FileCache * cache_,
     const FileSegment::Key & key_,
     const FileCacheUserInfo & user_,
+    size_t reserve_space_lock_wait_timeout_milliseconds_,
     std::shared_ptr<FilesystemCacheLog> cache_log_,
     const String & query_id_,
     const String & source_path_)
     : cache(cache_)
     , key(key_)
+    , user(user_)
+    , reserve_space_lock_wait_timeout_milliseconds(reserve_space_lock_wait_timeout_milliseconds_)
     , log(getLogger("FileSegmentRangeWriter"))
     , cache_log(cache_log_)
     , query_id(query_id_)
     , source_path(source_path_)
-    , user(user_)
 {
 }
 
@@ -89,7 +91,7 @@ bool FileSegmentRangeWriter::write(const char * data, size_t size, size_t offset
 
         size_t size_to_write = std::min(available_size, size);
 
-        bool reserved = file_segment->reserve(size_to_write);
+        bool reserved = file_segment->reserve(size_to_write, reserve_space_lock_wait_timeout_milliseconds);
         if (!reserved)
         {
             appendFilesystemCacheLog(*file_segment);
@@ -211,6 +213,7 @@ CachedOnDiskWriteBufferFromFile::CachedOnDiskWriteBufferFromFile(
     , key(key_)
     , query_id(query_id_)
     , user(user_)
+    , reserve_space_lock_wait_timeout_milliseconds(settings_.filesystem_cache_reserve_space_wait_lock_timeout_milliseconds)
     , throw_on_error_from_cache(settings_.throw_on_error_from_cache)
     , cache_log(!query_id_.empty() && settings_.enable_filesystem_cache_log ? cache_log_ : nullptr)
 {
@@ -251,7 +254,8 @@ void CachedOnDiskWriteBufferFromFile::cacheData(char * data, size_t size, bool t
 
     if (!cache_writer)
     {
-        cache_writer = std::make_unique<FileSegmentRangeWriter>(cache.get(), key, user, cache_log, query_id, source_path);
+        cache_writer = std::make_unique<FileSegmentRangeWriter>(
+            cache.get(), key, user, reserve_space_lock_wait_timeout_milliseconds, cache_log, query_id, source_path);
     }
 
     Stopwatch watch(CLOCK_MONOTONIC);
diff --git a/src/Disks/IO/CachedOnDiskWriteBufferFromFile.h b/src/Disks/IO/CachedOnDiskWriteBufferFromFile.h
index 59e0c76ca3d..ad4f6b5916d 100644
--- a/src/Disks/IO/CachedOnDiskWriteBufferFromFile.h
+++ b/src/Disks/IO/CachedOnDiskWriteBufferFromFile.h
@@ -30,6 +30,7 @@ public:
         FileCache * cache_,
         const FileSegment::Key & key_,
         const FileCacheUserInfo & user_,
+        size_t reserve_space_lock_wait_timeout_milliseconds_,
         std::shared_ptr<FilesystemCacheLog> cache_log_,
         const String & query_id_,
         const String & source_path_);
@@ -52,13 +53,14 @@ private:
     void completeFileSegment();
 
     FileCache * cache;
-    FileSegment::Key key;
+    const FileSegment::Key key;
+    const FileCacheUserInfo user;
+    const size_t reserve_space_lock_wait_timeout_milliseconds;
 
     LoggerPtr log;
     std::shared_ptr<FilesystemCacheLog> cache_log;
-    String query_id;
-    String source_path;
-    FileCacheUserInfo user;
+    const String query_id;
+    const String source_path;
 
     FileSegmentsHolderPtr file_segments;
 
@@ -99,11 +101,12 @@ private:
     String source_path;
     FileCacheKey key;
 
-    size_t current_download_offset = 0;
     const String query_id;
     const FileCacheUserInfo user;
+    const size_t reserve_space_lock_wait_timeout_milliseconds;
+    const bool throw_on_error_from_cache;
 
-    bool throw_on_error_from_cache;
+    size_t current_download_offset = 0;
     bool cache_in_error_state_or_disabled = false;
 
     std::unique_ptr<FileSegmentRangeWriter> cache_writer;
diff --git a/src/IO/ReadSettings.h b/src/IO/ReadSettings.h
index c0a63bf51b1..6a0cac35878 100644
--- a/src/IO/ReadSettings.h
+++ b/src/IO/ReadSettings.h
@@ -100,6 +100,7 @@ struct ReadSettings
     bool read_from_filesystem_cache_if_exists_otherwise_bypass_cache = false;
     bool enable_filesystem_cache_log = false;
     size_t filesystem_cache_segments_batch_size = 20;
+    size_t filesystem_cache_reserve_space_wait_lock_timeout_milliseconds = 1000;
 
     bool use_page_cache_for_disks_without_file_cache = false;
     bool read_from_page_cache_if_exists_otherwise_bypass_cache = false;
diff --git a/src/IO/WriteSettings.h b/src/IO/WriteSettings.h
index fcadf34f021..7d36677b468 100644
--- a/src/IO/WriteSettings.h
+++ b/src/IO/WriteSettings.h
@@ -20,6 +20,7 @@ struct WriteSettings
     bool enable_filesystem_cache_on_write_operations = false;
     bool enable_filesystem_cache_log = false;
     bool throw_on_error_from_cache = false;
+    size_t filesystem_cache_reserve_space_wait_lock_timeout_milliseconds = 1000;
 
     bool s3_allow_parallel_part_upload = true;
 
diff --git a/src/Interpreters/Cache/FileCache.cpp b/src/Interpreters/Cache/FileCache.cpp
index 5650b9ce44e..ea40ffcfa3c 100644
--- a/src/Interpreters/Cache/FileCache.cpp
+++ b/src/Interpreters/Cache/FileCache.cpp
@@ -777,12 +777,13 @@ bool FileCache::tryReserve(
     FileSegment & file_segment,
     const size_t size,
     FileCacheReserveStat & reserve_stat,
-    const UserInfo & user)
+    const UserInfo & user,
+    size_t lock_wait_timeout_milliseconds)
 {
     ProfileEventTimeIncrement<Microseconds> watch(ProfileEvents::FilesystemCacheReserveMicroseconds);
 
     assertInitialized();
-    auto cache_lock = tryLockCache(std::chrono::milliseconds(FILECACHE_TRY_RESERVE_LOCK_TIMEOUT_MILLISECONDS));
+    auto cache_lock = tryLockCache(std::chrono::milliseconds(lock_wait_timeout_milliseconds));
     if (!cache_lock)
     {
         ProfileEvents::increment(ProfileEvents::FilesystemCacheFailToReserveSpaceBecauseOfLockContention);
diff --git a/src/Interpreters/Cache/FileCache.h b/src/Interpreters/Cache/FileCache.h
index 7434b2ac78a..007c4fd9483 100644
--- a/src/Interpreters/Cache/FileCache.h
+++ b/src/Interpreters/Cache/FileCache.h
@@ -161,7 +161,8 @@ public:
         FileSegment & file_segment,
         size_t size,
         FileCacheReserveStat & stat,
-        const UserInfo & user);
+        const UserInfo & user,
+        size_t lock_wait_timeout_milliseconds);
 
     std::vector<FileSegment::Info> getFileSegmentInfos(const UserID & user_id);
 
diff --git a/src/Interpreters/Cache/FileCache_fwd.h b/src/Interpreters/Cache/FileCache_fwd.h
index eaed279e7fd..06261b19db7 100644
--- a/src/Interpreters/Cache/FileCache_fwd.h
+++ b/src/Interpreters/Cache/FileCache_fwd.h
@@ -12,7 +12,6 @@ static constexpr int FILECACHE_DEFAULT_LOAD_METADATA_THREADS = 16;
 static constexpr int FILECACHE_DEFAULT_MAX_ELEMENTS = 10000000;
 static constexpr int FILECACHE_DEFAULT_HITS_THRESHOLD = 0;
 static constexpr size_t FILECACHE_BYPASS_THRESHOLD = 256 * 1024 * 1024;
-static constexpr size_t FILECACHE_TRY_RESERVE_LOCK_TIMEOUT_MILLISECONDS = 1000; /// 1 sec.
 
 class FileCache;
 using FileCachePtr = std::shared_ptr<FileCache>;
diff --git a/src/Interpreters/Cache/FileSegment.cpp b/src/Interpreters/Cache/FileSegment.cpp
index 6b2d4a4bec8..9ec2b090dc7 100644
--- a/src/Interpreters/Cache/FileSegment.cpp
+++ b/src/Interpreters/Cache/FileSegment.cpp
@@ -497,7 +497,7 @@ LockedKeyPtr FileSegment::lockKeyMetadata(bool assert_exists) const
     return metadata->tryLock();
 }
 
-bool FileSegment::reserve(size_t size_to_reserve, FileCacheReserveStat * reserve_stat)
+bool FileSegment::reserve(size_t size_to_reserve, size_t lock_wait_timeout_milliseconds, FileCacheReserveStat * reserve_stat)
 {
     if (!size_to_reserve)
         throw Exception(ErrorCodes::LOGICAL_ERROR, "Zero space reservation is not allowed");
@@ -549,7 +549,7 @@ bool FileSegment::reserve(size_t size_to_reserve, FileCacheReserveStat * reserve
     if (!reserve_stat)
         reserve_stat = &dummy_stat;
 
-    bool reserved = cache->tryReserve(*this, size_to_reserve, *reserve_stat, getKeyMetadata()->user);
+    bool reserved = cache->tryReserve(*this, size_to_reserve, *reserve_stat, getKeyMetadata()->user, lock_wait_timeout_milliseconds);
 
     if (!reserved)
         setDownloadFailedUnlocked(lockFileSegment());
diff --git a/src/Interpreters/Cache/FileSegment.h b/src/Interpreters/Cache/FileSegment.h
index ea97a6b0157..c34ee064345 100644
--- a/src/Interpreters/Cache/FileSegment.h
+++ b/src/Interpreters/Cache/FileSegment.h
@@ -199,7 +199,7 @@ public:
 
     /// Try to reserve exactly `size` bytes (in addition to the getDownloadedSize() bytes already downloaded).
     /// Returns true if reservation was successful, false otherwise.
-    bool reserve(size_t size_to_reserve, FileCacheReserveStat * reserve_stat = nullptr);
+    bool reserve(size_t size_to_reserve, size_t lock_wait_timeout_milliseconds, FileCacheReserveStat * reserve_stat = nullptr);
 
     /// Write data into reserved space.
     void write(const char * from, size_t size, size_t offset);
diff --git a/src/Interpreters/Cache/Metadata.cpp b/src/Interpreters/Cache/Metadata.cpp
index 727f2762cca..b79605622b6 100644
--- a/src/Interpreters/Cache/Metadata.cpp
+++ b/src/Interpreters/Cache/Metadata.cpp
@@ -1,6 +1,7 @@
 #include <Interpreters/Cache/Metadata.h>
 #include <Interpreters/Cache/FileCache.h>
 #include <Interpreters/Cache/FileSegment.h>
+#include <Interpreters/Context.h>
 #include <Common/logger_useful.h>
 #include <Common/ElapsedTimeProfileEventIncrement.h>
 #include <filesystem>
@@ -693,6 +694,9 @@ void CacheMetadata::downloadImpl(FileSegment & file_segment, std::optional<Memor
         reader->set(memory->data(), memory->size());
     }
 
+    const auto reserve_space_lock_wait_timeout_milliseconds =
+        Context::getGlobalContextInstance()->getReadSettings().filesystem_cache_reserve_space_wait_lock_timeout_milliseconds;
+
     size_t offset = file_segment.getCurrentWriteOffset();
     if (offset != static_cast<size_t>(reader->getPosition()))
         reader->seek(offset, SEEK_SET);
@@ -701,7 +705,7 @@ void CacheMetadata::downloadImpl(FileSegment & file_segment, std::optional<Memor
     {
         auto size = reader->available();
 
-        if (!file_segment.reserve(size))
+        if (!file_segment.reserve(size, reserve_space_lock_wait_timeout_milliseconds))
         {
             LOG_TEST(
                 log, "Failed to reserve space during background download "
diff --git a/src/Interpreters/Cache/WriteBufferToFileSegment.cpp b/src/Interpreters/Cache/WriteBufferToFileSegment.cpp
index 7cd4e2d6e8d..759135722dc 100644
--- a/src/Interpreters/Cache/WriteBufferToFileSegment.cpp
+++ b/src/Interpreters/Cache/WriteBufferToFileSegment.cpp
@@ -1,6 +1,7 @@
 #include <Interpreters/Cache/WriteBufferToFileSegment.h>
 #include <Interpreters/Cache/FileSegment.h>
 #include <Interpreters/Cache/FileCache.h>
+#include <Interpreters/Context.h>
 #include <IO/SwapHelper.h>
 #include <IO/ReadBufferFromFile.h>
 
@@ -32,6 +33,11 @@ WriteBufferToFileSegment::WriteBufferToFileSegment(FileSegmentsHolderPtr segment
     , file_segment(&segment_holder_->front())
     , segment_holder(std::move(segment_holder_))
 {
+    auto query_context = CurrentThread::getQueryContext();
+    if (query_context)
+        reserve_space_lock_wait_timeout_milliseconds = query_context->getReadSettings().filesystem_cache_reserve_space_wait_lock_timeout_milliseconds;
+    else
+        reserve_space_lock_wait_timeout_milliseconds = Context::getGlobalContextInstance()->getReadSettings().filesystem_cache_reserve_space_wait_lock_timeout_milliseconds;
 }
 
 /// If it throws an exception, the file segment will be incomplete, so you should not use it in the future.
@@ -49,7 +55,7 @@ void WriteBufferToFileSegment::nextImpl()
     FileCacheReserveStat reserve_stat;
     /// In case of an error, we don't need to finalize the file segment
     /// because it will be deleted soon and completed in the holder's destructor.
-    bool ok = file_segment->reserve(bytes_to_write, &reserve_stat);
+    bool ok = file_segment->reserve(bytes_to_write, reserve_space_lock_wait_timeout_milliseconds, &reserve_stat);
 
     if (!ok)
     {
diff --git a/src/Interpreters/Cache/WriteBufferToFileSegment.h b/src/Interpreters/Cache/WriteBufferToFileSegment.h
index feb33472513..bff340d79b3 100644
--- a/src/Interpreters/Cache/WriteBufferToFileSegment.h
+++ b/src/Interpreters/Cache/WriteBufferToFileSegment.h
@@ -28,6 +28,8 @@ private:
 
     /// Empty if file_segment is not owned by this WriteBufferToFileSegment
     FileSegmentsHolderPtr segment_holder;
+
+    size_t reserve_space_lock_wait_timeout_milliseconds;
 };
 
 
diff --git a/src/Interpreters/Context.cpp b/src/Interpreters/Context.cpp
index d658fbe9920..6a0657a842c 100644
--- a/src/Interpreters/Context.cpp
+++ b/src/Interpreters/Context.cpp
@@ -5166,6 +5166,7 @@ ReadSettings Context::getReadSettings() const
     res.read_from_filesystem_cache_if_exists_otherwise_bypass_cache = settings.read_from_filesystem_cache_if_exists_otherwise_bypass_cache;
     res.enable_filesystem_cache_log = settings.enable_filesystem_cache_log;
     res.filesystem_cache_segments_batch_size = settings.filesystem_cache_segments_batch_size;
+    res.filesystem_cache_reserve_space_wait_lock_timeout_milliseconds = settings.filesystem_cache_reserve_space_wait_lock_timeout_milliseconds;
 
     res.filesystem_cache_max_download_size = settings.filesystem_cache_max_download_size;
     res.skip_download_if_exceeds_query_cache = settings.skip_download_if_exceeds_query_cache;
@@ -5214,6 +5215,7 @@ WriteSettings Context::getWriteSettings() const
     res.enable_filesystem_cache_on_write_operations = settings.enable_filesystem_cache_on_write_operations;
     res.enable_filesystem_cache_log = settings.enable_filesystem_cache_log;
     res.throw_on_error_from_cache = settings.throw_on_error_from_cache_on_write_operations;
+    res.filesystem_cache_reserve_space_wait_lock_timeout_milliseconds = settings.filesystem_cache_reserve_space_wait_lock_timeout_milliseconds;
 
     res.s3_allow_parallel_part_upload = settings.s3_allow_parallel_part_upload;
 

From ef796c668d187c80f13223dbf679dcc442907008 Mon Sep 17 00:00:00 2001
From: Nikolai Kochetov <nk@clickhouse.com>
Date: Wed, 13 Mar 2024 12:03:37 +0000
Subject: [PATCH 260/443] Fixing
 test_build_sets_from_multiple_threads/test.py::test_set

---
 src/Planner/Planner.cpp                     | 4 +++-
 tests/analyzer_integration_broken_tests.txt | 1 -
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/src/Planner/Planner.cpp b/src/Planner/Planner.cpp
index bc1fb30781d..861b12e3da2 100644
--- a/src/Planner/Planner.cpp
+++ b/src/Planner/Planner.cpp
@@ -1081,7 +1081,9 @@ void addBuildSubqueriesForSetsStepIfNeeded(
     for (auto & subquery : subqueries)
     {
         auto query_tree = subquery->detachQueryTree();
-        auto subquery_options = select_query_options.subquery();
+        /// I suppose it should be better to use all flags from select_query_options,
+        /// But for now it is done in the same way as in old analyzer.
+        auto subquery_options = SelectQueryOptions(QueryProcessingStage::Complete, select_query_options.subquery_depth).subquery();
         Planner subquery_planner(
             query_tree,
             subquery_options,
diff --git a/tests/analyzer_integration_broken_tests.txt b/tests/analyzer_integration_broken_tests.txt
index 31527dc3476..e71a047c215 100644
--- a/tests/analyzer_integration_broken_tests.txt
+++ b/tests/analyzer_integration_broken_tests.txt
@@ -1,4 +1,3 @@
-test_build_sets_from_multiple_threads/test.py::test_set
 test_concurrent_backups_s3/test.py::test_concurrent_backups
 test_distributed_type_object/test.py::test_distributed_type_object
 test_merge_table_over_distributed/test.py::test_global_in

From 1a47682c12a45acb888c89974302ec35bee5aaed Mon Sep 17 00:00:00 2001
From: Dmitry Novik <n0vik@clickhouse.com>
Date: Wed, 13 Mar 2024 13:08:12 +0100
Subject: [PATCH 261/443] Analyzer: Fix virtual columns in StorageMerge
 #ci_set_analyzer

---
 src/Interpreters/InterpreterSelectQueryAnalyzer.cpp | 1 +
 src/Storages/StorageMerge.cpp                       | 8 +++++---
 tests/analyzer_tech_debt.txt                        | 1 -
 3 files changed, 6 insertions(+), 4 deletions(-)

diff --git a/src/Interpreters/InterpreterSelectQueryAnalyzer.cpp b/src/Interpreters/InterpreterSelectQueryAnalyzer.cpp
index 922f4a99b4a..539d7a59f6f 100644
--- a/src/Interpreters/InterpreterSelectQueryAnalyzer.cpp
+++ b/src/Interpreters/InterpreterSelectQueryAnalyzer.cpp
@@ -86,6 +86,7 @@ void replaceStorageInQueryTree(QueryTreeNodePtr & query_tree, const ContextPtr &
             continue;
 
         auto replacement_table_expression = std::make_shared<TableNode>(storage, context);
+        replacement_table_expression->setAlias(node->getAlias());
 
         if (auto table_expression_modifiers = table_node.getTableExpressionModifiers())
             replacement_table_expression->setTableExpressionModifiers(*table_expression_modifiers);
diff --git a/src/Storages/StorageMerge.cpp b/src/Storages/StorageMerge.cpp
index d5fd0f51a62..fab4b2e5146 100644
--- a/src/Storages/StorageMerge.cpp
+++ b/src/Storages/StorageMerge.cpp
@@ -1135,13 +1135,15 @@ QueryPlan ReadFromMerge::createPlanForTable(
     bool allow_experimental_analyzer = modified_context->getSettingsRef().allow_experimental_analyzer;
 
     auto storage_stage = storage->getQueryProcessingStage(modified_context,
-        QueryProcessingStage::Complete,
+        processed_stage,
         storage_snapshot_,
         modified_query_info);
 
+    LOG_DEBUG(&Poco::Logger::get("createPlanForTable"), "Storage: {}", toString(storage_stage));
+
     QueryPlan plan;
 
-    if (processed_stage <= storage_stage || (allow_experimental_analyzer && processed_stage == QueryProcessingStage::FetchColumns))
+    if (processed_stage <= storage_stage)
     {
         /// If there are only virtual columns in query, you must request at least one other column.
         if (real_column_names.empty())
@@ -1186,7 +1188,7 @@ QueryPlan ReadFromMerge::createPlanForTable(
                 row_policy_data_opt->addStorageFilter(source_step_with_filter);
         }
     }
-    else if (processed_stage > storage_stage || (allow_experimental_analyzer && processed_stage != QueryProcessingStage::FetchColumns))
+    else if (processed_stage > storage_stage || allow_experimental_analyzer)
     {
         /// Maximum permissible parallelism is streams_num
         modified_context->setSetting("max_threads", streams_num);
diff --git a/tests/analyzer_tech_debt.txt b/tests/analyzer_tech_debt.txt
index dc6284d20c5..cee3cff8cd5 100644
--- a/tests/analyzer_tech_debt.txt
+++ b/tests/analyzer_tech_debt.txt
@@ -1,5 +1,4 @@
 00223_shard_distributed_aggregation_memory_efficient
-00717_merge_and_distributed
 00725_memory_tracking
 01062_pm_all_join_with_block_continuation
 01083_expressions_in_engine_arguments

From 9da03607282b8fc338be4356322c67229686fd39 Mon Sep 17 00:00:00 2001
From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com>
Date: Wed, 13 Mar 2024 13:10:48 +0100
Subject: [PATCH 262/443] Reload CI


From 6203d45b966c91a18b2aee6b6da6e3dd3d36db62 Mon Sep 17 00:00:00 2001
From: alesapin <alesapin@gmail.com>
Date: Wed, 13 Mar 2024 13:12:48 +0100
Subject: [PATCH 263/443] working version with comments

---
 programs/server/Server.cpp                    |  6 +-
 src/Common/Config/ConfigProcessor.cpp         | 71 ++++++++++++++-----
 src/Core/BaseSettings.cpp                     |  4 +-
 .../configs/config_zk_include_test.xml        |  6 +-
 .../test_config_substitutions/test.py         | 40 ++++++++++-
 5 files changed, 104 insertions(+), 23 deletions(-)

diff --git a/programs/server/Server.cpp b/programs/server/Server.cpp
index 7636f221ab5..9d7dd02a235 100644
--- a/programs/server/Server.cpp
+++ b/programs/server/Server.cpp
@@ -733,8 +733,6 @@ try
     LOG_INFO(log, "Available CPU instruction sets: {}", cpu_info);
 #endif
 
-    sanityChecks(*this);
-
     // Initialize global thread pool. Do it before we fetch configs from zookeeper
     // nodes (`from_zk`), because ZooKeeper interface uses the pool. We will
     // ignore `max_thread_pool_size` in configs we fetch from ZK, but oh well.
@@ -904,6 +902,7 @@ try
         config_processor.savePreprocessedConfig(loaded_config, config().getString("path", DBMS_DEFAULT_PATH));
         config().removeConfiguration(old_configuration.get());
         config().add(loaded_config.configuration.duplicate(), PRIO_DEFAULT, false);
+        global_context->setConfig(loaded_config.configuration);
     }
 
     Settings::checkNoSettingNamesAtTopLevel(config(), config_path);
@@ -911,6 +910,9 @@ try
     /// We need to reload server settings because config could be updated via zookeeper.
     server_settings.loadSettingsFromConfig(config());
 
+    /// NOTE: Do sanity checks after we loaded all possible substituions from ZK
+    sanityChecks(*this);
+
 #if defined(OS_LINUX)
     std::string executable_path = getExecutablePath();
 
diff --git a/src/Common/Config/ConfigProcessor.cpp b/src/Common/Config/ConfigProcessor.cpp
index 57e9f9efc10..75a973f582f 100644
--- a/src/Common/Config/ConfigProcessor.cpp
+++ b/src/Common/Config/ConfigProcessor.cpp
@@ -272,6 +272,7 @@ void ConfigProcessor::hideRecursive(Poco::XML::Node * config_root)
 
 void ConfigProcessor::mergeRecursive(XMLDocumentPtr config, Node * config_root, const Node * with_root)
 {
+    //LOG_DEBUG(log, "WITH ROOT {}", with_root->nodeName());
     const NodeListPtr with_nodes = with_root->childNodes();
     using ElementsByIdentifier = std::multimap<ElementIdentifier, Node *>;
     ElementsByIdentifier config_element_by_id;
@@ -287,7 +288,7 @@ void ConfigProcessor::mergeRecursive(XMLDocumentPtr config, Node * config_root,
         }
         else if (node->nodeType() == Node::ELEMENT_NODE)
         {
-            std::cerr << "NODES IN SOURCE: " << node->nodeName() << std::endl;
+            //LOG_DEBUG(log, "NODES IN SOURCE: {}", node->nodeName());
             config_element_by_id.insert(ElementsByIdentifier::value_type(getElementIdentifier(node), node));
         }
         node = next_node;
@@ -301,6 +302,7 @@ void ConfigProcessor::mergeRecursive(XMLDocumentPtr config, Node * config_root,
         bool remove = false;
         if (with_node->nodeType() == Node::ELEMENT_NODE)
         {
+            //LOG_DEBUG(log, "WITH NODE: {}", with_node->nodeName());
             //std::cerr << "WITH NODE: " << with_node->nodeName() << std::endl;
             Element & with_element = dynamic_cast<Element &>(*with_node);
             remove = with_element.hasAttribute("remove");
@@ -315,6 +317,7 @@ void ConfigProcessor::mergeRecursive(XMLDocumentPtr config, Node * config_root,
             if (it != config_element_by_id.end())
             {
                 Node * config_node = it->second;
+                //LOG_DEBUG(log, "SUBNODE NODE: {}", config_node->nodeName());
                 //std::cerr << "SUBNODE NODE: " << config_node->nodeName() << std::endl;
                 config_element_by_id.erase(it);
 
@@ -324,6 +327,7 @@ void ConfigProcessor::mergeRecursive(XMLDocumentPtr config, Node * config_root,
                 }
                 else if (replace)
                 {
+                    //LOG_DEBUG(log, "REPLACE: {}", config_node->nodeName());
                     //std::cerr << "REPLACE!!!" << std::endl;
                     with_element.removeAttribute("replace");
                     NodePtr new_node = config->importNode(with_node, true);
@@ -331,6 +335,7 @@ void ConfigProcessor::mergeRecursive(XMLDocumentPtr config, Node * config_root,
                 }
                 else
                 {
+                    //LOG_DEBUG(log, "SUBNODE NODE HERE: {}", config_node->nodeName());
                     //std::cerr << "SUBNODE NODE HERE: " << config_node->nodeName() << std::endl;
                     Element & config_element = dynamic_cast<Element &>(*config_node);
 
@@ -346,13 +351,15 @@ void ConfigProcessor::mergeRecursive(XMLDocumentPtr config, Node * config_root,
                 //std::cerr << "DONE\n";
                 merged = true;
             }
-        }
-        else
-        {
-            //std::cerr << "ELEMENT NOT FOUND\n";
+            //else
+            //{
+            //    LOG_DEBUG(log, "ELEMENT NOT FOUND");
+            //    //std::cerr << "ELEMENT NOT FOUND\n";
+            //}
         }
         if (!merged && !remove)
         {
+            //LOG_DEBUG(log, "NOTHING HAPPENED");
             //std::cerr << "NOTHING hAPPENED\n";
             /// Since we didn't find a pair to this node in default config, we will paste it as is.
             /// But it may have some child nodes which have attributes like "replace" or "remove".
@@ -443,13 +450,14 @@ void ConfigProcessor::doIncludesRecursive(
 
     /// Replace the original contents, not add to it.
     bool replace = attributes->getNamedItem("replace");
+    bool merge = attributes->getNamedItem("merge");
 
     bool included_something = false;
 
     auto process_include = [&](const Node * include_attr, const std::function<const Node * (const std::string &)> & get_node, const char * error_msg)
     {
         const std::string & name = include_attr->getNodeValue();
-        LOG_DEBUG(log, "PROCESS INCLUDE {}", name);
+        //LOG_DEBUG(log, "PROCESS INCLUDE {}", name);
         const Node * node_to_include = get_node(name);
         if (!node_to_include)
         {
@@ -467,26 +475,38 @@ void ConfigProcessor::doIncludesRecursive(
         }
         else
         {
+
+            Element & element = dynamic_cast<Element &>(*node);
             /// Replace the whole node not just contents.
             if (node->nodeName() == "include")
             {
-                LOG_DEBUG(log, "Include here for node {}", name);
+                //LOG_DEBUG(log, "Include here for node {}", name);
                 const NodeListPtr children = node_to_include->childNodes();
                 Node * next_child = nullptr;
+
                 for (Node * child = children->item(0); child; child = next_child)
                 {
                     next_child = child->nextSibling();
-                    NodePtr new_node = config->importNode(child, true);
-                    //node->parentNode()->insertBefore(new_node, node);
-                    mergeRecursive(config, node->parentNode(), new_node);
+
+                    //LOG_DEBUG(log, "MERGEEEE {} PARENT NODE {} NODE {}", merge, node->parentNode()->nodeName(), node->nodeName());
+                    if (merge)
+                    {
+                        //LOG_DEBUG(log, "MERGEEEE");
+                        NodePtr new_node = config->importNode(child->parentNode(), true);
+                        //LOG_DEBUG(log, "CHILD {} NEW NODE NAME {}", child->nodeName(), new_node->nodeName());
+                        mergeRecursive(config, node->parentNode(), new_node);
+                    }
+                    else
+                    {
+                        NodePtr new_node = config->importNode(child, true);
+                        node->parentNode()->insertBefore(new_node, node);
+                    }
                 }
 
                 node->parentNode()->removeChild(node);
             }
             else
             {
-                Element & element = dynamic_cast<Element &>(*node);
-
                 for (const auto & attr_name : SUBSTITUTION_ATTRS)
                     element.removeAttribute(attr_name);
 
@@ -792,13 +812,19 @@ ConfigProcessor::LoadedConfig ConfigProcessor::loadConfig(bool allow_zk_includes
 
     ConfigurationPtr configuration(new Poco::Util::XMLConfiguration(config_xml));
 
+    //LOG_DEBUG(log, "MIN BYTES FOR WIDE PART {}", configuration->getUInt64("merge_tree.min_bytes_for_wide_part"));
+    //Poco::Util::AbstractConfiguration::Keys config_keys;
+    //configuration->keys("merge_tree", config_keys);
+    //for (const String & key : config_keys)
+    //{
+    //    LOG_DEBUG(log, "CONFIG KEY {} LEFT in merge_tree before ZK", key);
+    //}
+
     return LoadedConfig{configuration, has_zk_includes, /* loaded_from_preprocessed = */ false, config_xml, path};
 }
 
 ConfigProcessor::LoadedConfig ConfigProcessor::loadConfigWithZooKeeperIncludes(
-        zkutil::ZooKeeperNodeCache & zk_node_cache,
-        const zkutil::EventPtr & zk_changed_event,
-        bool fallback_to_preprocessed)
+    zkutil::ZooKeeperNodeCache & zk_node_cache, const zkutil::EventPtr & zk_changed_event, bool fallback_to_preprocessed)
 {
     XMLDocumentPtr config_xml;
     bool has_zk_includes;
@@ -817,13 +843,26 @@ ConfigProcessor::LoadedConfig ConfigProcessor::loadConfigWithZooKeeperIncludes(
         if (!zk_exception)
             throw;
 
-        LOG_WARNING(log, "Error while processing from_zk config includes: {}. Config will be loaded from preprocessed file: {}", zk_exception->message(), preprocessed_path);
+        LOG_WARNING(
+            log,
+            "Error while processing from_zk config includes: {}. Config will be loaded from preprocessed file: {}",
+            zk_exception->message(),
+            preprocessed_path);
 
         config_xml = dom_parser.parse(preprocessed_path);
     }
 
     ConfigurationPtr configuration(new Poco::Util::XMLConfiguration(config_xml));
 
+    //LOG_DEBUG(log, "MIN BYTES FOR WIDE PART {} WITH ZK INCLUDE", configuration->getUInt64("merge_tree.min_bytes_for_wide_part"));
+
+    //Poco::Util::AbstractConfiguration::Keys config_keys;
+    //configuration->keys("merge_tree", config_keys);
+    //for (const String & key : config_keys)
+    //{
+    //    LOG_DEBUG(log, "CONFIG KEY {} LEFT in merge_tree after ZK", key);
+    //}
+
     return LoadedConfig{configuration, has_zk_includes, !processed_successfully, config_xml, path};
 }
 
diff --git a/src/Core/BaseSettings.cpp b/src/Core/BaseSettings.cpp
index a7e1ab99af7..c535b9ce65e 100644
--- a/src/Core/BaseSettings.cpp
+++ b/src/Core/BaseSettings.cpp
@@ -41,13 +41,13 @@ BaseSettingsHelpers::Flags BaseSettingsHelpers::readFlags(ReadBuffer & in)
 
 void BaseSettingsHelpers::throwSettingNotFound(std::string_view name)
 {
-    throw Exception(ErrorCodes::UNKNOWN_SETTING, "Unknown setting {}", String{name});
+    throw Exception(ErrorCodes::UNKNOWN_SETTING, "Unknown setting '{}'", String{name});
 }
 
 
 void BaseSettingsHelpers::warningSettingNotFound(std::string_view name)
 {
-    LOG_WARNING(getLogger("Settings"), "Unknown setting {}, skipping", name);
+    LOG_WARNING(getLogger("Settings"), "Unknown setting '{}', skipping", name);
 }
 
 }
diff --git a/tests/integration/test_config_substitutions/configs/config_zk_include_test.xml b/tests/integration/test_config_substitutions/configs/config_zk_include_test.xml
index 4a0ed623c4e..743770c3024 100644
--- a/tests/integration/test_config_substitutions/configs/config_zk_include_test.xml
+++ b/tests/integration/test_config_substitutions/configs/config_zk_include_test.xml
@@ -1,8 +1,12 @@
 <?xml version="1.0" encoding="utf-8"?>
 <clickhouse>
+  <background_pool_size>44</background_pool_size>
   <merge_tree>
+    <include from_zk="/merge_max_block_size" merge="true"/>
+    <merge_max_block_size>99</merge_max_block_size>
     <min_bytes_for_wide_part>1</min_bytes_for_wide_part>
+    <min_rows_for_wide_part>1111</min_rows_for_wide_part>
   </merge_tree>
 
-  <include from_zk="/min_bytes_for_wide_part"/>
+  <include from_zk="/min_bytes_for_wide_part" merge="true"/>
  </clickhouse>
diff --git a/tests/integration/test_config_substitutions/test.py b/tests/integration/test_config_substitutions/test.py
index 08834e20423..304506148b0 100644
--- a/tests/integration/test_config_substitutions/test.py
+++ b/tests/integration/test_config_substitutions/test.py
@@ -67,6 +67,12 @@ def start_cluster():
                 value=b"<merge_tree><min_bytes_for_wide_part>33</min_bytes_for_wide_part></merge_tree>",
                 makepath=True,
             )
+            zk.create(
+                path="/merge_max_block_size",
+                value=b"<merge_max_block_size>8888</merge_max_block_size>",
+                makepath=True,
+            )
+
 
         cluster.add_zookeeper_startup_command(create_zk_roots)
 
@@ -244,5 +250,35 @@ def test_allow_databases(start_cluster):
     )
 
 def test_config_multiple_zk_substitutions(start_cluster):
-    #print(node3.query("SELECT * FROM system.merge_tree_settings"))
-    print(node3.query("SELECT * FROM system.merge_tree_settings WHERE changed=1"))
+    assert node3.query("SELECT value FROM system.merge_tree_settings WHERE name='min_bytes_for_wide_part'") == "33\n"
+    assert node3.query("SELECT value FROM system.merge_tree_settings WHERE name='min_rows_for_wide_part'") == "1111\n"
+    assert node3.query("SELECT value FROM system.merge_tree_settings WHERE name='merge_max_block_size'") == "8888\n"
+    assert node3.query("SELECT value FROM system.server_settings WHERE name='background_pool_size'") == "44\n"
+
+    zk = cluster.get_kazoo_client("zoo1")
+    zk.create(
+        path="/background_pool_size",
+        value=b"<background_pool_size>72</background_pool_size>",
+        makepath=True,
+    )
+
+    node3.replace_config(
+        "/etc/clickhouse-server/config.d/config_zk_include_test.xml",
+        """
+<clickhouse>
+  <include from_zk="/background_pool_size" merge="true"/>
+  <background_pool_size>44</background_pool_size>
+  <merge_tree>
+    <include from_zk="/merge_max_block_size" merge="true"/>
+    <min_bytes_for_wide_part>1</min_bytes_for_wide_part>
+    <min_rows_for_wide_part>1111</min_rows_for_wide_part>
+  </merge_tree>
+
+  <include from_zk="/min_bytes_for_wide_part" merge="true"/>
+ </clickhouse>
+""",
+    )
+
+    node3.query("SYSTEM RELOAD CONFIG")
+
+    assert node3.query("SELECT value FROM system.server_settings WHERE name='background_pool_size'") == "72\n"

From eacf33b445f6b93abda804ef2a9aed2928c82e56 Mon Sep 17 00:00:00 2001
From: alesapin <alesapin@gmail.com>
Date: Wed, 13 Mar 2024 13:13:29 +0100
Subject: [PATCH 264/443] Remove comments

---
 src/Common/Config/ConfigProcessor.cpp | 21 ---------------------
 1 file changed, 21 deletions(-)

diff --git a/src/Common/Config/ConfigProcessor.cpp b/src/Common/Config/ConfigProcessor.cpp
index 75a973f582f..1aa41558923 100644
--- a/src/Common/Config/ConfigProcessor.cpp
+++ b/src/Common/Config/ConfigProcessor.cpp
@@ -173,12 +173,10 @@ static void deleteAttributesRecursive(Node * root)
 static void mergeAttributes(Element & config_element, Element & with_element)
 {
     auto * with_element_attributes = with_element.attributes();
-    //std::cerr << "MERGE ATTRIBUTES:" << with_element_attributes->length() << "\n";
 
     for (size_t i = 0; i < with_element_attributes->length(); ++i)
     {
         auto * attr = with_element_attributes->item(i);
-        //std::cerr << "ATTR NAME:" << attr->nodeName() << std::endl;
         config_element.setAttribute(attr->nodeName(), attr->getNodeValue());
     }
 
@@ -279,8 +277,6 @@ void ConfigProcessor::mergeRecursive(XMLDocumentPtr config, Node * config_root,
     for (Node * node = config_root->firstChild(); node;)
     {
         Node * next_node = node->nextSibling();
-        //if (next_node)
-        //    std::cerr << "NEXT NODE:"  << next_node->nodeName() << std::endl;
         /// Remove text from the original config node.
         if (node->nodeType() == Node::TEXT_NODE && !allWhitespace(node->getNodeValue()))
         {
@@ -288,7 +284,6 @@ void ConfigProcessor::mergeRecursive(XMLDocumentPtr config, Node * config_root,
         }
         else if (node->nodeType() == Node::ELEMENT_NODE)
         {
-            //LOG_DEBUG(log, "NODES IN SOURCE: {}", node->nodeName());
             config_element_by_id.insert(ElementsByIdentifier::value_type(getElementIdentifier(node), node));
         }
         node = next_node;
@@ -302,8 +297,6 @@ void ConfigProcessor::mergeRecursive(XMLDocumentPtr config, Node * config_root,
         bool remove = false;
         if (with_node->nodeType() == Node::ELEMENT_NODE)
         {
-            //LOG_DEBUG(log, "WITH NODE: {}", with_node->nodeName());
-            //std::cerr << "WITH NODE: " << with_node->nodeName() << std::endl;
             Element & with_element = dynamic_cast<Element &>(*with_node);
             remove = with_element.hasAttribute("remove");
             bool replace = with_element.hasAttribute("replace");
@@ -317,8 +310,6 @@ void ConfigProcessor::mergeRecursive(XMLDocumentPtr config, Node * config_root,
             if (it != config_element_by_id.end())
             {
                 Node * config_node = it->second;
-                //LOG_DEBUG(log, "SUBNODE NODE: {}", config_node->nodeName());
-                //std::cerr << "SUBNODE NODE: " << config_node->nodeName() << std::endl;
                 config_element_by_id.erase(it);
 
                 if (remove)
@@ -327,16 +318,12 @@ void ConfigProcessor::mergeRecursive(XMLDocumentPtr config, Node * config_root,
                 }
                 else if (replace)
                 {
-                    //LOG_DEBUG(log, "REPLACE: {}", config_node->nodeName());
-                    //std::cerr << "REPLACE!!!" << std::endl;
                     with_element.removeAttribute("replace");
                     NodePtr new_node = config->importNode(with_node, true);
                     config_root->replaceChild(new_node, config_node);
                 }
                 else
                 {
-                    //LOG_DEBUG(log, "SUBNODE NODE HERE: {}", config_node->nodeName());
-                    //std::cerr << "SUBNODE NODE HERE: " << config_node->nodeName() << std::endl;
                     Element & config_element = dynamic_cast<Element &>(*config_node);
 
                     /// Remove substitution attributes from the merge target node if source node already has a value
@@ -348,19 +335,11 @@ void ConfigProcessor::mergeRecursive(XMLDocumentPtr config, Node * config_root,
                     mergeAttributes(config_element, with_element);
                     mergeRecursive(config, config_node, with_node);
                 }
-                //std::cerr << "DONE\n";
                 merged = true;
             }
-            //else
-            //{
-            //    LOG_DEBUG(log, "ELEMENT NOT FOUND");
-            //    //std::cerr << "ELEMENT NOT FOUND\n";
-            //}
         }
         if (!merged && !remove)
         {
-            //LOG_DEBUG(log, "NOTHING HAPPENED");
-            //std::cerr << "NOTHING hAPPENED\n";
             /// Since we didn't find a pair to this node in default config, we will paste it as is.
             /// But it may have some child nodes which have attributes like "replace" or "remove".
             /// They are useless in preprocessed configuration.

From 21ccabd603cde420b761736966a172a6312de6f9 Mon Sep 17 00:00:00 2001
From: alesapin <alesapin@gmail.com>
Date: Wed, 13 Mar 2024 13:19:49 +0100
Subject: [PATCH 265/443] Remove redundant diff

---
 src/Common/Config/ConfigProcessor.cpp | 23 -----------------------
 1 file changed, 23 deletions(-)

diff --git a/src/Common/Config/ConfigProcessor.cpp b/src/Common/Config/ConfigProcessor.cpp
index 1aa41558923..d639c243639 100644
--- a/src/Common/Config/ConfigProcessor.cpp
+++ b/src/Common/Config/ConfigProcessor.cpp
@@ -270,7 +270,6 @@ void ConfigProcessor::hideRecursive(Poco::XML::Node * config_root)
 
 void ConfigProcessor::mergeRecursive(XMLDocumentPtr config, Node * config_root, const Node * with_root)
 {
-    //LOG_DEBUG(log, "WITH ROOT {}", with_root->nodeName());
     const NodeListPtr with_nodes = with_root->childNodes();
     using ElementsByIdentifier = std::multimap<ElementIdentifier, Node *>;
     ElementsByIdentifier config_element_by_id;
@@ -436,7 +435,6 @@ void ConfigProcessor::doIncludesRecursive(
     auto process_include = [&](const Node * include_attr, const std::function<const Node * (const std::string &)> & get_node, const char * error_msg)
     {
         const std::string & name = include_attr->getNodeValue();
-        //LOG_DEBUG(log, "PROCESS INCLUDE {}", name);
         const Node * node_to_include = get_node(name);
         if (!node_to_include)
         {
@@ -459,7 +457,6 @@ void ConfigProcessor::doIncludesRecursive(
             /// Replace the whole node not just contents.
             if (node->nodeName() == "include")
             {
-                //LOG_DEBUG(log, "Include here for node {}", name);
                 const NodeListPtr children = node_to_include->childNodes();
                 Node * next_child = nullptr;
 
@@ -467,12 +464,9 @@ void ConfigProcessor::doIncludesRecursive(
                 {
                     next_child = child->nextSibling();
 
-                    //LOG_DEBUG(log, "MERGEEEE {} PARENT NODE {} NODE {}", merge, node->parentNode()->nodeName(), node->nodeName());
                     if (merge)
                     {
-                        //LOG_DEBUG(log, "MERGEEEE");
                         NodePtr new_node = config->importNode(child->parentNode(), true);
-                        //LOG_DEBUG(log, "CHILD {} NEW NODE NAME {}", child->nodeName(), new_node->nodeName());
                         mergeRecursive(config, node->parentNode(), new_node);
                     }
                     else
@@ -791,14 +785,6 @@ ConfigProcessor::LoadedConfig ConfigProcessor::loadConfig(bool allow_zk_includes
 
     ConfigurationPtr configuration(new Poco::Util::XMLConfiguration(config_xml));
 
-    //LOG_DEBUG(log, "MIN BYTES FOR WIDE PART {}", configuration->getUInt64("merge_tree.min_bytes_for_wide_part"));
-    //Poco::Util::AbstractConfiguration::Keys config_keys;
-    //configuration->keys("merge_tree", config_keys);
-    //for (const String & key : config_keys)
-    //{
-    //    LOG_DEBUG(log, "CONFIG KEY {} LEFT in merge_tree before ZK", key);
-    //}
-
     return LoadedConfig{configuration, has_zk_includes, /* loaded_from_preprocessed = */ false, config_xml, path};
 }
 
@@ -833,15 +819,6 @@ ConfigProcessor::LoadedConfig ConfigProcessor::loadConfigWithZooKeeperIncludes(
 
     ConfigurationPtr configuration(new Poco::Util::XMLConfiguration(config_xml));
 
-    //LOG_DEBUG(log, "MIN BYTES FOR WIDE PART {} WITH ZK INCLUDE", configuration->getUInt64("merge_tree.min_bytes_for_wide_part"));
-
-    //Poco::Util::AbstractConfiguration::Keys config_keys;
-    //configuration->keys("merge_tree", config_keys);
-    //for (const String & key : config_keys)
-    //{
-    //    LOG_DEBUG(log, "CONFIG KEY {} LEFT in merge_tree after ZK", key);
-    //}
-
     return LoadedConfig{configuration, has_zk_includes, !processed_successfully, config_xml, path};
 }
 

From 62681225173b5c2044e33b83b691ca9ffb9245e5 Mon Sep 17 00:00:00 2001
From: alesapin <alesapin@gmail.com>
Date: Wed, 13 Mar 2024 13:21:06 +0100
Subject: [PATCH 266/443] Revert unit test

---
 src/Common/tests/gtest_config_processor.cpp | 49 ---------------------
 1 file changed, 49 deletions(-)

diff --git a/src/Common/tests/gtest_config_processor.cpp b/src/Common/tests/gtest_config_processor.cpp
index d5397dd54c2..f01460d515b 100644
--- a/src/Common/tests/gtest_config_processor.cpp
+++ b/src/Common/tests/gtest_config_processor.cpp
@@ -2,7 +2,6 @@
 #include <IO/WriteBufferFromFile.h>
 #include <IO/WriteHelpers.h>
 #include <Poco/Timestamp.h>
-#include <Common/XMLUtils.h>
 #include <Poco/Util/XMLConfiguration.h>
 #include <base/scope_guard.h>
 #include <gtest/gtest.h>
@@ -64,51 +63,3 @@ TEST(Common, ConfigProcessorManyElements)
     /// More that 5 min is way too slow
     ASSERT_LE(enumerate_elapsed_ms, 300*1000);
 }
-
-TEST(Common, ConfigProcessorMerge)
-{
-    namespace fs = std::filesystem;
-
-    auto path = fs::path("/tmp/test_config_processor/");
-
-    fs::remove_all(path);
-    fs::create_directories(path);
-    fs::create_directories(path / "config.d");
-    //SCOPE_EXIT({ fs::remove_all(path); });
-
-    auto config_file = std::make_unique<Poco::File>(path / "config.xml");
-    {
-        DB::WriteBufferFromFile out(config_file->path());
-        writeString("<clickhouse>\n", out);
-        writeString("<merge_tree><min_bytes_for_wide_part>1</min_bytes_for_wide_part></merge_tree>\n", out);
-        writeString("</clickhouse>\n", out);
-    }
-    DB::XMLDocumentPtr config_xml;
-    DB::ConfigProcessor processor(config_file->path(), /* throw_on_bad_incl = */ false, /* log_to_console = */ false);
-    {
-        bool has_zk_includes;
-        config_xml = processor.processConfig(&has_zk_includes);
-    }
-
-    auto small_part_file = std::make_unique<Poco::File>(path / "config.d" / "part.xml");
-    {
-        DB::WriteBufferFromFile out(small_part_file->path());
-        writeString("<merge_tree><min_bytes_for_wide_part>33</min_bytes_for_wide_part></merge_tree>\n", out);
-    }
-    DB::XMLDocumentPtr part_xml;
-
-    {
-        DB::ConfigProcessor tiny(small_part_file->path(), /* throw_on_bad_incl = */ false, /* log_to_console = */ false);
-        bool has_zk_includes;
-        part_xml = tiny.processConfig(&has_zk_includes);
-    }
-
-    auto * root_node = DB::XMLUtils::getRootNode(config_xml);
-    auto * part_node = DB::XMLUtils::getRootNode(part_xml);
-    auto * new_node = config_xml->importNode(part_node, true);
-    auto * deep_root = root_node->getNodeByPath("merge_tree");
-    processor.mergeRecursive(config_xml, deep_root, new_node);
-    DB::ConfigurationPtr configuration(new Poco::Util::XMLConfiguration(config_xml));
-
-    ASSERT_EQ(configuration->getUInt64("merge_tree.min_bytes_for_wide_part"), 33);
-}

From ef1f3c1af8c09ea94004857c0582de8e0137a1b4 Mon Sep 17 00:00:00 2001
From: alesapin <alesapin@gmail.com>
Date: Wed, 13 Mar 2024 13:21:51 +0100
Subject: [PATCH 267/443] Revert config processor.h

---
 src/Common/Config/ConfigProcessor.h | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)

diff --git a/src/Common/Config/ConfigProcessor.h b/src/Common/Config/ConfigProcessor.h
index 9201ec7b77d..5712c36d737 100644
--- a/src/Common/Config/ConfigProcessor.h
+++ b/src/Common/Config/ConfigProcessor.h
@@ -119,11 +119,6 @@ public:
 
     static inline const auto SUBSTITUTION_ATTRS = {"incl", "from_zk", "from_env"};
 
-    /// If config root node name is not 'clickhouse' and merging config's root node names doesn't match, bypasses merging and returns false.
-    /// For compatibility root node 'yandex' considered equal to 'clickhouse'.
-    bool merge(XMLDocumentPtr config, XMLDocumentPtr with);
-    void mergeRecursive(XMLDocumentPtr config, Poco::XML::Node * config_root, const Poco::XML::Node * with_root);
-
 private:
     const std::string path;
     std::string preprocessed_path;
@@ -150,7 +145,11 @@ private:
     void hideRecursive(Poco::XML::Node * config_root);
     XMLDocumentPtr hideElements(XMLDocumentPtr xml_tree);
 
-    //void mergeRecursive(XMLDocumentPtr config, Poco::XML::Node * config_root, const Poco::XML::Node * with_root);
+    void mergeRecursive(XMLDocumentPtr config, Poco::XML::Node * config_root, const Poco::XML::Node * with_root);
+
+    /// If config root node name is not 'clickhouse' and merging config's root node names doesn't match, bypasses merging and returns false.
+    /// For compatibility root node 'yandex' considered equal to 'clickhouse'.
+    bool merge(XMLDocumentPtr config, XMLDocumentPtr with);
 
     void doIncludesRecursive(
             XMLDocumentPtr config,

From 8af9262e43435528bebd528eaed39ae103fa229c Mon Sep 17 00:00:00 2001
From: alesapin <alesapin@gmail.com>
Date: Wed, 13 Mar 2024 13:26:25 +0100
Subject: [PATCH 268/443] Add docs

---
 docs/en/operations/configuration-files.md | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/docs/en/operations/configuration-files.md b/docs/en/operations/configuration-files.md
index 9f17f4af1e8..fbac2407640 100644
--- a/docs/en/operations/configuration-files.md
+++ b/docs/en/operations/configuration-files.md
@@ -116,6 +116,8 @@ XML substitution example:
 
 Substitutions can also be performed from ZooKeeper. To do this, specify the attribute `from_zk = "/path/to/node"`. The element value is replaced with the contents of the node at `/path/to/node` in ZooKeeper. You can also put an entire XML subtree on the ZooKeeper node, and it will be fully inserted into the source element.
 
+If you want to merge substitution content with existing configuration instead of append you can use attribute `merge="true`, for example: `<include from_zk="/some_path" merge="true">`. In this case existing configuration will be merged with content from substituion and existing configuration settings will be replaced with values from substituion.
+
 ## Encrypting and Hiding Configuration {#encryption}
 
 You can use symmetric encryption to encrypt a configuration element, for example, a plaintext password or private key. To do so, first configure the [encryption codec](../sql-reference/statements/create/table.md#encryption-codecs), then add attribute `encrypted_by` with the name of the encryption codec as value to the element to encrypt.

From 479b58ec70ec3c2563a5cdc358bfd2674f58325d Mon Sep 17 00:00:00 2001
From: alesapin <alesapin@gmail.com>
Date: Wed, 13 Mar 2024 13:28:10 +0100
Subject: [PATCH 269/443] Remove redundant changes

---
 src/Common/Config/ConfigProcessor.cpp | 13 ++++---------
 1 file changed, 4 insertions(+), 9 deletions(-)

diff --git a/src/Common/Config/ConfigProcessor.cpp b/src/Common/Config/ConfigProcessor.cpp
index d639c243639..333d4d45a8f 100644
--- a/src/Common/Config/ConfigProcessor.cpp
+++ b/src/Common/Config/ConfigProcessor.cpp
@@ -452,9 +452,6 @@ void ConfigProcessor::doIncludesRecursive(
         }
         else
         {
-
-            Element & element = dynamic_cast<Element &>(*node);
-            /// Replace the whole node not just contents.
             if (node->nodeName() == "include")
             {
                 const NodeListPtr children = node_to_include->childNodes();
@@ -789,7 +786,9 @@ ConfigProcessor::LoadedConfig ConfigProcessor::loadConfig(bool allow_zk_includes
 }
 
 ConfigProcessor::LoadedConfig ConfigProcessor::loadConfigWithZooKeeperIncludes(
-    zkutil::ZooKeeperNodeCache & zk_node_cache, const zkutil::EventPtr & zk_changed_event, bool fallback_to_preprocessed)
+    zkutil::ZooKeeperNodeCache & zk_node_cache,
+    const zkutil::EventPtr & zk_changed_event,
+    bool fallback_to_preprocessed)
 {
     XMLDocumentPtr config_xml;
     bool has_zk_includes;
@@ -808,11 +807,7 @@ ConfigProcessor::LoadedConfig ConfigProcessor::loadConfigWithZooKeeperIncludes(
         if (!zk_exception)
             throw;
 
-        LOG_WARNING(
-            log,
-            "Error while processing from_zk config includes: {}. Config will be loaded from preprocessed file: {}",
-            zk_exception->message(),
-            preprocessed_path);
+        LOG_WARNING(log, "Error while processing from_zk config includes: {}. Config will be loaded from preprocessed file: {}", zk_exception->message(), preprocessed_path);
 
         config_xml = dom_parser.parse(preprocessed_path);
     }

From ab8276f23c5c09b63e3537ad29043b5a4a9f13a2 Mon Sep 17 00:00:00 2001
From: alesapin <alesapin@gmail.com>
Date: Wed, 13 Mar 2024 13:29:04 +0100
Subject: [PATCH 270/443] Buildable code

---
 src/Common/Config/ConfigProcessor.cpp | 1 +
 1 file changed, 1 insertion(+)

diff --git a/src/Common/Config/ConfigProcessor.cpp b/src/Common/Config/ConfigProcessor.cpp
index 333d4d45a8f..deac1039372 100644
--- a/src/Common/Config/ConfigProcessor.cpp
+++ b/src/Common/Config/ConfigProcessor.cpp
@@ -477,6 +477,7 @@ void ConfigProcessor::doIncludesRecursive(
             }
             else
             {
+                Element & element = dynamic_cast<Element &>(*node);
                 for (const auto & attr_name : SUBSTITUTION_ATTRS)
                     element.removeAttribute(attr_name);
 

From 24626d431cac2e5c76ea14c5dd40afe5617ae1d0 Mon Sep 17 00:00:00 2001
From: robot-clickhouse <robot-clickhouse@users.noreply.github.com>
Date: Wed, 13 Mar 2024 12:35:27 +0000
Subject: [PATCH 271/443] Automatic style fix

---
 .../test_config_substitutions/test.py         | 44 ++++++++++++++++---
 1 file changed, 37 insertions(+), 7 deletions(-)

diff --git a/tests/integration/test_config_substitutions/test.py b/tests/integration/test_config_substitutions/test.py
index 304506148b0..ac75771cb9c 100644
--- a/tests/integration/test_config_substitutions/test.py
+++ b/tests/integration/test_config_substitutions/test.py
@@ -13,7 +13,12 @@ node2 = cluster.add_instance(
     env_variables={"MAX_QUERY_SIZE": "55555"},
 )
 node3 = cluster.add_instance(
-    "node3", user_configs=["configs/config_zk.xml",], main_configs=["configs/config_zk_include_test.xml"], with_zookeeper=True
+    "node3",
+    user_configs=[
+        "configs/config_zk.xml",
+    ],
+    main_configs=["configs/config_zk_include_test.xml"],
+    with_zookeeper=True,
 )
 node4 = cluster.add_instance(
     "node4",
@@ -73,7 +78,6 @@ def start_cluster():
                 makepath=True,
             )
 
-
         cluster.add_zookeeper_startup_command(create_zk_roots)
 
         cluster.start()
@@ -249,11 +253,32 @@ def test_allow_databases(start_cluster):
         == ""
     )
 
+
 def test_config_multiple_zk_substitutions(start_cluster):
-    assert node3.query("SELECT value FROM system.merge_tree_settings WHERE name='min_bytes_for_wide_part'") == "33\n"
-    assert node3.query("SELECT value FROM system.merge_tree_settings WHERE name='min_rows_for_wide_part'") == "1111\n"
-    assert node3.query("SELECT value FROM system.merge_tree_settings WHERE name='merge_max_block_size'") == "8888\n"
-    assert node3.query("SELECT value FROM system.server_settings WHERE name='background_pool_size'") == "44\n"
+    assert (
+        node3.query(
+            "SELECT value FROM system.merge_tree_settings WHERE name='min_bytes_for_wide_part'"
+        )
+        == "33\n"
+    )
+    assert (
+        node3.query(
+            "SELECT value FROM system.merge_tree_settings WHERE name='min_rows_for_wide_part'"
+        )
+        == "1111\n"
+    )
+    assert (
+        node3.query(
+            "SELECT value FROM system.merge_tree_settings WHERE name='merge_max_block_size'"
+        )
+        == "8888\n"
+    )
+    assert (
+        node3.query(
+            "SELECT value FROM system.server_settings WHERE name='background_pool_size'"
+        )
+        == "44\n"
+    )
 
     zk = cluster.get_kazoo_client("zoo1")
     zk.create(
@@ -281,4 +306,9 @@ def test_config_multiple_zk_substitutions(start_cluster):
 
     node3.query("SYSTEM RELOAD CONFIG")
 
-    assert node3.query("SELECT value FROM system.server_settings WHERE name='background_pool_size'") == "72\n"
+    assert (
+        node3.query(
+            "SELECT value FROM system.server_settings WHERE name='background_pool_size'"
+        )
+        == "72\n"
+    )

From c234e1bca623106f625b4789513cc6ef20d7986f Mon Sep 17 00:00:00 2001
From: alesapin <alesapin@gmail.com>
Date: Wed, 13 Mar 2024 13:35:37 +0100
Subject: [PATCH 272/443] Less diff

---
 src/Common/Config/ConfigProcessor.cpp | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/src/Common/Config/ConfigProcessor.cpp b/src/Common/Config/ConfigProcessor.cpp
index deac1039372..d524db883eb 100644
--- a/src/Common/Config/ConfigProcessor.cpp
+++ b/src/Common/Config/ConfigProcessor.cpp
@@ -303,7 +303,6 @@ void ConfigProcessor::mergeRecursive(XMLDocumentPtr config, Node * config_root,
             if (remove && replace)
                 throw Poco::Exception("both remove and replace attributes set for element <" + with_node->nodeName() + ">");
 
-
             ElementsByIdentifier::iterator it = config_element_by_id.find(getElementIdentifier(with_node));
 
             if (it != config_element_by_id.end())
@@ -456,17 +455,17 @@ void ConfigProcessor::doIncludesRecursive(
             {
                 const NodeListPtr children = node_to_include->childNodes();
                 Node * next_child = nullptr;
-
                 for (Node * child = children->item(0); child; child = next_child)
                 {
                     next_child = child->nextSibling();
 
+                    /// Recursively replace existing nodes in merge mode
                     if (merge)
                     {
                         NodePtr new_node = config->importNode(child->parentNode(), true);
                         mergeRecursive(config, node->parentNode(), new_node);
                     }
-                    else
+                    else  /// Append to existing node by default
                     {
                         NodePtr new_node = config->importNode(child, true);
                         node->parentNode()->insertBefore(new_node, node);
@@ -478,6 +477,7 @@ void ConfigProcessor::doIncludesRecursive(
             else
             {
                 Element & element = dynamic_cast<Element &>(*node);
+
                 for (const auto & attr_name : SUBSTITUTION_ATTRS)
                     element.removeAttribute(attr_name);
 

From 457578627103b0eb0028f51b739c3911de278bf4 Mon Sep 17 00:00:00 2001
From: Nikolai Kochetov <nk@clickhouse.com>
Date: Wed, 13 Mar 2024 12:38:17 +0000
Subject: [PATCH 273/443] Fix 01952_optimize_distributed_group_by_sharding_key
 with analyzer.

---
 ...istributed_group_by_sharding_key.reference | 78 +++++++++++++++++++
 ...mize_distributed_group_by_sharding_key.sql | 14 ++++
 2 files changed, 92 insertions(+)

diff --git a/tests/queries/0_stateless/01952_optimize_distributed_group_by_sharding_key.reference b/tests/queries/0_stateless/01952_optimize_distributed_group_by_sharding_key.reference
index ddfa6929d69..212dd348edb 100644
--- a/tests/queries/0_stateless/01952_optimize_distributed_group_by_sharding_key.reference
+++ b/tests/queries/0_stateless/01952_optimize_distributed_group_by_sharding_key.reference
@@ -71,3 +71,81 @@ Expression (Projection)
             Expression ((Before ORDER BY + (Convert VIEW subquery result to VIEW table structure + (Materialize constants after VIEW subquery + (Projection + Before ORDER BY)))))
               ReadFromSystemNumbers
       ReadFromRemote (Read from remote replica)
+set allow_experimental_analyzer = 1;
+explain select distinct k1 from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)); -- not optimized
+Expression (Project names)
+  Distinct (DISTINCT)
+    Union
+      Distinct (Preliminary DISTINCT)
+        Expression ((Projection + (Change column names to column identifiers + (Convert VIEW subquery result to VIEW table structure + (Materialize constants after VIEW subquery + (Project names + (Projection + Change column names to column identifiers)))))))
+          ReadFromSystemNumbers
+      ReadFromRemote (Read from remote replica)
+explain select distinct k1, k2 from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)); -- optimized
+Union
+  Expression (Project names)
+    Distinct (DISTINCT)
+      Distinct (Preliminary DISTINCT)
+        Expression ((Projection + (Change column names to column identifiers + (Convert VIEW subquery result to VIEW table structure + (Materialize constants after VIEW subquery + (Project names + (Projection + Change column names to column identifiers)))))))
+          ReadFromSystemNumbers
+  ReadFromRemote (Read from remote replica)
+explain select distinct on (k1) k2 from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)); -- not optimized
+Expression (Project names)
+  LimitBy
+    Union
+      Expression (Before LIMIT BY)
+        LimitBy
+          Expression ((Before LIMIT BY + (Projection + (Change column names to column identifiers + (Convert VIEW subquery result to VIEW table structure + (Materialize constants after VIEW subquery + (Project names + (Projection + Change column names to column identifiers))))))))
+            ReadFromSystemNumbers
+      Expression
+        ReadFromRemote (Read from remote replica)
+explain select distinct on (k1, k2) v from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)); -- optimized
+Union
+  Expression (Project names)
+    LimitBy
+      Expression ((Before LIMIT BY + (Projection + (Change column names to column identifiers + (Convert VIEW subquery result to VIEW table structure + (Materialize constants after VIEW subquery + (Project names + (Projection + Change column names to column identifiers))))))))
+        ReadFromSystemNumbers
+  ReadFromRemote (Read from remote replica)
+explain select distinct k1 from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)) order by v; -- not optimized
+Expression (Project names)
+  Distinct (DISTINCT)
+    Sorting (Merge sorted streams for ORDER BY, without aggregation)
+      Union
+        Sorting (Sorting for ORDER BY)
+          Expression (Before ORDER BY)
+            Distinct (Preliminary DISTINCT)
+              Expression ((Projection + (Change column names to column identifiers + (Convert VIEW subquery result to VIEW table structure + (Materialize constants after VIEW subquery + (Project names + (Projection + Change column names to column identifiers)))))))
+                ReadFromSystemNumbers
+        ReadFromRemote (Read from remote replica)
+explain select distinct k1, k2 from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)) order by v; -- optimized
+Expression (Project names)
+  Sorting (Merge sorted streams after aggregation stage for ORDER BY)
+    Union
+      Distinct (DISTINCT)
+        Sorting (Sorting for ORDER BY)
+          Expression (Before ORDER BY)
+            Distinct (Preliminary DISTINCT)
+              Expression ((Projection + (Change column names to column identifiers + (Convert VIEW subquery result to VIEW table structure + (Materialize constants after VIEW subquery + (Project names + (Projection + Change column names to column identifiers)))))))
+                ReadFromSystemNumbers
+      ReadFromRemote (Read from remote replica)
+explain select distinct on (k1) k2 from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)) order by v; -- not optimized
+Expression (Project names)
+  LimitBy
+    Expression (Before LIMIT BY)
+      Sorting (Merge sorted streams for ORDER BY, without aggregation)
+        Union
+          LimitBy
+            Expression ((Before LIMIT BY + (Before ORDER BY + (Projection + (Change column names to column identifiers + (Convert VIEW subquery result to VIEW table structure + (Materialize constants after VIEW subquery + (Project names + (Projection + Change column names to column identifiers))))))) [lifted up part]))
+              Sorting (Sorting for ORDER BY)
+                Expression ((Before ORDER BY + (Projection + (Change column names to column identifiers + (Convert VIEW subquery result to VIEW table structure + (Materialize constants after VIEW subquery + (Project names + (Projection + Change column names to column identifiers))))))))
+                  ReadFromSystemNumbers
+          ReadFromRemote (Read from remote replica)
+explain select distinct on (k1, k2) v from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)) order by v; -- optimized
+Expression (Project names)
+  Sorting (Merge sorted streams after aggregation stage for ORDER BY)
+    Union
+      LimitBy
+        Expression ((Before LIMIT BY + (Before ORDER BY + (Projection + (Change column names to column identifiers + (Convert VIEW subquery result to VIEW table structure + (Materialize constants after VIEW subquery + (Project names + (Projection + Change column names to column identifiers))))))) [lifted up part]))
+          Sorting (Sorting for ORDER BY)
+            Expression ((Before ORDER BY + (Projection + (Change column names to column identifiers + (Convert VIEW subquery result to VIEW table structure + (Materialize constants after VIEW subquery + (Project names + (Projection + Change column names to column identifiers))))))))
+              ReadFromSystemNumbers
+      ReadFromRemote (Read from remote replica)
diff --git a/tests/queries/0_stateless/01952_optimize_distributed_group_by_sharding_key.sql b/tests/queries/0_stateless/01952_optimize_distributed_group_by_sharding_key.sql
index 74b55b95315..adf55a9dd7f 100644
--- a/tests/queries/0_stateless/01952_optimize_distributed_group_by_sharding_key.sql
+++ b/tests/queries/0_stateless/01952_optimize_distributed_group_by_sharding_key.sql
@@ -4,6 +4,8 @@ set optimize_skip_unused_shards=1;
 set optimize_distributed_group_by_sharding_key=1;
 set prefer_localhost_replica=1;
 
+set allow_experimental_analyzer = 0;
+
 -- { echo }
 explain select distinct k1 from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)); -- not optimized
 explain select distinct k1, k2 from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)); -- optimized
@@ -14,3 +16,15 @@ explain select distinct k1 from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v
 explain select distinct k1, k2 from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)) order by v; -- optimized
 explain select distinct on (k1) k2 from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)) order by v; -- not optimized
 explain select distinct on (k1, k2) v from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)) order by v; -- optimized
+
+set allow_experimental_analyzer = 1;
+
+explain select distinct k1 from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)); -- not optimized
+explain select distinct k1, k2 from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)); -- optimized
+explain select distinct on (k1) k2 from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)); -- not optimized
+explain select distinct on (k1, k2) v from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)); -- optimized
+
+explain select distinct k1 from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)) order by v; -- not optimized
+explain select distinct k1, k2 from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)) order by v; -- optimized
+explain select distinct on (k1) k2 from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)) order by v; -- not optimized
+explain select distinct on (k1, k2) v from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)) order by v; -- optimized

From ffdb84b8df7120292543da077227ef049552dccf Mon Sep 17 00:00:00 2001
From: Dmitry Novik <n0vik@clickhouse.com>
Date: Wed, 13 Mar 2024 13:38:23 +0100
Subject: [PATCH 274/443] Fix style

---
 src/Storages/StorageMerge.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Storages/StorageMerge.cpp b/src/Storages/StorageMerge.cpp
index fab4b2e5146..e695594873b 100644
--- a/src/Storages/StorageMerge.cpp
+++ b/src/Storages/StorageMerge.cpp
@@ -1063,7 +1063,7 @@ QueryPipelineBuilderPtr ReadFromMerge::createSources(
             get_column_options.withSubcolumns();
 
         LOG_DEBUG(&Poco::Logger::get("createSources"), "Processed:{}\nStorage:{}", toString(processed_stage), toString(storage_stage));
-        
+
         String table_alias;
         if (allow_experimental_analyzer)
             table_alias = modified_query_info.query_tree->as<QueryNode>()->getJoinTree()->as<TableNode>()->getAlias();

From 4954bde599dd1bdcdf56957e17b0b9a661aa17f6 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Wed, 13 Mar 2024 13:38:35 +0100
Subject: [PATCH 275/443] Update
 docs/en/sql-reference/functions/string-functions.md

Co-authored-by: Johnny <9611008+johnnymatthews@users.noreply.github.com>
---
 docs/en/sql-reference/functions/string-functions.md | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/docs/en/sql-reference/functions/string-functions.md b/docs/en/sql-reference/functions/string-functions.md
index f9c3f91a12b..a9b7cc9566d 100644
--- a/docs/en/sql-reference/functions/string-functions.md
+++ b/docs/en/sql-reference/functions/string-functions.md
@@ -615,7 +615,9 @@ Assumes that the string contains valid UTF-8 encoded text. If this assumption is
 **Example**
 
 ```sql
-SELECT 'database' AS string, substringUTF8(string, 5), substringUTF8(string, 5, 1)
+SELECT 'Täglich grüßt das Murmeltier.' AS str,
+       substringUTF8(str, 9),
+       substringUTF8(str, 9, 5)
 ```
 
 ```response

From b4953f35b4a8ca3eca816557c080ff612062b482 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Wed, 13 Mar 2024 13:39:03 +0100
Subject: [PATCH 276/443] Update
 docs/en/sql-reference/functions/string-functions.md

Co-authored-by: Johnny <9611008+johnnymatthews@users.noreply.github.com>
---
 docs/en/sql-reference/functions/string-functions.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/en/sql-reference/functions/string-functions.md b/docs/en/sql-reference/functions/string-functions.md
index a9b7cc9566d..25a0c7e38d8 100644
--- a/docs/en/sql-reference/functions/string-functions.md
+++ b/docs/en/sql-reference/functions/string-functions.md
@@ -686,7 +686,7 @@ Assumes that the string contains valid UTF-8 encoded text. If this assumption is
 **Example**
 
 ```sql
-SELECT substringIndexUTF8('www.clickhouse.com', '.', 2)
+SELECT substringIndexUTF8('www.straßen-in-europa.de', '.', 2)
 ```
 
 ```response

From 6ca4fc26f4bca8f787b3a575d5496ffd75ee0c55 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Wed, 13 Mar 2024 13:39:14 +0100
Subject: [PATCH 277/443] Update
 docs/en/sql-reference/functions/string-functions.md

Co-authored-by: Johnny <9611008+johnnymatthews@users.noreply.github.com>
---
 docs/en/sql-reference/functions/string-functions.md | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/docs/en/sql-reference/functions/string-functions.md b/docs/en/sql-reference/functions/string-functions.md
index 25a0c7e38d8..01a583e1713 100644
--- a/docs/en/sql-reference/functions/string-functions.md
+++ b/docs/en/sql-reference/functions/string-functions.md
@@ -621,9 +621,7 @@ SELECT 'Täglich grüßt das Murmeltier.' AS str,
 ```
 
 ```response
-┌─string───┬─substringUTF8('database', 5)─┬─substringUTF8('database', 5, 1)─┐
-│ database │ base                         │ b                               │
-└──────────┴──────────────────────────────┴─────────────────────────────────┘
+Täglich grüßt das Murmeltier.	grüßt das Murmeltier.	grüßt
 ```
 
 ## substringIndex

From 1e536251a20a0fdbac08b0a99e420a8e74886bcd Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Wed, 13 Mar 2024 13:41:56 +0100
Subject: [PATCH 278/443] Update string-functions.md

---
 docs/en/sql-reference/functions/string-functions.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/en/sql-reference/functions/string-functions.md b/docs/en/sql-reference/functions/string-functions.md
index 01a583e1713..b4e2adbed3c 100644
--- a/docs/en/sql-reference/functions/string-functions.md
+++ b/docs/en/sql-reference/functions/string-functions.md
@@ -688,7 +688,7 @@ SELECT substringIndexUTF8('www.straßen-in-europa.de', '.', 2)
 ```
 
 ```response
-www.clickhouse
+www.straßen-in-europa
 ```
 
 ## appendTrailingCharIfAbsent

From e5e632ec3362d2106adca2e02ae2a4ea1862ee3c Mon Sep 17 00:00:00 2001
From: Nikolai Kochetov <nk@clickhouse.com>
Date: Wed, 13 Mar 2024 12:43:10 +0000
Subject: [PATCH 279/443] Update analyzer_tech_debt.txt

---
 tests/analyzer_tech_debt.txt | 1 -
 1 file changed, 1 deletion(-)

diff --git a/tests/analyzer_tech_debt.txt b/tests/analyzer_tech_debt.txt
index dbd216ea7be..42aa579658e 100644
--- a/tests/analyzer_tech_debt.txt
+++ b/tests/analyzer_tech_debt.txt
@@ -9,7 +9,6 @@
 01747_join_view_filter_dictionary
 01761_cast_to_enum_nullable
 01925_join_materialized_columns
-01952_optimize_distributed_group_by_sharding_key
 02354_annoy
 # Check after constants refactoring
 02901_parallel_replicas_rollup

From cca96e05cf7be69f53a479db13414824552b7ca0 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Mon, 11 Mar 2024 17:57:24 +0100
Subject: [PATCH 280/443] Bring clickhouse-test changes from private

---
 tests/clickhouse-test | 376 ++++++++++++++++++++++++++++++++++++++----
 1 file changed, 345 insertions(+), 31 deletions(-)

diff --git a/tests/clickhouse-test b/tests/clickhouse-test
index ce0feadf050..057502379ed 100755
--- a/tests/clickhouse-test
+++ b/tests/clickhouse-test
@@ -4,9 +4,11 @@
 # pylint: disable=global-variable-not-assigned
 # pylint: disable=too-many-lines
 # pylint: disable=anomalous-backslash-in-string
+# pylint: disable=protected-access
 
 import copy
 import enum
+import tempfile
 import glob
 
 # Not requests, to avoid requiring extra dependency.
@@ -68,6 +70,144 @@ TEST_FILE_EXTENSIONS = [".sql", ".sql.j2", ".sh", ".py", ".expect"]
 VERSION_PATTERN = r"^((\d+\.)?(\d+\.)?(\d+\.)?\d+)$"
 
 
+class SharedEngineReplacer:
+    ENGINES_NON_REPLICATED_REGEXP = r"[ =]((Collapsing|VersionedCollapsing|Summing|Replacing|Aggregating|)MergeTree\(?\)?)"
+    ENGINES_MAPPING_REPLICATED = [
+        ("ReplicatedMergeTree", "SharedMergeTree"),
+        ("ReplicatedCollapsingMergeTree", "SharedCollapsingMergeTree"),
+        (
+            "ReplicatedVersionedCollapsingMergeTree",
+            "SharedVersionedCollapsingMergeTree",
+        ),
+        ("ReplicatedSummingMergeTree", "SharedSummingMergeTree"),
+        ("ReplicatedReplacingMergeTree", "SharedReplacingMergeTree"),
+        ("ReplicatedAggregatingMergeTree", "SharedAggregatingMergeTree"),
+    ]
+    NEW_SYNTAX_REPLICATED_MERGE_TREE_RE = (
+        r"Replicated[a-zA-Z]*MergeTree\((\\?'.*\\?')?,?(\\?'.*\\?')?[a-zA-Z, _}{]*\)"
+    )
+    OLD_SYNTAX_OR_ARGUMENTS_RE = r"Tree\(.*[0-9]+.*\)"
+
+    def _check_replicad_new_syntax(self, line):
+        return re.search(self.NEW_SYNTAX_REPLICATED_MERGE_TREE_RE, line) is not None
+
+    def _check_old_syntax_or_arguments(self, line):
+        return re.search(self.OLD_SYNTAX_OR_ARGUMENTS_RE, line) is not None
+
+    @staticmethod
+    def _is_comment_line(line):
+        return line.startswith("SELECT") or line.startswith("select")
+
+    @staticmethod
+    def _is_create_query(line):
+        return (
+            line.startswith("CREATE")
+            or line.startswith("create")
+            or line.startswith("ENGINE")
+            or line.startswith("engine")
+        )
+
+    def _replace_non_replicated(self, line, escape_quotes, use_random_path):
+        groups = re.search(self.ENGINES_NON_REPLICATED_REGEXP, line)
+        if groups is not None and not self._check_old_syntax_or_arguments(line):
+            non_replicated_engine = groups.groups()[0]
+            basename_no_ext = os.path.splitext(os.path.basename(self.file_name))[0]
+            if use_random_path:
+                shared_path = "/" + os.path.join(
+                    basename_no_ext.replace("_", "/"),
+                    str(os.getpid()),
+                    str(random.randint(1, 1000)),
+                )
+            else:
+                shared_path = "/" + os.path.join(
+                    basename_no_ext.replace("_", "/"), str(os.getpid())
+                )
+
+            if escape_quotes:
+                shared_engine = (
+                    "Shared"
+                    + non_replicated_engine.replace("()", "")
+                    + f"(\\'{shared_path}\\', \\'1\\')"
+                )
+            else:
+                shared_engine = (
+                    "Shared"
+                    + non_replicated_engine.replace("()", "")
+                    + f"('{shared_path}', '1')"
+                )
+            return line.replace(non_replicated_engine, shared_engine)
+
+        return line
+
+    def _need_to_replace_something(self):
+        return (
+            self.replace_replicated or self.replace_non_replicated
+        ) and "shared_merge_tree" not in self.file_name
+
+    def _has_show_create_table(self):
+        with open(self.file_name, "r", encoding="utf-8") as f:
+            return re.search("show create table", f.read(), re.IGNORECASE)
+
+    def __init__(
+        self, file_name, replace_replicated, replace_non_replicated, reference_file
+    ):
+        self.file_name = file_name
+        self.temp_file_path = get_temp_file_path()
+        self.replace_replicated = replace_replicated
+        self.replace_non_replicated = replace_non_replicated
+
+        use_random_path = not reference_file and not self._has_show_create_table()
+
+        if not self._need_to_replace_something():
+            return
+
+        shutil.copyfile(self.file_name, self.temp_file_path)
+        shutil.copymode(self.file_name, self.temp_file_path)
+
+        with open(self.file_name, "w", newline="", encoding="utf-8") as modified:
+            with open(self.temp_file_path, "r", newline="", encoding="utf-8") as source:
+                for line in source:
+                    if self._is_comment_line(line) or (
+                        reference_file and not self._is_create_query(line)
+                    ):
+                        modified.write(line)
+                        continue
+
+                    if self.replace_replicated:
+                        for (
+                            engine_from,
+                            engine_to,
+                        ) in SharedEngineReplacer.ENGINES_MAPPING_REPLICATED:
+                            if engine_from in line and (
+                                self._check_replicad_new_syntax(line)
+                                or engine_from + " " in line
+                                or engine_from + ";" in line
+                            ):
+                                line = line.replace(engine_from, engine_to)
+                                break
+
+                    if self.replace_non_replicated:
+                        line = self._replace_non_replicated(
+                            line, reference_file, use_random_path
+                        )
+
+                    modified.write(line)
+
+    def __enter__(self):
+        return self
+
+    def __exit__(self, exc_type, exc_value, exc_tb):
+        if not self._need_to_replace_something():
+            return
+        shutil.move(self.temp_file_path, self.file_name)
+
+
+def get_temp_file_path():
+    return os.path.join(
+        tempfile._get_default_tempdir(), next(tempfile._get_candidate_names())
+    )
+
+
 def stringhash(s: str) -> int:
     # default hash() function consistent
     # only during process invocation https://stackoverflow.com/a/42089311
@@ -92,6 +232,16 @@ def trim_for_log(s):
         return "\n".join(lines)
 
 
+def is_valid_utf_8(fname):
+    try:
+        with open(fname, "rb") as f:
+            contents = f.read()
+            contents.decode("utf-8")
+            return True
+    except UnicodeDecodeError:
+        return False
+
+
 class TestException(Exception):
     pass
 
@@ -536,6 +686,8 @@ class FailureReason(enum.Enum):
     INTERNAL_QUERY_FAIL = "Internal query (CREATE/DROP DATABASE) failed:"
 
     # SKIPPED reasons
+    NOT_SUPPORTED_IN_CLOUD = "not supported in cloud environment"
+    NOT_SUPPORTED_IN_PRIVATE = "not supported in private build"
     DISABLED = "disabled"
     SKIP = "skip"
     NO_JINJA = "no jinja"
@@ -548,6 +700,7 @@ class FailureReason(enum.Enum):
     S3_STORAGE = "s3-storage"
     BUILD = "not running for current build"
     NO_PARALLEL_REPLICAS = "smth in not supported with parallel replicas"
+    SHARED_MERGE_TREE = "no-shared-merge-tree"
 
     # UNKNOWN reasons
     NO_REFERENCE = "no reference file"
@@ -606,8 +759,6 @@ class SettingsRandomizer:
         "read_in_order_two_level_merge_threshold": lambda: random.randint(0, 100),
         "optimize_aggregation_in_order": lambda: random.randint(0, 1),
         "aggregation_in_order_max_block_bytes": lambda: random.randint(0, 50000000),
-        "min_compress_block_size": lambda: random.randint(1, 1048576 * 3),
-        "max_compress_block_size": lambda: random.randint(1, 1048576 * 3),
         "use_uncompressed_cache": lambda: random.randint(0, 1),
         "min_bytes_to_use_direct_io": threshold_generator(
             0.2, 0.5, 1, 10 * 1024 * 1024 * 1024
@@ -659,6 +810,11 @@ class SettingsRandomizer:
             0.3, 0.5, 1, 10 * 1024 * 1024 * 1024
         ),
         "max_bytes_before_remerge_sort": lambda: random.randint(1, 3000000000),
+        "min_compress_block_size": lambda: random.randint(1, 1048576 * 3),
+        "max_compress_block_size": lambda: random.randint(1, 1048576 * 3),
+        "merge_tree_compact_parts_min_granules_to_multibuffer_read": lambda: random.randint(
+            1, 128
+        ),
         "optimize_sorting_by_input_stream_properties": lambda: random.randint(0, 1),
         "http_response_buffer_size": lambda: random.randint(0, 10 * 1048576),
         "http_wait_end_of_query": lambda: random.random() > 0.5,
@@ -684,6 +840,7 @@ class SettingsRandomizer:
                 get_localzone(),
             ]
         ),
+        "prefer_warmed_unmerged_parts_seconds": lambda: random.randint(0, 10),
         "use_page_cache_for_disks_without_file_cache": lambda: random.random() < 0.7,
         "page_cache_inject_eviction": lambda: random.random() < 0.5,
     }
@@ -733,6 +890,17 @@ class MergeTreeSettingsRandomizer:
         "primary_key_compress_block_size": lambda: random.randint(8000, 100000),
         "replace_long_file_name_to_hash": lambda: random.randint(0, 1),
         "max_file_name_length": threshold_generator(0.3, 0.3, 0, 128),
+        "min_bytes_for_full_part_storage": threshold_generator(
+            0.3, 0.3, 0, 512 * 1024 * 1024
+        ),
+        "compact_parts_max_bytes_to_buffer": lambda: random.randint(
+            1024, 512 * 1024 * 1024
+        ),
+        "compact_parts_max_granules_to_buffer": threshold_generator(0.15, 0.15, 1, 256),
+        "compact_parts_merge_max_bytes_to_prefetch_part": lambda: random.randint(
+            1, 32 * 1024 * 1024
+        ),
+        "cache_populated_by_fetch": lambda: random.randint(0, 1),
     }
 
     @staticmethod
@@ -744,6 +912,10 @@ class MergeTreeSettingsRandomizer:
         return random_settings
 
 
+def replace_in_file(filename, what, with_what):
+    os.system(f"LC_ALL=C sed -i -e 's|{what}|{with_what}|g' {filename}")
+
+
 class TestResult:
     def __init__(
         self,
@@ -972,6 +1144,15 @@ class TestCase:
         if tags and ("disabled" in tags) and not args.disabled:
             return FailureReason.DISABLED
 
+        elif args.private and self.name in suite.private_skip_list:
+            return FailureReason.NOT_SUPPORTED_IN_PRIVATE
+
+        elif args.cloud and ("no-replicated-database" in tags):
+            return FailureReason.REPLICATED_DB
+
+        elif args.cloud and self.name in suite.cloud_skip_list:
+            return FailureReason.NOT_SUPPORTED_IN_CLOUD
+
         elif (
             os.path.exists(os.path.join(suite.suite_path, self.name) + ".disabled")
             and not args.disabled
@@ -1022,6 +1203,13 @@ class TestCase:
         ):
             return FailureReason.NON_ATOMIC_DB
 
+        elif (
+            tags
+            and ("no-shared-merge-tree" in tags)
+            and args.replace_replicated_with_shared
+        ):
+            return FailureReason.SHARED_MERGE_TREE
+
         elif tags and ("no-s3-storage" in tags) and args.s3_storage:
             return FailureReason.S3_STORAGE
         elif (
@@ -1051,7 +1239,8 @@ class TestCase:
     ):
         description = ""
 
-        debug_log = trim_for_log(debug_log)
+        if debug_log:
+            debug_log = "\n".join(debug_log.splitlines()[:100])
 
         if proc:
             if proc.returncode is None:
@@ -1136,6 +1325,7 @@ class TestCase:
             description += "\nstdout:\n"
             description += trim_for_log(stdout)
             description += "\n"
+
             if debug_log:
                 description += "\n"
                 description += debug_log
@@ -1148,9 +1338,7 @@ class TestCase:
             )
 
         if "Exception" in stdout:
-            description += "\n"
-            description += trim_for_log(stdout)
-            description += "\n"
+            description += "\n{}\n".format("\n".join(stdout.splitlines()[:100]))
             if debug_log:
                 description += "\n"
                 description += debug_log
@@ -1358,7 +1546,13 @@ class TestCase:
         # because there are also output of per test database creation
         pattern = "{test} > {stdout} 2> {stderr}"
 
-        if self.ext == ".sql":
+        if self.ext == ".sql" and args.cloud:
+            # Get at least some logs, because we don't have access to system.text_log and pods...
+            pattern = (
+                "{client} --send_logs_level={logs_level} {secure} --multiquery {options}"
+                " --send_logs_level=trace < {test} > {stdout}  2>> /test_output/some_logs_from_server.log"
+            )
+        elif self.ext == ".sql" and not args.cloud:
             pattern = (
                 "{client} --send_logs_level={logs_level} {secure} --multiquery {options} < "
                 + pattern
@@ -1396,17 +1590,15 @@ class TestCase:
         total_time = (datetime.now() - start_time).total_seconds()
 
         # Normalize randomized database names in stdout, stderr files.
-        os.system(f"LC_ALL=C sed -i -e 's/{database}/default/g' {self.stdout_file}")
+        replace_in_file(self.stdout_file, database, "default")
         if args.hide_db_name:
-            os.system(f"LC_ALL=C sed -i -e 's/{database}/default/g' {self.stderr_file}")
+            replace_in_file(self.stderr_file, database, "default")
         if args.replicated_database:
-            os.system(f"LC_ALL=C sed -i -e 's|/auto_{{shard}}||g' {self.stdout_file}")
-            os.system(f"LC_ALL=C sed -i -e 's|auto_{{replica}}||g' {self.stdout_file}")
+            replace_in_file(self.stdout_file, "/auto_{shard}", "")
+            replace_in_file(self.stdout_file, "auto_{replica}", "")
 
         # Normalize hostname in stdout file.
-        os.system(
-            f"LC_ALL=C sed -i -e 's/{socket.gethostname()}/localhost/g' {self.stdout_file}"
-        )
+        replace_in_file(self.stdout_file, socket.gethostname(), "localhost")
 
         stdout = ""
         if os.path.exists(self.stdout_file):
@@ -1444,18 +1636,51 @@ class TestCase:
             self.testcase_args = self.configure_testcase_args(
                 args, self.case_file, suite.suite_tmp_path
             )
+
             client_options = self.add_random_settings(client_options)
-            proc, stdout, stderr, debug_log, total_time = self.run_single_test(
-                server_logs_level, client_options
-            )
 
-            result = self.process_result_impl(
-                proc, stdout, stderr, debug_log, total_time
-            )
-            result.check_if_need_retry(args, stdout, stderr, self.runs_count)
-            # to avoid breaking CSV parser
-            result.description = result.description.replace("\0", "")
+            if not is_valid_utf_8(self.case_file) or not is_valid_utf_8(
+                self.reference_file
+            ):
+                proc, stdout, stderr, debug_log, total_time = self.run_single_test(
+                    server_logs_level, client_options
+                )
 
+                result = self.process_result_impl(
+                    proc, stdout, stderr, debug_log, total_time
+                )
+                result.check_if_need_retry(args, stdout, stderr, self.runs_count)
+                # to avoid breaking CSV parser
+                result.description = result.description.replace("\0", "")
+            else:
+                with SharedEngineReplacer(
+                    self.case_file,
+                    args.replace_replicated_with_shared,
+                    args.replace_non_replicated_with_shared,
+                    False,
+                ):
+                    with SharedEngineReplacer(
+                        self.reference_file,
+                        args.replace_replicated_with_shared,
+                        args.replace_non_replicated_with_shared,
+                        True,
+                    ):
+                        (
+                            proc,
+                            stdout,
+                            stderr,
+                            debug_log,
+                            total_time,
+                        ) = self.run_single_test(server_logs_level, client_options)
+
+                        result = self.process_result_impl(
+                            proc, stdout, stderr, debug_log, total_time
+                        )
+                        result.check_if_need_retry(
+                            args, stdout, stderr, self.runs_count
+                        )
+                        # to avoid breaking CSV parser
+                        result.description = result.description.replace("\0", "")
             if result.status == TestStatus.FAIL:
                 result.description = self.add_info_about_settings(result.description)
 
@@ -1688,6 +1913,8 @@ class TestSuite:
         self.suite_path: str = suite_path
         self.suite_tmp_path: str = suite_tmp_path
         self.suite: str = suite
+        self.cloud_skip_list: List[str] = []
+        self.private_skip_list: List[str] = []
 
         if args.run_by_hash_num is not None and args.run_by_hash_total is not None:
             if args.run_by_hash_num > args.run_by_hash_total:
@@ -1987,10 +2214,16 @@ def check_server_started(args):
 
     sys.stdout.flush()
     retry_count = args.server_check_retries
+    query = "SELECT version(), arrayStringConcat(groupArray(value), ' ') FROM system.build_options WHERE name IN ('GIT_HASH', 'GIT_BRANCH')"
     while retry_count > 0:
         try:
-            clickhouse_execute(args, "SELECT 1", max_http_retries=1)
+            res = (
+                str(clickhouse_execute(args, query).decode())
+                .strip()
+                .replace("\t", " @ ")
+            )
             print(" OK")
+            print(f"Connected to server {res}")
             sys.stdout.flush()
             return True
         except (ConnectionError, http.client.ImproperConnectionState) as e:
@@ -2412,6 +2645,23 @@ def reportLogStats(args):
     print("\n")
 
 
+def try_get_skip_list(base_dir, name):
+    test_names_to_skip = []
+    skip_list_path = os.path.join(base_dir, name)
+    if not os.path.exists(skip_list_path):
+        return test_names_to_skip
+
+    with open(skip_list_path, "r", encoding="utf-8") as fd:
+        for line in fd.read().split("\n"):
+            if line == "" or line[0] == " ":
+                continue
+            test_name = line.split()[0].strip()
+            if test_name != "":
+                test_names_to_skip.append(test_name)
+
+    return test_names_to_skip
+
+
 def main(args):
     global server_died
     global stop_time
@@ -2430,18 +2680,18 @@ def main(args):
 
     args.build_flags = collect_build_flags(args)
     args.changed_merge_tree_settings = collect_changed_merge_tree_settings(args)
-    args.suppport_system_processes_is_all_data_sent = check_table_column(
-        args, "system", "processes", "is_all_data_sent"
-    )
 
-    if args.s3_storage and (
-        BuildFlags.THREAD in args.build_flags or BuildFlags.DEBUG in args.build_flags
-    ):
+    if args.s3_storage and (BuildFlags.RELEASE not in args.build_flags):
         args.no_random_settings = True
 
     if args.skip:
         args.skip = set(args.skip)
 
+    if args.replace_replicated_with_shared:
+        if not args.skip:
+            args.skip = set([])
+        args.skip = set(args.skip)
+
     base_dir = os.path.abspath(args.queries)
 
     # Keep same default values as in queries/shell_config.sh
@@ -2516,6 +2766,8 @@ def main(args):
         )
 
     total_tests_run = 0
+    cloud_skip_list = try_get_skip_list(base_dir, "../queries-no-cloud-tests.txt")
+    private_skip_list = try_get_skip_list(base_dir, "../queries-no-private-tests.txt")
 
     for suite in sorted(os.listdir(base_dir), key=suite_key_func):
         if server_died.is_set():
@@ -2525,6 +2777,8 @@ def main(args):
         if test_suite is None:
             continue
 
+        test_suite.cloud_skip_list = cloud_skip_list
+        test_suite.private_skip_list = private_skip_list
         total_tests_run += do_run_tests(args.jobs, test_suite, args.parallel)
 
     if server_died.is_set():
@@ -2644,7 +2898,14 @@ def find_clickhouse_command(binary, command):
 
 def get_additional_client_options(args):
     if args.client_option:
-        return " ".join("--" + option for option in args.client_option)
+        client_options = " ".join("--" + option for option in args.client_option)
+        if "CLICKHOUSE_CLIENT_OPT" in os.environ:
+            return os.environ["CLICKHOUSE_CLIENT_OPT"] + client_options
+        else:
+            return client_options
+    else:
+        if "CLICKHOUSE_CLIENT_OPT" in os.environ:
+            return os.environ["CLICKHOUSE_CLIENT_OPT"]
     return ""
 
 
@@ -2839,6 +3100,43 @@ def parse_args():
         help="Display $ characters after line with trailing whitespaces in diff output",
     )
 
+    group = parser.add_mutually_exclusive_group(required=False)
+    group.add_argument(
+        "--cloud",
+        action="store_true",
+        default=None,
+        dest="cloud",
+        help="Run only tests that are supported in ClickHouse Cloud environment",
+    )
+
+    group.add_argument(
+        "--no-cloud",
+        action="store_false",
+        default=None,
+        dest="cloud",
+        help="Run all the tests, including the ones not supported in ClickHouse Cloud environment",
+    )
+    parser.set_defaults(cloud=False)
+
+    group = parser.add_mutually_exclusive_group(required=False)
+    group.add_argument(
+        "--private",
+        action="store_true",
+        default=None,
+        dest="private",
+        help="Run only tests that are supported in the private build",
+    )
+
+    group.add_argument(
+        "--no-private",
+        action="store_false",
+        default=None,
+        dest="private",
+        help="Run all the tests, including the ones not supported in the private build",
+    )
+    # Only used to skip tests via "../queries-no-private-tests.txt", so it's fine to keep it enabled by default
+    parser.set_defaults(private=True)
+
     group = parser.add_mutually_exclusive_group(required=False)
     group.add_argument(
         "--zookeeper",
@@ -2920,6 +3218,18 @@ def parse_args():
         default=False,
         help="Do not include tests that are not supported with parallel replicas feature",
     )
+    parser.add_argument(
+        "--replace-replicated-with-shared",
+        action="store_true",
+        default=os.environ.get("USE_META_IN_KEEPER_FOR_MERGE_TREE", False),
+        help="Replace ReplicatedMergeTree engine with SharedMergeTree",
+    )
+    parser.add_argument(
+        "--replace-non-replicated-with-shared",
+        action="store_true",
+        default=False,
+        help="Replace ordinary MergeTree engine with SharedMergeTree",
+    )
 
     return parser.parse_args()
 
@@ -3062,6 +3372,7 @@ if __name__ == "__main__":
 
         client_options_query_str = get_additional_client_options_url(args)
         args.client_options_query_str = client_options_query_str + "&"
+        args.client_options_query_str += os.environ["CLICKHOUSE_URL_PARAMS"]
         os.environ["CLICKHOUSE_URL_PARAMS"] += client_options_query_str
     else:
         args.client_options_query_str = ""
@@ -3072,4 +3383,7 @@ if __name__ == "__main__":
     if args.db_engine and args.db_engine == "Ordinary":
         MESSAGES_TO_RETRY.append(" locking attempt on ")
 
+    if args.replace_replicated_with_shared:
+        args.s3_storage = True
+
     main(args)

From 80723134d6957f72606a419040c8101ec60c05e9 Mon Sep 17 00:00:00 2001
From: Dmitry Novik <n0vik@clickhouse.com>
Date: Wed, 13 Mar 2024 14:00:57 +0100
Subject: [PATCH 281/443] Fix fast test #ci_set_analyzer

---
 src/Storages/StorageMerge.cpp | 89 +++++++++++++++++++++++------------
 1 file changed, 58 insertions(+), 31 deletions(-)

diff --git a/src/Storages/StorageMerge.cpp b/src/Storages/StorageMerge.cpp
index e695594873b..f0b9d58f3dd 100644
--- a/src/Storages/StorageMerge.cpp
+++ b/src/Storages/StorageMerge.cpp
@@ -1058,47 +1058,74 @@ QueryPipelineBuilderPtr ReadFromMerge::createSources(
 
         Block pipe_header = builder->getHeader();
 
-        auto get_column_options = GetColumnsOptions(GetColumnsOptions::All).withExtendedObjects().withVirtuals();
-        if (storage_snapshot_->storage.supportsSubcolumns())
-            get_column_options.withSubcolumns();
-
         LOG_DEBUG(&Poco::Logger::get("createSources"), "Processed:{}\nStorage:{}", toString(processed_stage), toString(storage_stage));
 
-        String table_alias;
         if (allow_experimental_analyzer)
-            table_alias = modified_query_info.query_tree->as<QueryNode>()->getJoinTree()->as<TableNode>()->getAlias();
-
-        String database_column = table_alias.empty() || processed_stage == QueryProcessingStage::FetchColumns ? "_database" : table_alias + "._database";
-        String table_column = table_alias.empty() || processed_stage == QueryProcessingStage::FetchColumns ? "_table" : table_alias + "._table";
-
-        if (has_database_virtual_column && common_header.has(database_column) && (storage_stage == QueryProcessingStage::FetchColumns || dynamic_cast<const StorageDistributed *>(&storage_snapshot_->storage) != nullptr))
         {
-            ColumnWithTypeAndName column;
-            column.name = database_column;
-            column.type = std::make_shared<DataTypeLowCardinality>(std::make_shared<DataTypeString>());
-            column.column = column.type->createColumnConst(0, Field(database_name));
+            String table_alias = modified_query_info.query_tree->as<QueryNode>()->getJoinTree()->as<TableNode>()->getAlias();
 
-            auto adding_column_dag = ActionsDAG::makeAddingColumnActions(std::move(column));
-            auto adding_column_actions = std::make_shared<ExpressionActions>(
-                std::move(adding_column_dag), ExpressionActionsSettings::fromContext(context, CompileExpressions::yes));
+            String database_column = table_alias.empty() || processed_stage == QueryProcessingStage::FetchColumns ? "_database" : table_alias + "._database";
+            String table_column = table_alias.empty() || processed_stage == QueryProcessingStage::FetchColumns ? "_table" : table_alias + "._table";
 
-            builder->addSimpleTransform([&](const Block & stream_header)
-                                        { return std::make_shared<ExpressionTransform>(stream_header, adding_column_actions); });
+            if (has_database_virtual_column && common_header.has(database_column) && (storage_stage == QueryProcessingStage::FetchColumns || dynamic_cast<const StorageDistributed *>(&storage_snapshot_->storage) != nullptr))
+            {
+                ColumnWithTypeAndName column;
+                column.name = database_column;
+                column.type = std::make_shared<DataTypeLowCardinality>(std::make_shared<DataTypeString>());
+                column.column = column.type->createColumnConst(0, Field(database_name));
+
+                auto adding_column_dag = ActionsDAG::makeAddingColumnActions(std::move(column));
+                auto adding_column_actions = std::make_shared<ExpressionActions>(
+                    std::move(adding_column_dag), ExpressionActionsSettings::fromContext(context, CompileExpressions::yes));
+
+                builder->addSimpleTransform([&](const Block & stream_header)
+                                            { return std::make_shared<ExpressionTransform>(stream_header, adding_column_actions); });
+            }
+
+            if (has_table_virtual_column && common_header.has(table_column) && (storage_stage == QueryProcessingStage::FetchColumns || dynamic_cast<const StorageDistributed *>(&storage_snapshot_->storage) != nullptr))
+            {
+                ColumnWithTypeAndName column;
+                column.name = table_column;
+                column.type = std::make_shared<DataTypeLowCardinality>(std::make_shared<DataTypeString>());
+                column.column = column.type->createColumnConst(0, Field(table_name));
+
+                auto adding_column_dag = ActionsDAG::makeAddingColumnActions(std::move(column));
+                auto adding_column_actions = std::make_shared<ExpressionActions>(
+                    std::move(adding_column_dag), ExpressionActionsSettings::fromContext(context, CompileExpressions::yes));
+
+                builder->addSimpleTransform([&](const Block & stream_header)
+                                            { return std::make_shared<ExpressionTransform>(stream_header, adding_column_actions); });
+            }
         }
-
-        if (has_table_virtual_column && common_header.has(table_column) && (storage_stage == QueryProcessingStage::FetchColumns || dynamic_cast<const StorageDistributed *>(&storage_snapshot_->storage) != nullptr))
+        else
         {
-            ColumnWithTypeAndName column;
-            column.name = table_column;
-            column.type = std::make_shared<DataTypeLowCardinality>(std::make_shared<DataTypeString>());
-            column.column = column.type->createColumnConst(0, Field(table_name));
+            if (has_database_virtual_column && common_header.has("_database") && !pipe_header.has("_database"))
+            {
+                ColumnWithTypeAndName column;
+                column.name = "_database";
+                column.type = std::make_shared<DataTypeLowCardinality>(std::make_shared<DataTypeString>());
+                column.column = column.type->createColumnConst(0, Field(database_name));
 
-            auto adding_column_dag = ActionsDAG::makeAddingColumnActions(std::move(column));
-            auto adding_column_actions = std::make_shared<ExpressionActions>(
-                std::move(adding_column_dag), ExpressionActionsSettings::fromContext(context, CompileExpressions::yes));
+                auto adding_column_dag = ActionsDAG::makeAddingColumnActions(std::move(column));
+                auto adding_column_actions = std::make_shared<ExpressionActions>(
+                    std::move(adding_column_dag), ExpressionActionsSettings::fromContext(context, CompileExpressions::yes));
+                builder->addSimpleTransform([&](const Block & stream_header)
+                                            { return std::make_shared<ExpressionTransform>(stream_header, adding_column_actions); });
+            }
 
-            builder->addSimpleTransform([&](const Block & stream_header)
-                                        { return std::make_shared<ExpressionTransform>(stream_header, adding_column_actions); });
+            if (has_table_virtual_column && common_header.has("_table") && !pipe_header.has("_table"))
+            {
+                ColumnWithTypeAndName column;
+                column.name = "_table";
+                column.type = std::make_shared<DataTypeLowCardinality>(std::make_shared<DataTypeString>());
+                column.column = column.type->createColumnConst(0, Field(table_name));
+
+                auto adding_column_dag = ActionsDAG::makeAddingColumnActions(std::move(column));
+                auto adding_column_actions = std::make_shared<ExpressionActions>(
+                    std::move(adding_column_dag), ExpressionActionsSettings::fromContext(context, CompileExpressions::yes));
+                builder->addSimpleTransform([&](const Block & stream_header)
+                                            { return std::make_shared<ExpressionTransform>(stream_header, adding_column_actions); });
+            }
         }
 
         /// Subordinary tables could have different but convertible types, like numeric types of different width.

From d8c5008280aaf19bd481d436099afd89019a81c4 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Wed, 13 Mar 2024 13:07:14 +0000
Subject: [PATCH 282/443] Follow up to #61258

---
 .../functions/other-functions.md              |  6 ++---
 src/Functions/sleep.h                         | 25 ++++---------------
 2 files changed, 8 insertions(+), 23 deletions(-)

diff --git a/docs/en/sql-reference/functions/other-functions.md b/docs/en/sql-reference/functions/other-functions.md
index 288905c83da..e7fca31483a 100644
--- a/docs/en/sql-reference/functions/other-functions.md
+++ b/docs/en/sql-reference/functions/other-functions.md
@@ -298,7 +298,7 @@ Full columns and constants are represented differently in memory. Functions usua
 Accepts any arguments, including `NULL` and does nothing. Always returns 0.
 The argument is internally still evaluated. Useful e.g. for benchmarks.
 
-## sleep(seconds)
+## sleep
 
 Used to introduce a delay or pause in the execution of a query. It is primarily used for testing and debugging purposes.
 
@@ -310,7 +310,7 @@ sleep(seconds)
 
 **Arguments**
 
-- `seconds`: [Int](../../sql-reference/data-types/int-uint.md) The number of seconds to pause the query execution to a maximum of 3 seconds. It can be a floating-point value to specify fractional seconds.
+- `seconds`: [UInt*](../../sql-reference/data-types/int-uint.md) or [Float](../../sql-reference/data-types/float.md) The number of seconds to pause the query execution to a maximum of 3 seconds. It can be a floating-point value to specify fractional seconds.
 
 **Returned value**
 
@@ -360,7 +360,7 @@ sleepEachRow(seconds)
 
 **Arguments**
 
-- `seconds`: [Int](../../sql-reference/data-types/int-uint.md) The number of seconds to pause the query execution for each row in the result set to a maximum of 3 seconds. It can be a floating-point value to specify fractional seconds.
+- `seconds`: [UInt*](../../sql-reference/data-types/int-uint.md) or [Float*](../../sql-reference/data-types/float.md) The number of seconds to pause the query execution for each row in the result set to a maximum of 3 seconds. It can be a floating-point value to specify fractional seconds.
 
 **Returned value**
 
diff --git a/src/Functions/sleep.h b/src/Functions/sleep.h
index 73d58ca6b5b..84f08dd5440 100644
--- a/src/Functions/sleep.h
+++ b/src/Functions/sleep.h
@@ -62,32 +62,17 @@ public:
     {
     }
 
-    /// Get the name of the function.
-    String getName() const override
-    {
-        return name;
-    }
-
-    /// Do not sleep during query analysis.
-    bool isSuitableForConstantFolding() const override
-    {
-        return false;
-    }
-
-    size_t getNumberOfArguments() const override
-    {
-        return 1;
-    }
-
+    String getName() const override { return name; }
+    bool isSuitableForConstantFolding() const override { return false; } /// Do not sleep during query analysis.
+    size_t getNumberOfArguments() const override { return 1; }
     bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return true; }
 
     DataTypePtr getReturnTypeImpl(const DataTypes & arguments) const override
     {
         WhichDataType which(arguments[0]);
 
-        if (!which.isFloat()
-            && !which.isNativeUInt())
-            throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of argument of function {}, expected Float64",
+        if (!which.isFloat() && !which.isNativeUInt())
+            throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of argument of function {}, expected UInt* or Float*",
                 arguments[0]->getName(), getName());
 
         return std::make_shared<DataTypeUInt8>();

From f6e09fd02ccae269e5c54f817f6c36d7693c4376 Mon Sep 17 00:00:00 2001
From: Han Fei <hanfei19910905@gmail.com>
Date: Wed, 13 Mar 2024 14:11:50 +0100
Subject: [PATCH 283/443] address comments

---
 src/Common/CgroupsMemoryUsageObserver.cpp | 9 +++++----
 src/Common/CgroupsMemoryUsageObserver.h   | 4 ++--
 2 files changed, 7 insertions(+), 6 deletions(-)

diff --git a/src/Common/CgroupsMemoryUsageObserver.cpp b/src/Common/CgroupsMemoryUsageObserver.cpp
index e9121754230..c8379aed248 100644
--- a/src/Common/CgroupsMemoryUsageObserver.cpp
+++ b/src/Common/CgroupsMemoryUsageObserver.cpp
@@ -49,7 +49,7 @@ CgroupsMemoryUsageObserver::~CgroupsMemoryUsageObserver()
 
 void CgroupsMemoryUsageObserver::setLimits(uint64_t hard_limit_, uint64_t soft_limit_)
 {
-    std::lock_guard<std::mutex> lock(set_limit_mutex);
+    std::lock_guard<std::mutex> lock(limit_mutex);
     if (hard_limit_ == hard_limit && soft_limit_ == soft_limit)
         return;
 
@@ -99,7 +99,7 @@ void CgroupsMemoryUsageObserver::setLimits(uint64_t hard_limit_, uint64_t soft_l
 
 void CgroupsMemoryUsageObserver::setOnMemoryLimitUpdate(UpdateMemLimitCallbackFn on_memory_limit_update_)
 {
-    std::lock_guard<std::mutex> set_limit_lock(set_limit_mutex);
+    std::lock_guard<std::mutex> set_limit_lock(limit_mutex);
     on_memory_limit_update = on_memory_limit_update_;
 }
 
@@ -298,11 +298,12 @@ void CgroupsMemoryUsageObserver::runThread()
             {
                 LOG_INFO(log, "Find memory amount change, old limit is {} bytes, new limit is {} bytes", last_process_memory_amount, process_memory_limit);
                 last_process_memory_amount = process_memory_limit;
-                /// if we find memory amount changes, we just reload config.
+                /// If the available memory for the process changes (typically because the limit was adjusted via cgroups), then we just reload the config.
                 /// Reloading config will check the memory amount again and calculate soft/hard limit again.
+                std::lock_guard<std::mutex> set_limit_lock(limit_mutex);
                 on_memory_limit_update();
             }
-            std::lock_guard<std::mutex> set_limit_lock(set_limit_mutex);
+            std::lock_guard<std::mutex> set_limit_lock(limit_mutex);
             if (soft_limit > 0 && hard_limit > 0)
             {
                 uint64_t memory_usage = file.readMemoryUsage();
diff --git a/src/Common/CgroupsMemoryUsageObserver.h b/src/Common/CgroupsMemoryUsageObserver.h
index 24acc0d8f40..e1978af8d28 100644
--- a/src/Common/CgroupsMemoryUsageObserver.h
+++ b/src/Common/CgroupsMemoryUsageObserver.h
@@ -44,10 +44,11 @@ private:
 
     const std::chrono::seconds wait_time;
 
+    std::mutex limit_mutex;
     using CallbackFn = std::function<void(bool)>;
     CallbackFn on_hard_limit;
     CallbackFn on_soft_limit;
-    UpdateMemLimitCallbackFn on_memory_limit_update;
+    UpdateMemLimitCallbackFn on_memory_limit_update TSA_GUARDED_BY(limit_mutex);
 
     uint64_t last_usage = 0;
 
@@ -74,7 +75,6 @@ private:
     void processMemoryUsage(uint64_t usage);
 
     std::mutex thread_mutex;
-    std::mutex set_limit_mutex;
     std::condition_variable cond;
     ThreadFromGlobalPool thread;
     bool quit = false;

From e6af636a549f808730c87ab69a6b76531d3dbc95 Mon Sep 17 00:00:00 2001
From: Sema Checherinda <Sema.Checherinda@clickhouse.com>
Date: Wed, 13 Mar 2024 14:07:49 +0100
Subject: [PATCH 284/443] fix data race in poco tcp server

---
 base/poco/Net/src/TCPServerDispatcher.cpp  |  4 +++-
 src/Common/tests/gtest_connection_pool.cpp | 11 +++--------
 2 files changed, 6 insertions(+), 9 deletions(-)

diff --git a/base/poco/Net/src/TCPServerDispatcher.cpp b/base/poco/Net/src/TCPServerDispatcher.cpp
index 20a1ffe1b4f..7f9f9a20ee7 100644
--- a/base/poco/Net/src/TCPServerDispatcher.cpp
+++ b/base/poco/Net/src/TCPServerDispatcher.cpp
@@ -93,7 +93,7 @@ void TCPServerDispatcher::release()
 
 void TCPServerDispatcher::run()
 {
-	AutoPtr<TCPServerDispatcher> guard(this, true); // ensure object stays alive
+	AutoPtr<TCPServerDispatcher> guard(this); // ensure object stays alive
 
 	int idleTime = (int) _pParams->getThreadIdleTime().totalMilliseconds();
 
@@ -149,11 +149,13 @@ void TCPServerDispatcher::enqueue(const StreamSocket& socket)
 		{
 			try
 			{
+                this->duplicate();
 				_threadPool.startWithPriority(_pParams->getThreadPriority(), *this, threadName);
 				++_currentThreads;
 			}
 			catch (Poco::Exception& exc)
 			{
+                this->release();
 				++_refusedConnections;
 				std::cerr << "Got exception while starting thread for connection. Error code: "
 						  << exc.code() << ", message: '" << exc.displayText() << "'" << std::endl;
diff --git a/src/Common/tests/gtest_connection_pool.cpp b/src/Common/tests/gtest_connection_pool.cpp
index c271cc0e2ec..dcc3c11fd52 100644
--- a/src/Common/tests/gtest_connection_pool.cpp
+++ b/src/Common/tests/gtest_connection_pool.cpp
@@ -123,17 +123,15 @@ protected:
 
     std::string getServerUrl() const
     {
-        return "http://" + server_data.socket->address().toString();
+        return "http://" + server_data.server->socket().address().toString();
     }
 
     void startServer()
     {
         server_data.reset();
-        server_data.params = new Poco::Net::HTTPServerParams();
-        server_data.socket = std::make_unique<Poco::Net::ServerSocket>(server_data.port);
         server_data.handler_factory = new HTTPRequestHandlerFactory(slowdown_receive);
         server_data.server =  std::make_unique<Poco::Net::HTTPServer>(
-            server_data.handler_factory, *server_data.socket, server_data.params);
+            server_data.handler_factory, server_data.port);
 
         server_data.server->start();
     }
@@ -155,8 +153,7 @@ protected:
     {
         // just some port to avoid collisions with others tests
         UInt16 port = 9871;
-        Poco::Net::HTTPServerParams::Ptr params;
-        std::unique_ptr<Poco::Net::ServerSocket> socket;
+
         HTTPRequestHandlerFactory::Ptr handler_factory;
         std::unique_ptr<Poco::Net::HTTPServer> server;
 
@@ -171,8 +168,6 @@ protected:
 
             server = nullptr;
             handler_factory = nullptr;
-            socket = nullptr;
-            params = nullptr;
         }
 
         ~ServerData() {

From 9b1d8a2f52f2dd67a242eef0451b127c86066be5 Mon Sep 17 00:00:00 2001
From: alesapin <alesapin@gmail.com>
Date: Wed, 13 Mar 2024 14:15:34 +0100
Subject: [PATCH 285/443] fix style

---
 docs/en/operations/configuration-files.md | 2 +-
 programs/server/Server.cpp                | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/docs/en/operations/configuration-files.md b/docs/en/operations/configuration-files.md
index fbac2407640..6b2dd309623 100644
--- a/docs/en/operations/configuration-files.md
+++ b/docs/en/operations/configuration-files.md
@@ -116,7 +116,7 @@ XML substitution example:
 
 Substitutions can also be performed from ZooKeeper. To do this, specify the attribute `from_zk = "/path/to/node"`. The element value is replaced with the contents of the node at `/path/to/node` in ZooKeeper. You can also put an entire XML subtree on the ZooKeeper node, and it will be fully inserted into the source element.
 
-If you want to merge substitution content with existing configuration instead of append you can use attribute `merge="true`, for example: `<include from_zk="/some_path" merge="true">`. In this case existing configuration will be merged with content from substituion and existing configuration settings will be replaced with values from substituion.
+If you want to merge substitution content with existing configuration instead of append you can use attribute `merge="true`, for example: `<include from_zk="/some_path" merge="true">`. In this case existing configuration will be merged with content from substitution and existing configuration settings will be replaced with values from substituion.
 
 ## Encrypting and Hiding Configuration {#encryption}
 
diff --git a/programs/server/Server.cpp b/programs/server/Server.cpp
index 9d7dd02a235..3be34341173 100644
--- a/programs/server/Server.cpp
+++ b/programs/server/Server.cpp
@@ -910,7 +910,7 @@ try
     /// We need to reload server settings because config could be updated via zookeeper.
     server_settings.loadSettingsFromConfig(config());
 
-    /// NOTE: Do sanity checks after we loaded all possible substituions from ZK
+    /// NOTE: Do sanity checks after we loaded all possible substitutions from ZK
     sanityChecks(*this);
 
 #if defined(OS_LINUX)

From fefee44540bd029eb2d354706f61a1d96ed0e272 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Wed, 13 Mar 2024 14:27:47 +0100
Subject: [PATCH 286/443] Update settings changes history

---
 src/Core/SettingsChangesHistory.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/src/Core/SettingsChangesHistory.h b/src/Core/SettingsChangesHistory.h
index e680c02671a..d7b0669f64f 100644
--- a/src/Core/SettingsChangesHistory.h
+++ b/src/Core/SettingsChangesHistory.h
@@ -93,6 +93,7 @@ static std::map<ClickHouseVersion, SettingsChangesHistory::SettingsChanges> sett
               {"input_format_json_use_string_type_for_ambiguous_paths_in_named_tuples_inference_from_objects", false, false, "Allow to use String type for ambiguous paths during named tuple inference from JSON objects"},
               {"throw_if_deduplication_in_dependent_materialized_views_enabled_with_async_insert", false, true, "Deduplication is dependent materialized view cannot work together with async inserts."},
               {"parallel_replicas_allow_in_with_subquery", false, true, "If true, subquery for IN will be executed on every follower replica"},
+              {"filesystem_cache_reserve_space_wait_lock_timeout_milliseconds", 1000, 1000, "Wait time to lock cache for sapce reservation in filesystem cache"},
               }},
     {"24.2", {{"allow_suspicious_variant_types", true, false, "Don't allow creating Variant type with suspicious variants by default"},
               {"validate_experimental_and_suspicious_types_inside_nested_types", false, true, "Validate usage of experimental and suspicious types inside nested types"},

From 9bb71291d50d29dd0c401580402adc12290224bb Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Wed, 13 Mar 2024 14:42:57 +0100
Subject: [PATCH 287/443] Fix unit test

---
 src/Interpreters/tests/gtest_filecache.cpp | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/src/Interpreters/tests/gtest_filecache.cpp b/src/Interpreters/tests/gtest_filecache.cpp
index b596ccb0285..2679d1b8d18 100644
--- a/src/Interpreters/tests/gtest_filecache.cpp
+++ b/src/Interpreters/tests/gtest_filecache.cpp
@@ -245,7 +245,7 @@ void download(FileSegment & file_segment)
     ASSERT_EQ(file_segment.state(), State::DOWNLOADING);
     ASSERT_EQ(file_segment.getDownloadedSize(), 0);
 
-    ASSERT_TRUE(file_segment.reserve(file_segment.range().size()));
+    ASSERT_TRUE(file_segment.reserve(file_segment.range().size(), 1000));
     download(cache_base_path, file_segment);
     ASSERT_EQ(file_segment.state(), State::DOWNLOADING);
 
@@ -257,7 +257,7 @@ void assertDownloadFails(FileSegment & file_segment)
 {
     ASSERT_EQ(file_segment.getOrSetDownloader(), FileSegment::getCallerId());
     ASSERT_EQ(file_segment.getDownloadedSize(), 0);
-    ASSERT_FALSE(file_segment.reserve(file_segment.range().size()));
+    ASSERT_FALSE(file_segment.reserve(file_segment.range().size(), 1000));
     file_segment.complete();
 }
 
@@ -956,7 +956,7 @@ TEST_F(FileCacheTest, temporaryData)
         for (auto & segment : *some_data_holder)
         {
             ASSERT_TRUE(segment->getOrSetDownloader() == DB::FileSegment::getCallerId());
-            ASSERT_TRUE(segment->reserve(segment->range().size()));
+            ASSERT_TRUE(segment->reserve(segment->range().size(), 1000));
             download(*segment);
             segment->complete();
         }

From 4f2be003521b00c9a9087e17fcffdf08cabcd5f1 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Wed, 13 Mar 2024 13:55:32 +0000
Subject: [PATCH 288/443] Some fixups

---
 docs/en/sql-reference/functions/functions-for-nulls.md | 10 +++++-----
 docs/en/sql-reference/operators/index.md               |  4 ++--
 2 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/docs/en/sql-reference/functions/functions-for-nulls.md b/docs/en/sql-reference/functions/functions-for-nulls.md
index e73d6c899e7..61da9a191a1 100644
--- a/docs/en/sql-reference/functions/functions-for-nulls.md
+++ b/docs/en/sql-reference/functions/functions-for-nulls.md
@@ -10,11 +10,13 @@ sidebar_label: Nullable
 
 Returns whether the argument is [NULL](../../sql-reference/syntax.md#null).
 
+See also operator [`IS NULL`](../operators/index.md#is_null).
+
 ``` sql
 isNull(x)
 ```
 
-Alias: `IS NULL`.
+Alias: `ISNULL`
 
 **Arguments**
 
@@ -54,12 +56,12 @@ Result:
 
 Returns whether the argument is not [NULL](../../sql-reference/syntax.md#null-literal).
 
+See also operator [`IS NOT NULL`](../operators/index.md#is_not_null).
+
 ``` sql
 isNotNull(x)
 ```
 
-Alias: `IS NOT NULL`.
-
 **Arguments:**
 
 - `x` — A value of non-compound data type.
@@ -102,8 +104,6 @@ Returns whether the argument is 0 (zero) or [NULL](../../sql-reference/syntax.md
 isZeroOrNull(x)
 ```
 
-Alias: `x = 0 OR x IS NULL`.
-
 **Arguments:**
 
 - `x` — A value of non-compound data type.
diff --git a/docs/en/sql-reference/operators/index.md b/docs/en/sql-reference/operators/index.md
index 120e464e009..31bf43e8b35 100644
--- a/docs/en/sql-reference/operators/index.md
+++ b/docs/en/sql-reference/operators/index.md
@@ -353,7 +353,7 @@ For efficiency, the `and` and `or` functions accept any number of arguments. The
 
 ClickHouse supports the `IS NULL` and `IS NOT NULL` operators.
 
-### IS NULL
+### IS NULL {#is_null}
 
 - For [Nullable](../../sql-reference/data-types/nullable.md) type values, the `IS NULL` operator returns:
     - `1`, if the value is `NULL`.
@@ -374,7 +374,7 @@ SELECT x+100 FROM t_null WHERE y IS NULL
 └──────────────┘
 ```
 
-### IS NOT NULL
+### IS NOT NULL {#is_not_null}
 
 - For [Nullable](../../sql-reference/data-types/nullable.md) type values, the `IS NOT NULL` operator returns:
     - `0`, if the value is `NULL`.

From 6bbf9eb5400206c326a4e453a38612c8beb6ef89 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Wed, 13 Mar 2024 13:57:55 +0000
Subject: [PATCH 289/443] Fixup fixups

---
 docs/en/sql-reference/functions/functions-for-nulls.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/en/sql-reference/functions/functions-for-nulls.md b/docs/en/sql-reference/functions/functions-for-nulls.md
index 61da9a191a1..4dfbf4262ed 100644
--- a/docs/en/sql-reference/functions/functions-for-nulls.md
+++ b/docs/en/sql-reference/functions/functions-for-nulls.md
@@ -16,7 +16,7 @@ See also operator [`IS NULL`](../operators/index.md#is_null).
 isNull(x)
 ```
 
-Alias: `ISNULL`
+Alias: `ISNULL`.
 
 **Arguments**
 

From 9338b681efa7ea67e09e2ee37636ca0d551590a6 Mon Sep 17 00:00:00 2001
From: alesapin <alesapin@gmail.com>
Date: Wed, 13 Mar 2024 15:12:36 +0100
Subject: [PATCH 290/443] I will bypass style check

---
 docs/en/operations/configuration-files.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/en/operations/configuration-files.md b/docs/en/operations/configuration-files.md
index 6b2dd309623..30f65261322 100644
--- a/docs/en/operations/configuration-files.md
+++ b/docs/en/operations/configuration-files.md
@@ -116,7 +116,7 @@ XML substitution example:
 
 Substitutions can also be performed from ZooKeeper. To do this, specify the attribute `from_zk = "/path/to/node"`. The element value is replaced with the contents of the node at `/path/to/node` in ZooKeeper. You can also put an entire XML subtree on the ZooKeeper node, and it will be fully inserted into the source element.
 
-If you want to merge substitution content with existing configuration instead of append you can use attribute `merge="true`, for example: `<include from_zk="/some_path" merge="true">`. In this case existing configuration will be merged with content from substitution and existing configuration settings will be replaced with values from substituion.
+If you want to merge substitution content with existing configuration instead of append you can use attribute `merge="true`, for example: `<include from_zk="/some_path" merge="true">`. In this case existing configuration will be merged with content from substitution and existing configuration settings will be replaced with values from substitution.
 
 ## Encrypting and Hiding Configuration {#encryption}
 

From ce665eae3012e330d0fb5186712c0e9c7086a4fa Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Wed, 13 Mar 2024 15:13:17 +0000
Subject: [PATCH 291/443] Doc fixups

---
 docs/en/operations/configuration-files.md | 10 +++++-----
 programs/server/Server.cpp                |  2 +-
 src/Common/Config/ConfigProcessor.cpp     |  1 +
 3 files changed, 7 insertions(+), 6 deletions(-)

diff --git a/docs/en/operations/configuration-files.md b/docs/en/operations/configuration-files.md
index 30f65261322..089704705d0 100644
--- a/docs/en/operations/configuration-files.md
+++ b/docs/en/operations/configuration-files.md
@@ -95,9 +95,11 @@ which is equal to
 
 ## Substituting Configuration {#substitution}
 
-The config can also define “substitutions”. If an element has the `incl` attribute, the corresponding substitution from the file will be used as the value. By default, the path to the file with substitutions is `/etc/metrika.xml`. This can be changed in the [include_from](../operations/server-configuration-parameters/settings.md#server_configuration_parameters-include_from) element in the server config. The substitution values are specified in `/clickhouse/substitution_name` elements in this file. If a substitution specified in `incl` does not exist, it is recorded in the log. To prevent ClickHouse from logging missing substitutions, specify the `optional="true"` attribute (for example, settings for [macros](../operations/server-configuration-parameters/settings.md#macros)).
+The config can define substitutions. There are two types of substitutions:
 
-If you want to replace an entire element with a substitution use `include` as the element name.
+- If an element has the `incl` attribute, the corresponding substitution from the file will be used as the value. By default, the path to the file with substitutions is `/etc/metrika.xml`. This can be changed in the [include_from](../operations/server-configuration-parameters/settings.md#server_configuration_parameters-include_from) element in the server config. The substitution values are specified in `/clickhouse/substitution_name` elements in this file. If a substitution specified in `incl` does not exist, it is recorded in the log. To prevent ClickHouse from logging missing substitutions, specify the `optional="true"` attribute (for example, settings for [macros](../operations/server-configuration-parameters/settings.md#macros)).
+
+- If you want to replace an entire element with a substitution, use `include` as the element name. Substitutions can also be performed from ZooKeeper by specifying attribute `from_zk = "/path/to/node"`. In this case, the element value is replaced with the contents of the Zookeeper node at `/path/to/node`. This also works with you store an entire XML subtree as a Zookeeper node, it will be fully inserted into the source element.
 
 XML substitution example:
 
@@ -114,9 +116,7 @@ XML substitution example:
 </clickhouse>
 ```
 
-Substitutions can also be performed from ZooKeeper. To do this, specify the attribute `from_zk = "/path/to/node"`. The element value is replaced with the contents of the node at `/path/to/node` in ZooKeeper. You can also put an entire XML subtree on the ZooKeeper node, and it will be fully inserted into the source element.
-
-If you want to merge substitution content with existing configuration instead of append you can use attribute `merge="true`, for example: `<include from_zk="/some_path" merge="true">`. In this case existing configuration will be merged with content from substitution and existing configuration settings will be replaced with values from substitution.
+If you want to merge the substituting content with the existing configuration instead of appending you can use attribute `merge="true"`, for example: `<include from_zk="/some_path" merge="true">`. In this case, the existing configuration will be merged with the content from the substitution and the existing configuration settings will be replaced with values from substitution.
 
 ## Encrypting and Hiding Configuration {#encryption}
 
diff --git a/programs/server/Server.cpp b/programs/server/Server.cpp
index 3be34341173..d7030e3b0aa 100644
--- a/programs/server/Server.cpp
+++ b/programs/server/Server.cpp
@@ -910,7 +910,7 @@ try
     /// We need to reload server settings because config could be updated via zookeeper.
     server_settings.loadSettingsFromConfig(config());
 
-    /// NOTE: Do sanity checks after we loaded all possible substitutions from ZK
+    /// NOTE: Do sanity checks after we loaded all possible substitutions (for the configuration) from ZK
     sanityChecks(*this);
 
 #if defined(OS_LINUX)
diff --git a/src/Common/Config/ConfigProcessor.cpp b/src/Common/Config/ConfigProcessor.cpp
index d524db883eb..60407c6a174 100644
--- a/src/Common/Config/ConfigProcessor.cpp
+++ b/src/Common/Config/ConfigProcessor.cpp
@@ -427,6 +427,7 @@ void ConfigProcessor::doIncludesRecursive(
 
     /// Replace the original contents, not add to it.
     bool replace = attributes->getNamedItem("replace");
+    /// Merge with the original contents
     bool merge = attributes->getNamedItem("merge");
 
     bool included_something = false;

From b8abd57b0f1e4f7b059cb119ccc0a7ad1d723d2c Mon Sep 17 00:00:00 2001
From: qaziqarta <96023488+qaziqarta@users.noreply.github.com>
Date: Wed, 13 Mar 2024 21:18:05 +0600
Subject: [PATCH 292/443] Updated references to format settings in datetime.md

Updated references to date_time_input_format and date_time_output_format.
---
 docs/en/sql-reference/data-types/datetime.md | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/docs/en/sql-reference/data-types/datetime.md b/docs/en/sql-reference/data-types/datetime.md
index 1adff18f598..a465106c2ff 100644
--- a/docs/en/sql-reference/data-types/datetime.md
+++ b/docs/en/sql-reference/data-types/datetime.md
@@ -36,9 +36,9 @@ You can explicitly set a time zone for `DateTime`-type columns when creating a t
 
 The [clickhouse-client](../../interfaces/cli.md) applies the server time zone by default if a time zone isn’t explicitly set when initializing the data type. To use the client time zone, run `clickhouse-client` with the `--use_client_time_zone` parameter.
 
-ClickHouse outputs values depending on the value of the [date_time_output_format](../../operations/settings/settings.md#settings-date_time_output_format) setting. `YYYY-MM-DD hh:mm:ss` text format by default. Additionally, you can change the output with the [formatDateTime](../../sql-reference/functions/date-time-functions.md#formatdatetime) function.
+ClickHouse outputs values depending on the value of the [date_time_output_format](../../operations/settings/settings-formats.md#date_time_output_format) setting. `YYYY-MM-DD hh:mm:ss` text format by default. Additionally, you can change the output with the [formatDateTime](../../sql-reference/functions/date-time-functions.md#formatdatetime) function.
 
-When inserting data into ClickHouse, you can use different formats of date and time strings, depending on the value of the [date_time_input_format](../../operations/settings/settings.md#settings-date_time_input_format) setting.
+When inserting data into ClickHouse, you can use different formats of date and time strings, depending on the value of the [date_time_input_format](../../operations/settings/settings-formats.md#date_time_input_format) setting.
 
 ## Examples
 
@@ -147,8 +147,8 @@ Time shifts for multiple days. Some pacific islands changed their timezone offse
 - [Type conversion functions](../../sql-reference/functions/type-conversion-functions.md)
 - [Functions for working with dates and times](../../sql-reference/functions/date-time-functions.md)
 - [Functions for working with arrays](../../sql-reference/functions/array-functions.md)
-- [The `date_time_input_format` setting](../../operations/settings/settings-formats.md#settings-date_time_input_format)
-- [The `date_time_output_format` setting](../../operations/settings/settings-formats.md#settings-date_time_output_format)
+- [The `date_time_input_format` setting](../../operations/settings/settings-formats.md#date_time_input_format)
+- [The `date_time_output_format` setting](../../operations/settings/settings-formats.md#date_time_output_format)
 - [The `timezone` server configuration parameter](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-timezone)
 - [The `session_timezone` setting](../../operations/settings/settings.md#session_timezone)
 - [Operators for working with dates and times](../../sql-reference/operators/index.md#operators-datetime)

From cb28c84a93709ab12fc32171bf880c0e911ec0d5 Mon Sep 17 00:00:00 2001
From: Sergei Trifonov <sergei@clickhouse.com>
Date: Wed, 13 Mar 2024 16:25:58 +0100
Subject: [PATCH 293/443] Fix `forget_partition` test (#61237)

---
 ...et_partition.sql => 02995_forget_partition.sh} | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)
 rename tests/queries/0_stateless/{02995_forget_partition.sql => 02995_forget_partition.sh} (63%)
 mode change 100644 => 100755

diff --git a/tests/queries/0_stateless/02995_forget_partition.sql b/tests/queries/0_stateless/02995_forget_partition.sh
old mode 100644
new mode 100755
similarity index 63%
rename from tests/queries/0_stateless/02995_forget_partition.sql
rename to tests/queries/0_stateless/02995_forget_partition.sh
index 269f7932ea4..8ece8d3ddb3
--- a/tests/queries/0_stateless/02995_forget_partition.sql
+++ b/tests/queries/0_stateless/02995_forget_partition.sh
@@ -1,5 +1,12 @@
--- Tags: zookeeper, no-replicated-database
+#!/usr/bin/env bash
+# Tags: zookeeper, no-replicated-database
 
+CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
+# shellcheck source=../shell_config.sh
+. "$CUR_DIR"/../shell_config.sh
+
+
+${CLICKHOUSE_CLIENT} --multiline --multiquery -q """
 drop table if exists forget_partition;
 
 create table forget_partition
@@ -16,7 +23,12 @@ insert into forget_partition select number, '2024-01-01' + interval number day,
 
 alter table forget_partition drop partition '20240101';
 alter table forget_partition drop partition '20240102';
+"""
 
+# DROP PARTITION do not wait for a part to be removed from memory due to possible concurrent SELECTs, so we have to do wait manually here
+while [[ $(${CLICKHOUSE_CLIENT} -q "select count() from system.parts where database=currentDatabase() and table='forget_partition' and partition='20240101'") != 0 ]]; do sleep 0.1; done
+
+${CLICKHOUSE_CLIENT} --multiline --multiquery -q """
 set allow_unrestricted_reads_from_keeper=1;
 
 select '---before---';
@@ -31,3 +43,4 @@ select '---after---';
 select name from system.zookeeper where path = '/test/02995/' || currentDatabase() || '/rmt/block_numbers' order by name;
 
 drop table forget_partition;
+"""

From a366acf59c91fca4df1262c812ac8c58ee7643e2 Mon Sep 17 00:00:00 2001
From: avogar <avogar@clickhouse.com>
Date: Wed, 13 Mar 2024 15:36:53 +0000
Subject: [PATCH 294/443] Don't use default cluster in test
 test_distibuted_settings

---
 .../test_distributed_config/configs/clusters.xml     | 12 ++++++++++++
 tests/integration/test_distributed_config/test.py    |  6 +++---
 2 files changed, 15 insertions(+), 3 deletions(-)
 create mode 100644 tests/integration/test_distributed_config/configs/clusters.xml

diff --git a/tests/integration/test_distributed_config/configs/clusters.xml b/tests/integration/test_distributed_config/configs/clusters.xml
new file mode 100644
index 00000000000..754d765f23f
--- /dev/null
+++ b/tests/integration/test_distributed_config/configs/clusters.xml
@@ -0,0 +1,12 @@
+<clickhouse>
+    <remote_servers>
+        <localhost_cluster>
+            <shard>
+                <replica>
+                    <host>localhost</host>
+                    <port>9000</port>
+                </replica>
+            </shard>
+        </localhost_cluster>
+    </remote_servers>
+</clickhouse>
diff --git a/tests/integration/test_distributed_config/test.py b/tests/integration/test_distributed_config/test.py
index 500e9ecdeed..c08334985b1 100644
--- a/tests/integration/test_distributed_config/test.py
+++ b/tests/integration/test_distributed_config/test.py
@@ -3,7 +3,7 @@ from helpers.cluster import ClickHouseCluster
 import logging
 
 cluster = ClickHouseCluster(__file__)
-node = cluster.add_instance("node", main_configs=["configs/overrides.xml"])
+node = cluster.add_instance("node", main_configs=["configs/overrides.xml", "configs/clusters.xml"])
 
 
 @pytest.fixture(scope="module")
@@ -23,7 +23,7 @@ def test_distibuted_settings(start_cluster):
     node.query(
         """
         CREATE TABLE data_1 (key Int) ENGINE Memory();
-        CREATE TABLE dist_1 as data_1 ENGINE Distributed(default, default, data_1) SETTINGS flush_on_detach = true;
+        CREATE TABLE dist_1 as data_1 ENGINE Distributed(localhost_cluster, default, data_1) SETTINGS flush_on_detach = true;
         SYSTEM STOP DISTRIBUTED SENDS dist_1;
         INSERT INTO dist_1 SETTINGS prefer_localhost_replica=0 VALUES (1);
         DETACH TABLE dist_1;
@@ -36,7 +36,7 @@ def test_distibuted_settings(start_cluster):
     node.query(
         """
         CREATE TABLE data_2 (key Int) ENGINE Memory();
-        CREATE TABLE dist_2 as data_2 ENGINE Distributed(default, default, data_2);
+        CREATE TABLE dist_2 as data_2 ENGINE Distributed(localhost_cluster, default, data_2);
         SYSTEM STOP DISTRIBUTED SENDS dist_2;
         INSERT INTO dist_2 SETTINGS prefer_localhost_replica=0 VALUES (2);
         DETACH TABLE dist_2;

From a4a859ba31c731d8d5163fe5759a166c8c95601c Mon Sep 17 00:00:00 2001
From: qaziqarta <96023488+qaziqarta@users.noreply.github.com>
Date: Wed, 13 Mar 2024 21:37:50 +0600
Subject: [PATCH 295/443] Updated ru portion of previous commit

---
 docs/ru/sql-reference/data-types/datetime.md | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/docs/ru/sql-reference/data-types/datetime.md b/docs/ru/sql-reference/data-types/datetime.md
index 57f24786bb7..25e87794147 100644
--- a/docs/ru/sql-reference/data-types/datetime.md
+++ b/docs/ru/sql-reference/data-types/datetime.md
@@ -27,9 +27,9 @@ DateTime([timezone])
 
 Консольный клиент ClickHouse по умолчанию использует часовой пояс сервера, если для значения `DateTime` часовой пояс не был задан в явном виде при инициализации типа данных. Чтобы использовать часовой пояс клиента, запустите [clickhouse-client](../../interfaces/cli.md) с параметром `--use_client_time_zone`.
 
-ClickHouse отображает значения в зависимости от значения параметра [date\_time\_output\_format](../../operations/settings/index.md#settings-date_time_output_format). Текстовый формат по умолчанию `YYYY-MM-DD hh:mm:ss`. Кроме того, вы можете поменять отображение с помощью функции [formatDateTime](../../sql-reference/functions/date-time-functions.md#formatdatetime).
+ClickHouse отображает значения в зависимости от значения параметра [date\_time\_output\_format](../../operations/settings/settings-formats.md#date_time_output_format). Текстовый формат по умолчанию `YYYY-MM-DD hh:mm:ss`. Кроме того, вы можете поменять отображение с помощью функции [formatDateTime](../../sql-reference/functions/date-time-functions.md#formatdatetime).
 
-При вставке данных в ClickHouse, можно использовать различные форматы даты и времени в зависимости от значения настройки [date_time_input_format](../../operations/settings/index.md#settings-date_time_input_format).
+При вставке данных в ClickHouse, можно использовать различные форматы даты и времени в зависимости от значения настройки [date_time_input_format](../../operations/settings/settings-formats.md#date_time_input_format).
 
 ## Примеры {#primery}
 
@@ -119,8 +119,8 @@ FROM dt
 -   [Функции преобразования типов](../../sql-reference/functions/type-conversion-functions.md)
 -   [Функции для работы с датой и временем](../../sql-reference/functions/date-time-functions.md)
 -   [Функции для работы с массивами](../../sql-reference/functions/array-functions.md)
--   [Настройка `date_time_input_format`](../../operations/settings/index.md#settings-date_time_input_format)
--   [Настройка `date_time_output_format`](../../operations/settings/index.md)
+-   [Настройка `date_time_input_format`](../../operations/settings/settings-formats.md#date_time_input_format)
+-   [Настройка `date_time_output_format`](../../operations/settings/settings-formats.md#date_time_output_format)
 -   [Конфигурационный параметр сервера `timezone`](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-timezone)
 -   [Параметр `session_timezone`](../../operations/settings/settings.md#session_timezone)
 -   [Операторы для работы с датой и временем](../../sql-reference/operators/index.md#operators-datetime)

From 868d24415b295da1c0d325e837823bfb7cde9253 Mon Sep 17 00:00:00 2001
From: avogar <avogar@clickhouse.com>
Date: Wed, 13 Mar 2024 15:48:29 +0000
Subject: [PATCH 296/443] Fix style

---
 tests/integration/test_distributed_config/test.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/tests/integration/test_distributed_config/test.py b/tests/integration/test_distributed_config/test.py
index c08334985b1..bf4bb5a4335 100644
--- a/tests/integration/test_distributed_config/test.py
+++ b/tests/integration/test_distributed_config/test.py
@@ -3,7 +3,9 @@ from helpers.cluster import ClickHouseCluster
 import logging
 
 cluster = ClickHouseCluster(__file__)
-node = cluster.add_instance("node", main_configs=["configs/overrides.xml", "configs/clusters.xml"])
+node = cluster.add_instance(
+    "node", main_configs=["configs/overrides.xml", "configs/clusters.xml"]
+)
 
 
 @pytest.fixture(scope="module")

From 51ccb520fb5747a00b7c56adae197467284c98d5 Mon Sep 17 00:00:00 2001
From: Nikolai Kochetov <nk@clickhouse.com>
Date: Wed, 13 Mar 2024 15:54:06 +0000
Subject: [PATCH 297/443] Change only ignore_limits setting

---
 src/Planner/Planner.cpp | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/src/Planner/Planner.cpp b/src/Planner/Planner.cpp
index 861b12e3da2..fde9f110d09 100644
--- a/src/Planner/Planner.cpp
+++ b/src/Planner/Planner.cpp
@@ -1081,9 +1081,10 @@ void addBuildSubqueriesForSetsStepIfNeeded(
     for (auto & subquery : subqueries)
     {
         auto query_tree = subquery->detachQueryTree();
-        /// I suppose it should be better to use all flags from select_query_options,
+        auto subquery_options = select_query_options.subquery();
+        /// I don't know if this is a good decision,
         /// But for now it is done in the same way as in old analyzer.
-        auto subquery_options = SelectQueryOptions(QueryProcessingStage::Complete, select_query_options.subquery_depth).subquery();
+        subquery_options.ignore_limits = false;
         Planner subquery_planner(
             query_tree,
             subquery_options,

From 617138cf482f9a55487d0a6c2e63dacdbbdafe8d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Wed, 13 Mar 2024 16:54:28 +0100
Subject: [PATCH 298/443] Teach the fuzzer to use other numeric types

---
 src/Client/QueryFuzzer.cpp | 46 +++++++++++++++++++++++++++++++++++++-
 1 file changed, 45 insertions(+), 1 deletion(-)

diff --git a/src/Client/QueryFuzzer.cpp b/src/Client/QueryFuzzer.cpp
index 7f1dce4b29a..38e78157096 100644
--- a/src/Client/QueryFuzzer.cpp
+++ b/src/Client/QueryFuzzer.cpp
@@ -914,6 +914,38 @@ ASTPtr QueryFuzzer::fuzzLiteralUnderExpressionList(ASTPtr child)
         child = makeASTFunction(
             "toFixedString", std::make_shared<ASTLiteral>(value), std::make_shared<ASTLiteral>(static_cast<UInt64>(value.size())));
     }
+    else if (type == Field::Types::Which::UInt64 && fuzz_rand() % 7 == 0)
+    {
+        child = makeASTFunction(fuzz_rand() % 2 == 0 ? "toUInt128" : "toUInt256", std::make_shared<ASTLiteral>(l->value.get<UInt64>()));
+    }
+    else if (type == Field::Types::Which::Int64 && fuzz_rand() % 7 == 0)
+    {
+        child = makeASTFunction(fuzz_rand() % 2 == 0 ? "toInt128" : "toInt256", std::make_shared<ASTLiteral>(l->value.get<Int64>()));
+    }
+    else if (type == Field::Types::Which::Float64 && fuzz_rand() % 7 == 0)
+    {
+        int decimal = fuzz_rand() % 4;
+        if (decimal == 0)
+            child = makeASTFunction(
+                "toDecimal32",
+                std::make_shared<ASTLiteral>(l->value.get<Float64>()),
+                std::make_shared<ASTLiteral>(static_cast<UInt64>(fuzz_rand() % 9)));
+        else if (decimal == 1)
+            child = makeASTFunction(
+                "toDecimal64",
+                std::make_shared<ASTLiteral>(l->value.get<Float64>()),
+                std::make_shared<ASTLiteral>(static_cast<UInt64>(fuzz_rand() % 18)));
+        else if (decimal == 2)
+            child = makeASTFunction(
+                "toDecimal128",
+                std::make_shared<ASTLiteral>(l->value.get<Float64>()),
+                std::make_shared<ASTLiteral>(static_cast<UInt64>(fuzz_rand() % 38)));
+        else
+            child = makeASTFunction(
+                "toDecimal256",
+                std::make_shared<ASTLiteral>(l->value.get<Float64>()),
+                std::make_shared<ASTLiteral>(static_cast<UInt64>(fuzz_rand() % 76)));
+    }
 
     if (fuzz_rand() % 7 == 0)
         child = makeASTFunction("toNullable", child);
@@ -933,7 +965,19 @@ ASTPtr QueryFuzzer::reverseLiteralFuzzing(ASTPtr child)
 {
     if (auto * function = child.get()->as<ASTFunction>())
     {
-        std::unordered_set<String> can_be_reverted{"toNullable", "toLowCardinality", "materialize"};
+        const std::unordered_set<String> can_be_reverted{
+            "materialize",
+            "toDecimal32", /// Keeping the first parameter only should be ok (valid query most of the time)
+            "toDecimal64",
+            "toDecimal128",
+            "toDecimal256",
+            "toFixedString", /// Same as toDecimal
+            "toInt128",
+            "toInt256",
+            "toLowCardinality"
+            "toNullable",
+            "toUInt128",
+            "toUInt256"};
         if (can_be_reverted.contains(function->name) && function->children.size() == 1)
         {
             if (fuzz_rand() % 7 == 0)

From aecc135f5dd76c55b99205af170beafc06b9ee62 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Wed, 13 Mar 2024 17:03:18 +0100
Subject: [PATCH 299/443] Add more details about how NULLs are processed in
 aggregations

---
 docs/en/sql-reference/aggregate-functions/index.md | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/docs/en/sql-reference/aggregate-functions/index.md b/docs/en/sql-reference/aggregate-functions/index.md
index 5d2229fbcce..e97db436271 100644
--- a/docs/en/sql-reference/aggregate-functions/index.md
+++ b/docs/en/sql-reference/aggregate-functions/index.md
@@ -18,6 +18,10 @@ ClickHouse also supports:
 
 During aggregation, all `NULL`s are skipped. If the aggregation has several parameters it will ignore any row in which one or more of the parameters are NULL.
 
+There are a few exceptions to this rule:
+  - Both [`first_value`](../../sql-reference/aggregate-functions/reference/first_value.md) and [`last_value`](../../sql-reference/aggregate-functions/reference/last_value.md) support modifiers that respect NULLs (`first_value(b) ignore nulls`).
+  - [`count`](../../sql-reference/aggregate-functions/reference/count.md) without parameters (`count()`) or with constant ones (`count(1)`) will count NULL rows too. With a column as parameter, it will count only not null values.
+
 **Examples:**
 
 Consider this table:

From 5b15ec6ae19fe1caa3800dfa333b61570cfc92b4 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Wed, 13 Mar 2024 17:10:55 +0100
Subject: [PATCH 300/443] Move test from stateless to integration

---
 .../config.d/storage_conf.xml                 |  4 +
 .../integration/test_filesystem_cache/test.py | 75 +++++++++++++++++++
 ...810_system_sync_filesystem_cache.reference |  3 -
 .../02810_system_sync_filesystem_cache.sh     | 69 -----------------
 4 files changed, 79 insertions(+), 72 deletions(-)
 delete mode 100644 tests/queries/0_stateless/02810_system_sync_filesystem_cache.reference
 delete mode 100755 tests/queries/0_stateless/02810_system_sync_filesystem_cache.sh

diff --git a/tests/integration/test_filesystem_cache/config.d/storage_conf.xml b/tests/integration/test_filesystem_cache/config.d/storage_conf.xml
index b614815b34f..a8e4f9f8a99 100644
--- a/tests/integration/test_filesystem_cache/config.d/storage_conf.xml
+++ b/tests/integration/test_filesystem_cache/config.d/storage_conf.xml
@@ -7,4 +7,8 @@
             </hdd_blob>
         </disks>
     </storage_configuration>
+    <filesystem_cache_log>
+            <database>system</database>
+            <table>filesystem_cache_log</table>
+    </filesystem_cache_log>
 </clickhouse>
diff --git a/tests/integration/test_filesystem_cache/test.py b/tests/integration/test_filesystem_cache/test.py
index c44d817c57c..dfab462732a 100644
--- a/tests/integration/test_filesystem_cache/test.py
+++ b/tests/integration/test_filesystem_cache/test.py
@@ -426,3 +426,78 @@ def test_force_filesystem_cache_on_merges(cluster):
     test(node, True)
     node = cluster.instances["node"]
     test(node, False)
+
+
+def test_system_sync_filesystem_cache(cluster):
+    node = cluster.instances["node"]
+    node.query(
+        """
+DROP TABLE IF EXISTS test;
+
+CREATE TABLE test (a Int32, b String)
+ENGINE = MergeTree() ORDER BY tuple()
+SETTINGS disk = disk(type = cache,
+            max_size = '100Ki',
+            path = "test_system_sync_filesystem_cache",
+            delayed_cleanup_interval_ms = 10000000, disk = hdd_blob),
+        min_bytes_for_wide_part = 10485760;
+
+INSERT INTO test SELECT 1, 'test';
+    """
+    )
+
+    query_id = "system_sync_filesystem_cache_1"
+    node.query(
+        "SELECT * FROM test FORMAT Null SETTINGS enable_filesystem_cache_log = 1",
+        query_id=query_id,
+    )
+
+    key, offset = (
+        node.query(
+            f"""
+    SYSTEM FLUSH LOGS;
+    SELECT key, offset FROM system.filesystem_cache_log WHERE query_id = '{query_id}' ORDER BY size DESC LIMIT 1;
+    """
+        )
+        .strip()
+        .split("\t")
+    )
+
+    cache_path = node.query(
+        f"SELECT cache_path FROM system.filesystem_cache WHERE key = '{key}' and file_segment_range_begin = {offset}"
+    )
+
+    node.exec_in_container(["bash", "-c", f"rm {cache_path}"])
+
+    assert key in node.query("SYSTEM SYNC FILESYSTEM CACHE")
+
+    node.query("SELECT * FROM test FORMAT Null")
+    assert key not in node.query("SYSTEM SYNC FILESYSTEM CACHE")
+
+    query_id = "system_sync_filesystem_cache_2"
+    node.query(
+        "SELECT * FROM test FORMAT Null SETTINGS enable_filesystem_cache_log = 1",
+        query_id=query_id,
+    )
+
+    key, offset = (
+        node.query(
+            f"""
+    SYSTEM FLUSH LOGS;
+    SELECT key, offset FROM system.filesystem_cache_log WHERE query_id = '{query_id}' ORDER BY size DESC LIMIT 1;
+    """
+        )
+        .strip()
+        .split("\t")
+    )
+    cache_path = node.query(
+        f"SELECT cache_path FROM system.filesystem_cache WHERE key = '{key}' and file_segment_range_begin = {offset}"
+    )
+
+    node.exec_in_container(["bash", "-c", f"echo -n 'fff' > {cache_path}"])
+
+    assert key in node.query("SYSTEM SYNC FILESYSTEM CACHE")
+
+    node.query("SELECT * FROM test FORMAT Null")
+
+    assert key not in node.query("SYSTEM SYNC FILESYSTEM CACHE")
diff --git a/tests/queries/0_stateless/02810_system_sync_filesystem_cache.reference b/tests/queries/0_stateless/02810_system_sync_filesystem_cache.reference
deleted file mode 100644
index 7614df8ec46..00000000000
--- a/tests/queries/0_stateless/02810_system_sync_filesystem_cache.reference
+++ /dev/null
@@ -1,3 +0,0 @@
-ok
-ok
-ok
diff --git a/tests/queries/0_stateless/02810_system_sync_filesystem_cache.sh b/tests/queries/0_stateless/02810_system_sync_filesystem_cache.sh
deleted file mode 100755
index c88ba4d5a74..00000000000
--- a/tests/queries/0_stateless/02810_system_sync_filesystem_cache.sh
+++ /dev/null
@@ -1,69 +0,0 @@
-#!/usr/bin/env bash
-# Tags: no-fasttest, no-parallel, no-s3-storage, no-random-settings
-
-# set -x
-
-CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
-# shellcheck source=../shell_config.sh
-. "$CUR_DIR"/../shell_config.sh
-
-
-$CLICKHOUSE_CLIENT -nm --query """
-DROP TABLE IF EXISTS test;
-
-CREATE TABLE test (a Int32, b String)
-ENGINE = MergeTree() ORDER BY tuple()
-SETTINGS disk = disk(type = cache, max_size = '100Ki', path = ${CLICKHOUSE_TEST_UNIQUE_NAME}, delayed_cleanup_interval_ms = 10000000, disk = s3_disk), min_bytes_for_wide_part = 10485760;
-
-INSERT INTO test SELECT 1, 'test';
-"""
-
-query_id=$RANDOM
-
-$CLICKHOUSE_CLIENT --query_id "$query_id" --query "SELECT * FROM test FORMAT Null SETTINGS enable_filesystem_cache_log = 1"
-
-${CLICKHOUSE_CLIENT} -q "system flush logs"
-
-key=$($CLICKHOUSE_CLIENT -nm --query """
-SELECT key FROM system.filesystem_cache_log WHERE query_id = '$query_id' ORDER BY size DESC LIMIT 1;
-""")
-
-offset=$($CLICKHOUSE_CLIENT -nm --query """
-SELECT offset FROM system.filesystem_cache_log WHERE query_id = '$query_id' ORDER BY size DESC LIMIT 1;
-""")
-
-path=$($CLICKHOUSE_CLIENT -nm --query """
-SELECT cache_path FROM system.filesystem_cache WHERE key = '$key' AND file_segment_range_begin = $offset;
-""")
-
-rm $path
-
-$CLICKHOUSE_CLIENT  --query "SELECT * FROM test FORMAT Null SETTINGS enable_filesystem_cache_log = 1" 2>&1 | grep -F -e "No such file or directory" > /dev/null && echo "ok" || echo "fail"
-
-CLICKHOUSE_CLIENT=$(echo ${CLICKHOUSE_CLIENT} | sed 's/'"--send_logs_level=${CLICKHOUSE_CLIENT_SERVER_LOGS_LEVEL}"'/--send_logs_level=fatal/g')
-
-$CLICKHOUSE_CLIENT --query "SYSTEM SYNC FILESYSTEM CACHE" 2>&1 | grep -q "$key" && echo 'ok' || echo 'fail'
-
-$CLICKHOUSE_CLIENT --query "SELECT * FROM test FORMAT Null"
-
-key=$($CLICKHOUSE_CLIENT -nm --query """
-SELECT key FROM system.filesystem_cache_log WHERE query_id = '$query_id' ORDER BY size DESC LIMIT 1;
-""")
-
-offset=$($CLICKHOUSE_CLIENT -nm --query """
-SELECT offset FROM system.filesystem_cache_log WHERE query_id = '$query_id' ORDER BY size DESC LIMIT 1;
-""")
-
-path=$($CLICKHOUSE_CLIENT -nm --query """
-SELECT cache_path FROM system.filesystem_cache WHERE key = '$key' AND file_segment_range_begin = $offset;
-""")
-
-echo -n 'fff' > $path
-
-#cat $path
-
-$CLICKHOUSE_CLIENT --query "SYSTEM SYNC FILESYSTEM CACHE" 2>&1 | grep -q "$key" && echo 'ok' || echo 'fail'
-
-$CLICKHOUSE_CLIENT --query "SELECT * FROM test FORMAT Null"
-
-$CLICKHOUSE_CLIENT --query "SYSTEM SYNC FILESYSTEM CACHE"

From b2e067b3daee284c4a97289dc1b4dac1f920c3e6 Mon Sep 17 00:00:00 2001
From: vdimir <vdimir@clickhouse.com>
Date: Wed, 13 Mar 2024 16:41:35 +0000
Subject: [PATCH 301/443] Fix logical error in RabbitMQ storage with
 MATERIALIZED columns

---
 .../table-engines/integrations/rabbitmq.md    |  8 ++-
 src/Interpreters/InterpreterInsertQuery.cpp   | 23 +++++--
 src/Interpreters/InterpreterInsertQuery.h     |  2 +-
 src/Storages/RabbitMQ/RabbitMQSource.cpp      | 14 +++-
 src/Storages/RabbitMQ/StorageRabbitMQ.cpp     | 35 ++++++----
 .../integration/test_storage_rabbitmq/test.py | 64 ++++++++++++++-----
 .../test_rabbitmq_json.reference              | 50 ---------------
 7 files changed, 106 insertions(+), 90 deletions(-)
 delete mode 100644 tests/integration/test_storage_rabbitmq/test_rabbitmq_json.reference

diff --git a/docs/en/engines/table-engines/integrations/rabbitmq.md b/docs/en/engines/table-engines/integrations/rabbitmq.md
index 0f3fef3d6fb..a4d0cf78066 100644
--- a/docs/en/engines/table-engines/integrations/rabbitmq.md
+++ b/docs/en/engines/table-engines/integrations/rabbitmq.md
@@ -18,8 +18,8 @@ This engine allows integrating ClickHouse with [RabbitMQ](https://www.rabbitmq.c
 ``` sql
 CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster]
 (
-    name1 [type1] [DEFAULT|MATERIALIZED|ALIAS expr1],
-    name2 [type2] [DEFAULT|MATERIALIZED|ALIAS expr2],
+    name1 [type1],
+    name2 [type2],
     ...
 ) ENGINE = RabbitMQ SETTINGS
     rabbitmq_host_port = 'host:port' [or rabbitmq_address = 'amqp(s)://guest:guest@localhost/vhost'],
@@ -198,6 +198,10 @@ Additional virtual columns when `kafka_handle_error_mode='stream'`:
 
 Note: `_raw_message` and `_error` virtual columns are filled only in case of exception during parsing, they are always `NULL` when message was parsed successfully.
 
+## Caveats {#caveats}
+
+Even though you may specify [default column expressions](/docs/en/sql-reference/statements/create/table.md/#default_values) (such as `DEFAULT`, `MATERIALIZED`, `ALIAS`) in the table definition, these will be ignored. Instead, the columns will be filled with their respective default values for their types.
+
 ## Data formats support {#data-formats-support}
 
 RabbitMQ engine supports all [formats](../../../interfaces/formats.md) supported in ClickHouse.
diff --git a/src/Interpreters/InterpreterInsertQuery.cpp b/src/Interpreters/InterpreterInsertQuery.cpp
index 70f9e0c51da..3e8bb268fe7 100644
--- a/src/Interpreters/InterpreterInsertQuery.cpp
+++ b/src/Interpreters/InterpreterInsertQuery.cpp
@@ -151,7 +151,7 @@ Block InterpreterInsertQuery::getSampleBlock(
         names.emplace_back(std::move(current_name));
     }
 
-    return getSampleBlock(names, table, metadata_snapshot, allow_materialized);
+    return getSampleBlockImpl(names, table, metadata_snapshot, no_destination, allow_materialized);
 }
 
 std::optional<Names> InterpreterInsertQuery::getInsertColumnNames() const
@@ -173,13 +173,18 @@ std::optional<Names> InterpreterInsertQuery::getInsertColumnNames() const
     return names;
 }
 
-Block InterpreterInsertQuery::getSampleBlock(
+Block InterpreterInsertQuery::getSampleBlockImpl(
     const Names & names,
     const StoragePtr & table,
     const StorageMetadataPtr & metadata_snapshot,
+    bool no_destination,
     bool allow_materialized)
 {
     Block table_sample_physical = metadata_snapshot->getSampleBlock();
+    Block table_sample_virtuals;
+    if (no_destination)
+        table_sample_virtuals = table->getVirtualsHeader();
+
     Block table_sample_insertable = metadata_snapshot->getSampleBlockInsertable();
     Block res;
     for (const auto & current_name : names)
@@ -194,13 +199,19 @@ Block InterpreterInsertQuery::getSampleBlock(
             if (table_sample_physical.has(current_name))
             {
                 if (!allow_materialized)
-                    throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Cannot insert column {}, because it is MATERIALIZED column.",
-                        current_name);
+                    throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Cannot insert column {}, because it is MATERIALIZED column", current_name);
                 res.insert(ColumnWithTypeAndName(table_sample_physical.getByName(current_name).type, current_name));
             }
-            else /// The table does not have a column with that name
+            else if (table_sample_virtuals.has(current_name))
+            {
+                res.insert(ColumnWithTypeAndName(table_sample_virtuals.getByName(current_name).type, current_name));
+            }
+            else
+            {
+                /// The table does not have a column with that name
                 throw Exception(ErrorCodes::NO_SUCH_COLUMN_IN_TABLE, "No such column {} in table {}",
                     current_name, table->getStorageID().getNameForLogs());
+            }
         }
         else
             res.insert(ColumnWithTypeAndName(table_sample_insertable.getByName(current_name).type, current_name));
@@ -276,7 +287,7 @@ Chain InterpreterInsertQuery::buildChain(
     if (!running_group)
         running_group = std::make_shared<ThreadGroup>(getContext());
 
-    auto sample = getSampleBlock(columns, table, metadata_snapshot, allow_materialized);
+    auto sample = getSampleBlockImpl(columns, table, metadata_snapshot, no_destination, allow_materialized);
     if (check_access)
         getContext()->checkAccess(AccessType::INSERT, table->getStorageID(), sample.getNames());
 
diff --git a/src/Interpreters/InterpreterInsertQuery.h b/src/Interpreters/InterpreterInsertQuery.h
index 3647126afb9..bf73fb2a319 100644
--- a/src/Interpreters/InterpreterInsertQuery.h
+++ b/src/Interpreters/InterpreterInsertQuery.h
@@ -69,7 +69,7 @@ public:
     bool shouldAddSquashingFroStorage(const StoragePtr & table) const;
 
 private:
-    static Block getSampleBlock(const Names & names, const StoragePtr & table, const StorageMetadataPtr & metadata_snapshot, bool allow_materialized);
+    static Block getSampleBlockImpl(const Names & names, const StoragePtr & table, const StorageMetadataPtr & metadata_snapshot, bool no_destination, bool allow_materialized);
 
     ASTPtr query_ptr;
     const bool allow_materialized;
diff --git a/src/Storages/RabbitMQ/RabbitMQSource.cpp b/src/Storages/RabbitMQ/RabbitMQSource.cpp
index 4dc257074f3..09c1bf1b2e7 100644
--- a/src/Storages/RabbitMQ/RabbitMQSource.cpp
+++ b/src/Storages/RabbitMQ/RabbitMQSource.cpp
@@ -11,11 +11,21 @@
 namespace DB
 {
 
-static std::pair<Block, Block> getHeaders(const StorageSnapshotPtr & storage_snapshot)
+static std::pair<Block, Block> getHeaders(const StorageSnapshotPtr & storage_snapshot, const Names & column_names)
 {
+    auto all_columns_header = storage_snapshot->metadata->getSampleBlock();
+
     auto non_virtual_header = storage_snapshot->metadata->getSampleBlockNonMaterialized();
     auto virtual_header = storage_snapshot->virtual_columns->getSampleBlock();
 
+    for (const auto & column_name : column_names)
+    {
+        if (non_virtual_header.has(column_name) || virtual_header.has(column_name))
+            continue;
+        const auto & column = all_columns_header.getByName(column_name);
+        non_virtual_header.insert(column);
+    }
+
     return {non_virtual_header, virtual_header};
 }
 
@@ -40,7 +50,7 @@ RabbitMQSource::RabbitMQSource(
     : RabbitMQSource(
         storage_,
         storage_snapshot_,
-        getHeaders(storage_snapshot_),
+        getHeaders(storage_snapshot_, columns),
         context_,
         columns,
         max_block_size_,
diff --git a/src/Storages/RabbitMQ/StorageRabbitMQ.cpp b/src/Storages/RabbitMQ/StorageRabbitMQ.cpp
index 980fccd307e..b882fd2728c 100644
--- a/src/Storages/RabbitMQ/StorageRabbitMQ.cpp
+++ b/src/Storages/RabbitMQ/StorageRabbitMQ.cpp
@@ -8,6 +8,7 @@
 #include <Parsers/ASTCreateQuery.h>
 #include <Parsers/ASTExpressionList.h>
 #include <Parsers/ASTInsertQuery.h>
+#include <Parsers/ASTIdentifier.h>
 #include <Processors/Executors/CompletedPipelineExecutor.h>
 #include <Processors/Executors/PushingPipelineExecutor.h>
 #include <Processors/Transforms/ExpressionTransform.h>
@@ -133,6 +134,9 @@ StorageRabbitMQ::StorageRabbitMQ(
     if (configuration.secure)
         SSL_library_init();
 
+    if (!columns_.getMaterialized().empty() || !columns_.getAliases().empty() || !columns_.getDefaults().empty() || !columns_.getEphemeral().empty())
+        context_->addWarningMessage("RabbitMQ table engine doesn't support ALIAS, DEFAULT or MATERIALIZED columns. They will be ignored and filled with default values");
+
     StorageInMemoryMetadata storage_metadata;
     storage_metadata.setColumns(columns_);
     setInMemoryMetadata(storage_metadata);
@@ -1055,18 +1059,7 @@ bool StorageRabbitMQ::tryStreamToViews()
     if (!table)
         throw Exception(ErrorCodes::LOGICAL_ERROR, "Engine table {} doesn't exist.", table_id.getNameForLogs());
 
-    // Create an INSERT query for streaming data
-    auto insert = std::make_shared<ASTInsertQuery>();
-    insert->table_id = table_id;
-
-    // Only insert into dependent views and expect that input blocks contain virtual columns
-    InterpreterInsertQuery interpreter(insert, rabbitmq_context, false, true, true);
-    auto block_io = interpreter.execute();
-
     auto storage_snapshot = getStorageSnapshot(getInMemoryMetadataPtr(), getContext());
-    auto column_names = block_io.pipeline.getHeader().getNames();
-    auto sample_block = storage_snapshot->getSampleBlockForColumns(column_names);
-
     auto block_size = getMaxBlockSize();
 
     // Create a stream for each consumer and join them in a union stream
@@ -1082,13 +1075,29 @@ bool StorageRabbitMQ::tryStreamToViews()
     for (size_t i = 0; i < num_created_consumers; ++i)
     {
         auto source = std::make_shared<RabbitMQSource>(
-            *this, storage_snapshot, rabbitmq_context, column_names, block_size,
-            max_execution_time_ms, rabbitmq_settings->rabbitmq_handle_error_mode, false);
+            *this, storage_snapshot, rabbitmq_context, Names{}, block_size,
+            max_execution_time_ms, rabbitmq_settings->rabbitmq_handle_error_mode);
 
         sources.emplace_back(source);
         pipes.emplace_back(source);
     }
 
+    // Create an INSERT query for streaming data
+    auto insert = std::make_shared<ASTInsertQuery>();
+    insert->table_id = table_id;
+    if (!sources.empty())
+    {
+        auto column_list = std::make_shared<ASTExpressionList>();
+        const auto & header = sources[0]->getPort().getHeader();
+        for (const auto & column : header)
+            column_list->children.emplace_back(std::make_shared<ASTIdentifier>(column.name));
+        insert->columns = std::move(column_list);
+    }
+
+    // Only insert into dependent views and expect that input blocks contain virtual columns
+    InterpreterInsertQuery interpreter(insert, rabbitmq_context, /* allow_materialized_ */ false, /* no_squash_ */ true, /* no_destination_ */ true);
+    auto block_io = interpreter.execute();
+
     block_io.pipeline.complete(Pipe::unitePipes(std::move(pipes)));
 
     std::atomic_size_t rows = 0;
diff --git a/tests/integration/test_storage_rabbitmq/test.py b/tests/integration/test_storage_rabbitmq/test.py
index 280ce230921..0f1c5eb17dd 100644
--- a/tests/integration/test_storage_rabbitmq/test.py
+++ b/tests/integration/test_storage_rabbitmq/test.py
@@ -53,13 +53,13 @@ instance3 = cluster.add_instance(
 # Helpers
 
 
-def rabbitmq_check_result(result, check=False, ref_file="test_rabbitmq_json.reference"):
-    fpath = p.join(p.dirname(__file__), ref_file)
-    with open(fpath) as reference:
-        if check:
-            assert TSV(result) == TSV(reference)
-        else:
-            return TSV(result) == TSV(reference)
+def rabbitmq_check_result(result, check=False, reference=None):
+    if reference is None:
+        reference = "\n".join([f"{i}\t{i}" for i in range(50)])
+    if check:
+        assert TSV(result) == TSV(reference)
+    else:
+        return TSV(result) == TSV(reference)
 
 
 def wait_rabbitmq_to_start(rabbitmq_docker_id, cookie, timeout=180):
@@ -133,9 +133,10 @@ def test_rabbitmq_select(rabbitmq_cluster, secure):
     if secure:
         port = cluster.rabbitmq_secure_port
 
+    # MATERIALIZED and ALIAS columns are not supported in RabbitMQ engine, but we can test that it does not fail
     instance.query(
         """
-        CREATE TABLE test.rabbitmq (key UInt64, value UInt64)
+        CREATE TABLE test.rabbitmq (key UInt64, value UInt64, value2 ALIAS value + 1, value3 MATERIALIZED value + 1)
             ENGINE = RabbitMQ
             SETTINGS rabbitmq_host_port = '{}:{}',
                      rabbitmq_exchange_name = 'select',
@@ -148,6 +149,11 @@ def test_rabbitmq_select(rabbitmq_cluster, secure):
         )
     )
 
+    assert (
+        "RabbitMQ table engine doesn\\'t support ALIAS, DEFAULT or MATERIALIZED columns"
+        in instance.query("SELECT * FROM system.warnings")
+    )
+
     credentials = pika.PlainCredentials("root", "clickhouse")
     parameters = pika.ConnectionParameters(
         rabbitmq_cluster.rabbitmq_ip, rabbitmq_cluster.rabbitmq_port, "/", credentials
@@ -379,7 +385,7 @@ def test_rabbitmq_macros(rabbitmq_cluster):
 def test_rabbitmq_materialized_view(rabbitmq_cluster):
     instance.query(
         """
-        CREATE TABLE test.rabbitmq (key UInt64, value UInt64)
+        CREATE TABLE test.rabbitmq (key UInt64, value UInt64, dt1 DateTime MATERIALIZED now(), value2 ALIAS value + 1)
             ENGINE = RabbitMQ
             SETTINGS rabbitmq_host_port = 'rabbitmq1:5672',
                      rabbitmq_exchange_name = 'mv',
@@ -484,9 +490,11 @@ def test_rabbitmq_many_materialized_views(rabbitmq_cluster):
         """
         DROP TABLE IF EXISTS test.view1;
         DROP TABLE IF EXISTS test.view2;
+        DROP TABLE IF EXISTS test.view3;
         DROP TABLE IF EXISTS test.consumer1;
         DROP TABLE IF EXISTS test.consumer2;
-        CREATE TABLE test.rabbitmq (key UInt64, value UInt64)
+        DROP TABLE IF EXISTS test.consumer3;
+        CREATE TABLE test.rabbitmq (key UInt64, value UInt64, value2 ALIAS value + 1, value3 MATERIALIZED value + 1, value4 DEFAULT 1)
             ENGINE = RabbitMQ
             SETTINGS rabbitmq_host_port = 'rabbitmq1:5672',
                      rabbitmq_exchange_name = 'mmv',
@@ -497,13 +505,18 @@ def test_rabbitmq_many_materialized_views(rabbitmq_cluster):
         CREATE TABLE test.view1 (key UInt64, value UInt64)
             ENGINE = MergeTree()
             ORDER BY key;
-        CREATE TABLE test.view2 (key UInt64, value UInt64)
+        CREATE TABLE test.view2 (key UInt64, value UInt64, value2 UInt64, value3 UInt64, value4 UInt64)
+            ENGINE = MergeTree()
+            ORDER BY key;
+        CREATE TABLE test.view3 (key UInt64)
             ENGINE = MergeTree()
             ORDER BY key;
         CREATE MATERIALIZED VIEW test.consumer1 TO test.view1 AS
             SELECT * FROM test.rabbitmq;
         CREATE MATERIALIZED VIEW test.consumer2 TO test.view2 AS
             SELECT * FROM test.rabbitmq;
+        CREATE MATERIALIZED VIEW test.consumer3 TO test.view3 AS
+            SELECT * FROM test.rabbitmq;
     """
     )
 
@@ -514,7 +527,7 @@ def test_rabbitmq_many_materialized_views(rabbitmq_cluster):
     connection = pika.BlockingConnection(parameters)
     channel = connection.channel()
 
-    instance.wait_for_log_line("Started streaming to 2 attached views")
+    instance.wait_for_log_line("Started streaming to 3 attached views")
 
     messages = []
     for i in range(50):
@@ -522,24 +535,43 @@ def test_rabbitmq_many_materialized_views(rabbitmq_cluster):
     for message in messages:
         channel.basic_publish(exchange="mmv", routing_key="", body=message)
 
-    while True:
+    is_check_passed = False
+    deadline = time.monotonic() + 60
+    while time.monotonic() < deadline:
         result1 = instance.query("SELECT * FROM test.view1 ORDER BY key")
         result2 = instance.query("SELECT * FROM test.view2 ORDER BY key")
-        if rabbitmq_check_result(result1) and rabbitmq_check_result(result2):
+        result3 = instance.query("SELECT * FROM test.view3 ORDER BY key")
+        # Note that for view2 result is `i i 0 0 0`, but not `i i i+1 i+1 1` as expected, ALIAS/MATERIALIZED/DEFAULT columns are not supported in RabbitMQ engine
+        # We onlt check that at least it do not fail
+        if (
+            rabbitmq_check_result(result1)
+            and rabbitmq_check_result(
+                result2, reference="\n".join([f"{i}\t{i}\t0\t0\t0" for i in range(50)])
+            )
+            and rabbitmq_check_result(
+                result3, reference="\n".join([str(i) for i in range(50)])
+            )
+        ):
+            is_check_passed = True
             break
+        time.sleep(0.1)
+
+    assert (
+        is_check_passed
+    ), f"References are not equal to results, result1: {result1}, result2: {result2}, result3: {result3}"
 
     instance.query(
         """
         DROP TABLE test.consumer1;
         DROP TABLE test.consumer2;
+        DROP TABLE test.consumer3;
         DROP TABLE test.view1;
         DROP TABLE test.view2;
+        DROP TABLE test.view3;
     """
     )
 
     connection.close()
-    rabbitmq_check_result(result1, True)
-    rabbitmq_check_result(result2, True)
 
 
 def test_rabbitmq_big_message(rabbitmq_cluster):
diff --git a/tests/integration/test_storage_rabbitmq/test_rabbitmq_json.reference b/tests/integration/test_storage_rabbitmq/test_rabbitmq_json.reference
deleted file mode 100644
index 959bb2aad74..00000000000
--- a/tests/integration/test_storage_rabbitmq/test_rabbitmq_json.reference
+++ /dev/null
@@ -1,50 +0,0 @@
-0	0
-1	1
-2	2
-3	3
-4	4
-5	5
-6	6
-7	7
-8	8
-9	9
-10	10
-11	11
-12	12
-13	13
-14	14
-15	15
-16	16
-17	17
-18	18
-19	19
-20	20
-21	21
-22	22
-23	23
-24	24
-25	25
-26	26
-27	27
-28	28
-29	29
-30	30
-31	31
-32	32
-33	33
-34	34
-35	35
-36	36
-37	37
-38	38
-39	39
-40	40
-41	41
-42	42
-43	43
-44	44
-45	45
-46	46
-47	47
-48	48
-49	49

From 4e12b5dc9197db89699a1a420471d31a4aeb03bc Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Wed, 13 Mar 2024 16:54:20 +0000
Subject: [PATCH 302/443] Some fixups

---
 base/base/getMemoryAmount.cpp             |   3 -
 base/base/getMemoryAmount.h               |   9 +-
 programs/keeper/Keeper.cpp                |  19 ++--
 programs/server/Server.cpp                |   6 +-
 src/Common/CgroupsMemoryUsageObserver.cpp | 100 ++++++++++------------
 src/Common/CgroupsMemoryUsageObserver.h   |  60 ++++++-------
 6 files changed, 93 insertions(+), 104 deletions(-)

diff --git a/base/base/getMemoryAmount.cpp b/base/base/getMemoryAmount.cpp
index 0311238caed..3d01e301f45 100644
--- a/base/base/getMemoryAmount.cpp
+++ b/base/base/getMemoryAmount.cpp
@@ -50,9 +50,6 @@ std::optional<uint64_t> getCgroupsV2MemoryLimit()
 
 }
 
-/** Returns the size of physical memory (RAM) in bytes.
-  * Returns 0 on unsupported platform
-  */
 uint64_t getMemoryAmountOrZero()
 {
     int64_t num_pages = sysconf(_SC_PHYS_PAGES);
diff --git a/base/base/getMemoryAmount.h b/base/base/getMemoryAmount.h
index 7ebd92a8bcf..37ee0ebe7c6 100644
--- a/base/base/getMemoryAmount.h
+++ b/base/base/getMemoryAmount.h
@@ -2,11 +2,10 @@
 
 #include <cstdint>
 
-/** Returns the size of physical memory (RAM) in bytes.
-  * Returns 0 on unsupported platform or if it cannot determine the size of physical memory.
-  */
+/// Returns the size in bytes of physical memory (RAM) available to the process. The value can
+/// be smaller than the total available RAM available to the system due to cgroups settings.
+/// Returns 0 on unsupported platform or if it cannot determine the size of physical memory.
 uint64_t getMemoryAmountOrZero();
 
-/** Throws exception if it cannot determine the size of physical memory.
-  */
+/// Throws exception if it cannot determine the size of physical memory.
 uint64_t getMemoryAmount();
diff --git a/programs/keeper/Keeper.cpp b/programs/keeper/Keeper.cpp
index 5ec1d2a6568..671dd50215b 100644
--- a/programs/keeper/Keeper.cpp
+++ b/programs/keeper/Keeper.cpp
@@ -625,23 +625,22 @@ try
     main_config_reloader->start();
 
     std::optional<CgroupsMemoryUsageObserver> observer;
-    auto cgroups_memory_observer_wait_time = config().getUInt64("keeper_server.cgroups_memory_observer_wait_time", 15);
-    if (cgroups_memory_observer_wait_time != 0)
+    try
     {
-        try
+        auto wait_time = config().getUInt64("keeper_server.cgroups_memory_observer_wait_time", 15);
+        if (wait_time != 0)
         {
-            observer.emplace(std::chrono::seconds(cgroups_memory_observer_wait_time));
-            observer->setOnMemoryLimitUpdate([&]()
+            observer.emplace(std::chrono::seconds(wait_time));
+            observer->setOnMemoryAmountAvailableChanged([&]()
             {
                 main_config_reloader->reload();
             });
             observer->startThread();
         }
-        catch (Exception &)
-        {
-            tryLogCurrentException(log, "Disabling cgroup memory observer because of an error during initialization");
-        }
-
+    }
+    catch (Exception &)
+    {
+        tryLogCurrentException(log, "Disabling cgroup memory observer because of an error during initialization");
     }
 
 
diff --git a/programs/server/Server.cpp b/programs/server/Server.cpp
index 8e672f3f73e..78d95d8340f 100644
--- a/programs/server/Server.cpp
+++ b/programs/server/Server.cpp
@@ -1293,7 +1293,7 @@ try
     std::optional<CgroupsMemoryUsageObserver> cgroups_memory_usage_observer;
     try
     {
-        UInt64 wait_time = server_settings.cgroups_memory_usage_observer_wait_time;
+        auto wait_time = server_settings.cgroups_memory_usage_observer_wait_time;
         if (wait_time != 0)
             cgroups_memory_usage_observer.emplace(std::chrono::seconds(wait_time));
     }
@@ -1359,7 +1359,7 @@ try
             {
                 double hard_limit_ratio = new_server_settings.cgroup_memory_watcher_hard_limit_ratio;
                 double soft_limit_ratio = new_server_settings.cgroup_memory_watcher_soft_limit_ratio;
-                cgroups_memory_usage_observer->setLimits(
+                cgroups_memory_usage_observer->setMemoryUsageLimits(
                     static_cast<uint64_t>(max_server_memory_usage * hard_limit_ratio),
                     static_cast<uint64_t>(max_server_memory_usage * soft_limit_ratio));
             }
@@ -1702,7 +1702,7 @@ try
 
     if (cgroups_memory_usage_observer)
     {
-        cgroups_memory_usage_observer->setOnMemoryLimitUpdate([&]()
+        cgroups_memory_usage_observer->setOnMemoryAmountAvailableChanged([&]()
         {
             main_config_reloader->reload();
         });
diff --git a/src/Common/CgroupsMemoryUsageObserver.cpp b/src/Common/CgroupsMemoryUsageObserver.cpp
index c8379aed248..e55fd44a524 100644
--- a/src/Common/CgroupsMemoryUsageObserver.cpp
+++ b/src/Common/CgroupsMemoryUsageObserver.cpp
@@ -37,7 +37,7 @@ namespace ErrorCodes
 CgroupsMemoryUsageObserver::CgroupsMemoryUsageObserver(std::chrono::seconds wait_time_)
     : log(getLogger("CgroupsMemoryUsageObserver"))
     , wait_time(wait_time_)
-    , file(log)
+    , memory_usage_file(log)
 {
     LOG_INFO(log, "Initialized cgroups memory limit observer, wait time is {} sec", wait_time.count());
 }
@@ -47,9 +47,10 @@ CgroupsMemoryUsageObserver::~CgroupsMemoryUsageObserver()
     stopThread();
 }
 
-void CgroupsMemoryUsageObserver::setLimits(uint64_t hard_limit_, uint64_t soft_limit_)
+void CgroupsMemoryUsageObserver::setMemoryUsageLimits(uint64_t hard_limit_, uint64_t soft_limit_)
 {
-    std::lock_guard<std::mutex> lock(limit_mutex);
+    std::lock_guard<std::mutex> limit_lock(limit_mutex);
+
     if (hard_limit_ == hard_limit && soft_limit_ == soft_limit)
         return;
 
@@ -83,10 +84,10 @@ void CgroupsMemoryUsageObserver::setLimits(uint64_t hard_limit_, uint64_t soft_l
             mallctl("arena." STRINGIFY(MALLCTL_ARENAS_ALL) ".purge", nullptr, nullptr, nullptr, 0);
 #endif
             /// Reset current usage in memory tracker. Expect zero for free_memory_in_allocator_arenas as we just purged them.
-            uint64_t current_usage = readMemoryUsage();
-            MemoryTracker::setRSS(current_usage, 0);
+            uint64_t memory_usage = readMemoryUsage();
+            MemoryTracker::setRSS(memory_usage, 0);
 
-            LOG_INFO(log, "Purged jemalloc arenas. Current memory usage is {}", ReadableSize(current_usage));
+            LOG_INFO(log, "Purged jemalloc arenas. Current memory usage is {}", ReadableSize(memory_usage));
         }
         else
         {
@@ -97,15 +98,15 @@ void CgroupsMemoryUsageObserver::setLimits(uint64_t hard_limit_, uint64_t soft_l
     LOG_INFO(log, "Set new limits, soft limit: {}, hard limit: {}", ReadableSize(soft_limit_), ReadableSize(hard_limit_));
 }
 
-void CgroupsMemoryUsageObserver::setOnMemoryLimitUpdate(UpdateMemLimitCallbackFn on_memory_limit_update_)
+void CgroupsMemoryUsageObserver::setOnMemoryAmountAvailableChanged(OnMemoryAmountAvailableChangedFn on_memory_amount_available_changed_)
 {
-    std::lock_guard<std::mutex> set_limit_lock(limit_mutex);
-    on_memory_limit_update = on_memory_limit_update_;
+    std::lock_guard<std::mutex> limit_lock(limit_mutex);
+    on_memory_amount_available_changed = on_memory_amount_available_changed_;
 }
 
 uint64_t CgroupsMemoryUsageObserver::readMemoryUsage() const
 {
-    return file.readMemoryUsage();
+    return memory_usage_file.readMemoryUsage();
 }
 
 namespace
@@ -167,7 +168,7 @@ std::pair<std::string, CgroupsMemoryUsageObserver::CgroupsVersion> getCgroupsFil
 
 }
 
-CgroupsMemoryUsageObserver::File::File(LoggerPtr log_)
+CgroupsMemoryUsageObserver::MemoryUsageFile::MemoryUsageFile(LoggerPtr log_)
     : log(log_)
 {
     std::tie(file_name, version) = getCgroupsFileName();
@@ -181,7 +182,7 @@ CgroupsMemoryUsageObserver::File::File(LoggerPtr log_)
             file_name, "Cannot open file '{}'", file_name);
 }
 
-CgroupsMemoryUsageObserver::File::~File()
+CgroupsMemoryUsageObserver::MemoryUsageFile::~MemoryUsageFile()
 {
     assert(fd != -1);
     if (::close(fd) != 0)
@@ -199,7 +200,7 @@ CgroupsMemoryUsageObserver::File::~File()
     }
 }
 
-uint64_t CgroupsMemoryUsageObserver::File::readMemoryUsage() const
+uint64_t CgroupsMemoryUsageObserver::MemoryUsageFile::readMemoryUsage() const
 {
     /// File read is probably not read is thread-safe, just to be sure
     std::lock_guard lock(mutex);
@@ -282,8 +283,8 @@ void CgroupsMemoryUsageObserver::runThread()
 {
     setThreadName("CgrpMemUsgObsr");
 
-    last_process_memory_amount = getMemoryAmount();
-    LOG_INFO(log, "Init memory amount is {} bytes", last_process_memory_amount);
+    last_available_memory_amount = getMemoryAmount();
+    LOG_INFO(log, "Memory amount initially available to the process is {}", ReadableSize(last_available_memory_amount));
 
     std::unique_lock lock(thread_mutex);
     while (true)
@@ -293,21 +294,41 @@ void CgroupsMemoryUsageObserver::runThread()
 
         try
         {
-            uint64_t process_memory_limit = getMemoryAmount();
-            if (process_memory_limit != last_process_memory_amount)
+            uint64_t available_memory_amount = getMemoryAmount();
+            if (available_memory_amount != last_available_memory_amount)
             {
-                LOG_INFO(log, "Find memory amount change, old limit is {} bytes, new limit is {} bytes", last_process_memory_amount, process_memory_limit);
-                last_process_memory_amount = process_memory_limit;
-                /// If the available memory for the process changes (typically because the limit was adjusted via cgroups), then we just reload the config.
-                /// Reloading config will check the memory amount again and calculate soft/hard limit again.
-                std::lock_guard<std::mutex> set_limit_lock(limit_mutex);
-                on_memory_limit_update();
+                LOG_INFO(log, "Memory amount available to the process changed from {} to {}", ReadableSize(last_available_memory_amount), ReadableSize(available_memory_amount));
+                last_available_memory_amount = available_memory_amount;
+                std::lock_guard<std::mutex> limit_lock(limit_mutex);
+                on_memory_amount_available_changed();
             }
-            std::lock_guard<std::mutex> set_limit_lock(limit_mutex);
+
+            std::lock_guard<std::mutex> limit_lock(limit_mutex);
             if (soft_limit > 0 && hard_limit > 0)
             {
-                uint64_t memory_usage = file.readMemoryUsage();
-                processMemoryUsage(memory_usage);
+                uint64_t memory_usage = memory_usage_file.readMemoryUsage();
+                if (memory_usage > hard_limit)
+                {
+                    if (last_memory_usage <= hard_limit)
+                        on_hard_limit(true);
+                }
+                else
+                {
+                    if (last_memory_usage > hard_limit)
+                        on_hard_limit(false);
+                }
+
+                if (memory_usage > soft_limit)
+                {
+                    if (last_memory_usage <= soft_limit)
+                        on_soft_limit(true);
+                }
+                else
+                {
+                    if (last_memory_usage > soft_limit)
+                        on_soft_limit(false);
+                }
+                last_memory_usage = memory_usage;
             }
         }
         catch (...)
@@ -317,33 +338,6 @@ void CgroupsMemoryUsageObserver::runThread()
     }
 }
 
-void CgroupsMemoryUsageObserver::processMemoryUsage(uint64_t current_usage)
-{
-    if (current_usage > hard_limit)
-    {
-        if (last_usage <= hard_limit)
-            on_hard_limit(true);
-    }
-    else
-    {
-        if (last_usage > hard_limit)
-            on_hard_limit(false);
-    }
-
-    if (current_usage > soft_limit)
-    {
-        if (last_usage <= soft_limit)
-            on_soft_limit(true);
-    }
-    else
-    {
-        if (last_usage > soft_limit)
-            on_soft_limit(false);
-    }
-
-    last_usage = current_usage;
-}
-
 }
 
 #endif
diff --git a/src/Common/CgroupsMemoryUsageObserver.h b/src/Common/CgroupsMemoryUsageObserver.h
index e1978af8d28..eb755d04187 100644
--- a/src/Common/CgroupsMemoryUsageObserver.h
+++ b/src/Common/CgroupsMemoryUsageObserver.h
@@ -9,55 +9,59 @@
 namespace DB
 {
 
-/// Periodically reads the current memory usage from Linux cgroups.
-/// You can specify soft or hard memory limits:
-/// - When the soft memory limit is hit, drop jemalloc cache.
-/// - When the hard memory limit is hit, update MemoryTracking metric to throw memory exceptions faster.
+/// Does two things:
+/// 1. Periodically reads the memory usage of the process from Linux cgroups.
+///    You can specify soft or hard memory limits:
+///    - When the soft memory limit is hit, drop jemalloc cache.
+///    - When the hard memory limit is hit, update MemoryTracking metric to throw memory exceptions faster.
+/// 2. Periodically reads the the maximum memory available to the process (which can change due to cgroups settings).
+///    You can specify a callback to react on changes. The callback typically performs circular logic: It reloads the
+///    configuration (e.g. server configuration file), which will check the memory amount again and re-calculate
+///    soft/hard limits (see 1.).
 #if defined(OS_LINUX)
 class CgroupsMemoryUsageObserver
 {
 public:
+    using OnMemoryLimitFn = std::function<void(bool)>;
+    using OnMemoryAmountAvailableChangedFn = std::function<void()>;
+
     enum class CgroupsVersion
     {
         V1,
         V2
-
     };
+
     explicit CgroupsMemoryUsageObserver(std::chrono::seconds wait_time_);
     ~CgroupsMemoryUsageObserver();
 
-    void setLimits(uint64_t hard_limit_, uint64_t soft_limit_);
-    using UpdateMemLimitCallbackFn = std::function<void()>;
-    void setOnMemoryLimitUpdate(UpdateMemLimitCallbackFn on_memory_limit_update_);
-    void startThread();
+    void setMemoryUsageLimits(uint64_t hard_limit_, uint64_t soft_limit_);
+    void setOnMemoryAmountAvailableChanged(OnMemoryAmountAvailableChangedFn on_memory_amount_available_changed);
 
-    size_t getHardLimit() const { return hard_limit; }
-    size_t getSoftLimit() const { return soft_limit; }
+    void startThread();
 
     uint64_t readMemoryUsage() const;
 
 private:
     LoggerPtr log;
 
-    std::atomic<size_t> hard_limit = 0;
-    std::atomic<size_t> soft_limit = 0;
-
     const std::chrono::seconds wait_time;
 
     std::mutex limit_mutex;
-    using CallbackFn = std::function<void(bool)>;
-    CallbackFn on_hard_limit;
-    CallbackFn on_soft_limit;
-    UpdateMemLimitCallbackFn on_memory_limit_update TSA_GUARDED_BY(limit_mutex);
+    size_t hard_limit TSA_GUARDED_BY(limit_mutex) = 0;
+    size_t soft_limit TSA_GUARDED_BY(limit_mutex) = 0;
+    OnMemoryLimitFn on_hard_limit TSA_GUARDED_BY(limit_mutex);
+    OnMemoryLimitFn on_soft_limit TSA_GUARDED_BY(limit_mutex);
+    OnMemoryAmountAvailableChangedFn on_memory_amount_available_changed TSA_GUARDED_BY(limit_mutex);
 
-    uint64_t last_usage = 0;
+    uint64_t last_memory_usage = 0;        /// how much memory does the process use
+    uint64_t last_available_memory_amount; /// how much memory can the process use
 
     /// Represents the cgroup virtual file that shows the memory consumption of the process's cgroup.
-    struct File
+    struct MemoryUsageFile
     {
     public:
-        explicit File(LoggerPtr log_);
-        ~File();
+        explicit MemoryUsageFile(LoggerPtr log_);
+        ~MemoryUsageFile();
         uint64_t readMemoryUsage() const;
     private:
         LoggerPtr log;
@@ -67,18 +71,16 @@ private:
         std::string file_name;
     };
 
-    File file;
+    MemoryUsageFile memory_usage_file;
 
     void stopThread();
 
     void runThread();
-    void processMemoryUsage(uint64_t usage);
 
     std::mutex thread_mutex;
     std::condition_variable cond;
     ThreadFromGlobalPool thread;
     bool quit = false;
-    uint64_t last_process_memory_amount;
 };
 
 #else
@@ -87,12 +89,10 @@ class CgroupsMemoryUsageObserver
 public:
     explicit CgroupsMemoryUsageObserver(std::chrono::seconds) {}
 
-    void setLimits(uint64_t, uint64_t) {}
-    size_t readMemoryUsage() { return 0; }
+    void setMemoryUsageLimits(uint64_t, uint64_t) {}
+    void setOnMemoryAmountAvailableChanged(OnMemoryAmountAvailableChangedFn) {}
     void startThread() {}
-    size_t getHardLimit() { return 0; }
-    size_t getSoftLimit() { return 0; }
-    void setOnMemoryLimitUpdate(std::function<void()>) {}
+    size_t readMemoryUsage() { return 0; }
 };
 #endif
 

From 4670f055649f5f8f216acd42947881038dedbdbd Mon Sep 17 00:00:00 2001
From: avogar <avogar@clickhouse.com>
Date: Wed, 13 Mar 2024 17:18:13 +0000
Subject: [PATCH 303/443] Fix test
 test_input_format_parallel_parsing_memory_tracking

---
 .../test.py                                                   | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/integration/test_input_format_parallel_parsing_memory_tracking/test.py b/tests/integration/test_input_format_parallel_parsing_memory_tracking/test.py
index c95bbfda708..a89cb619350 100644
--- a/tests/integration/test_input_format_parallel_parsing_memory_tracking/test.py
+++ b/tests/integration/test_input_format_parallel_parsing_memory_tracking/test.py
@@ -41,7 +41,7 @@ def test_memory_tracking_total():
         [
             "bash",
             "-c",
-            "clickhouse local -q \"SELECT arrayStringConcat(arrayMap(x->toString(cityHash64(x)), range(1000)), ' ') from numbers(10000)\" > data.json",
+            "clickhouse local -q \"SELECT arrayStringConcat(arrayMap(x->toString(cityHash64(x)), range(1000)), ' ') from numbers(10000)\" > data.jsonl",
         ]
     )
 
@@ -56,7 +56,7 @@ def test_memory_tracking_total():
                     "--show-error",
                     "--data-binary",
                     "@data.json",
-                    "http://127.1:8123/?query=INSERT%20INTO%20null%20FORMAT%20TSV",
+                    "http://127.1:8123/?query=INSERT%20INTO%20null%20FORMAT%20JSONEachRow",
                 ]
             )
             == ""

From c73e299820ffef1142c3b6cc9ef89b7dc69e3525 Mon Sep 17 00:00:00 2001
From: Jake Bamrah <login@bamrah.sh>
Date: Thu, 29 Feb 2024 23:42:23 +0100
Subject: [PATCH 304/443] add size cap to memory tables

---
 src/Core/Settings.h                           |  4 ++
 src/Core/SettingsChangesHistory.h             |  4 ++
 src/Storages/StorageMemory.cpp                | 40 ++++++++++++++--
 ...storage_memory_circ_buffer_usage.reference |  8 ++++
 ...03001_storage_memory_circ_buffer_usage.sql | 47 +++++++++++++++++++
 5 files changed, 98 insertions(+), 5 deletions(-)
 create mode 100644 tests/queries/0_stateless/03001_storage_memory_circ_buffer_usage.reference
 create mode 100644 tests/queries/0_stateless/03001_storage_memory_circ_buffer_usage.sql

diff --git a/src/Core/Settings.h b/src/Core/Settings.h
index d70a6cf51c5..500157081a3 100644
--- a/src/Core/Settings.h
+++ b/src/Core/Settings.h
@@ -885,6 +885,10 @@ class IColumn;
     M(Int64, ignore_cold_parts_seconds, 0, "Only available in ClickHouse Cloud. Exclude new data parts from SELECT queries until they're either pre-warmed (see cache_populated_by_fetch) or this many seconds old. Only for Replicated-/SharedMergeTree.", 0) \
     M(Int64, prefer_warmed_unmerged_parts_seconds, 0, "Only available in ClickHouse Cloud. If a merged part is less than this many seconds old and is not pre-warmed (see cache_populated_by_fetch), but all its source parts are available and pre-warmed, SELECT queries will read from those parts instead. Only for ReplicatedMergeTree. Note that this only checks whether CacheWarmer processed the part; if the part was fetched into cache by something else, it'll still be considered cold until CacheWarmer gets to it; if it was warmed, then evicted from cache, it'll still be considered warm.", 0) \
     M(Bool, iceberg_engine_ignore_schema_evolution, false, "Ignore schema evolution in Iceberg table engine and read all data using latest schema saved on table creation. Note that it can lead to incorrect result", 0) \
+    M(UInt64, min_rows_to_keep, 0, "Minimum block size (in rows) to retain in Memory table buffer.", 0) \
+    M(UInt64, max_rows_to_keep, 0, "Maximum block size (in rows) to retain in Memory table buffer.", 0) \
+    M(UInt64, min_bytes_to_keep, 0, "Minimum block size (in bytes) to retain in Memory table buffer.", 0) \
+    M(UInt64, max_bytes_to_keep, 0, "Maximum block size (in bytes) to retain in Memory table buffer.", 0) \
 
 // End of COMMON_SETTINGS
 // Please add settings related to formats into the FORMAT_FACTORY_SETTINGS, move obsolete settings to OBSOLETE_SETTINGS and obsolete format settings to OBSOLETE_FORMAT_SETTINGS.
diff --git a/src/Core/SettingsChangesHistory.h b/src/Core/SettingsChangesHistory.h
index e680c02671a..436c6963fb8 100644
--- a/src/Core/SettingsChangesHistory.h
+++ b/src/Core/SettingsChangesHistory.h
@@ -93,6 +93,10 @@ static std::map<ClickHouseVersion, SettingsChangesHistory::SettingsChanges> sett
               {"input_format_json_use_string_type_for_ambiguous_paths_in_named_tuples_inference_from_objects", false, false, "Allow to use String type for ambiguous paths during named tuple inference from JSON objects"},
               {"throw_if_deduplication_in_dependent_materialized_views_enabled_with_async_insert", false, true, "Deduplication is dependent materialized view cannot work together with async inserts."},
               {"parallel_replicas_allow_in_with_subquery", false, true, "If true, subquery for IN will be executed on every follower replica"},
+              {"max_rows_to_keep", 0, 0, "Introducing new feature for memory tables."},
+              {"min_rows_to_keep", 0, 0, "Introducing new feature for memory tables."},
+              {"max_bytes_to_keep", 0, 0, "Introducing new feature for memory tables."},
+              {"min_bytes_to_keep", 0, 0, "Introducing new feature for memory tables."},
               }},
     {"24.2", {{"allow_suspicious_variant_types", true, false, "Don't allow creating Variant type with suspicious variants by default"},
               {"validate_experimental_and_suspicious_types_inside_nested_types", false, true, "Validate usage of experimental and suspicious types inside nested types"},
diff --git a/src/Storages/StorageMemory.cpp b/src/Storages/StorageMemory.cpp
index 7a8fb9feeda..518b9f5b212 100644
--- a/src/Storages/StorageMemory.cpp
+++ b/src/Storages/StorageMemory.cpp
@@ -46,6 +46,7 @@ namespace ErrorCodes
     extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH;
     extern const int CANNOT_RESTORE_TABLE;
     extern const int NOT_IMPLEMENTED;
+    extern const int SETTING_CONSTRAINT_VIOLATION;
 }
 
 class MemorySink : public SinkToStorage
@@ -54,10 +55,11 @@ public:
     MemorySink(
         StorageMemory & storage_,
         const StorageMetadataPtr & metadata_snapshot_,
-        ContextPtr context)
+        ContextPtr context_)
         : SinkToStorage(metadata_snapshot_->getSampleBlock())
         , storage(storage_)
-        , storage_snapshot(storage_.getStorageSnapshot(metadata_snapshot_, context))
+        , storage_snapshot(storage_.getStorageSnapshot(metadata_snapshot_, context_))
+        , context(context_)
     {
     }
 
@@ -100,21 +102,49 @@ public:
             inserted_rows += block.rows();
         }
 
+        Settings settings = context->getSettings();
+        if ((settings.min_bytes_to_keep && settings.min_bytes_to_keep > settings.max_bytes_to_keep)
+            || (settings.min_rows_to_keep && settings.min_rows_to_keep > settings.max_rows_to_keep)) {
+            throw Exception(ErrorCodes::SETTING_CONSTRAINT_VIOLATION, "Min. bytes / rows must be set with a max.");
+        }
+
         std::lock_guard lock(storage.mutex);
 
         auto new_data = std::make_unique<Blocks>(*(storage.data.get()));
+        UInt64 new_total_rows = storage.total_size_rows.load(std::memory_order_relaxed) + inserted_rows;
+        UInt64 new_total_bytes = storage.total_size_bytes.load(std::memory_order_relaxed) + inserted_bytes;
+        while (!new_data->empty()
+               && ((settings.max_bytes_to_keep && new_total_bytes > settings.max_bytes_to_keep)
+                   || (settings.max_rows_to_keep && new_total_rows > settings.max_rows_to_keep)))
+        {
+            Block oldest_block = new_data->front();
+            UInt64 rows_to_remove = oldest_block.rows();
+            UInt64 bytes_to_remove = oldest_block.allocatedBytes();
+            if (new_total_bytes - bytes_to_remove < settings.min_bytes_to_keep
+                || new_total_rows - rows_to_remove < settings.min_rows_to_keep)
+            {
+                break; // stop - removing next block will put us under min_bytes / min_rows threshold
+            }
+
+            // delete old block from current storage table
+            new_total_rows -= rows_to_remove;
+            new_total_bytes -= bytes_to_remove;
+            new_data->erase(new_data->begin());
+        }
+
+        // finally - append new data to modified storage table and commit
         new_data->insert(new_data->end(), new_blocks.begin(), new_blocks.end());
 
         storage.data.set(std::move(new_data));
-        storage.total_size_bytes.fetch_add(inserted_bytes, std::memory_order_relaxed);
-        storage.total_size_rows.fetch_add(inserted_rows, std::memory_order_relaxed);
+        storage.total_size_rows.store(new_total_rows, std::memory_order_relaxed);
+        storage.total_size_bytes.store(new_total_bytes, std::memory_order_relaxed);
     }
 
 private:
     Blocks new_blocks;
-
     StorageMemory & storage;
     StorageSnapshotPtr storage_snapshot;
+    ContextPtr context;
 };
 
 
diff --git a/tests/queries/0_stateless/03001_storage_memory_circ_buffer_usage.reference b/tests/queries/0_stateless/03001_storage_memory_circ_buffer_usage.reference
new file mode 100644
index 00000000000..e9e710a535b
--- /dev/null
+++ b/tests/queries/0_stateless/03001_storage_memory_circ_buffer_usage.reference
@@ -0,0 +1,8 @@
+8192
+9216
+9216
+65536
+50
+1000
+1020
+1100
diff --git a/tests/queries/0_stateless/03001_storage_memory_circ_buffer_usage.sql b/tests/queries/0_stateless/03001_storage_memory_circ_buffer_usage.sql
new file mode 100644
index 00000000000..13ddc428415
--- /dev/null
+++ b/tests/queries/0_stateless/03001_storage_memory_circ_buffer_usage.sql
@@ -0,0 +1,47 @@
+DROP TABLE IF EXISTS memory;
+CREATE TABLE memory (i UInt32) ENGINE = Memory;
+SET min_bytes_to_keep = 4096, max_bytes_to_keep = 16384;
+
+/* TESTING BYTES */
+/* 1. testing oldest block doesn't get deleted because of min-threshold */
+INSERT INTO memory SELECT * FROM numbers(0, 1600);
+SELECT total_bytes FROM system.tables WHERE name = 'memory' and database = currentDatabase();
+
+/* 2. adding block that doesn't get deleted */
+INSERT INTO memory SELECT * FROM numbers(1000, 100);
+SELECT total_bytes FROM system.tables WHERE name = 'memory' and database = currentDatabase();
+
+/* 3. testing oldest block gets deleted - 9216 bytes - 1100 */
+INSERT INTO memory SELECT * FROM numbers(9000, 1000);
+SELECT total_bytes FROM system.tables WHERE name = 'memory' and database = currentDatabase();
+
+/* 4.check large block over-writes all bytes / rows */
+INSERT INTO memory SELECT * FROM numbers(9000, 10000);
+SELECT total_bytes FROM system.tables WHERE name = 'memory' and database = currentDatabase();
+
+
+truncate memory;
+SET min_rows_to_keep = 100, max_rows_to_keep = 1000;
+
+/* TESTING ROWS */
+/* 1. add normal number of rows */
+INSERT INTO memory SELECT * FROM numbers(0, 50);
+SELECT total_rows FROM system.tables WHERE name = 'memory' and database = currentDatabase();
+
+/* 2. table should have 1000 */
+INSERT INTO memory SELECT * FROM numbers(50, 950);
+SELECT total_rows FROM system.tables WHERE name = 'memory' and database = currentDatabase();
+
+/* 3. table should have 1020 - removed first 50 */
+INSERT INTO memory SELECT * FROM numbers(2000, 70);
+SELECT total_rows FROM system.tables WHERE name = 'memory' and database = currentDatabase();
+
+/* 4. check large block over-writes all rows */
+INSERT INTO memory SELECT * FROM numbers(3000, 1100);
+SELECT total_rows FROM system.tables WHERE name = 'memory' and database = currentDatabase();
+
+/* test invalid settings */
+SET min_bytes_to_keep = 4096, max_bytes_to_keep = 0;
+INSERT INTO memory SELECT * FROM numbers(3000, 1100); -- { serverError 452 }
+
+DROP TABLE memory;
\ No newline at end of file

From 000719aa2ab71947b7ad67e92c10e1a8663e9e2c Mon Sep 17 00:00:00 2001
From: Jake Bamrah <45361366+JakeBamrah@users.noreply.github.com>
Date: Thu, 29 Feb 2024 23:54:55 +0100
Subject: [PATCH 305/443] update memory table docs with size cap changes

---
 .../engines/table-engines/special/memory.md   | 53 +++++++++++++++++++
 1 file changed, 53 insertions(+)

diff --git a/docs/en/engines/table-engines/special/memory.md b/docs/en/engines/table-engines/special/memory.md
index 0d552a69804..d6bcce68a0e 100644
--- a/docs/en/engines/table-engines/special/memory.md
+++ b/docs/en/engines/table-engines/special/memory.md
@@ -21,3 +21,56 @@ When restarting a server, data disappears from the table and the table becomes e
 Normally, using this table engine is not justified. However, it can be used for tests, and for tasks where maximum speed is required on a relatively small number of rows (up to approximately 100,000,000).
 
 The Memory engine is used by the system for temporary tables with external query data (see the section “External data for processing a query”), and for implementing `GLOBAL IN` (see the section “IN operators”).
+
+## Engine Parameters
+
+- `min_bytes_to_keep` — Minimum bytes to keep when memory table is size-capped.
+  - Default value: `0`
+  - Requires `max_bytes_to_keep`
+- `max_bytes_to_keep` — Maximum bytes to keep within memory table where oldest rows are deleted on each insertion (i.e circular buffer). Max bytes can exceed the stated limit if the oldest batch of rows to remove falls under the `min_bytes_to_keep` limit when adding a large block.
+  - Default value: `0`
+- `min_rows_to_keep` — Minimum rows to keep when memory table is size-capped.
+  - Default value: `0`
+  - Requires `max_rows_to_keep`
+- `max_rows_to_keep` — Maximum rows to keep within memory table where oldest rows are deleted on each insertion (i.e circular buffer). Max rows can exceed the stated limit if the oldest batch of rows to remove falls under the `min_rows_to_keep` limit when adding a large block.
+  - Default value: `0`
+
+## Usage {#usage}
+
+``` sql
+CREATE TABLE memory (i UInt32) ENGINE = Memory;
+```
+
+## Examples {#examples}
+
+**Example 1: Setting parameters**
+``` sql
+SET min_bytes_to_keep = 4096, max_bytes_to_keep = 16384;
+```
+
+**Note:** Both `bytes` and `rows` capping parameters can be set at the same time, however, the lower bounds of `max` and `min` will be adhered to.
+
+**Example 2: Basic usage**
+``` sql
+CREATE TABLE memory (i UInt32) ENGINE = Memory;
+
+SET min_bytes_to_keep = 4096, max_bytes_to_keep = 16384;
+
+/* 1. testing oldest block doesn't get deleted due to min-threshold - 3000 rows */
+INSERT INTO memory SELECT * FROM numbers(0, 1600);
+
+/* 2. adding block that doesn't get deleted */
+INSERT INTO memory SELECT * FROM numbers(1000, 100);
+
+/* 3. testing oldest block gets deleted - 9216 bytes - 1100 */
+INSERT INTO memory SELECT * FROM numbers(9000, 1000);
+
+/* 4. checking a very large block overrides all */
+INSERT INTO memory SELECT * FROM numbers(9000, 10000);
+```
+
+``` text
+┌─total_bytes─┬─total_rows─┐
+│       65536 │      10000 │
+└─────────────┴────────────┘
+```

From 77223d682d9d4ba7dcb7d3c36d323c43cd224bdd Mon Sep 17 00:00:00 2001
From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com>
Date: Fri, 1 Mar 2024 02:01:40 +0100
Subject: [PATCH 306/443] reload tests

---
 src/Storages/StorageMemory.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Storages/StorageMemory.cpp b/src/Storages/StorageMemory.cpp
index 518b9f5b212..1a7a76ee285 100644
--- a/src/Storages/StorageMemory.cpp
+++ b/src/Storages/StorageMemory.cpp
@@ -132,7 +132,7 @@ public:
             new_data->erase(new_data->begin());
         }
 
-        // finally - append new data to modified storage table and commit
+        // append new data to modified storage table and commit
         new_data->insert(new_data->end(), new_blocks.begin(), new_blocks.end());
 
         storage.data.set(std::move(new_data));

From 4aab1170ef6c167df867a0565968744f9eb79852 Mon Sep 17 00:00:00 2001
From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com>
Date: Thu, 7 Mar 2024 12:34:53 +0100
Subject: [PATCH 307/443] fix style

---
 src/Storages/StorageMemory.cpp | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/src/Storages/StorageMemory.cpp b/src/Storages/StorageMemory.cpp
index 1a7a76ee285..6afe3331054 100644
--- a/src/Storages/StorageMemory.cpp
+++ b/src/Storages/StorageMemory.cpp
@@ -104,9 +104,8 @@ public:
 
         Settings settings = context->getSettings();
         if ((settings.min_bytes_to_keep && settings.min_bytes_to_keep > settings.max_bytes_to_keep)
-            || (settings.min_rows_to_keep && settings.min_rows_to_keep > settings.max_rows_to_keep)) {
+            || (settings.min_rows_to_keep && settings.min_rows_to_keep > settings.max_rows_to_keep))
             throw Exception(ErrorCodes::SETTING_CONSTRAINT_VIOLATION, "Min. bytes / rows must be set with a max.");
-        }
 
         std::lock_guard lock(storage.mutex);
 

From e53d052dccf6c2a1e913224b5d4d3a00a86be921 Mon Sep 17 00:00:00 2001
From: Jake Bamrah <login@bamrah.sh>
Date: Sun, 10 Mar 2024 22:38:05 +0100
Subject: [PATCH 308/443] update docs to include circular buffer behaviour

---
 docs/en/engines/table-engines/special/memory.md | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/docs/en/engines/table-engines/special/memory.md b/docs/en/engines/table-engines/special/memory.md
index d6bcce68a0e..e008ed8619d 100644
--- a/docs/en/engines/table-engines/special/memory.md
+++ b/docs/en/engines/table-engines/special/memory.md
@@ -22,7 +22,10 @@ Normally, using this table engine is not justified. However, it can be used for
 
 The Memory engine is used by the system for temporary tables with external query data (see the section “External data for processing a query”), and for implementing `GLOBAL IN` (see the section “IN operators”).
 
-## Engine Parameters
+Upper and lower bounds can be specified to limit Memory engine table size, effectively allowing it to act as a circular
+buffer (see [Engine Parameters](#engine-parameters)).
+
+## Engine Parameters {#engine-parameters}
 
 - `min_bytes_to_keep` — Minimum bytes to keep when memory table is size-capped.
   - Default value: `0`

From a9e054e1c8c1b6435cb3f152eb6e31d3086b04b4 Mon Sep 17 00:00:00 2001
From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com>
Date: Mon, 11 Mar 2024 13:17:33 +0100
Subject: [PATCH 309/443] Reload CI


From 5e66cd7154e36cc9ffbbfaf123c3aeb81d2a14de Mon Sep 17 00:00:00 2001
From: Jake Bamrah <login@bamrah.sh>
Date: Tue, 12 Mar 2024 23:41:00 +0100
Subject: [PATCH 310/443] fix settings seg-fault and expand test case coverage

---
 .../engines/table-engines/special/memory.md   | 14 ++++-----
 src/Core/Settings.h                           |  4 ---
 src/Core/SettingsChangesHistory.h             |  4 ---
 src/Storages/MemorySettings.h                 |  4 +++
 src/Storages/StorageMemory.cpp                | 31 +++++++++----------
 src/Storages/StorageMemory.h                  |  8 ++++-
 ...torage_memory_circ_buffer_usage.reference} |  4 +++
 ...3009_storage_memory_circ_buffer_usage.sql} | 30 +++++++++++++-----
 8 files changed, 58 insertions(+), 41 deletions(-)
 rename tests/queries/0_stateless/{03001_storage_memory_circ_buffer_usage.reference => 03009_storage_memory_circ_buffer_usage.reference} (63%)
 rename tests/queries/0_stateless/{03001_storage_memory_circ_buffer_usage.sql => 03009_storage_memory_circ_buffer_usage.sql} (59%)

diff --git a/docs/en/engines/table-engines/special/memory.md b/docs/en/engines/table-engines/special/memory.md
index e008ed8619d..8ebdd57ae67 100644
--- a/docs/en/engines/table-engines/special/memory.md
+++ b/docs/en/engines/table-engines/special/memory.md
@@ -44,20 +44,16 @@ buffer (see [Engine Parameters](#engine-parameters)).
 CREATE TABLE memory (i UInt32) ENGINE = Memory;
 ```
 
-## Examples {#examples}
-
-**Example 1: Setting parameters**
+**Initialize settings**
 ``` sql
-SET min_bytes_to_keep = 4096, max_bytes_to_keep = 16384;
+CREATE TABLE memory (i UInt32) ENGINE = Memory SETTINGS min_rows_to_keep = 100, max_rows_to_keep = 1000;
 ```
 
 **Note:** Both `bytes` and `rows` capping parameters can be set at the same time, however, the lower bounds of `max` and `min` will be adhered to.
 
-**Example 2: Basic usage**
+## Examples {#examples}
 ``` sql
-CREATE TABLE memory (i UInt32) ENGINE = Memory;
-
-SET min_bytes_to_keep = 4096, max_bytes_to_keep = 16384;
+CREATE TABLE memory (i UInt32) ENGINE = Memory SETTINGS min_bytes_to_keep = 4096, max_bytes_to_keep = 16384;
 
 /* 1. testing oldest block doesn't get deleted due to min-threshold - 3000 rows */
 INSERT INTO memory SELECT * FROM numbers(0, 1600);
@@ -70,6 +66,8 @@ INSERT INTO memory SELECT * FROM numbers(9000, 1000);
 
 /* 4. checking a very large block overrides all */
 INSERT INTO memory SELECT * FROM numbers(9000, 10000);
+
+SELECT total_bytes, total_rows FROM system.tables WHERE name = 'memory' and database = currentDatabase();
 ```
 
 ``` text
diff --git a/src/Core/Settings.h b/src/Core/Settings.h
index 500157081a3..d70a6cf51c5 100644
--- a/src/Core/Settings.h
+++ b/src/Core/Settings.h
@@ -885,10 +885,6 @@ class IColumn;
     M(Int64, ignore_cold_parts_seconds, 0, "Only available in ClickHouse Cloud. Exclude new data parts from SELECT queries until they're either pre-warmed (see cache_populated_by_fetch) or this many seconds old. Only for Replicated-/SharedMergeTree.", 0) \
     M(Int64, prefer_warmed_unmerged_parts_seconds, 0, "Only available in ClickHouse Cloud. If a merged part is less than this many seconds old and is not pre-warmed (see cache_populated_by_fetch), but all its source parts are available and pre-warmed, SELECT queries will read from those parts instead. Only for ReplicatedMergeTree. Note that this only checks whether CacheWarmer processed the part; if the part was fetched into cache by something else, it'll still be considered cold until CacheWarmer gets to it; if it was warmed, then evicted from cache, it'll still be considered warm.", 0) \
     M(Bool, iceberg_engine_ignore_schema_evolution, false, "Ignore schema evolution in Iceberg table engine and read all data using latest schema saved on table creation. Note that it can lead to incorrect result", 0) \
-    M(UInt64, min_rows_to_keep, 0, "Minimum block size (in rows) to retain in Memory table buffer.", 0) \
-    M(UInt64, max_rows_to_keep, 0, "Maximum block size (in rows) to retain in Memory table buffer.", 0) \
-    M(UInt64, min_bytes_to_keep, 0, "Minimum block size (in bytes) to retain in Memory table buffer.", 0) \
-    M(UInt64, max_bytes_to_keep, 0, "Maximum block size (in bytes) to retain in Memory table buffer.", 0) \
 
 // End of COMMON_SETTINGS
 // Please add settings related to formats into the FORMAT_FACTORY_SETTINGS, move obsolete settings to OBSOLETE_SETTINGS and obsolete format settings to OBSOLETE_FORMAT_SETTINGS.
diff --git a/src/Core/SettingsChangesHistory.h b/src/Core/SettingsChangesHistory.h
index 436c6963fb8..e680c02671a 100644
--- a/src/Core/SettingsChangesHistory.h
+++ b/src/Core/SettingsChangesHistory.h
@@ -93,10 +93,6 @@ static std::map<ClickHouseVersion, SettingsChangesHistory::SettingsChanges> sett
               {"input_format_json_use_string_type_for_ambiguous_paths_in_named_tuples_inference_from_objects", false, false, "Allow to use String type for ambiguous paths during named tuple inference from JSON objects"},
               {"throw_if_deduplication_in_dependent_materialized_views_enabled_with_async_insert", false, true, "Deduplication is dependent materialized view cannot work together with async inserts."},
               {"parallel_replicas_allow_in_with_subquery", false, true, "If true, subquery for IN will be executed on every follower replica"},
-              {"max_rows_to_keep", 0, 0, "Introducing new feature for memory tables."},
-              {"min_rows_to_keep", 0, 0, "Introducing new feature for memory tables."},
-              {"max_bytes_to_keep", 0, 0, "Introducing new feature for memory tables."},
-              {"min_bytes_to_keep", 0, 0, "Introducing new feature for memory tables."},
               }},
     {"24.2", {{"allow_suspicious_variant_types", true, false, "Don't allow creating Variant type with suspicious variants by default"},
               {"validate_experimental_and_suspicious_types_inside_nested_types", false, true, "Validate usage of experimental and suspicious types inside nested types"},
diff --git a/src/Storages/MemorySettings.h b/src/Storages/MemorySettings.h
index 9e1a8db3595..ac6cdf73329 100644
--- a/src/Storages/MemorySettings.h
+++ b/src/Storages/MemorySettings.h
@@ -10,6 +10,10 @@ class ASTStorage;
 
 #define MEMORY_SETTINGS(M, ALIAS) \
     M(Bool, compress, false, "Compress data in memory", 0) \
+    M(UInt64, min_rows_to_keep, 0, "Minimum block size (in rows) to retain in Memory table buffer.", 0) \
+    M(UInt64, max_rows_to_keep, 0, "Maximum block size (in rows) to retain in Memory table buffer.", 0) \
+    M(UInt64, min_bytes_to_keep, 0, "Minimum block size (in bytes) to retain in Memory table buffer.", 0) \
+    M(UInt64, max_bytes_to_keep, 0, "Maximum block size (in bytes) to retain in Memory table buffer.", 0) \
 
 DECLARE_SETTINGS_TRAITS(memorySettingsTraits, MEMORY_SETTINGS)
 
diff --git a/src/Storages/StorageMemory.cpp b/src/Storages/StorageMemory.cpp
index 6afe3331054..c6222d2124e 100644
--- a/src/Storages/StorageMemory.cpp
+++ b/src/Storages/StorageMemory.cpp
@@ -55,11 +55,10 @@ public:
     MemorySink(
         StorageMemory & storage_,
         const StorageMetadataPtr & metadata_snapshot_,
-        ContextPtr context_)
+        ContextPtr context)
         : SinkToStorage(metadata_snapshot_->getSampleBlock())
         , storage(storage_)
-        , storage_snapshot(storage_.getStorageSnapshot(metadata_snapshot_, context_))
-        , context(context_)
+        , storage_snapshot(storage_.getStorageSnapshot(metadata_snapshot_, context))
     {
     }
 
@@ -102,25 +101,20 @@ public:
             inserted_rows += block.rows();
         }
 
-        Settings settings = context->getSettings();
-        if ((settings.min_bytes_to_keep && settings.min_bytes_to_keep > settings.max_bytes_to_keep)
-            || (settings.min_rows_to_keep && settings.min_rows_to_keep > settings.max_rows_to_keep))
-            throw Exception(ErrorCodes::SETTING_CONSTRAINT_VIOLATION, "Min. bytes / rows must be set with a max.");
-
         std::lock_guard lock(storage.mutex);
 
         auto new_data = std::make_unique<Blocks>(*(storage.data.get()));
         UInt64 new_total_rows = storage.total_size_rows.load(std::memory_order_relaxed) + inserted_rows;
         UInt64 new_total_bytes = storage.total_size_bytes.load(std::memory_order_relaxed) + inserted_bytes;
         while (!new_data->empty()
-               && ((settings.max_bytes_to_keep && new_total_bytes > settings.max_bytes_to_keep)
-                   || (settings.max_rows_to_keep && new_total_rows > settings.max_rows_to_keep)))
+               && ((storage.max_bytes_to_keep && new_total_bytes > storage.max_bytes_to_keep)
+                   || (storage.max_rows_to_keep && new_total_rows > storage.max_rows_to_keep)))
         {
             Block oldest_block = new_data->front();
             UInt64 rows_to_remove = oldest_block.rows();
             UInt64 bytes_to_remove = oldest_block.allocatedBytes();
-            if (new_total_bytes - bytes_to_remove < settings.min_bytes_to_keep
-                || new_total_rows - rows_to_remove < settings.min_rows_to_keep)
+            if (new_total_bytes - bytes_to_remove < storage.min_bytes_to_keep
+                || new_total_rows - rows_to_remove < storage.min_rows_to_keep)
             {
                 break; // stop - removing next block will put us under min_bytes / min_rows threshold
             }
@@ -143,7 +137,6 @@ private:
     Blocks new_blocks;
     StorageMemory & storage;
     StorageSnapshotPtr storage_snapshot;
-    ContextPtr context;
 };
 
 
@@ -152,8 +145,10 @@ StorageMemory::StorageMemory(
     ColumnsDescription columns_description_,
     ConstraintsDescription constraints_,
     const String & comment,
-    bool compress_)
-    : IStorage(table_id_), data(std::make_unique<const Blocks>()), compress(compress_)
+    const MemorySettings & settings)
+    : IStorage(table_id_), data(std::make_unique<const Blocks>()), compress(settings.compress),
+    min_rows_to_keep(settings.min_rows_to_keep), max_rows_to_keep(settings.max_rows_to_keep),
+    min_bytes_to_keep(settings.min_bytes_to_keep), max_bytes_to_keep(settings.max_bytes_to_keep)
 {
     StorageInMemoryMetadata storage_metadata;
     storage_metadata.setColumns(std::move(columns_description_));
@@ -571,7 +566,11 @@ void registerStorageMemory(StorageFactory & factory)
         if (has_settings)
             settings.loadFromQuery(*args.storage_def);
 
-        return std::make_shared<StorageMemory>(args.table_id, args.columns, args.constraints, args.comment, settings.compress);
+        if (settings.min_bytes_to_keep > settings.max_bytes_to_keep
+            || settings.min_rows_to_keep > settings.max_rows_to_keep)
+            throw Exception(ErrorCodes::SETTING_CONSTRAINT_VIOLATION, "Min. bytes / rows must be set with a max.");
+
+        return std::make_shared<StorageMemory>(args.table_id, args.columns, args.constraints, args.comment, settings);
     },
     {
         .supports_settings = true,
diff --git a/src/Storages/StorageMemory.h b/src/Storages/StorageMemory.h
index 3293e5e4fe5..13f1c971d82 100644
--- a/src/Storages/StorageMemory.h
+++ b/src/Storages/StorageMemory.h
@@ -7,6 +7,7 @@
 #include <Core/NamesAndTypes.h>
 #include <Interpreters/DatabaseCatalog.h>
 #include <Storages/IStorage.h>
+#include <Storages/MemorySettings.h>
 
 #include <Common/MultiVersion.h>
 
@@ -30,7 +31,7 @@ public:
         ColumnsDescription columns_description_,
         ConstraintsDescription constraints_,
         const String & comment,
-        bool compress_ = false);
+        const MemorySettings & settings = MemorySettings());
 
     String getName() const override { return "Memory"; }
 
@@ -134,6 +135,11 @@ private:
     std::atomic<size_t> total_size_rows = 0;
 
     bool compress;
+    UInt64 min_rows_to_keep;
+    UInt64 max_rows_to_keep;
+    UInt64 min_bytes_to_keep;
+    UInt64 max_bytes_to_keep;
+
 
     friend class ReadFromMemoryStorageStep;
 };
diff --git a/tests/queries/0_stateless/03001_storage_memory_circ_buffer_usage.reference b/tests/queries/0_stateless/03009_storage_memory_circ_buffer_usage.reference
similarity index 63%
rename from tests/queries/0_stateless/03001_storage_memory_circ_buffer_usage.reference
rename to tests/queries/0_stateless/03009_storage_memory_circ_buffer_usage.reference
index e9e710a535b..0bcc6fe4434 100644
--- a/tests/queries/0_stateless/03001_storage_memory_circ_buffer_usage.reference
+++ b/tests/queries/0_stateless/03009_storage_memory_circ_buffer_usage.reference
@@ -6,3 +6,7 @@
 1000
 1020
 1100
+8192
+9216
+17408
+82944
diff --git a/tests/queries/0_stateless/03001_storage_memory_circ_buffer_usage.sql b/tests/queries/0_stateless/03009_storage_memory_circ_buffer_usage.sql
similarity index 59%
rename from tests/queries/0_stateless/03001_storage_memory_circ_buffer_usage.sql
rename to tests/queries/0_stateless/03009_storage_memory_circ_buffer_usage.sql
index 13ddc428415..77aa1d0befc 100644
--- a/tests/queries/0_stateless/03001_storage_memory_circ_buffer_usage.sql
+++ b/tests/queries/0_stateless/03009_storage_memory_circ_buffer_usage.sql
@@ -1,6 +1,5 @@
 DROP TABLE IF EXISTS memory;
-CREATE TABLE memory (i UInt32) ENGINE = Memory;
-SET min_bytes_to_keep = 4096, max_bytes_to_keep = 16384;
+CREATE TABLE memory (i UInt32) ENGINE = Memory SETTINGS min_bytes_to_keep = 4096, max_bytes_to_keep = 16384;
 
 /* TESTING BYTES */
 /* 1. testing oldest block doesn't get deleted because of min-threshold */
@@ -19,9 +18,8 @@ SELECT total_bytes FROM system.tables WHERE name = 'memory' and database = curre
 INSERT INTO memory SELECT * FROM numbers(9000, 10000);
 SELECT total_bytes FROM system.tables WHERE name = 'memory' and database = currentDatabase();
 
-
-truncate memory;
-SET min_rows_to_keep = 100, max_rows_to_keep = 1000;
+DROP TABLE IF EXISTS memory;
+CREATE TABLE memory (i UInt32) ENGINE = Memory SETTINGS min_rows_to_keep = 100, max_rows_to_keep = 1000;
 
 /* TESTING ROWS */
 /* 1. add normal number of rows */
@@ -40,8 +38,24 @@ SELECT total_rows FROM system.tables WHERE name = 'memory' and database = curren
 INSERT INTO memory SELECT * FROM numbers(3000, 1100);
 SELECT total_rows FROM system.tables WHERE name = 'memory' and database = currentDatabase();
 
-/* test invalid settings */
-SET min_bytes_to_keep = 4096, max_bytes_to_keep = 0;
-INSERT INTO memory SELECT * FROM numbers(3000, 1100); -- { serverError 452 }
+/* TESTING NO CIRCULAR-BUFFER */
+DROP TABLE IF EXISTS memory;
+CREATE TABLE memory (i UInt32) ENGINE = Memory;
+
+INSERT INTO memory SELECT * FROM numbers(0, 1600);
+SELECT total_bytes FROM system.tables WHERE name = 'memory' and database = currentDatabase();
+
+INSERT INTO memory SELECT * FROM numbers(1000, 100);
+SELECT total_bytes FROM system.tables WHERE name = 'memory' and database = currentDatabase();
+
+INSERT INTO memory SELECT * FROM numbers(9000, 1000);
+SELECT total_bytes FROM system.tables WHERE name = 'memory' and database = currentDatabase();
+
+INSERT INTO memory SELECT * FROM numbers(9000, 10000);
+SELECT total_bytes FROM system.tables WHERE name = 'memory' and database = currentDatabase();
+
+/* TESTING INVALID SETTINGS */
+CREATE TABLE faulty_memory (i UInt32) ENGINE = Memory SETTINGS min_rows_to_keep = 100;  -- { serverError 452 }
+CREATE TABLE faulty_memory (i UInt32) ENGINE = Memory SETTINGS min_bytes_to_keep = 100; -- { serverError 452 }
 
 DROP TABLE memory;
\ No newline at end of file

From a915dace2752947e8d8d4e256342f59f8a287fff Mon Sep 17 00:00:00 2001
From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com>
Date: Wed, 13 Mar 2024 16:19:44 +0100
Subject: [PATCH 311/443] reload CI


From 208a91a2f7ae97795e03c62c1cce121775ee58ff Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Wed, 13 Mar 2024 19:03:29 +0100
Subject: [PATCH 312/443] Add retries exists check in loadOutdatedDataParts

---
 src/Storages/StorageReplicatedMergeTree.cpp | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/src/Storages/StorageReplicatedMergeTree.cpp b/src/Storages/StorageReplicatedMergeTree.cpp
index 73c7e484bcb..71ac1fc4965 100644
--- a/src/Storages/StorageReplicatedMergeTree.cpp
+++ b/src/Storages/StorageReplicatedMergeTree.cpp
@@ -7680,9 +7680,13 @@ void StorageReplicatedMergeTree::forcefullyRemoveBrokenOutdatedPartFromZooKeeper
     /// It could became Outdated because it was merged/mutated (so we have a covering part) or because it was dropped.
     /// But DROP [PART]ITION waits for all Outdated parts to be loaded, so it's not the case.
 
-    auto zookeeper = getZooKeeper();
+    bool exists = false;
     String part_path = replica_path + "/parts/" + part_name;
-    if (!zookeeper->exists(part_path))
+    ZooKeeperRetriesInfo retries_info{/* max_retries */5, /* initial_backoff_ms */100, /* max_backoff_ms */100};
+    ZooKeeperRetriesControl retries_ctl("outdatedPartExists", log.load(), retries_info, nullptr);
+
+    retries_ctl.retryLoop([&]() { exists = getZooKeeper()->exists(part_path); });
+    if (!exists)
         return;
 
     auto part = getActiveContainingPart(part_name);

From 42a1cc47e6e24c3e854346bf9dec522677bb0d09 Mon Sep 17 00:00:00 2001
From: Nikolai Kochetov <nk@clickhouse.com>
Date: Wed, 13 Mar 2024 18:05:08 +0000
Subject: [PATCH 313/443] Fix 01761_cast_to_enum_nullable with analyzer.

---
 src/DataTypes/DataTypeNullable.cpp | 25 +++++++++++++++++++++++++
 src/DataTypes/DataTypeNullable.h   |  1 +
 tests/analyzer_tech_debt.txt       |  1 -
 3 files changed, 26 insertions(+), 1 deletion(-)

diff --git a/src/DataTypes/DataTypeNullable.cpp b/src/DataTypes/DataTypeNullable.cpp
index 16d5d41e5e5..db252659d41 100644
--- a/src/DataTypes/DataTypeNullable.cpp
+++ b/src/DataTypes/DataTypeNullable.cpp
@@ -4,6 +4,7 @@
 #include <DataTypes/Serializations/SerializationNullable.h>
 #include <DataTypes/DataTypeLowCardinality.h>
 #include <Columns/ColumnNullable.h>
+#include <Columns/ColumnConst.h>
 #include <Core/Field.h>
 #include <Parsers/IAST.h>
 #include <Common/typeid_cast.h>
@@ -56,6 +57,30 @@ bool DataTypeNullable::equals(const IDataType & rhs) const
     return rhs.isNullable() && nested_data_type->equals(*static_cast<const DataTypeNullable &>(rhs).nested_data_type);
 }
 
+ColumnPtr DataTypeNullable::createColumnConst(size_t size, const Field & field) const
+{
+    if (onlyNull())
+    {
+        auto column = createColumn();
+        column->insert(field);
+        return ColumnConst::create(std::move(column), size);
+    }
+
+    auto column = nested_data_type->createColumn();
+    bool is_null = field.isNull();
+
+    if (is_null)
+        nested_data_type->insertDefaultInto(*column);
+    else
+        column->insert(field);
+
+    auto null_mask = ColumnUInt8::create();
+    null_mask->getData().push_back(is_null ? 1 : 0);
+
+    auto res = ColumnNullable::create(std::move(column), std::move(null_mask));
+    return ColumnConst::create(std::move(res), size);
+}
+
 SerializationPtr DataTypeNullable::doGetDefaultSerialization() const
 {
     return std::make_shared<SerializationNullable>(nested_data_type->getDefaultSerialization());
diff --git a/src/DataTypes/DataTypeNullable.h b/src/DataTypes/DataTypeNullable.h
index b102c767993..71abe48c151 100644
--- a/src/DataTypes/DataTypeNullable.h
+++ b/src/DataTypes/DataTypeNullable.h
@@ -41,6 +41,7 @@ public:
     bool onlyNull() const override;
     bool canBeInsideLowCardinality() const override { return nested_data_type->canBeInsideLowCardinality(); }
     bool canBePromoted() const override { return nested_data_type->canBePromoted(); }
+    ColumnPtr createColumnConst(size_t size, const Field & field) const override;
 
     const DataTypePtr & getNestedType() const { return nested_data_type; }
 
diff --git a/tests/analyzer_tech_debt.txt b/tests/analyzer_tech_debt.txt
index dbd216ea7be..b4cf3cf288b 100644
--- a/tests/analyzer_tech_debt.txt
+++ b/tests/analyzer_tech_debt.txt
@@ -7,7 +7,6 @@
 01584_distributed_buffer_cannot_find_column
 01624_soft_constraints
 01747_join_view_filter_dictionary
-01761_cast_to_enum_nullable
 01925_join_materialized_columns
 01952_optimize_distributed_group_by_sharding_key
 02354_annoy

From 6e6a67a2fb37ef22807d22aac1a8958bc618dc4d Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Wed, 13 Mar 2024 19:13:39 +0100
Subject: [PATCH 314/443] Fix unit test

---
 .../Cache/WriteBufferToFileSegment.cpp        | 19 ++++++++++++++-----
 .../Cache/WriteBufferToFileSegment.h          |  2 +-
 2 files changed, 15 insertions(+), 6 deletions(-)

diff --git a/src/Interpreters/Cache/WriteBufferToFileSegment.cpp b/src/Interpreters/Cache/WriteBufferToFileSegment.cpp
index 759135722dc..51914c0a14e 100644
--- a/src/Interpreters/Cache/WriteBufferToFileSegment.cpp
+++ b/src/Interpreters/Cache/WriteBufferToFileSegment.cpp
@@ -19,9 +19,22 @@ namespace ErrorCodes
     extern const int NOT_ENOUGH_SPACE;
 }
 
+namespace
+{
+    size_t getCacheLockWaitTimeout()
+    {
+        auto query_context = CurrentThread::getQueryContext();
+        if (query_context)
+            return query_context->getReadSettings().filesystem_cache_reserve_space_wait_lock_timeout_milliseconds;
+        else
+            return Context::getGlobalContextInstance()->getReadSettings().filesystem_cache_reserve_space_wait_lock_timeout_milliseconds;
+    }
+}
+
 WriteBufferToFileSegment::WriteBufferToFileSegment(FileSegment * file_segment_)
     : WriteBufferFromFileDecorator(std::make_unique<WriteBufferFromFile>(file_segment_->getPath()))
     , file_segment(file_segment_)
+    , reserve_space_lock_wait_timeout_milliseconds(getCacheLockWaitTimeout())
 {
 }
 
@@ -32,12 +45,8 @@ WriteBufferToFileSegment::WriteBufferToFileSegment(FileSegmentsHolderPtr segment
         : throw Exception(ErrorCodes::LOGICAL_ERROR, "WriteBufferToFileSegment can be created only from single segment"))
     , file_segment(&segment_holder_->front())
     , segment_holder(std::move(segment_holder_))
+    , reserve_space_lock_wait_timeout_milliseconds(getCacheLockWaitTimeout())
 {
-    auto query_context = CurrentThread::getQueryContext();
-    if (query_context)
-        reserve_space_lock_wait_timeout_milliseconds = query_context->getReadSettings().filesystem_cache_reserve_space_wait_lock_timeout_milliseconds;
-    else
-        reserve_space_lock_wait_timeout_milliseconds = Context::getGlobalContextInstance()->getReadSettings().filesystem_cache_reserve_space_wait_lock_timeout_milliseconds;
 }
 
 /// If it throws an exception, the file segment will be incomplete, so you should not use it in the future.
diff --git a/src/Interpreters/Cache/WriteBufferToFileSegment.h b/src/Interpreters/Cache/WriteBufferToFileSegment.h
index bff340d79b3..822488ceb48 100644
--- a/src/Interpreters/Cache/WriteBufferToFileSegment.h
+++ b/src/Interpreters/Cache/WriteBufferToFileSegment.h
@@ -29,7 +29,7 @@ private:
     /// Empty if file_segment is not owned by this WriteBufferToFileSegment
     FileSegmentsHolderPtr segment_holder;
 
-    size_t reserve_space_lock_wait_timeout_milliseconds;
+    const size_t reserve_space_lock_wait_timeout_milliseconds;
 };
 
 

From 2cb1fa79d9741100063dd29a8033b0888c863dad Mon Sep 17 00:00:00 2001
From: alesapin <alesapin@gmail.com>
Date: Wed, 13 Mar 2024 20:01:54 +0100
Subject: [PATCH 315/443] Bump CI


From 89de338676a0233dac4563d9f4ba7e4a16b54135 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Wed, 13 Mar 2024 17:33:39 +0100
Subject: [PATCH 316/443] Better

---
 .../aggregate-functions/index.md              | 53 +++++++++++++++++--
 1 file changed, 49 insertions(+), 4 deletions(-)

diff --git a/docs/en/sql-reference/aggregate-functions/index.md b/docs/en/sql-reference/aggregate-functions/index.md
index e97db436271..96bf0c5d93b 100644
--- a/docs/en/sql-reference/aggregate-functions/index.md
+++ b/docs/en/sql-reference/aggregate-functions/index.md
@@ -16,11 +16,9 @@ ClickHouse also supports:
 
 ## NULL Processing
 
-During aggregation, all `NULL`s are skipped. If the aggregation has several parameters it will ignore any row in which one or more of the parameters are NULL.
+During aggregation, all `NULL` arguments are skipped. If the aggregation has several arguments it will ignore any row in which one or more of them are NULL.
 
-There are a few exceptions to this rule:
-  - Both [`first_value`](../../sql-reference/aggregate-functions/reference/first_value.md) and [`last_value`](../../sql-reference/aggregate-functions/reference/last_value.md) support modifiers that respect NULLs (`first_value(b) ignore nulls`).
-  - [`count`](../../sql-reference/aggregate-functions/reference/count.md) without parameters (`count()`) or with constant ones (`count(1)`) will count NULL rows too. With a column as parameter, it will count only not null values.
+There is an exception to this rule, which are the functions [`first_value`](../../sql-reference/aggregate-functions/reference/first_value.md), [`last_value`](../../sql-reference/aggregate-functions/reference/last_value.md) and their aliases when followed by the modifier `RESPECT NULLS`: `FIRST_VALUE(b) RESPECT NULLS`.
 
 **Examples:**
 
@@ -89,3 +87,50 @@ FROM t_null_big;
 │ [2,2,3]       │ [2,NULL,2,3,NULL]                     │
 └───────────────┴───────────────────────────────────────┘
 ```
+
+Note that aggregations are skipped when the columns are used as arguments to an aggregated function.  For example [`count`](../../sql-reference/aggregate-functions/reference/count.md) without parameters (`count()`) or with constant ones (`count(1)`) will count all rows in the block (independently of the value of the GROUP BY column as it's not an argument), while `count(column)` will only return the number of rows where column is not NULL.
+
+```sql
+SELECT
+    v,
+    count(1),
+    count(v)
+FROM
+(
+    SELECT if(number < 10, NULL, number % 3) AS v
+    FROM numbers(15)
+)
+GROUP BY v
+
+┌────v─┬─count()─┬─count(v)─┐
+│ ᴺᵁᴸᴸ │      10 │        0 │
+│    0 │       1 │        1 │
+│    1 │       2 │        2 │
+│    2 │       2 │        2 │
+└──────┴─────────┴──────────┘
+```
+
+And here is an example of of first_value with `RESPECT NULLS` where we can see that NULL inputs are respected and it will return the first value read, whether it's NULL or not:
+
+```sql
+SELECT
+    col || '_' || ((col + 1) * 5 - 1) as range,
+    first_value(odd_or_null) as first,
+    first_value(odd_or_null) IGNORE NULLS as first_ignore_null,
+    first_value(odd_or_null) RESPECT NULLS as first_respect_nulls
+FROM
+(
+    SELECT
+        intDiv(number, 5) AS col,
+        if(number % 2 == 0, NULL, number) as odd_or_null
+    FROM numbers(15)
+)
+GROUP BY col
+ORDER BY col
+
+┌─range─┬─first─┬─first_ignore_null─┬─first_respect_nulls─┐
+│ 0_4   │     1 │                 1 │                ᴺᵁᴸᴸ │
+│ 1_9   │     5 │                 5 │                   5 │
+│ 2_14  │    11 │                11 │                ᴺᵁᴸᴸ │
+└───────┴───────┴───────────────────┴─────────────────────┘
+```

From 31ebee41d040f3af8d4266623ba25a37d7df1fbd Mon Sep 17 00:00:00 2001
From: alesapin <alesapin@gmail.com>
Date: Wed, 13 Mar 2024 20:06:15 +0100
Subject: [PATCH 317/443] Fxi

---
 .../0_stateless/02151_invalid_setting_with_hints_in_query.sh    | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/queries/0_stateless/02151_invalid_setting_with_hints_in_query.sh b/tests/queries/0_stateless/02151_invalid_setting_with_hints_in_query.sh
index 2faaa3bb1b6..367d0163497 100755
--- a/tests/queries/0_stateless/02151_invalid_setting_with_hints_in_query.sh
+++ b/tests/queries/0_stateless/02151_invalid_setting_with_hints_in_query.sh
@@ -4,4 +4,4 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
 # shellcheck source=../shell_config.sh
 . "$CURDIR"/../shell_config.sh
 
-$CLICKHOUSE_LOCAL --query="SET input_format_with_names_use_headers = 1" 2>&1 | grep -qF "Code: 115. DB::Exception: Unknown setting input_format_with_names_use_headers: Maybe you meant ['input_format_with_names_use_header','input_format_with_types_use_header']. (UNKNOWN_SETTING)" && echo 'OK' || echo 'FAIL' ||:
+$CLICKHOUSE_LOCAL --query="SET input_format_with_names_use_headers = 1" 2>&1 | grep -qF "Code: 115. DB::Exception: Unknown setting 'input_format_with_names_use_headers': Maybe you meant ['input_format_with_names_use_header','input_format_with_types_use_header']. (UNKNOWN_SETTING)" && echo 'OK' || echo 'FAIL' ||:

From bd6bed161cc3beddd3966842385df66a0fb71e1a Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Wed, 13 Mar 2024 20:38:55 +0000
Subject: [PATCH 318/443] Try to fix flaky test_undrop_query

---
 .../settings.md                               |  4 +-
 docs/en/sql-reference/statements/undrop.md    | 48 +++++++------------
 src/Interpreters/DatabaseCatalog.cpp          |  4 +-
 src/Interpreters/InterpreterUndropQuery.cpp   |  6 ++-
 tests/integration/test_undrop_query/test.py   | 25 ++++++----
 .../0_stateless/02681_undrop_query.sql        |  2 +-
 6 files changed, 43 insertions(+), 46 deletions(-)

diff --git a/docs/en/operations/server-configuration-parameters/settings.md b/docs/en/operations/server-configuration-parameters/settings.md
index 07c9a2b88ab..f20dcb9025e 100644
--- a/docs/en/operations/server-configuration-parameters/settings.md
+++ b/docs/en/operations/server-configuration-parameters/settings.md
@@ -933,9 +933,9 @@ Hard limit is configured via system tools
 
 ## database_atomic_delay_before_drop_table_sec {#database_atomic_delay_before_drop_table_sec}
 
-Sets the delay before remove table data in seconds. If the query has `SYNC` modifier, this setting is ignored.
+The delay before a table data is dropped in seconds. If the `DROP TABLE` query has a `SYNC` modifier, this setting is ignored.
 
-Default value: `480` (8 minute).
+Default value: `480` (8 minutes).
 
 ## database_catalog_unused_dir_hide_timeout_sec {#database_catalog_unused_dir_hide_timeout_sec}
 
diff --git a/docs/en/sql-reference/statements/undrop.md b/docs/en/sql-reference/statements/undrop.md
index 40ac1ab4f99..4b138bfe679 100644
--- a/docs/en/sql-reference/statements/undrop.md
+++ b/docs/en/sql-reference/statements/undrop.md
@@ -13,13 +13,6 @@ a system table called `system.dropped_tables`.
 
 If you have a materialized view without a `TO` clause associated with the dropped table, then you will also have to UNDROP the inner table of that view.
 
-:::note
-UNDROP TABLE is experimental.  To use it add this setting: 
-```sql
-set allow_experimental_undrop_table_query = 1;
-```
-:::
-
 :::tip
 Also see [DROP TABLE](/docs/en/sql-reference/statements/drop.md)
 :::
@@ -32,60 +25,53 @@ UNDROP TABLE [db.]name [UUID '<uuid>'] [ON CLUSTER cluster]
 
 **Example**
 
-``` sql
-set allow_experimental_undrop_table_query = 1;
-```
-
 ```sql
-CREATE TABLE undropMe
+CREATE TABLE tab
 (
     `id` UInt8
 )
 ENGINE = MergeTree
-ORDER BY id
-```
+ORDER BY id;
+
+DROP TABLE tab;
 
-```sql
-DROP TABLE undropMe
-```
-```sql
 SELECT *
 FROM system.dropped_tables
-FORMAT Vertical
+FORMAT Vertical;
 ```
+
 ```response
 Row 1:
 ──────
 index:                 0
 database:              default
-table:                 undropMe
+table:                 tab
 uuid:                  aa696a1a-1d70-4e60-a841-4c80827706cc
 engine:                MergeTree
-metadata_dropped_path: /var/lib/clickhouse/metadata_dropped/default.undropMe.aa696a1a-1d70-4e60-a841-4c80827706cc.sql
+metadata_dropped_path: /var/lib/clickhouse/metadata_dropped/default.tab.aa696a1a-1d70-4e60-a841-4c80827706cc.sql
 table_dropped_time:    2023-04-05 14:12:12
 
 1 row in set. Elapsed: 0.001 sec. 
 ```
+
 ```sql
-UNDROP TABLE undropMe
-```
-```response
-Ok.
-```
-```sql
+UNDROP TABLE tab;
+
 SELECT *
 FROM system.dropped_tables
-FORMAT Vertical
-```
+FORMAT Vertical;
+
 ```response
 Ok.
 
 0 rows in set. Elapsed: 0.001 sec. 
 ```
+
 ```sql
-DESCRIBE TABLE undropMe
-FORMAT Vertical
+DESCRIBE TABLE tab
+FORMAT Vertical;
 ```
+
 ```response
 Row 1:
 ──────
diff --git a/src/Interpreters/DatabaseCatalog.cpp b/src/Interpreters/DatabaseCatalog.cpp
index a9fd5c852ba..9d9f418934f 100644
--- a/src/Interpreters/DatabaseCatalog.cpp
+++ b/src/Interpreters/DatabaseCatalog.cpp
@@ -1143,7 +1143,7 @@ void DatabaseCatalog::dequeueDroppedTableCleanup(StorageID table_id)
     TableMarkedAsDropped dropped_table;
     {
         std::lock_guard lock(tables_marked_dropped_mutex);
-        time_t latest_drop_time = std::numeric_limits<time_t>::min();
+        auto latest_drop_time = std::numeric_limits<time_t>::min();
         auto it_dropped_table = tables_marked_dropped.end();
         for (auto it = tables_marked_dropped.begin(); it != tables_marked_dropped.end(); ++it)
         {
@@ -1168,7 +1168,7 @@ void DatabaseCatalog::dequeueDroppedTableCleanup(StorageID table_id)
         }
         if (it_dropped_table == tables_marked_dropped.end())
             throw Exception(ErrorCodes::UNKNOWN_TABLE,
-                "The drop task of table {} is in progress, has been dropped or the database engine doesn't support it",
+                "Table {} is being dropped, has been dropped, or the database engine does not support UNDROP",
                 table_id.getNameForLogs());
         latest_metadata_dropped_path = it_dropped_table->metadata_path;
         String table_metadata_path = getPathForMetadata(it_dropped_table->table_id);
diff --git a/src/Interpreters/InterpreterUndropQuery.cpp b/src/Interpreters/InterpreterUndropQuery.cpp
index 8401c47df6b..f628a656947 100644
--- a/src/Interpreters/InterpreterUndropQuery.cpp
+++ b/src/Interpreters/InterpreterUndropQuery.cpp
@@ -17,14 +17,16 @@ namespace ErrorCodes
     extern const int SUPPORT_IS_DISABLED;
 }
 
-InterpreterUndropQuery::InterpreterUndropQuery(const ASTPtr & query_ptr_, ContextMutablePtr context_) : WithMutableContext(context_), query_ptr(query_ptr_)
+InterpreterUndropQuery::InterpreterUndropQuery(const ASTPtr & query_ptr_, ContextMutablePtr context_)
+    : WithMutableContext(context_)
+    , query_ptr(query_ptr_)
 {
 }
 
-
 BlockIO InterpreterUndropQuery::execute()
 {
     getContext()->checkAccess(AccessType::UNDROP_TABLE);
+
     auto & undrop = query_ptr->as<ASTUndropQuery &>();
     if (!undrop.cluster.empty() && !maybeRemoveOnCluster(query_ptr, getContext()))
     {
diff --git a/tests/integration/test_undrop_query/test.py b/tests/integration/test_undrop_query/test.py
index 590a5690e55..d57aa8c2dc7 100644
--- a/tests/integration/test_undrop_query/test.py
+++ b/tests/integration/test_undrop_query/test.py
@@ -29,30 +29,39 @@ def test_undrop_drop_and_undrop_loop(started_cluster):
         logging.info(
             "random_sec: " + random_sec.__str__() + ", table_uuid: " + table_uuid
         )
+
         node.query(
-            "create table test_undrop_loop"
+            "CREATE TABLE test_undrop_loop"
             + count.__str__()
             + " UUID '"
             + table_uuid
-            + "' (id Int32) Engine=MergeTree() order by id;"
+            + "' (id Int32) ENGINE = MergeTree() ORDER BY id;"
         )
-        node.query("drop table test_undrop_loop" + count.__str__() + ";")
+
+        node.query("DROP TABLE test_undrop_loop" + count.__str__() + ";")
+
         time.sleep(random_sec)
+
         if random_sec >= 5:
             error = node.query_and_get_error(
-                "undrop table test_undrop_loop"
+                "UNDROP TABLE test_undrop_loop"
                 + count.__str__()
-                + " uuid '"
+                + " UUID '"
                 + table_uuid
                 + "';"
             )
             assert "UNKNOWN_TABLE" in error
-        else:
+        elif random_sec <= 3:
+            # (*)
             node.query(
-                "undrop table test_undrop_loop"
+                "UNDROP TABLE test_undrop_loop"
                 + count.__str__()
-                + " uuid '"
+                + " UUID '"
                 + table_uuid
                 + "';"
             )
             count = count + 1
+        else:
+            pass
+            # ignore random_sec = 4 to account for communication delay with the database.
+            # if we don't do that, then the second case (*) may find the table already dropped and receive an unexpected exception from the database (Bug #55167)
diff --git a/tests/queries/0_stateless/02681_undrop_query.sql b/tests/queries/0_stateless/02681_undrop_query.sql
index 66447fc6c44..d038a383690 100644
--- a/tests/queries/0_stateless/02681_undrop_query.sql
+++ b/tests/queries/0_stateless/02681_undrop_query.sql
@@ -85,5 +85,5 @@ drop table 02681_undrop_multiple;
 select table from system.dropped_tables where table = '02681_undrop_multiple' limit 1;
 undrop table 02681_undrop_multiple;
 select * from 02681_undrop_multiple order by id;
-undrop table 02681_undrop_multiple; -- { serverError 57 }
+undrop table 02681_undrop_multiple; -- { serverError TABLE_ALREADY_EXISTS }
 drop table 02681_undrop_multiple sync;

From be9a77b6ca7aa35b78857bc812ccacbe1d22bfa3 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Thu, 14 Mar 2024 03:54:07 +0300
Subject: [PATCH 319/443] Update SettingsChangesHistory.h

---
 src/Core/SettingsChangesHistory.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Core/SettingsChangesHistory.h b/src/Core/SettingsChangesHistory.h
index 129a89bfa23..04b6add0e56 100644
--- a/src/Core/SettingsChangesHistory.h
+++ b/src/Core/SettingsChangesHistory.h
@@ -88,6 +88,7 @@ static std::map<ClickHouseVersion, SettingsChangesHistory::SettingsChanges> sett
     {"24.3", {{"use_page_cache_for_disks_without_file_cache", false, false, "Added userspace page cache"},
               {"read_from_page_cache_if_exists_otherwise_bypass_cache", false, false, "Added userspace page cache"},
               {"page_cache_inject_eviction", false, false, "Added userspace page cache"},
+              {"default_table_engine", "None", "MergeTree", "Set default table engine to MergeTree for better usability"},
               }},
     {"24.2", {{"allow_suspicious_variant_types", true, false, "Don't allow creating Variant type with suspicious variants by default"},
               {"validate_experimental_and_suspicious_types_inside_nested_types", false, true, "Validate usage of experimental and suspicious types inside nested types"},
@@ -136,7 +137,6 @@ static std::map<ClickHouseVersion, SettingsChangesHistory::SettingsChanges> sett
               {"function_visible_width_behavior", 0, 1, "We changed the default behavior of `visibleWidth` to be more precise"},
               {"max_estimated_execution_time", 0, 0, "Separate max_execution_time and max_estimated_execution_time"},
               {"iceberg_engine_ignore_schema_evolution", false, false, "Allow to ignore schema evolution in Iceberg table engine"},
-              {"default_table_engine", "None", "MergeTree", "Set default table engine to MergeTree for better usability"},
               {"optimize_injective_functions_in_group_by", false, true, "Replace injective functions by it's arguments in GROUP BY section in analyzer"},
               {"update_insert_deduplication_token_in_dependent_materialized_views", false, false, "Allow to update insert deduplication token with table identifier during insert in dependent materialized views"},
               {"azure_max_unexpected_write_error_retries", 4, 4, "The maximum number of retries in case of unexpected errors during Azure blob storage write"},

From 0b522f5fb34a774ce6cf6092dc97d5360fde7ea6 Mon Sep 17 00:00:00 2001
From: peter279k <peter279k@gmail.com>
Date: Thu, 14 Mar 2024 13:53:26 +0800
Subject: [PATCH 320/443] Fix issue #61351

---
 docs/en/getting-started/example-datasets/laion.md | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/docs/en/getting-started/example-datasets/laion.md b/docs/en/getting-started/example-datasets/laion.md
index 0dbaceffc13..327c1796d11 100644
--- a/docs/en/getting-started/example-datasets/laion.md
+++ b/docs/en/getting-started/example-datasets/laion.md
@@ -10,10 +10,14 @@ The embeddings and the metadata are stored in separate files in the raw data. A
 converts them to CSV and imports them into ClickHouse. You can use the following `download.sh` script for that:
 
 ```bash
-wget --tries=100 https://deploy.laion.ai/8f83b608504d46bb81708ec86e912220/embeddings/img_emb/img_emb_${1}.npy          # download image embedding
-wget --tries=100 https://deploy.laion.ai/8f83b608504d46bb81708ec86e912220/embeddings/text_emb/text_emb_${1}.npy        # download text embedding
-wget --tries=100 https://deploy.laion.ai/8f83b608504d46bb81708ec86e912220/embeddings/metadata/metadata_${1}.parquet    # download metadata
-python3 process.py ${1} # merge files and convert to CSV
+number=${1}
+if [[ $number == '' ]]; then
+    number=1
+fi;
+wget --tries=100 https://deploy.laion.ai/8f83b608504d46bb81708ec86e912220/embeddings/img_emb/img_emb_${number}.npy          # download image embedding
+wget --tries=100 https://deploy.laion.ai/8f83b608504d46bb81708ec86e912220/embeddings/text_emb/text_emb_${number}.npy        # download text embedding
+wget --tries=100 https://deploy.laion.ai/8f83b608504d46bb81708ec86e912220/embeddings/metadata/metadata_${number}.parquet    # download metadata
+python3 process.py $number # merge files and convert to CSV
 ```
 Script `process.py` is defined as follows:
 

From ff976520389c277cfcb034466aed514fe96acc6d Mon Sep 17 00:00:00 2001
From: serxa <sergei@clickhouse.com>
Date: Thu, 14 Mar 2024 09:53:18 +0000
Subject: [PATCH 321/443] fix spell check

---
 utils/check-style/aspell-ignore/en/aspell-dict.txt | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/utils/check-style/aspell-ignore/en/aspell-dict.txt b/utils/check-style/aspell-ignore/en/aspell-dict.txt
index c7eb213bff2..1706d44bc8a 100644
--- a/utils/check-style/aspell-ignore/en/aspell-dict.txt
+++ b/utils/check-style/aspell-ignore/en/aspell-dict.txt
@@ -2687,7 +2687,9 @@ userver
 utils
 uuid
 varPop
+varPopStable
 varSamp
+varSampStable
 variadic
 variantElement
 variantType

From c6f0b2a5a458648c179456a6843aae8c9c3fb53a Mon Sep 17 00:00:00 2001
From: Antonio Andelic <antonio2368@users.noreply.github.com>
Date: Thu, 14 Mar 2024 11:57:03 +0100
Subject: [PATCH 322/443] Revert "Fix usage of session_token in S3 engine"

---
 src/Coordination/KeeperSnapshotManagerS3.cpp | 3 +--
 src/Storages/StorageS3.cpp                   | 3 +--
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/src/Coordination/KeeperSnapshotManagerS3.cpp b/src/Coordination/KeeperSnapshotManagerS3.cpp
index 796506a07db..80345db2524 100644
--- a/src/Coordination/KeeperSnapshotManagerS3.cpp
+++ b/src/Coordination/KeeperSnapshotManagerS3.cpp
@@ -121,8 +121,7 @@ void KeeperSnapshotManagerS3::updateS3Configuration(const Poco::Util::AbstractCo
                 auth_settings.use_insecure_imds_request.value_or(false),
                 auth_settings.expiration_window_seconds.value_or(S3::DEFAULT_EXPIRATION_WINDOW_SECONDS),
                 auth_settings.no_sign_request.value_or(false),
-            },
-            credentials.GetSessionToken());
+            });
 
         auto new_client = std::make_shared<KeeperSnapshotManagerS3::S3Configuration>(std::move(new_uri), std::move(auth_settings), std::move(client));
 
diff --git a/src/Storages/StorageS3.cpp b/src/Storages/StorageS3.cpp
index ff055508aa6..11da394feec 100644
--- a/src/Storages/StorageS3.cpp
+++ b/src/Storages/StorageS3.cpp
@@ -1451,8 +1451,7 @@ void StorageS3::Configuration::connect(const ContextPtr & context)
             auth_settings.expiration_window_seconds.value_or(
                 context->getConfigRef().getUInt64("s3.expiration_window_seconds", S3::DEFAULT_EXPIRATION_WINDOW_SECONDS)),
             auth_settings.no_sign_request.value_or(context->getConfigRef().getBool("s3.no_sign_request", false)),
-        },
-        credentials.GetSessionToken());
+        });
 }
 
 void StorageS3::processNamedCollectionResult(StorageS3::Configuration & configuration, const NamedCollection & collection)

From 56220b5105005acef0ab4b09d4f7b48c6aac8d66 Mon Sep 17 00:00:00 2001
From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com>
Date: Thu, 14 Mar 2024 12:23:32 +0100
Subject: [PATCH 323/443] Revert "Revert "Fix usage of session_token in S3
 engine""

---
 src/Coordination/KeeperSnapshotManagerS3.cpp | 3 ++-
 src/Storages/StorageS3.cpp                   | 3 ++-
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/src/Coordination/KeeperSnapshotManagerS3.cpp b/src/Coordination/KeeperSnapshotManagerS3.cpp
index 80345db2524..796506a07db 100644
--- a/src/Coordination/KeeperSnapshotManagerS3.cpp
+++ b/src/Coordination/KeeperSnapshotManagerS3.cpp
@@ -121,7 +121,8 @@ void KeeperSnapshotManagerS3::updateS3Configuration(const Poco::Util::AbstractCo
                 auth_settings.use_insecure_imds_request.value_or(false),
                 auth_settings.expiration_window_seconds.value_or(S3::DEFAULT_EXPIRATION_WINDOW_SECONDS),
                 auth_settings.no_sign_request.value_or(false),
-            });
+            },
+            credentials.GetSessionToken());
 
         auto new_client = std::make_shared<KeeperSnapshotManagerS3::S3Configuration>(std::move(new_uri), std::move(auth_settings), std::move(client));
 
diff --git a/src/Storages/StorageS3.cpp b/src/Storages/StorageS3.cpp
index 11da394feec..ff055508aa6 100644
--- a/src/Storages/StorageS3.cpp
+++ b/src/Storages/StorageS3.cpp
@@ -1451,7 +1451,8 @@ void StorageS3::Configuration::connect(const ContextPtr & context)
             auth_settings.expiration_window_seconds.value_or(
                 context->getConfigRef().getUInt64("s3.expiration_window_seconds", S3::DEFAULT_EXPIRATION_WINDOW_SECONDS)),
             auth_settings.no_sign_request.value_or(context->getConfigRef().getBool("s3.no_sign_request", false)),
-        });
+        },
+        credentials.GetSessionToken());
 }
 
 void StorageS3::processNamedCollectionResult(StorageS3::Configuration & configuration, const NamedCollection & collection)

From c7e8ca789f8bffa4451765bd51882014ed1768c3 Mon Sep 17 00:00:00 2001
From: vdimir <vdimir@clickhouse.com>
Date: Tue, 12 Mar 2024 15:41:43 +0000
Subject: [PATCH 324/443] Followup fix ASTRenameQuery::clone  #61013

---
 src/Parsers/ASTRenameQuery.h | 25 ++++++++++++++++++++++++-
 1 file changed, 24 insertions(+), 1 deletion(-)

diff --git a/src/Parsers/ASTRenameQuery.h b/src/Parsers/ASTRenameQuery.h
index beaf93c4761..d51c382f374 100644
--- a/src/Parsers/ASTRenameQuery.h
+++ b/src/Parsers/ASTRenameQuery.h
@@ -94,7 +94,24 @@ public:
     ASTPtr clone() const override
     {
         auto res = std::make_shared<ASTRenameQuery>(*this);
-        res->cloneChildren();
+        res->children.clear();
+
+        auto clone_child = [&res](ASTPtr & node)
+        {
+            if (node)
+            {
+                node = node->clone();
+                res->children.push_back(node);
+            }
+        };
+
+        for (auto & elem : res->elements)
+        {
+            clone_child(elem.from.database);
+            clone_child(elem.from.table);
+            clone_child(elem.to.database);
+            clone_child(elem.to.table);
+        }
         cloneOutputOptions(*res);
         return res;
     }
@@ -108,9 +125,15 @@ public:
         for (Element & elem : query.elements)
         {
             if (!elem.from.database)
+            {
                 elem.from.database = std::make_shared<ASTIdentifier>(params.default_database);
+                query.children.push_back(elem.from.database);
+            }
             if (!elem.to.database)
+            {
                 elem.to.database = std::make_shared<ASTIdentifier>(params.default_database);
+                query.children.push_back(elem.to.database);
+            }
         }
 
         return query_ptr;

From 95b218e7d40cc6527a597f962947424cfe540a34 Mon Sep 17 00:00:00 2001
From: Han Fei <hanfei19910905@gmail.com>
Date: Thu, 14 Mar 2024 13:25:24 +0100
Subject: [PATCH 325/443] fix dead lock

---
 src/Common/CgroupsMemoryUsageObserver.cpp | 4 ++--
 src/Common/CgroupsMemoryUsageObserver.h   | 4 +++-
 2 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/src/Common/CgroupsMemoryUsageObserver.cpp b/src/Common/CgroupsMemoryUsageObserver.cpp
index e55fd44a524..661b80a6059 100644
--- a/src/Common/CgroupsMemoryUsageObserver.cpp
+++ b/src/Common/CgroupsMemoryUsageObserver.cpp
@@ -100,7 +100,7 @@ void CgroupsMemoryUsageObserver::setMemoryUsageLimits(uint64_t hard_limit_, uint
 
 void CgroupsMemoryUsageObserver::setOnMemoryAmountAvailableChanged(OnMemoryAmountAvailableChangedFn on_memory_amount_available_changed_)
 {
-    std::lock_guard<std::mutex> limit_lock(limit_mutex);
+    std::lock_guard<std::mutex> memory_amount_change_lock(memory_amount_change_mutex);
     on_memory_amount_available_changed = on_memory_amount_available_changed_;
 }
 
@@ -299,7 +299,7 @@ void CgroupsMemoryUsageObserver::runThread()
             {
                 LOG_INFO(log, "Memory amount available to the process changed from {} to {}", ReadableSize(last_available_memory_amount), ReadableSize(available_memory_amount));
                 last_available_memory_amount = available_memory_amount;
-                std::lock_guard<std::mutex> limit_lock(limit_mutex);
+                std::lock_guard<std::mutex> memory_amount_change_lock(memory_amount_change_mutex);
                 on_memory_amount_available_changed();
             }
 
diff --git a/src/Common/CgroupsMemoryUsageObserver.h b/src/Common/CgroupsMemoryUsageObserver.h
index eb755d04187..7ad0f882edd 100644
--- a/src/Common/CgroupsMemoryUsageObserver.h
+++ b/src/Common/CgroupsMemoryUsageObserver.h
@@ -51,7 +51,8 @@ private:
     size_t soft_limit TSA_GUARDED_BY(limit_mutex) = 0;
     OnMemoryLimitFn on_hard_limit TSA_GUARDED_BY(limit_mutex);
     OnMemoryLimitFn on_soft_limit TSA_GUARDED_BY(limit_mutex);
-    OnMemoryAmountAvailableChangedFn on_memory_amount_available_changed TSA_GUARDED_BY(limit_mutex);
+    std::mutex memory_amount_change_mutex;
+    OnMemoryAmountAvailableChangedFn on_memory_amount_available_changed TSA_GUARDED_BY(memory_amount_change_mutex);
 
     uint64_t last_memory_usage = 0;        /// how much memory does the process use
     uint64_t last_available_memory_amount; /// how much memory can the process use
@@ -86,6 +87,7 @@ private:
 #else
 class CgroupsMemoryUsageObserver
 {
+    using OnMemoryAmountAvailableChangedFn = std::function<void()>;
 public:
     explicit CgroupsMemoryUsageObserver(std::chrono::seconds) {}
 

From 8fdf3ae747877b1aa0cbab3a74290f5a94edea63 Mon Sep 17 00:00:00 2001
From: Nikolai Kochetov <nk@clickhouse.com>
Date: Thu, 14 Mar 2024 12:31:30 +0000
Subject: [PATCH 326/443] Add a comment.

---
 src/Planner/Planner.cpp | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/src/Planner/Planner.cpp b/src/Planner/Planner.cpp
index fde9f110d09..5624a911210 100644
--- a/src/Planner/Planner.cpp
+++ b/src/Planner/Planner.cpp
@@ -1084,6 +1084,8 @@ void addBuildSubqueriesForSetsStepIfNeeded(
         auto subquery_options = select_query_options.subquery();
         /// I don't know if this is a good decision,
         /// But for now it is done in the same way as in old analyzer.
+        /// This would not ignore limits for subqueries (affects mutations only).
+        /// See test_build_sets_from_multiple_threads-analyzer.
         subquery_options.ignore_limits = false;
         Planner subquery_planner(
             query_tree,

From b59680911c036eca84776247b9644b9f5475e25c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <benjamin.antal@clickhouse.com>
Date: Thu, 14 Mar 2024 12:32:13 +0000
Subject: [PATCH 327/443] Move `getItemsShortCircuitImpl` and `getItemsImpl`
 into separate files

---
 src/Dictionaries/DictionaryStructure.h       |  27 ++
 src/Dictionaries/RangeHashedDictionary.cpp   | 218 ----------------
 src/Dictionaries/RangeHashedDictionary.h     |  13 +-
 src/Dictionaries/RangeHashedDictionary_2.cpp | 250 +++++++++++++++++++
 src/Dictionaries/RangeHashedDictionary_3.cpp | 135 ++++++++++
 utils/check-style/check-large-objects.sh     |   1 +
 6 files changed, 422 insertions(+), 222 deletions(-)
 create mode 100644 src/Dictionaries/RangeHashedDictionary_2.cpp
 create mode 100644 src/Dictionaries/RangeHashedDictionary_3.cpp

diff --git a/src/Dictionaries/DictionaryStructure.h b/src/Dictionaries/DictionaryStructure.h
index 55060b1592f..56d11be9837 100644
--- a/src/Dictionaries/DictionaryStructure.h
+++ b/src/Dictionaries/DictionaryStructure.h
@@ -41,6 +41,33 @@ enum class AttributeUnderlyingType : TypeIndexUnderlying
 
 #undef map_item
 
+
+#define CALL_FOR_ALL_DICTIONARY_ATTRIBUTE_TYPES(M) \
+    M(UInt8) \
+    M(UInt16) \
+    M(UInt32) \
+    M(UInt64) \
+    M(UInt128) \
+    M(UInt256) \
+    M(Int8) \
+    M(Int16) \
+    M(Int32) \
+    M(Int64) \
+    M(Int128) \
+    M(Int256) \
+    M(Decimal32) \
+    M(Decimal64) \
+    M(Decimal128) \
+    M(Decimal256) \
+    M(DateTime64) \
+    M(Float32) \
+    M(Float64) \
+    M(UUID) \
+    M(IPv4) \
+    M(IPv6) \
+    M(String) \
+    M(Array)
+
 /// Min and max lifetimes for a dictionary or its entry
 using DictionaryLifetime = ExternalLoadableLifetime;
 
diff --git a/src/Dictionaries/RangeHashedDictionary.cpp b/src/Dictionaries/RangeHashedDictionary.cpp
index 203561fc23d..8299c3ad93a 100644
--- a/src/Dictionaries/RangeHashedDictionary.cpp
+++ b/src/Dictionaries/RangeHashedDictionary.cpp
@@ -211,224 +211,6 @@ ColumnPtr RangeHashedDictionary<dictionary_key_type>::getColumn(
     return result;
 }
 
-template <DictionaryKeyType dictionary_key_type>
-template <typename AttributeType, bool is_nullable, typename ValueSetter, typename DefaultValueExtractor>
-void RangeHashedDictionary<dictionary_key_type>::getItemsImpl(
-    const Attribute & attribute,
-    const Columns & key_columns,
-    ValueSetter && set_value,
-    DefaultValueExtractor & default_value_extractor) const
-{
-    const auto & attribute_container = std::get<AttributeContainerType<AttributeType>>(attribute.container);
-
-    size_t keys_found = 0;
-
-    const ColumnPtr & range_column = key_columns.back();
-    auto key_columns_copy = key_columns;
-    key_columns_copy.pop_back();
-
-    DictionaryKeysArenaHolder<dictionary_key_type> arena_holder;
-    DictionaryKeysExtractor<dictionary_key_type> keys_extractor(key_columns_copy, arena_holder.getComplexKeyArena());
-    const size_t keys_size = keys_extractor.getKeysSize();
-
-    callOnRangeType(dict_struct.range_min->type, [&](const auto & types)
-    {
-        using Types = std::decay_t<decltype(types)>;
-        using RangeColumnType = typename Types::LeftType;
-        using RangeStorageType = typename RangeColumnType::ValueType;
-        using RangeInterval = Interval<RangeStorageType>;
-
-        const auto * range_column_typed = typeid_cast<const RangeColumnType *>(range_column.get());
-        if (!range_column_typed)
-            throw Exception(ErrorCodes::TYPE_MISMATCH,
-                "Dictionary {} range column type should be equal to {}",
-                getFullName(),
-                dict_struct.range_min->type->getName());
-
-        const auto & range_column_data = range_column_typed->getData();
-
-        const auto & key_attribute_container = std::get<KeyAttributeContainerType<RangeStorageType>>(key_attribute.container);
-
-        for (size_t key_index = 0; key_index < keys_size; ++key_index)
-        {
-            auto key = keys_extractor.extractCurrentKey();
-            const auto it = key_attribute_container.find(key);
-
-            if (it)
-            {
-                const auto date = range_column_data[key_index];
-                const auto & interval_tree = it->getMapped();
-
-                size_t value_index = 0;
-                std::optional<RangeInterval> range;
-
-                interval_tree.find(date, [&](auto & interval, auto & interval_value_index)
-                {
-                    if (range)
-                    {
-                        if (likely(configuration.lookup_strategy == RangeHashedDictionaryLookupStrategy::min) && interval < *range)
-                        {
-                            range = interval;
-                            value_index = interval_value_index;
-                        }
-                        else if (configuration.lookup_strategy == RangeHashedDictionaryLookupStrategy::max && interval > * range)
-                        {
-                            range = interval;
-                            value_index = interval_value_index;
-                        }
-                    }
-                    else
-                    {
-                        range = interval;
-                        value_index = interval_value_index;
-                    }
-
-                    return true;
-                });
-
-                if (range.has_value())
-                {
-                    ++keys_found;
-
-                    AttributeType value = attribute_container[value_index];
-
-                    if constexpr (is_nullable)
-                    {
-                        bool is_null = (*attribute.is_value_nullable)[value_index];
-                        set_value(key_index, value, is_null);
-                    }
-                    else
-                    {
-                        set_value(key_index, value, false);
-                    }
-
-                    keys_extractor.rollbackCurrentKey();
-                    continue;
-                }
-            }
-
-            if constexpr (is_nullable)
-                set_value(key_index, default_value_extractor[key_index], default_value_extractor.isNullAt(key_index));
-            else
-                set_value(key_index, default_value_extractor[key_index], false);
-
-            keys_extractor.rollbackCurrentKey();
-        }
-    });
-
-    query_count.fetch_add(keys_size, std::memory_order_relaxed);
-    found_count.fetch_add(keys_found, std::memory_order_relaxed);
-}
-
-template <DictionaryKeyType dictionary_key_type>
-template <typename AttributeType, bool is_nullable, typename ValueSetter>
-size_t RangeHashedDictionary<dictionary_key_type>::getItemsShortCircuitImpl(
-    const Attribute & attribute,
-    const Columns & key_columns,
-    ValueSetter && set_value,
-    IColumn::Filter & default_mask) const
-{
-    const auto & attribute_container = std::get<AttributeContainerType<AttributeType>>(attribute.container);
-
-    size_t keys_found = 0;
-
-    const ColumnPtr & range_column = key_columns.back();
-    auto key_columns_copy = key_columns;
-    key_columns_copy.pop_back();
-
-    DictionaryKeysArenaHolder<dictionary_key_type> arena_holder;
-    DictionaryKeysExtractor<dictionary_key_type> keys_extractor(key_columns_copy, arena_holder.getComplexKeyArena());
-    const size_t keys_size = keys_extractor.getKeysSize();
-    default_mask.resize(keys_size);
-
-    callOnRangeType(dict_struct.range_min->type, [&](const auto & types)
-    {
-        using Types = std::decay_t<decltype(types)>;
-        using RangeColumnType = typename Types::LeftType;
-        using RangeStorageType = typename RangeColumnType::ValueType;
-        using RangeInterval = Interval<RangeStorageType>;
-
-        const auto * range_column_typed = typeid_cast<const RangeColumnType *>(range_column.get());
-        if (!range_column_typed)
-            throw Exception(ErrorCodes::TYPE_MISMATCH,
-                "Dictionary {} range column type should be equal to {}",
-                getFullName(),
-                dict_struct.range_min->type->getName());
-
-        const auto & range_column_data = range_column_typed->getData();
-
-        const auto & key_attribute_container = std::get<KeyAttributeContainerType<RangeStorageType>>(key_attribute.container);
-
-        for (size_t key_index = 0; key_index < keys_size; ++key_index)
-        {
-            auto key = keys_extractor.extractCurrentKey();
-            const auto it = key_attribute_container.find(key);
-
-            if (it)
-            {
-                const auto date = range_column_data[key_index];
-                const auto & interval_tree = it->getMapped();
-
-                size_t value_index = 0;
-                std::optional<RangeInterval> range;
-
-                interval_tree.find(date, [&](auto & interval, auto & interval_value_index)
-                {
-                    if (range)
-                    {
-                        if (likely(configuration.lookup_strategy == RangeHashedDictionaryLookupStrategy::min) && interval < *range)
-                        {
-                            range = interval;
-                            value_index = interval_value_index;
-                        }
-                        else if (configuration.lookup_strategy == RangeHashedDictionaryLookupStrategy::max && interval > * range)
-                        {
-                            range = interval;
-                            value_index = interval_value_index;
-                        }
-                    }
-                    else
-                    {
-                        range = interval;
-                        value_index = interval_value_index;
-                    }
-
-                    return true;
-                });
-
-                if (range.has_value())
-                {
-                    default_mask[key_index] = 0;
-                    ++keys_found;
-
-                    AttributeType value = attribute_container[value_index];
-
-                    if constexpr (is_nullable)
-                    {
-                        bool is_null = (*attribute.is_value_nullable)[value_index];
-                        set_value(key_index, value, is_null);
-                    }
-                    else
-                    {
-                        set_value(key_index, value, false);
-                    }
-
-                    keys_extractor.rollbackCurrentKey();
-                    continue;
-                }
-            }
-
-            default_mask[key_index] = 1;
-
-            keys_extractor.rollbackCurrentKey();
-        }
-    });
-
-    query_count.fetch_add(keys_size, std::memory_order_relaxed);
-    found_count.fetch_add(keys_found, std::memory_order_relaxed);
-    return keys_found;
-}
-
 template
 ColumnPtr RangeHashedDictionary<DictionaryKeyType::Simple>::getColumn(
     const std::string & attribute_name,
diff --git a/src/Dictionaries/RangeHashedDictionary.h b/src/Dictionaries/RangeHashedDictionary.h
index 4a8008b9051..a5dedae97c4 100644
--- a/src/Dictionaries/RangeHashedDictionary.h
+++ b/src/Dictionaries/RangeHashedDictionary.h
@@ -236,18 +236,23 @@ private:
 
     static Attribute createAttribute(const DictionaryAttribute & dictionary_attribute);
 
-    template <typename AttributeType, bool is_nullable, typename ValueSetter, typename DefaultValueExtractor>
+
+
+    template <typename ValueType>
+    using ValueSetterFunc = std::function<void(size_t, const ValueType &, bool)>;
+
+    template <typename ValueType, bool is_nullable, typename DefaultValueExtractor>
     void getItemsImpl(
         const Attribute & attribute,
         const Columns & key_columns,
-        ValueSetter && set_value,
+        ValueSetterFunc<ValueType> && set_value,
         DefaultValueExtractor & default_value_extractor) const;
 
-    template <typename AttributeType, bool is_nullable, typename ValueSetter>
+    template <typename ValueType, bool is_nullable>
     size_t getItemsShortCircuitImpl(
         const Attribute & attribute,
         const Columns & key_columns,
-        ValueSetter && set_value,
+        ValueSetterFunc<ValueType> && set_value,
         IColumn::Filter & default_mask) const;
 
     ColumnPtr getColumnInternal(
diff --git a/src/Dictionaries/RangeHashedDictionary_2.cpp b/src/Dictionaries/RangeHashedDictionary_2.cpp
new file mode 100644
index 00000000000..2329d621da4
--- /dev/null
+++ b/src/Dictionaries/RangeHashedDictionary_2.cpp
@@ -0,0 +1,250 @@
+#include <Dictionaries/RangeHashedDictionary.h>
+
+namespace DB
+{
+
+
+template <DictionaryKeyType dictionary_key_type>
+template <typename ValueType, bool is_nullable>
+size_t RangeHashedDictionary<dictionary_key_type>::getItemsShortCircuitImpl(
+    const Attribute & attribute,
+    const Columns & key_columns,
+    typename RangeHashedDictionary<dictionary_key_type>::ValueSetterFunc<ValueType> && set_value,
+    IColumn::Filter & default_mask) const
+{
+    const auto & attribute_container = std::get<AttributeContainerType<ValueType>>(attribute.container);
+
+    size_t keys_found = 0;
+
+    const ColumnPtr & range_column = key_columns.back();
+    auto key_columns_copy = key_columns;
+    key_columns_copy.pop_back();
+
+    DictionaryKeysArenaHolder<dictionary_key_type> arena_holder;
+    DictionaryKeysExtractor<dictionary_key_type> keys_extractor(key_columns_copy, arena_holder.getComplexKeyArena());
+    const size_t keys_size = keys_extractor.getKeysSize();
+    default_mask.resize(keys_size);
+
+    callOnRangeType(
+        dict_struct.range_min->type,
+        [&](const auto & types)
+        {
+            using Types = std::decay_t<decltype(types)>;
+            using RangeColumnType = typename Types::LeftType;
+            using RangeStorageType = typename RangeColumnType::ValueType;
+            using RangeInterval = Interval<RangeStorageType>;
+
+            const auto * range_column_typed = typeid_cast<const RangeColumnType *>(range_column.get());
+            if (!range_column_typed)
+                throw Exception(
+                    ErrorCodes::TYPE_MISMATCH,
+                    "Dictionary {} range column type should be equal to {}",
+                    getFullName(),
+                    dict_struct.range_min->type->getName());
+
+            const auto & range_column_data = range_column_typed->getData();
+
+            const auto & key_attribute_container = std::get<KeyAttributeContainerType<RangeStorageType>>(key_attribute.container);
+
+            for (size_t key_index = 0; key_index < keys_size; ++key_index)
+            {
+                auto key = keys_extractor.extractCurrentKey();
+                const auto it = key_attribute_container.find(key);
+
+                if (it)
+                {
+                    const auto date = range_column_data[key_index];
+                    const auto & interval_tree = it->getMapped();
+
+                    size_t value_index = 0;
+                    std::optional<RangeInterval> range;
+
+                    interval_tree.find(
+                        date,
+                        [&](auto & interval, auto & interval_value_index)
+                        {
+                            if (range)
+                            {
+                                if (likely(configuration.lookup_strategy == RangeHashedDictionaryLookupStrategy::min) && interval < *range)
+                                {
+                                    range = interval;
+                                    value_index = interval_value_index;
+                                }
+                                else if (configuration.lookup_strategy == RangeHashedDictionaryLookupStrategy::max && interval > *range)
+                                {
+                                    range = interval;
+                                    value_index = interval_value_index;
+                                }
+                            }
+                            else
+                            {
+                                range = interval;
+                                value_index = interval_value_index;
+                            }
+
+                            return true;
+                        });
+
+                    if (range.has_value())
+                    {
+                        default_mask[key_index] = 0;
+                        ++keys_found;
+
+                        ValueType value = attribute_container[value_index];
+
+                        if constexpr (is_nullable)
+                        {
+                            bool is_null = (*attribute.is_value_nullable)[value_index];
+                            set_value(key_index, value, is_null);
+                        }
+                        else
+                        {
+                            set_value(key_index, value, false);
+                        }
+
+                        keys_extractor.rollbackCurrentKey();
+                        continue;
+                    }
+                }
+
+                default_mask[key_index] = 1;
+
+                keys_extractor.rollbackCurrentKey();
+            }
+        });
+
+    query_count.fetch_add(keys_size, std::memory_order_relaxed);
+    found_count.fetch_add(keys_found, std::memory_order_relaxed);
+    return keys_found;
+}
+
+template <DictionaryKeyType dictionary_key_type>
+template <typename ValueType, bool is_nullable, typename DefaultValueExtractor>
+void RangeHashedDictionary<dictionary_key_type>::getItemsImpl(
+    const Attribute & attribute,
+    const Columns & key_columns,
+    typename RangeHashedDictionary<dictionary_key_type>::ValueSetterFunc<ValueType> && set_value,
+    DefaultValueExtractor & default_value_extractor) const
+{
+    const auto & attribute_container = std::get<AttributeContainerType<ValueType>>(attribute.container);
+
+    size_t keys_found = 0;
+
+    const ColumnPtr & range_column = key_columns.back();
+    auto key_columns_copy = key_columns;
+    key_columns_copy.pop_back();
+
+    DictionaryKeysArenaHolder<dictionary_key_type> arena_holder;
+    DictionaryKeysExtractor<dictionary_key_type> keys_extractor(key_columns_copy, arena_holder.getComplexKeyArena());
+    const size_t keys_size = keys_extractor.getKeysSize();
+
+    callOnRangeType(
+        dict_struct.range_min->type,
+        [&](const auto & types)
+        {
+            using Types = std::decay_t<decltype(types)>;
+            using RangeColumnType = typename Types::LeftType;
+            using RangeStorageType = typename RangeColumnType::ValueType;
+            using RangeInterval = Interval<RangeStorageType>;
+
+            const auto * range_column_typed = typeid_cast<const RangeColumnType *>(range_column.get());
+            if (!range_column_typed)
+                throw Exception(
+                    ErrorCodes::TYPE_MISMATCH,
+                    "Dictionary {} range column type should be equal to {}",
+                    getFullName(),
+                    dict_struct.range_min->type->getName());
+
+            const auto & range_column_data = range_column_typed->getData();
+
+            const auto & key_attribute_container = std::get<KeyAttributeContainerType<RangeStorageType>>(key_attribute.container);
+
+            for (size_t key_index = 0; key_index < keys_size; ++key_index)
+            {
+                auto key = keys_extractor.extractCurrentKey();
+                const auto it = key_attribute_container.find(key);
+
+                if (it)
+                {
+                    const auto date = range_column_data[key_index];
+                    const auto & interval_tree = it->getMapped();
+
+                    size_t value_index = 0;
+                    std::optional<RangeInterval> range;
+
+                    interval_tree.find(
+                        date,
+                        [&](auto & interval, auto & interval_value_index)
+                        {
+                            if (range)
+                            {
+                                if (likely(configuration.lookup_strategy == RangeHashedDictionaryLookupStrategy::min) && interval < *range)
+                                {
+                                    range = interval;
+                                    value_index = interval_value_index;
+                                }
+                                else if (configuration.lookup_strategy == RangeHashedDictionaryLookupStrategy::max && interval > *range)
+                                {
+                                    range = interval;
+                                    value_index = interval_value_index;
+                                }
+                            }
+                            else
+                            {
+                                range = interval;
+                                value_index = interval_value_index;
+                            }
+
+                            return true;
+                        });
+
+                    if (range.has_value())
+                    {
+                        ++keys_found;
+
+                        ValueType value = attribute_container[value_index];
+
+                        if constexpr (is_nullable)
+                        {
+                            bool is_null = (*attribute.is_value_nullable)[value_index];
+                            set_value(key_index, value, is_null);
+                        }
+                        else
+                        {
+                            set_value(key_index, value, false);
+                        }
+
+                        keys_extractor.rollbackCurrentKey();
+                        continue;
+                    }
+                }
+
+                if constexpr (is_nullable)
+                    set_value(key_index, default_value_extractor[key_index], default_value_extractor.isNullAt(key_index));
+                else
+                    set_value(key_index, default_value_extractor[key_index], false);
+
+                keys_extractor.rollbackCurrentKey();
+            }
+        });
+
+    query_count.fetch_add(keys_size, std::memory_order_relaxed);
+    found_count.fetch_add(keys_found, std::memory_order_relaxed);
+}
+
+#define INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL(DictionaryKeyType, IsNullable, ValueType) \
+    template size_t RangeHashedDictionary<DictionaryKeyType>::getItemsShortCircuitImpl<ValueType, IsNullable>( \
+        const Attribute & attribute, \
+        const Columns & key_columns, \
+        typename RangeHashedDictionary<DictionaryKeyType>::ValueSetterFunc<ValueType> && set_value, \
+        IColumn::Filter & default_mask) const;
+
+#define INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(AttributeType) \
+    INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL(DictionaryKeyType::Simple, true, DictionaryValueType<AttributeType>) \
+    INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL(DictionaryKeyType::Simple, false, DictionaryValueType<AttributeType>) \
+    INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL(DictionaryKeyType::Complex, true, DictionaryValueType<AttributeType>) \
+    INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL(DictionaryKeyType::Complex, false, DictionaryValueType<AttributeType>)
+
+CALL_FOR_ALL_DICTIONARY_ATTRIBUTE_TYPES(INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE)
+
+}
diff --git a/src/Dictionaries/RangeHashedDictionary_3.cpp b/src/Dictionaries/RangeHashedDictionary_3.cpp
new file mode 100644
index 00000000000..a3136d6f63d
--- /dev/null
+++ b/src/Dictionaries/RangeHashedDictionary_3.cpp
@@ -0,0 +1,135 @@
+#include <Dictionaries/RangeHashedDictionary.h>
+
+namespace DB
+{
+
+template <DictionaryKeyType dictionary_key_type>
+template <typename ValueType, bool is_nullable, typename DefaultValueExtractor>
+void RangeHashedDictionary<dictionary_key_type>::getItemsImpl(
+    const Attribute & attribute,
+    const Columns & key_columns,
+    typename RangeHashedDictionary<dictionary_key_type>::ValueSetterFunc<ValueType> && set_value,
+    DefaultValueExtractor & default_value_extractor) const
+{
+    const auto & attribute_container = std::get<AttributeContainerType<ValueType>>(attribute.container);
+
+    size_t keys_found = 0;
+
+    const ColumnPtr & range_column = key_columns.back();
+    auto key_columns_copy = key_columns;
+    key_columns_copy.pop_back();
+
+    DictionaryKeysArenaHolder<dictionary_key_type> arena_holder;
+    DictionaryKeysExtractor<dictionary_key_type> keys_extractor(key_columns_copy, arena_holder.getComplexKeyArena());
+    const size_t keys_size = keys_extractor.getKeysSize();
+
+    callOnRangeType(
+        dict_struct.range_min->type,
+        [&](const auto & types)
+        {
+            using Types = std::decay_t<decltype(types)>;
+            using RangeColumnType = typename Types::LeftType;
+            using RangeStorageType = typename RangeColumnType::ValueType;
+            using RangeInterval = Interval<RangeStorageType>;
+
+            const auto * range_column_typed = typeid_cast<const RangeColumnType *>(range_column.get());
+            if (!range_column_typed)
+                throw Exception(
+                    ErrorCodes::TYPE_MISMATCH,
+                    "Dictionary {} range column type should be equal to {}",
+                    getFullName(),
+                    dict_struct.range_min->type->getName());
+
+            const auto & range_column_data = range_column_typed->getData();
+
+            const auto & key_attribute_container = std::get<KeyAttributeContainerType<RangeStorageType>>(key_attribute.container);
+
+            for (size_t key_index = 0; key_index < keys_size; ++key_index)
+            {
+                auto key = keys_extractor.extractCurrentKey();
+                const auto it = key_attribute_container.find(key);
+
+                if (it)
+                {
+                    const auto date = range_column_data[key_index];
+                    const auto & interval_tree = it->getMapped();
+
+                    size_t value_index = 0;
+                    std::optional<RangeInterval> range;
+
+                    interval_tree.find(
+                        date,
+                        [&](auto & interval, auto & interval_value_index)
+                        {
+                            if (range)
+                            {
+                                if (likely(configuration.lookup_strategy == RangeHashedDictionaryLookupStrategy::min) && interval < *range)
+                                {
+                                    range = interval;
+                                    value_index = interval_value_index;
+                                }
+                                else if (configuration.lookup_strategy == RangeHashedDictionaryLookupStrategy::max && interval > *range)
+                                {
+                                    range = interval;
+                                    value_index = interval_value_index;
+                                }
+                            }
+                            else
+                            {
+                                range = interval;
+                                value_index = interval_value_index;
+                            }
+
+                            return true;
+                        });
+
+                    if (range.has_value())
+                    {
+                        ++keys_found;
+
+                        ValueType value = attribute_container[value_index];
+
+                        if constexpr (is_nullable)
+                        {
+                            bool is_null = (*attribute.is_value_nullable)[value_index];
+                            set_value(key_index, value, is_null);
+                        }
+                        else
+                        {
+                            set_value(key_index, value, false);
+                        }
+
+                        keys_extractor.rollbackCurrentKey();
+                        continue;
+                    }
+                }
+
+                if constexpr (is_nullable)
+                    set_value(key_index, default_value_extractor[key_index], default_value_extractor.isNullAt(key_index));
+                else
+                    set_value(key_index, default_value_extractor[key_index], false);
+
+                keys_extractor.rollbackCurrentKey();
+            }
+        });
+
+    query_count.fetch_add(keys_size, std::memory_order_relaxed);
+    found_count.fetch_add(keys_found, std::memory_order_relaxed);
+}
+
+#define INSTANTIATE_GET_ITEMS_IMPL(DictionaryKeyType, IsNullable, AttributeType, ValueType) \
+template void RangeHashedDictionary<DictionaryKeyType>::getItemsImpl<ValueType, IsNullable, DictionaryDefaultValueExtractor<AttributeType>>( \
+    const Attribute & attribute,\
+    const Columns & key_columns,\
+    typename RangeHashedDictionary<DictionaryKeyType>::ValueSetterFunc<ValueType> && set_value,\
+    DictionaryDefaultValueExtractor<AttributeType> & default_value_extractor) const;
+
+#define INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(AttributeType) \
+    INSTANTIATE_GET_ITEMS_IMPL(DictionaryKeyType::Simple, true, AttributeType, DictionaryValueType<AttributeType>) \
+    INSTANTIATE_GET_ITEMS_IMPL(DictionaryKeyType::Simple, false, AttributeType, DictionaryValueType<AttributeType>) \
+    INSTANTIATE_GET_ITEMS_IMPL(DictionaryKeyType::Complex, true, AttributeType, DictionaryValueType<AttributeType>) \
+    INSTANTIATE_GET_ITEMS_IMPL(DictionaryKeyType::Complex, false, AttributeType, DictionaryValueType<AttributeType>)
+
+CALL_FOR_ALL_DICTIONARY_ATTRIBUTE_TYPES(INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE)
+
+}
diff --git a/utils/check-style/check-large-objects.sh b/utils/check-style/check-large-objects.sh
index 5b0e8e88df5..5ef57ea4f6c 100755
--- a/utils/check-style/check-large-objects.sh
+++ b/utils/check-style/check-large-objects.sh
@@ -6,6 +6,7 @@ TU_EXCLUDES=(
     CastOverloadResolver
     AggregateFunctionUniq
     FunctionsConversion
+    RangeHashedDictionary_
 
     Aggregator
 )

From 85a79bc1cc8b9a374bfa2584dc4af87b24267cfd Mon Sep 17 00:00:00 2001
From: serxa <sergei@clickhouse.com>
Date: Thu, 14 Mar 2024 12:39:23 +0000
Subject: [PATCH 328/443] Fix the longest test unnecessary 3 minute wait

---
 .../01599_multiline_input_and_singleline_comments.sh  | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/tests/queries/0_stateless/01599_multiline_input_and_singleline_comments.sh b/tests/queries/0_stateless/01599_multiline_input_and_singleline_comments.sh
index 7f77f8bb403..a537dce2d92 100755
--- a/tests/queries/0_stateless/01599_multiline_input_and_singleline_comments.sh
+++ b/tests/queries/0_stateless/01599_multiline_input_and_singleline_comments.sh
@@ -1,9 +1,10 @@
 #!/usr/bin/expect -f
-# Tags: no-fasttest
-# Tag no-fasttest: 180 seconds running
 
 log_user 0
+
+# In some places `-timeout 1`` is used to avoid expect to always wait for the whole timeout
 set timeout 60
+
 match_max 100000
 
 if ![info exists env(CLICKHOUSE_PORT_TCP)] {set env(CLICKHOUSE_PORT_TCP) 9000}
@@ -13,11 +14,11 @@ expect ":) "
 
 # Make a query
 send -- "SELECT 1\r"
-expect ":-] "
+expect -timeout 1 ":-] "
 send -- "-- xxx\r"
-expect ":-] "
+expect -timeout 1 ":-] "
 send -- ", 2\r"
-expect ":-] "
+expect -timeout 1 ":-] "
 send -- ";\r"
 
 expect "│ 1 │ 2 │"

From 37913d94a3c9704775010cfc2f1d1fb9c705f7a9 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Thu, 14 Mar 2024 13:39:50 +0100
Subject: [PATCH 329/443] Merge with master

---
 src/Functions/FunctionsConversion.cpp | 18 ++++++++----------
 1 file changed, 8 insertions(+), 10 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 70cbf31bcb3..42056067f00 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -4471,20 +4471,20 @@ arguments, result_type, input_rows_count); \
 
                 if (from_low_cardinality)
                 {
-                    const auto * col_low_cardinality = typeid_cast<const ColumnLowCardinality *>(arguments[0].column.get());
+                    const auto & col_low_cardinality = typeid_cast<const ColumnLowCardinality &>(*arguments[0].column);
 
-                    if (skip_not_null_check && col_low_cardinality->containsNull())
+                    if (skip_not_null_check && col_low_cardinality.containsNull())
                         throw Exception(ErrorCodes::CANNOT_INSERT_NULL_IN_ORDINARY_COLUMN, "Cannot convert NULL value to non-Nullable type");
 
-                    arg.column = col_low_cardinality->getDictionary().getNestedColumn();
+                    arg.column = col_low_cardinality.getDictionary().getNestedColumn();
                     arg.type = from_low_cardinality->getDictionaryType();
 
                     /// TODO: Make map with defaults conversion.
                     src_converted_to_full_column = !removeNullable(arg.type)->equals(*removeNullable(res_type));
                     if (src_converted_to_full_column)
-                        arg.column = arg.column->index(col_low_cardinality->getIndexes(), 0);
+                        arg.column = arg.column->index(col_low_cardinality.getIndexes(), 0);
                     else
-                        res_indexes = col_low_cardinality->getIndexesPtr();
+                        res_indexes = col_low_cardinality.getIndexesPtr();
 
                     tmp_rows_count = arg.column->size();
                 }
@@ -4496,14 +4496,12 @@ arguments, result_type, input_rows_count); \
             if (to_low_cardinality)
             {
                 auto res_column = to_low_cardinality->createColumn();
-                auto * col_low_cardinality = typeid_cast<ColumnLowCardinality *>(res_column.get());
+                auto & col_low_cardinality = typeid_cast<ColumnLowCardinality &>(*res_column);
 
                 if (from_low_cardinality && !src_converted_to_full_column)
-                {
-                    col_low_cardinality->insertRangeFromDictionaryEncodedColumn(*converted_column, *res_indexes);
-                }
+                    col_low_cardinality.insertRangeFromDictionaryEncodedColumn(*converted_column, *res_indexes);
                 else
-                    col_low_cardinality->insertRangeFromFullColumn(*converted_column, 0, converted_column->size());
+                    col_low_cardinality.insertRangeFromFullColumn(*converted_column, 0, converted_column->size());
 
                 return res_column;
             }

From 107acf54c609147acc6bf84407f168e274079505 Mon Sep 17 00:00:00 2001
From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com>
Date: Thu, 14 Mar 2024 13:42:07 +0100
Subject: [PATCH 330/443] Fix tests

---
 tests/integration/test_storage_s3/test.py | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/tests/integration/test_storage_s3/test.py b/tests/integration/test_storage_s3/test.py
index dbbe670e8ca..6d5b84a8143 100644
--- a/tests/integration/test_storage_s3/test.py
+++ b/tests/integration/test_storage_s3/test.py
@@ -1414,10 +1414,10 @@ def test_signatures(started_cluster):
     )
     assert int(result) == 1
 
-    result = instance.query(
+    error = instance.query_and_get_error(
         f"select * from s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test.arrow', 'minio', 'minio123', '{session_token}')"
     )
-    assert int(result) == 1
+    assert "S3_ERROR" in error
 
     result = instance.query(
         f"select * from s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test.arrow', 'Arrow', 'x UInt64', 'auto')"
@@ -1429,20 +1429,20 @@ def test_signatures(started_cluster):
     )
     assert int(result) == 1
 
-    result = instance.query(
+    error = instance.query_and_get_error(
         f"select * from s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test.arrow', 'minio', 'minio123', '{session_token}', 'Arrow')"
     )
-    assert int(result) == 1
+    assert "S3_ERROR" in error
 
-    lt = instance.query(
+    error = instance.query_and_get_error(
         f"select * from s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test.arrow', 'minio', 'minio123', '{session_token}', 'Arrow', 'x UInt64')"
     )
-    assert int(result) == 1
+    assert "S3_ERROR" in error
 
-    lt = instance.query(
+    error = instance.query_and_get_error(
         f"select * from s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test.arrow', 'minio', 'minio123', '{session_token}', 'Arrow', 'x UInt64', 'auto')"
     )
-    assert int(result) == 1
+    assert "S3_ERROR" in error
 
 
 def test_select_columns(started_cluster):

From 047fb87a9ae20dd57ba370a277cfd4a43411876d Mon Sep 17 00:00:00 2001
From: serxa <sergei@clickhouse.com>
Date: Thu, 14 Mar 2024 12:46:45 +0000
Subject: [PATCH 331/443] typo

---
 .../01599_multiline_input_and_singleline_comments.sh            | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/queries/0_stateless/01599_multiline_input_and_singleline_comments.sh b/tests/queries/0_stateless/01599_multiline_input_and_singleline_comments.sh
index a537dce2d92..f1acd39136f 100755
--- a/tests/queries/0_stateless/01599_multiline_input_and_singleline_comments.sh
+++ b/tests/queries/0_stateless/01599_multiline_input_and_singleline_comments.sh
@@ -2,7 +2,7 @@
 
 log_user 0
 
-# In some places `-timeout 1`` is used to avoid expect to always wait for the whole timeout
+# In some places `-timeout 1` is used to avoid expect to always wait for the whole timeout
 set timeout 60
 
 match_max 100000

From bf9da768bfb090dec9b366d97240a2949b0e150e Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Thu, 14 Mar 2024 13:53:40 +0100
Subject: [PATCH 332/443] More detailed explanation

---
 docs/en/operations/storing-data.md | 42 +++++++++++++++++++++++++-----
 1 file changed, 36 insertions(+), 6 deletions(-)

diff --git a/docs/en/operations/storing-data.md b/docs/en/operations/storing-data.md
index 4f676904375..fd81bc197d1 100644
--- a/docs/en/operations/storing-data.md
+++ b/docs/en/operations/storing-data.md
@@ -10,7 +10,11 @@ Data, processed in ClickHouse, is usually stored in the local file system — on
 2. The Hadoop Distributed File System ([HDFS](https://hadoop.apache.org/docs/current/hadoop-project-dist/hadoop-hdfs/HdfsDesign.html))
 3. [Azure Blob Storage](https://azure.microsoft.com/en-us/products/storage/blobs).
 
-Note: to work with data stored on `Amazon S3` disks use [S3](/docs/en/engines/table-engines/integrations/s3.md) table engine, to work with data in the Hadoop Distributed File System — [HDFS](/docs/en/engines/table-engines/integrations/hdfs.md) table engine, and to work with data stored in Azure Blob Storage use [AzureBlobStorage](/docs/en/engines/table-engines/integrations/azureBlobStorage.md) table engine. They are different from external storage described on this page as they allow to read data stored in some general file format (like Parquet), while on this page we are describing storage configuration for ClickHouse `MergeTree` family or `Log` family tables.
+:::note ClickHouse also has support for external table engines, which are different from external storage option described on this page as they allow to read data stored in some general file format (like Parquet), while on this page we are describing storage configuration for ClickHouse `MergeTree` family or `Log` family tables.
+1. to work with data stored on `Amazon S3` disks, use [S3](/docs/en/engines/table-engines/integrations/s3.md) table engine.
+2. to work with data in the Hadoop Distributed File System — [HDFS](/docs/en/engines/table-engines/integrations/hdfs.md) table engine.
+3. to work with data stored in Azure Blob Storage use [AzureBlobStorage](/docs/en/engines/table-engines/integrations/azureBlobStorage.md) table engine.
+:::
 
 ## Configuring external storage {#configuring-external-storage}
 
@@ -23,8 +27,9 @@ Disk configuration requires:
 Starting from 24.1 clickhouse version, it is possible to use a new configuration option.
 It requires to specify:
 1. `type` equal to `object_storage`
-2. `object_storage_type`, equal to one of `s3`, `azure_blob_storage`, `hdfs`, `local_blob_storage`, `web`.
+2. `object_storage_type`, equal to one of `s3`, `azure_blob_storage` (or just `azure` from `24.3`), `hdfs`, `local_blob_storage` (or just `local` from `24.3`), `web`.
 Optionally, `metadata_type` can be specified (it is equal to `local` by default), but it can also be set to `plain`, `web`.
+Usage of `plain` metadata type is described in [plain storage section](/docs/en/operations/storing-data.md/#storing-data-on-webserver), `web` metadata type can be used only with `web` object storage type, `local` metadata type stores metadata files locally (each metadata files contains mapping to files in object storage and some additional meta information about them).
 
 E.g. configuration option
 ``` xml
@@ -143,7 +148,7 @@ SETTINGS disk = 's3';
 
 ## Dynamic Configuration {#dynamic-configuration}
 
-There is also a possibility to specify storage configuration without a predefined disk in configuration in a configuration file, but can be configured in the CREATE/ATTACH query settings.
+There is also a possibility to specify storage configuration without a predefined disk in configuration in a configuration file, but can be configured in the `CREATE`/`ATTACH` query settings.
 
 The following example query builds on the above dynamic disk configuration and shows how to use a local disk to cache data from a table stored at a URL.
 
@@ -306,10 +311,35 @@ Optional parameters:
 Google Cloud Storage (GCS) is also supported using the type `s3`. See [GCS backed MergeTree](/docs/en/integrations/gcs).
 :::
 
-### Using Plain Storage {#s3-storage}
+### Using Plain Storage {#plain-storage}
 
-There is a disk type `s3_plain`, which provides a write-once storage. Unlike `s3` disk type, it stores data as is, e.g. instead of randomly-generated blob names, it uses normal file names as clickhouse stores files on local disk. So this disk type allows to keeper a static version of the table and can also be used to create backups on it.
-Configuration parameters are the same as for `s3` disk type.
+In `22.10` a new disk type `s3_plain` was introduced, which provides a write-once storage. Configuration parameters are the same as for `s3` disk type.
+Unlike `s3` disk type, it stores data as is, e.g. instead of randomly-generated blob names, it uses normal file names (the same way as clickhouse stores files on local disk) and does not store any metadata locally, e.g. it is derived from data on `s3`.
+
+This disk type allows to keep a static version of the table, as it does not allow executing merges on the existing data and does not allow inserting of new data.
+A use case for this disk type is to create backups on it, which can be done via `BACKUP TABLE data TO Disk('plain_disk_name', 'backup_name')`. Afterwards you can do `RESTORE TABLE data AS data_restored FROM Disk('plain_disk_name', 'backup_name')` or using `ATTACH TABLE data (...) ENGINE = MergeTree() SETTINGS disk = 'plain_disk_name'`.
+
+Configuration:
+``` xml
+<s3_plain>
+    <type>s3_plain</type>
+    <endpoint>https://s3.eu-west-1.amazonaws.com/clickhouse-eu-west-1.clickhouse.com/data/</endpoint>
+    <use_invironment_credentials>1</use_invironment_credentials>
+</s3_plain>
+```
+
+Starting from `24.1` it is possible configure any object storage disk (`s3`, `azure`, `hdfs`, `local`) using `plain` metadata type.
+
+Configuration:
+``` xml
+<s3_plain>
+    <type>object_storage</type>
+    <object_storage_type>azure</object_storage_type>
+    <metadata_type>plain</metadata_type>
+    <endpoint>https://s3.eu-west-1.amazonaws.com/clickhouse-eu-west-1.clickhouse.com/data/</endpoint>
+    <use_invironment_credentials>1</use_invironment_credentials>
+</s3_plain>
+```
 
 ### Using Azure Blob Storage {#azure-blob-storage}
 

From c3aefb28e5eb11c171700dfe439c2371c6a3825e Mon Sep 17 00:00:00 2001
From: "Mikhail f. Shiryaev" <felixoid@clickhouse.com>
Date: Thu, 14 Mar 2024 14:13:15 +0100
Subject: [PATCH 333/443] Remove unnecessary layers from clickhouse/cctools

---
 docker/packager/cctools/Dockerfile | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/docker/packager/cctools/Dockerfile b/docker/packager/cctools/Dockerfile
index 1b8c675a5c5..d986c6a3c86 100644
--- a/docker/packager/cctools/Dockerfile
+++ b/docker/packager/cctools/Dockerfile
@@ -2,7 +2,7 @@
 # It's based on the assumption that we don't care of the cctools version so much
 # It event does not depend on the clickhouse/fasttest in the `docker/images.json`
 ARG FROM_TAG=latest
-FROM clickhouse/fasttest:$FROM_TAG
+FROM clickhouse/fasttest:$FROM_TAG as builder
 
 ENV CC=clang-${LLVM_VERSION}
 ENV CXX=clang++-${LLVM_VERSION}
@@ -29,3 +29,6 @@ RUN git clone https://github.com/tpoechtrager/cctools-port.git \
     && make install -j$(nproc) \
     && cd ../.. \
     && rm -rf cctools-port
+
+FROM scratch
+COPY --from=builder /cctools /cctools

From 9067c1ab9292de5064a3cb8547557798ace4ac99 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Thu, 14 Mar 2024 14:14:24 +0100
Subject: [PATCH 334/443] Merge with master

---
 src/Functions/FunctionsConversion.h | 4990 ---------------------------
 1 file changed, 4990 deletions(-)
 delete mode 100644 src/Functions/FunctionsConversion.h

diff --git a/src/Functions/FunctionsConversion.h b/src/Functions/FunctionsConversion.h
deleted file mode 100644
index f338af28240..00000000000
--- a/src/Functions/FunctionsConversion.h
+++ /dev/null
@@ -1,4990 +0,0 @@
-#pragma once
-
-#include <cstddef>
-#include <type_traits>
-
-#include <IO/WriteBufferFromVector.h>
-#include <IO/ReadBufferFromMemory.h>
-#include <IO/Operators.h>
-#include <IO/parseDateTimeBestEffort.h>
-#include <DataTypes/DataTypeFactory.h>
-#include <DataTypes/DataTypesNumber.h>
-#include <DataTypes/DataTypesDecimal.h>
-#include <DataTypes/DataTypeString.h>
-#include <DataTypes/DataTypeFixedString.h>
-#include <DataTypes/DataTypeDate.h>
-#include <DataTypes/DataTypeDate32.h>
-#include <DataTypes/DataTypeDateTime.h>
-#include <DataTypes/DataTypeDateTime64.h>
-#include <DataTypes/DataTypeEnum.h>
-#include <DataTypes/DataTypeArray.h>
-#include <DataTypes/DataTypeTuple.h>
-#include <DataTypes/DataTypeMap.h>
-#include <DataTypes/DataTypeNullable.h>
-#include <DataTypes/DataTypeNothing.h>
-#include <DataTypes/DataTypeUUID.h>
-#include <DataTypes/DataTypeInterval.h>
-#include <DataTypes/DataTypeAggregateFunction.h>
-#include <DataTypes/DataTypeObject.h>
-#include <DataTypes/ObjectUtils.h>
-#include <DataTypes/DataTypeNested.h>
-#include <DataTypes/DataTypeVariant.h>
-#include <DataTypes/Serializations/SerializationDecimal.h>
-#include <Formats/FormatSettings.h>
-#include <Columns/ColumnString.h>
-#include <Columns/ColumnFixedString.h>
-#include <Columns/ColumnConst.h>
-#include <Columns/ColumnAggregateFunction.h>
-#include <Columns/ColumnArray.h>
-#include <Columns/ColumnNullable.h>
-#include <Columns/ColumnTuple.h>
-#include <Columns/ColumnMap.h>
-#include <Columns/ColumnObject.h>
-#include <Columns/ColumnsCommon.h>
-#include <Columns/ColumnVariant.h>
-#include <Columns/ColumnStringHelpers.h>
-#include <Common/assert_cast.h>
-#include <Common/Concepts.h>
-#include <Common/quoteString.h>
-#include <Common/Exception.h>
-#include <Core/AccurateComparison.h>
-#include <Functions/IFunctionAdaptors.h>
-#include <Functions/FunctionHelpers.h>
-#include <Functions/DateTimeTransforms.h>
-#include <Functions/toFixedString.h>
-#include <Functions/TransformDateTime64.h>
-#include <Functions/FunctionsCodingIP.h>
-#include <Functions/CastOverloadResolver.h>
-#include <DataTypes/DataTypeLowCardinality.h>
-#include <Columns/ColumnLowCardinality.h>
-#include <Interpreters/Context.h>
-#include <Common/HashTable/HashMap.h>
-#include <DataTypes/DataTypeIPv4andIPv6.h>
-#include <Common/IPv6ToBinary.h>
-#include <Core/Types.h>
-
-
-namespace DB
-{
-
-namespace ErrorCodes
-{
-    extern const int ATTEMPT_TO_READ_AFTER_EOF;
-    extern const int CANNOT_PARSE_NUMBER;
-    extern const int CANNOT_READ_ARRAY_FROM_TEXT;
-    extern const int CANNOT_PARSE_INPUT_ASSERTION_FAILED;
-    extern const int CANNOT_PARSE_QUOTED_STRING;
-    extern const int CANNOT_PARSE_ESCAPE_SEQUENCE;
-    extern const int CANNOT_PARSE_DATE;
-    extern const int CANNOT_PARSE_DATETIME;
-    extern const int CANNOT_PARSE_TEXT;
-    extern const int CANNOT_PARSE_UUID;
-    extern const int CANNOT_PARSE_IPV4;
-    extern const int CANNOT_PARSE_IPV6;
-    extern const int TOO_FEW_ARGUMENTS_FOR_FUNCTION;
-    extern const int LOGICAL_ERROR;
-    extern const int TYPE_MISMATCH;
-    extern const int CANNOT_CONVERT_TYPE;
-    extern const int ILLEGAL_COLUMN;
-    extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH;
-    extern const int ILLEGAL_TYPE_OF_ARGUMENT;
-    extern const int NOT_IMPLEMENTED;
-    extern const int CANNOT_INSERT_NULL_IN_ORDINARY_COLUMN;
-    extern const int CANNOT_PARSE_BOOL;
-    extern const int VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE;
-}
-
-/** Type conversion functions.
-  * toType - conversion in "natural way";
-  */
-
-inline UInt32 extractToDecimalScale(const ColumnWithTypeAndName & named_column)
-{
-    const auto * arg_type = named_column.type.get();
-    bool ok = checkAndGetDataType<DataTypeUInt64>(arg_type)
-        || checkAndGetDataType<DataTypeUInt32>(arg_type)
-        || checkAndGetDataType<DataTypeUInt16>(arg_type)
-        || checkAndGetDataType<DataTypeUInt8>(arg_type);
-    if (!ok)
-        throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type of toDecimal() scale {}", named_column.type->getName());
-
-    Field field;
-    named_column.column->get(0, field);
-    return static_cast<UInt32>(field.get<UInt32>());
-}
-
-/// Function toUnixTimestamp has exactly the same implementation as toDateTime of String type.
-struct NameToUnixTimestamp { static constexpr auto name = "toUnixTimestamp"; };
-
-struct AccurateConvertStrategyAdditions
-{
-    UInt32 scale { 0 };
-};
-
-struct AccurateOrNullConvertStrategyAdditions
-{
-    UInt32 scale { 0 };
-};
-
-
-struct ConvertDefaultBehaviorTag {};
-struct ConvertReturnNullOnErrorTag {};
-struct ConvertReturnZeroOnErrorTag {};
-
-/** Conversion of number types to each other, enums to numbers, dates and datetimes to numbers and back: done by straight assignment.
-  *  (Date is represented internally as number of days from some day; DateTime - as unix timestamp)
-  */
-template <typename FromDataType, typename ToDataType, typename Name,
-    typename SpecialTag = ConvertDefaultBehaviorTag,
-    FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior = default_date_time_overflow_behavior>
-struct ConvertImpl
-{
-    using FromFieldType = typename FromDataType::FieldType;
-    using ToFieldType = typename ToDataType::FieldType;
-
-    template <typename Additions = void *>
-    static ColumnPtr NO_SANITIZE_UNDEFINED execute(
-        const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type [[maybe_unused]], size_t input_rows_count,
-        Additions additions [[maybe_unused]] = Additions())
-    {
-        const ColumnWithTypeAndName & named_from = arguments[0];
-
-        using ColVecFrom = typename FromDataType::ColumnType;
-        using ColVecTo = typename ToDataType::ColumnType;
-
-        if constexpr ((IsDataTypeDecimal<FromDataType> || IsDataTypeDecimal<ToDataType>)
-            && !(std::is_same_v<DataTypeDateTime64, FromDataType> || std::is_same_v<DataTypeDateTime64, ToDataType>))
-        {
-            if constexpr (!IsDataTypeDecimalOrNumber<FromDataType> || !IsDataTypeDecimalOrNumber<ToDataType>)
-            {
-                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                    named_from.column->getName(), Name::name);
-            }
-        }
-
-        if (const ColVecFrom * col_from = checkAndGetColumn<ColVecFrom>(named_from.column.get()))
-        {
-            typename ColVecTo::MutablePtr col_to = nullptr;
-
-            if constexpr (IsDataTypeDecimal<ToDataType>)
-            {
-                UInt32 scale;
-
-                if constexpr (std::is_same_v<Additions, AccurateConvertStrategyAdditions>
-                    || std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-                {
-                    scale = additions.scale;
-                }
-                else
-                {
-                    scale = additions;
-                }
-
-                col_to = ColVecTo::create(0, scale);
-            }
-            else
-                col_to = ColVecTo::create();
-
-            const auto & vec_from = col_from->getData();
-            auto & vec_to = col_to->getData();
-            vec_to.resize(input_rows_count);
-
-            ColumnUInt8::MutablePtr col_null_map_to;
-            ColumnUInt8::Container * vec_null_map_to [[maybe_unused]] = nullptr;
-            if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-            {
-                col_null_map_to = ColumnUInt8::create(input_rows_count, false);
-                vec_null_map_to = &col_null_map_to->getData();
-            }
-
-            bool result_is_bool = isBool(result_type);
-            for (size_t i = 0; i < input_rows_count; ++i)
-            {
-                if constexpr (std::is_same_v<ToDataType, DataTypeUInt8>)
-                {
-                    if (result_is_bool)
-                    {
-                        vec_to[i] = vec_from[i] != FromFieldType(0);
-                        continue;
-                    }
-                }
-
-                if constexpr (std::is_same_v<FromDataType, DataTypeUUID> && std::is_same_v<ToDataType, DataTypeUInt128>)
-                {
-                    static_assert(
-                        std::is_same_v<DataTypeUInt128::FieldType, DataTypeUUID::FieldType::UnderlyingType>,
-                        "UInt128 and UUID types must be same");
-
-                    vec_to[i].items[1] = vec_from[i].toUnderType().items[0];
-                    vec_to[i].items[0] = vec_from[i].toUnderType().items[1];
-
-                    continue;
-                }
-
-                if constexpr (std::is_same_v<FromDataType, DataTypeIPv6> && std::is_same_v<ToDataType, DataTypeUInt128>)
-                {
-                    static_assert(
-                        std::is_same_v<DataTypeUInt128::FieldType, DataTypeUUID::FieldType::UnderlyingType>,
-                        "UInt128 and IPv6 types must be same");
-
-                    vec_to[i].items[1] = std::byteswap(vec_from[i].toUnderType().items[0]);
-                    vec_to[i].items[0] = std::byteswap(vec_from[i].toUnderType().items[1]);
-
-                    continue;
-                }
-
-                if constexpr (std::is_same_v<FromDataType, DataTypeUUID> != std::is_same_v<ToDataType, DataTypeUUID>)
-                {
-                    throw Exception(ErrorCodes::NOT_IMPLEMENTED,
-                                    "Conversion between numeric types and UUID is not supported. "
-                                    "Probably the passed UUID is unquoted");
-                }
-                else if constexpr (
-                    (std::is_same_v<FromDataType, DataTypeIPv4> != std::is_same_v<ToDataType, DataTypeIPv4>)
-                    && !(is_any_of<FromDataType, DataTypeUInt8, DataTypeUInt16, DataTypeUInt32, DataTypeUInt64, DataTypeIPv6> || is_any_of<ToDataType, DataTypeUInt32, DataTypeUInt64, DataTypeUInt128, DataTypeUInt256, DataTypeIPv6>)
-                )
-                {
-                    throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Conversion from {} to {} is not supported",
-                                    TypeName<typename FromDataType::FieldType>, TypeName<typename ToDataType::FieldType>);
-                }
-                else if constexpr (std::is_same_v<FromDataType, DataTypeIPv6> != std::is_same_v<ToDataType, DataTypeIPv6> && !(std::is_same_v<ToDataType, DataTypeIPv4> || std::is_same_v<FromDataType, DataTypeIPv4>))
-                {
-                    throw Exception(ErrorCodes::NOT_IMPLEMENTED,
-                                    "Conversion between numeric types and IPv6 is not supported. "
-                                    "Probably the passed IPv6 is unquoted");
-                }
-                else
-                {
-                    if constexpr (IsDataTypeDecimal<FromDataType> || IsDataTypeDecimal<ToDataType>)
-                    {
-                        if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-                        {
-                            ToFieldType result;
-                            bool convert_result = false;
-
-                            if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeDecimal<ToDataType>)
-                                convert_result = tryConvertDecimals<FromDataType, ToDataType>(vec_from[i], col_from->getScale(), col_to->getScale(), result);
-                            else if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeNumber<ToDataType>)
-                                convert_result = tryConvertFromDecimal<FromDataType, ToDataType>(vec_from[i], col_from->getScale(), result);
-                            else if constexpr (IsDataTypeNumber<FromDataType> && IsDataTypeDecimal<ToDataType>)
-                                convert_result = tryConvertToDecimal<FromDataType, ToDataType>(vec_from[i], col_to->getScale(), result);
-
-                            if (convert_result)
-                                vec_to[i] = result;
-                            else
-                            {
-                                vec_to[i] = static_cast<ToFieldType>(0);
-                                (*vec_null_map_to)[i] = true;
-                            }
-                        }
-                        else
-                        {
-                            if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeDecimal<ToDataType>)
-                                vec_to[i] = convertDecimals<FromDataType, ToDataType>(vec_from[i], col_from->getScale(), col_to->getScale());
-                            else if constexpr (IsDataTypeDecimal<FromDataType> && IsDataTypeNumber<ToDataType>)
-                                vec_to[i] = convertFromDecimal<FromDataType, ToDataType>(vec_from[i], col_from->getScale());
-                            else if constexpr (IsDataTypeNumber<FromDataType> && IsDataTypeDecimal<ToDataType>)
-                                vec_to[i] = convertToDecimal<FromDataType, ToDataType>(vec_from[i], col_to->getScale());
-                            else
-                                throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Unsupported data type in conversion function");
-                        }
-                    }
-                    else
-                    {
-                        /// If From Data is Nan or Inf and we convert to integer type, throw exception
-                        if constexpr (std::is_floating_point_v<FromFieldType> && !std::is_floating_point_v<ToFieldType>)
-                        {
-                            if (!isFinite(vec_from[i]))
-                            {
-                                if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-                                {
-                                    vec_to[i] = 0;
-                                    (*vec_null_map_to)[i] = true;
-                                    continue;
-                                }
-                                else
-                                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Unexpected inf or nan to integer conversion");
-                            }
-                        }
-
-                        if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>
-                                || std::is_same_v<Additions, AccurateConvertStrategyAdditions>)
-                        {
-                            bool convert_result = accurate::convertNumeric(vec_from[i], vec_to[i]);
-
-                            if (!convert_result)
-                            {
-                                if (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-                                {
-                                    vec_to[i] = 0;
-                                    (*vec_null_map_to)[i] = true;
-                                }
-                                else
-                                {
-                                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Value in column {} cannot be safely converted into type {}",
-                                        named_from.column->getName(), result_type->getName());
-                                }
-                            }
-                        }
-                        else
-                        {
-                            if constexpr (std::is_same_v<ToDataType, DataTypeIPv4> && std::is_same_v<FromDataType, DataTypeIPv6>)
-                            {
-                                const uint8_t ip4_cidr[] {0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x00, 0x00, 0x00, 0x00};
-                                const uint8_t * src = reinterpret_cast<const uint8_t *>(&vec_from[i].toUnderType());
-                                if (!matchIPv6Subnet(src, ip4_cidr, 96))
-                                {
-                                    char addr[IPV6_MAX_TEXT_LENGTH + 1] {};
-                                    char * paddr = addr;
-                                    formatIPv6(src, paddr);
-
-                                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "IPv6 {} in column {} is not in IPv4 mapping block", addr, named_from.column->getName());
-                                }
-
-                                uint8_t * dst = reinterpret_cast<uint8_t *>(&vec_to[i].toUnderType());
-                                if constexpr (std::endian::native == std::endian::little)
-                                {
-                                    dst[0] = src[15];
-                                    dst[1] = src[14];
-                                    dst[2] = src[13];
-                                    dst[3] = src[12];
-                                }
-                                else
-                                {
-                                    dst[0] = src[12];
-                                    dst[1] = src[13];
-                                    dst[2] = src[14];
-                                    dst[3] = src[15];
-                                }
-                            }
-                            else if constexpr (std::is_same_v<ToDataType, DataTypeIPv6> && std::is_same_v<FromDataType, DataTypeIPv4>)
-                            {
-                                const uint8_t * src = reinterpret_cast<const uint8_t *>(&vec_from[i].toUnderType());
-                                uint8_t * dst = reinterpret_cast<uint8_t *>(&vec_to[i].toUnderType());
-                                std::memset(dst, '\0', IPV6_BINARY_LENGTH);
-                                dst[10] = dst[11] = 0xff;
-
-                                if constexpr (std::endian::native == std::endian::little)
-                                {
-                                    dst[12] = src[3];
-                                    dst[13] = src[2];
-                                    dst[14] = src[1];
-                                    dst[15] = src[0];
-                                }
-                                else
-                                {
-                                    dst[12] = src[0];
-                                    dst[13] = src[1];
-                                    dst[14] = src[2];
-                                    dst[15] = src[3];
-                                }
-                            }
-                            else if constexpr (std::is_same_v<ToDataType, DataTypeIPv4> && std::is_same_v<FromDataType, DataTypeUInt64>)
-                                vec_to[i] = static_cast<ToFieldType>(static_cast<IPv4::UnderlyingType>(vec_from[i]));
-                            else if constexpr (std::is_same_v<Name, NameToUnixTimestamp> && (std::is_same_v<FromDataType, DataTypeDate> || std::is_same_v<FromDataType, DataTypeDate32>))
-                                vec_to[i] = static_cast<ToFieldType>(vec_from[i] * DATE_SECONDS_PER_DAY);
-                            else
-                                vec_to[i] = static_cast<ToFieldType>(vec_from[i]);
-                        }
-                    }
-                }
-            }
-
-            if constexpr (std::is_same_v<Additions, AccurateOrNullConvertStrategyAdditions>)
-                return ColumnNullable::create(std::move(col_to), std::move(col_null_map_to));
-            else
-                return col_to;
-        }
-        else
-            throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                named_from.column->getName(), Name::name);
-    }
-};
-
-/** Conversion of DateTime to Date: throw off time component.
-  */
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDateTime, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDateTime, DataTypeDate, ToDateImpl<date_time_overflow_behavior>, false> {};
-
-/** Conversion of DateTime to Date32: throw off time component.
-  */
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDateTime, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDateTime, DataTypeDate32, ToDate32Impl, false> {};
-
-/** Conversion of Date to DateTime: adding 00:00:00 time component.
-  */
-template <FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior = default_date_time_overflow_behavior>
-struct ToDateTimeImpl
-{
-    static constexpr auto name = "toDateTime";
-
-    static UInt32 execute(UInt16 d, const DateLUTImpl & time_zone)
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (d > MAX_DATETIME_DAY_NUM) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Day number {} is out of bounds of type DateTime", d);
-        }
-        else if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Saturate)
-        {
-            if (d > MAX_DATETIME_DAY_NUM)
-                d = MAX_DATETIME_DAY_NUM;
-        }
-        return static_cast<UInt32>(time_zone.fromDayNum(DayNum(d)));
-    }
-
-    static UInt32 execute(Int32 d, const DateLUTImpl & time_zone)
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Saturate)
-        {
-            if (d < 0)
-                return 0;
-            else if (d > MAX_DATETIME_DAY_NUM)
-                d = MAX_DATETIME_DAY_NUM;
-        }
-        else if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (d < 0 || d > MAX_DATETIME_DAY_NUM) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type DateTime", d);
-        }
-        return static_cast<UInt32>(time_zone.fromDayNum(ExtendedDayNum(d)));
-    }
-
-    static UInt32 execute(UInt32 dt, const DateLUTImpl & /*time_zone*/)
-    {
-        return dt;
-    }
-
-    static UInt32 execute(Int64 dt64, const DateLUTImpl & /*time_zone*/)
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Ignore)
-            return static_cast<UInt32>(dt64);
-        else
-        {
-            if (dt64 < 0 || dt64 >= MAX_DATETIME_TIMESTAMP)
-            {
-                if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Saturate)
-                    return dt64 < 0 ? 0 : std::numeric_limits<UInt32>::max();
-                else
-                    throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type DateTime", dt64);
-            }
-            else
-                return static_cast<UInt32>(dt64);
-        }
-    }
-};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDate, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDate, DataTypeDateTime, ToDateTimeImpl<date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDate32, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDate32, DataTypeDateTime, ToDateTimeImpl<date_time_overflow_behavior>, false> {};
-
-/// Implementation of toDate function.
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTransform32Or64
-{
-    static constexpr auto name = "toDate";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from > MAX_DATETIME_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type Date", from);
-        }
-        /// if value is smaller (or equal) than maximum day value for Date, than treat it as day num,
-        /// otherwise treat it as unix timestamp. This is a bit weird, but we leave this behavior.
-        if (from <= DATE_LUT_MAX_DAY_NUM)
-            return from;
-        else
-            return time_zone.toDayNum(std::min(time_t(from), time_t(MAX_DATETIME_TIMESTAMP)));
-    }
-};
-
-/** Conversion of Date32 to Date.
-  */
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDate32, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDate32, DataTypeDate, ToDateImpl<date_time_overflow_behavior>, false> {};
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTransform32Or64Signed
-{
-    static constexpr auto name = "toDate";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
-    {
-        // TODO: decide narrow or extended range based on FromType
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from < 0 || from > MAX_DATE_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type Date", from);
-        }
-        else
-        {
-            if (from < 0)
-                return 0;
-        }
-        return (from <= DATE_LUT_MAX_DAY_NUM)
-            ? static_cast<ToType>(from)
-            : time_zone.toDayNum(std::min(time_t(from), time_t(MAX_DATE_TIMESTAMP)));
-    }
-};
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTransform8Or16Signed
-{
-    static constexpr auto name = "toDate";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
-    {
-        if (from < 0)
-        {
-            if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Value {} is out of bounds of type Date", from);
-            else
-                return 0;
-        }
-        return from;
-    }
-};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDateTime64, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-        : DateTimeTransformImpl<DataTypeDateTime64, DataTypeDate32, TransformDateTime64<ToDate32Impl>, false> {};
-
-/// Implementation of toDate32 function.
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDate32Transform32Or64
-{
-    static constexpr auto name = "toDate32";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
-    {
-        if (from < DATE_LUT_MAX_EXTEND_DAY_NUM)
-            return static_cast<ToType>(from);
-        else
-        {
-            if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-            {
-                if (from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
-                    throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type Date32", from);
-            }
-            return time_zone.toDayNum(std::min(time_t(from), time_t(MAX_DATETIME64_TIMESTAMP)));
-        }
-    }
-};
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDate32Transform32Or64Signed
-{
-    static constexpr auto name = "toDate32";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
-    {
-        static const Int32 daynum_min_offset = -static_cast<Int32>(time_zone.getDayNumOffsetEpoch());
-
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from < daynum_min_offset || from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type Date32", from);
-        }
-
-        if (from < daynum_min_offset)
-            return daynum_min_offset;
-
-        return (from < DATE_LUT_MAX_EXTEND_DAY_NUM)
-            ? static_cast<ToType>(from)
-            : time_zone.toDayNum(std::min(time_t(Int64(from)), time_t(MAX_DATETIME64_TIMESTAMP)));
-    }
-};
-
-template <typename FromType, typename ToType>
-struct ToDate32Transform8Or16Signed
-{
-    static constexpr auto name = "toDate32";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
-    {
-        return from;
-    }
-};
-
-/** Special case of converting Int8, Int16, (U)Int32 or (U)Int64 (and also, for convenience,
-  * Float32, Float64) to Date. If the
-  * number is less than 65536, then it is treated as DayNum, and if it's greater or equals to 65536,
-  * then treated as unix timestamp. If the number exceeds UInt32, saturate to MAX_UINT32 then as DayNum.
-  * It's a bit illogical, as we actually have two functions in one.
-  * But allows to support frequent case,
-  *  when user write toDate(UInt32), expecting conversion of unix timestamp to Date.
-  *  (otherwise such usage would be frequent mistake).
-  */
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeUInt32, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeUInt32, DataTypeDate, ToDateTransform32Or64<UInt32, UInt16, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeUInt64, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeUInt64, DataTypeDate, ToDateTransform32Or64<UInt64, UInt16, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt8, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt8, DataTypeDate, ToDateTransform8Or16Signed<Int8, UInt16, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt16, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt16, DataTypeDate, ToDateTransform8Or16Signed<Int16, UInt16, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt32, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt32, DataTypeDate, ToDateTransform32Or64Signed<Int32, UInt16, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt64, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt64, DataTypeDate, ToDateTransform32Or64Signed<Int64, UInt16, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat32, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat32, DataTypeDate, ToDateTransform32Or64Signed<Float32, UInt16, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat64, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat64, DataTypeDate, ToDateTransform32Or64Signed<Float64, UInt16, default_date_time_overflow_behavior>, false> {};
-
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeUInt32, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeUInt32, DataTypeDate32, ToDate32Transform32Or64<UInt32, Int32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeUInt64, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeUInt64, DataTypeDate32, ToDate32Transform32Or64<UInt64, Int32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt8, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt8, DataTypeDate32, ToDate32Transform8Or16Signed<Int8, Int32>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt16, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt16, DataTypeDate32, ToDate32Transform8Or16Signed<Int16, Int32>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt32, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt32, DataTypeDate32, ToDate32Transform32Or64Signed<Int32, Int32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt64, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt64, DataTypeDate32, ToDate32Transform32Or64Signed<Int64, Int32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat32, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat32, DataTypeDate32, ToDate32Transform32Or64Signed<Float32, Int32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat64, DataTypeDate32, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat64, DataTypeDate32, ToDate32Transform32Or64Signed<Float64, Int32, default_date_time_overflow_behavior>, false> {};
-
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTimeTransform64
-{
-    static constexpr auto name = "toDateTime";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from > MAX_DATETIME_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime", from);
-        }
-        return static_cast<ToType>(std::min(time_t(from), time_t(MAX_DATETIME_TIMESTAMP)));
-    }
-};
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTimeTransformSigned
-{
-    static constexpr auto name = "toDateTime";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
-    {
-        if (from < 0)
-        {
-            if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime", from);
-            else
-                return 0;
-        }
-        return from;
-    }
-};
-
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTimeTransform64Signed
-{
-    static constexpr auto name = "toDateTime";
-
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from < 0 || from > MAX_DATETIME_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime", from);
-        }
-
-        if (from < 0)
-            return 0;
-        return static_cast<ToType>(std::min(time_t(from), time_t(MAX_DATETIME_TIMESTAMP)));
-    }
-};
-
-/// Special case of converting Int8, Int16, Int32 or (U)Int64 (and also, for convenience, Float32, Float64) to DateTime.
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt8, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt8, DataTypeDateTime, ToDateTimeTransformSigned<Int8, UInt32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt16, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt16, DataTypeDateTime, ToDateTimeTransformSigned<Int16, UInt32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt32, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt32, DataTypeDateTime, ToDateTimeTransformSigned<Int32, UInt32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt64, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt64, DataTypeDateTime, ToDateTimeTransform64Signed<Int64, UInt32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeUInt64, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeUInt64, DataTypeDateTime, ToDateTimeTransform64<UInt64, UInt32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat32, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat32, DataTypeDateTime, ToDateTimeTransform64Signed<Float32, UInt32, default_date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat64, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat64, DataTypeDateTime, ToDateTimeTransform64Signed<Float64, UInt32, default_date_time_overflow_behavior>, false> {};
-
-/** Conversion of numeric to DateTime64
-  */
-
-template <typename FromType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTime64TransformUnsigned
-{
-    static constexpr auto name = "toDateTime64";
-
-    const DateTime64::NativeType scale_multiplier = 1;
-
-    ToDateTime64TransformUnsigned(UInt32 scale = 0) /// NOLINT
-        : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
-    {}
-
-    NO_SANITIZE_UNDEFINED DateTime64::NativeType execute(FromType from, const DateLUTImpl &) const
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime64", from);
-            else
-                return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(from, 0, scale_multiplier);
-        }
-        else
-            return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(std::min<time_t>(from, MAX_DATETIME64_TIMESTAMP), 0, scale_multiplier);
-    }
-};
-template <typename FromType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTime64TransformSigned
-{
-    static constexpr auto name = "toDateTime64";
-
-    const DateTime64::NativeType scale_multiplier = 1;
-
-    ToDateTime64TransformSigned(UInt32 scale = 0) /// NOLINT
-        : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
-    {}
-
-    NO_SANITIZE_UNDEFINED DateTime64::NativeType execute(FromType from, const DateLUTImpl &) const
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from < MIN_DATETIME64_TIMESTAMP || from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime64", from);
-        }
-        from = static_cast<FromType>(std::max<time_t>(from, MIN_DATETIME64_TIMESTAMP));
-        from = static_cast<FromType>(std::min<time_t>(from, MAX_DATETIME64_TIMESTAMP));
-
-        return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(from, 0, scale_multiplier);
-    }
-};
-template <typename FromDataType, typename FromType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ToDateTime64TransformFloat
-{
-    static constexpr auto name = "toDateTime64";
-
-    const UInt32 scale = 1;
-
-    ToDateTime64TransformFloat(UInt32 scale_ = 0) /// NOLINT
-        : scale(scale_)
-    {}
-
-    NO_SANITIZE_UNDEFINED DateTime64::NativeType execute(FromType from, const DateLUTImpl &) const
-    {
-        if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
-        {
-            if (from < MIN_DATETIME64_TIMESTAMP || from > MAX_DATETIME64_TIMESTAMP) [[unlikely]]
-                throw Exception(ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, "Timestamp value {} is out of bounds of type DateTime64", from);
-        }
-
-        from = std::max(from, static_cast<FromType>(MIN_DATETIME64_TIMESTAMP));
-        from = std::min(from, static_cast<FromType>(MAX_DATETIME64_TIMESTAMP));
-        return convertToDecimal<FromDataType, DataTypeDateTime64>(from, scale);
-    }
-};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt8, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt8, DataTypeDateTime64, ToDateTime64TransformSigned<Int8, date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt16, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt16, DataTypeDateTime64, ToDateTime64TransformSigned<Int16, date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt32, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt32, DataTypeDateTime64, ToDateTime64TransformSigned<Int32, date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeInt64, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeInt64, DataTypeDateTime64, ToDateTime64TransformSigned<Int64, date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeUInt64, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeUInt64, DataTypeDateTime64, ToDateTime64TransformUnsigned<UInt64, date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat32, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat32, DataTypeDateTime64, ToDateTime64TransformFloat<DataTypeFloat32, Float32, date_time_overflow_behavior>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeFloat64, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeFloat64, DataTypeDateTime64, ToDateTime64TransformFloat<DataTypeFloat64, Float64, date_time_overflow_behavior>, false> {};
-
-
-/** Conversion of DateTime64 to Date or DateTime: discards fractional part.
- */
-template <typename Transform>
-struct FromDateTime64Transform
-{
-    static constexpr auto name = Transform::name;
-
-    const DateTime64::NativeType scale_multiplier = 1;
-
-    FromDateTime64Transform(UInt32 scale) /// NOLINT
-        : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
-    {}
-
-    auto execute(DateTime64::NativeType dt, const DateLUTImpl & time_zone) const
-    {
-        const auto c = DecimalUtils::splitWithScaleMultiplier(DateTime64(dt), scale_multiplier);
-        return Transform::execute(static_cast<UInt32>(c.whole), time_zone);
-    }
-};
-
-/** Conversion of DateTime64 to Date or DateTime: discards fractional part.
- */
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDateTime64, DataTypeDate, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDateTime64, DataTypeDate, TransformDateTime64<ToDateImpl<date_time_overflow_behavior>>, false> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDateTime64, DataTypeDateTime, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDateTime64, DataTypeDateTime, TransformDateTime64<ToDateTimeImpl<date_time_overflow_behavior>>, false> {};
-
-struct ToDateTime64Transform
-{
-    static constexpr auto name = "toDateTime64";
-
-    const DateTime64::NativeType scale_multiplier = 1;
-
-    ToDateTime64Transform(UInt32 scale = 0) /// NOLINT
-        : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
-    {}
-
-    DateTime64::NativeType execute(UInt16 d, const DateLUTImpl & time_zone) const
-    {
-        const auto dt = ToDateTimeImpl<>::execute(d, time_zone);
-        return execute(dt, time_zone);
-    }
-
-    DateTime64::NativeType execute(Int32 d, const DateLUTImpl & time_zone) const
-    {
-        Int64 dt = static_cast<Int64>(time_zone.fromDayNum(ExtendedDayNum(d)));
-        return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(dt, 0, scale_multiplier);
-    }
-
-    DateTime64::NativeType execute(UInt32 dt, const DateLUTImpl & /*time_zone*/) const
-    {
-        return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(dt, 0, scale_multiplier);
-    }
-};
-
-/** Conversion of Date or DateTime to DateTime64: add zero sub-second part.
-  */
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDate, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDate, DataTypeDateTime64, ToDateTime64Transform> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDate32, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDate32, DataTypeDateTime64, ToDateTime64Transform> {};
-
-template <typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct ConvertImpl<DataTypeDateTime, DataTypeDateTime64, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : DateTimeTransformImpl<DataTypeDateTime, DataTypeDateTime64, ToDateTime64Transform> {};
-
-
-/** Transformation of numbers, dates, datetimes to strings: through formatting.
-  */
-template <typename DataType>
-struct FormatImpl
-{
-    template <typename ReturnType = void>
-    static ReturnType execute(const typename DataType::FieldType x, WriteBuffer & wb, const DataType *, const DateLUTImpl *)
-    {
-        writeText(x, wb);
-        return ReturnType(true);
-    }
-};
-
-template <>
-struct FormatImpl<DataTypeDate>
-{
-    template <typename ReturnType = void>
-    static ReturnType execute(const DataTypeDate::FieldType x, WriteBuffer & wb, const DataTypeDate *, const DateLUTImpl * time_zone)
-    {
-        writeDateText(DayNum(x), wb, *time_zone);
-        return ReturnType(true);
-    }
-};
-
-template <>
-struct FormatImpl<DataTypeDate32>
-{
-    template <typename ReturnType = void>
-    static ReturnType execute(const DataTypeDate32::FieldType x, WriteBuffer & wb, const DataTypeDate32 *, const DateLUTImpl * time_zone)
-    {
-        writeDateText(ExtendedDayNum(x), wb, *time_zone);
-        return ReturnType(true);
-    }
-};
-
-template <>
-struct FormatImpl<DataTypeDateTime>
-{
-    template <typename ReturnType = void>
-    static ReturnType execute(const DataTypeDateTime::FieldType x, WriteBuffer & wb, const DataTypeDateTime *, const DateLUTImpl * time_zone)
-    {
-        writeDateTimeText(x, wb, *time_zone);
-        return ReturnType(true);
-    }
-};
-
-template <>
-struct FormatImpl<DataTypeDateTime64>
-{
-    template <typename ReturnType = void>
-    static ReturnType execute(const DataTypeDateTime64::FieldType x, WriteBuffer & wb, const DataTypeDateTime64 * type, const DateLUTImpl * time_zone)
-    {
-        writeDateTimeText(DateTime64(x), type->getScale(), wb, *time_zone);
-        return ReturnType(true);
-    }
-};
-
-
-template <typename FieldType>
-struct FormatImpl<DataTypeEnum<FieldType>>
-{
-    template <typename ReturnType = void>
-    static ReturnType execute(const FieldType x, WriteBuffer & wb, const DataTypeEnum<FieldType> * type, const DateLUTImpl *)
-    {
-        static constexpr bool throw_exception = std::is_same_v<ReturnType, void>;
-
-        if constexpr (throw_exception)
-        {
-            writeString(type->getNameForValue(x), wb);
-        }
-        else
-        {
-            StringRef res;
-            bool is_ok = type->getNameForValue(x, res);
-            if (is_ok)
-                writeString(res, wb);
-            return ReturnType(is_ok);
-        }
-    }
-};
-
-template <typename FieldType>
-struct FormatImpl<DataTypeDecimal<FieldType>>
-{
-    template <typename ReturnType = void>
-    static ReturnType execute(const FieldType x, WriteBuffer & wb, const DataTypeDecimal<FieldType> * type, const DateLUTImpl *)
-    {
-        writeText(x, type->getScale(), wb, false);
-        return ReturnType(true);
-    }
-};
-
-
-/// DataTypeEnum<T> to DataType<T> free conversion
-template <typename FieldType, typename Name>
-struct ConvertImpl<DataTypeEnum<FieldType>, DataTypeNumber<FieldType>, Name, ConvertDefaultBehaviorTag>
-{
-    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/)
-    {
-        return arguments[0].column;
-    }
-};
-
-static inline ColumnUInt8::MutablePtr copyNullMap(ColumnPtr col)
-{
-    ColumnUInt8::MutablePtr null_map = nullptr;
-    if (const auto * col_null = checkAndGetColumn<ColumnNullable>(col.get()))
-    {
-        null_map = ColumnUInt8::create();
-        null_map->insertRangeFrom(col_null->getNullMapColumn(), 0, col_null->size());
-    }
-    return null_map;
-}
-
-template <typename FromDataType, typename Name>
-requires (!std::is_same_v<FromDataType, DataTypeString>)
-struct ConvertImpl<FromDataType, DataTypeString, Name, ConvertDefaultBehaviorTag>
-{
-    using FromFieldType = typename FromDataType::FieldType;
-    using ColVecType = ColumnVectorOrDecimal<FromFieldType>;
-
-    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/)
-    {
-        if constexpr (IsDataTypeDateOrDateTime<FromDataType>)
-        {
-            auto datetime_arg = arguments[0];
-
-            const DateLUTImpl * time_zone = nullptr;
-            const ColumnConst * time_zone_column = nullptr;
-
-            if (arguments.size() == 1)
-            {
-                auto non_null_args = createBlockWithNestedColumns(arguments);
-                time_zone = &extractTimeZoneFromFunctionArguments(non_null_args, 1, 0);
-            }
-            else /// When we have a column for timezone
-            {
-                datetime_arg.column = datetime_arg.column->convertToFullColumnIfConst();
-
-                if constexpr (std::is_same_v<FromDataType, DataTypeDate> || std::is_same_v<FromDataType, DataTypeDate32>)
-                    time_zone = &DateLUT::instance();
-                /// For argument of Date or DateTime type, second argument with time zone could be specified.
-                if constexpr (std::is_same_v<FromDataType, DataTypeDateTime> || std::is_same_v<FromDataType, DataTypeDateTime64>)
-                {
-                    if ((time_zone_column = checkAndGetColumnConst<ColumnString>(arguments[1].column.get())))
-                    {
-                        auto non_null_args = createBlockWithNestedColumns(arguments);
-                        time_zone = &extractTimeZoneFromFunctionArguments(non_null_args, 1, 0);
-                    }
-                }
-            }
-            const auto & col_with_type_and_name = columnGetNested(datetime_arg);
-
-            if (const auto col_from = checkAndGetColumn<ColVecType>(col_with_type_and_name.column.get()))
-            {
-                auto col_to = ColumnString::create();
-
-                const typename ColVecType::Container & vec_from = col_from->getData();
-                ColumnString::Chars & data_to = col_to->getChars();
-                ColumnString::Offsets & offsets_to = col_to->getOffsets();
-                size_t size = vec_from.size();
-
-                if constexpr (std::is_same_v<FromDataType, DataTypeDate>)
-                    data_to.resize(size * (strlen("YYYY-MM-DD") + 1));
-                else if constexpr (std::is_same_v<FromDataType, DataTypeDate32>)
-                    data_to.resize(size * (strlen("YYYY-MM-DD") + 1));
-                else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime>)
-                    data_to.resize(size * (strlen("YYYY-MM-DD hh:mm:ss") + 1));
-                else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime64>)
-                    data_to.resize(size * (strlen("YYYY-MM-DD hh:mm:ss.") + col_from->getScale() + 1));
-                else
-                    data_to.resize(size * 3);   /// Arbitrary
-
-                offsets_to.resize(size);
-
-                WriteBufferFromVector<ColumnString::Chars> write_buffer(data_to);
-                const auto & type = static_cast<const FromDataType &>(*col_with_type_and_name.type);
-
-                ColumnUInt8::MutablePtr null_map = copyNullMap(datetime_arg.column);
-
-                if (!null_map && arguments.size() > 1)
-                    null_map = copyNullMap(arguments[1].column->convertToFullColumnIfConst());
-
-                if (null_map)
-                {
-                    for (size_t i = 0; i < size; ++i)
-                    {
-                        if (!time_zone_column && arguments.size() > 1)
-                        {
-                            if (!arguments[1].column.get()->getDataAt(i).toString().empty())
-                                time_zone = &DateLUT::instance(arguments[1].column.get()->getDataAt(i).toString());
-                            else
-                                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Provided time zone must be non-empty");
-                        }
-                        bool is_ok = FormatImpl<FromDataType>::template execute<bool>(vec_from[i], write_buffer, &type, time_zone);
-                        null_map->getData()[i] |= !is_ok;
-                        writeChar(0, write_buffer);
-                        offsets_to[i] = write_buffer.count();
-                    }
-                }
-                else
-                {
-                    for (size_t i = 0; i < size; ++i)
-                    {
-                        if (!time_zone_column && arguments.size() > 1)
-                        {
-                            if (!arguments[1].column.get()->getDataAt(i).toString().empty())
-                                time_zone = &DateLUT::instance(arguments[1].column.get()->getDataAt(i).toString());
-                            else
-                                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Provided time zone must be non-empty");
-                        }
-                        FormatImpl<FromDataType>::template execute<void>(vec_from[i], write_buffer, &type, time_zone);
-                        writeChar(0, write_buffer);
-                        offsets_to[i] = write_buffer.count();
-                    }
-                }
-
-                write_buffer.finalize();
-
-                if (null_map)
-                    return ColumnNullable::create(std::move(col_to), std::move(null_map));
-                return col_to;
-            }
-            else
-                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                        arguments[0].column->getName(), Name::name);
-        }
-        else
-        {
-            ColumnUInt8::MutablePtr null_map = copyNullMap(arguments[0].column);
-
-            const auto & col_with_type_and_name = columnGetNested(arguments[0]);
-            const auto & type = static_cast<const FromDataType &>(*col_with_type_and_name.type);
-
-            if (const auto col_from = checkAndGetColumn<ColVecType>(col_with_type_and_name.column.get()))
-            {
-                auto col_to = ColumnString::create();
-
-                const typename ColVecType::Container & vec_from = col_from->getData();
-                ColumnString::Chars & data_to = col_to->getChars();
-                ColumnString::Offsets & offsets_to = col_to->getOffsets();
-                size_t size = vec_from.size();
-
-                data_to.resize(size * 3);
-                offsets_to.resize(size);
-
-                WriteBufferFromVector<ColumnString::Chars> write_buffer(data_to);
-
-                if (null_map)
-                {
-                    for (size_t i = 0; i < size; ++i)
-                    {
-                        bool is_ok = FormatImpl<FromDataType>::template execute<bool>(vec_from[i], write_buffer, &type, nullptr);
-                        /// We don't use timezones in this branch
-                        null_map->getData()[i] |= !is_ok;
-                        writeChar(0, write_buffer);
-                        offsets_to[i] = write_buffer.count();
-                    }
-                }
-                else
-                {
-                    for (size_t i = 0; i < size; ++i)
-                    {
-                        FormatImpl<FromDataType>::template execute<void>(vec_from[i], write_buffer, &type, nullptr);
-                        writeChar(0, write_buffer);
-                        offsets_to[i] = write_buffer.count();
-                    }
-                }
-
-                write_buffer.finalize();
-
-                if (null_map)
-                    return ColumnNullable::create(std::move(col_to), std::move(null_map));
-                return col_to;
-            }
-            else
-                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                        arguments[0].column->getName(), Name::name);
-        }
-    }
-};
-
-
-/// Generic conversion of any type to String or FixedString via serialization to text.
-template <typename StringColumnType>
-struct ConvertImplGenericToString
-{
-    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t /*input_rows_count*/)
-    {
-        static_assert(std::is_same_v<StringColumnType, ColumnString> || std::is_same_v<StringColumnType, ColumnFixedString>,
-                "Can be used only to serialize to ColumnString or ColumnFixedString");
-
-        ColumnUInt8::MutablePtr null_map = copyNullMap(arguments[0].column);
-
-        const auto & col_with_type_and_name = columnGetNested(arguments[0]);
-        const IDataType & type = *col_with_type_and_name.type;
-        const IColumn & col_from = *col_with_type_and_name.column;
-
-        size_t size = col_from.size();
-        auto col_to = removeNullable(result_type)->createColumn();
-
-        {
-            ColumnStringHelpers::WriteHelper write_helper(
-                    assert_cast<StringColumnType &>(*col_to),
-                    size);
-
-            auto & write_buffer = write_helper.getWriteBuffer();
-
-            FormatSettings format_settings;
-            auto serialization = type.getDefaultSerialization();
-            for (size_t row = 0; row < size; ++row)
-            {
-                serialization->serializeText(col_from, row, write_buffer, format_settings);
-                write_helper.rowWritten();
-            }
-
-            write_helper.finalize();
-        }
-
-        if (result_type->isNullable() && null_map)
-            return ColumnNullable::create(std::move(col_to), std::move(null_map));
-        return col_to;
-    }
-};
-
-/** Conversion of time_t to UInt16, Int32, UInt32
-  */
-template <typename DataType>
-void convertFromTime(typename DataType::FieldType & x, time_t & time)
-{
-    x = time;
-}
-
-template <>
-inline void convertFromTime<DataTypeDate>(DataTypeDate::FieldType & x, time_t & time)
-{
-    if (unlikely(time < 0))
-        x = 0;
-    else if (unlikely(time > 0xFFFF))
-        x = 0xFFFF;
-    else
-        x = time;
-}
-
-template <>
-inline void convertFromTime<DataTypeDate32>(DataTypeDate32::FieldType & x, time_t & time)
-{
-    x = static_cast<UInt32>(time);
-}
-
-template <>
-inline void convertFromTime<DataTypeDateTime>(DataTypeDateTime::FieldType & x, time_t & time)
-{
-    if (unlikely(time < 0))
-        x = 0;
-    else if (unlikely(time > MAX_DATETIME_TIMESTAMP))
-        x = MAX_DATETIME_TIMESTAMP;
-    else
-        x = static_cast<UInt32>(time);
-}
-
-/** Conversion of strings to numbers, dates, datetimes: through parsing.
-  */
-template <typename DataType>
-void parseImpl(typename DataType::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool precise_float_parsing)
-{
-    if constexpr (std::is_floating_point_v<typename DataType::FieldType>)
-    {
-        if (precise_float_parsing)
-            readFloatTextPrecise(x, rb);
-        else
-            readFloatTextFast(x, rb);
-    }
-    else
-        readText(x, rb);
-}
-
-template <>
-inline void parseImpl<DataTypeDate>(DataTypeDate::FieldType & x, ReadBuffer & rb, const DateLUTImpl * time_zone, bool)
-{
-    DayNum tmp(0);
-    readDateText(tmp, rb, *time_zone);
-    x = tmp;
-}
-
-template <>
-inline void parseImpl<DataTypeDate32>(DataTypeDate32::FieldType & x, ReadBuffer & rb, const DateLUTImpl * time_zone, bool)
-{
-    ExtendedDayNum tmp(0);
-    readDateText(tmp, rb, *time_zone);
-    x = tmp;
-}
-
-
-// NOTE: no need of extra overload of DateTime64, since readDateTimeText64 has different signature and that case is explicitly handled in the calling code.
-template <>
-inline void parseImpl<DataTypeDateTime>(DataTypeDateTime::FieldType & x, ReadBuffer & rb, const DateLUTImpl * time_zone, bool)
-{
-    time_t time = 0;
-    readDateTimeText(time, rb, *time_zone);
-    convertFromTime<DataTypeDateTime>(x, time);
-}
-
-template <>
-inline void parseImpl<DataTypeUUID>(DataTypeUUID::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool)
-{
-    UUID tmp;
-    readUUIDText(tmp, rb);
-    x = tmp.toUnderType();
-}
-
-template <>
-inline void parseImpl<DataTypeIPv4>(DataTypeIPv4::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool)
-{
-    IPv4 tmp;
-    readIPv4Text(tmp, rb);
-    x = tmp.toUnderType();
-}
-
-template <>
-inline void parseImpl<DataTypeIPv6>(DataTypeIPv6::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool)
-{
-    IPv6 tmp;
-    readIPv6Text(tmp, rb);
-    x = tmp;
-}
-
-template <typename DataType>
-bool tryParseImpl(typename DataType::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool precise_float_parsing)
-{
-    if constexpr (std::is_floating_point_v<typename DataType::FieldType>)
-    {
-        if (precise_float_parsing)
-            return tryReadFloatTextPrecise(x, rb);
-        else
-            return tryReadFloatTextFast(x, rb);
-    }
-    else /*if constexpr (is_integer_v<typename DataType::FieldType>)*/
-        return tryReadIntText(x, rb);
-}
-
-template <>
-inline bool tryParseImpl<DataTypeDate>(DataTypeDate::FieldType & x, ReadBuffer & rb, const DateLUTImpl * time_zone, bool)
-{
-    DayNum tmp(0);
-    if (!tryReadDateText(tmp, rb, *time_zone))
-        return false;
-    x = tmp;
-    return true;
-}
-
-template <>
-inline bool tryParseImpl<DataTypeDate32>(DataTypeDate32::FieldType & x, ReadBuffer & rb, const DateLUTImpl * time_zone, bool)
-{
-    ExtendedDayNum tmp(0);
-    if (!tryReadDateText(tmp, rb, *time_zone))
-        return false;
-    x = tmp;
-    return true;
-}
-
-template <>
-inline bool tryParseImpl<DataTypeDateTime>(DataTypeDateTime::FieldType & x, ReadBuffer & rb, const DateLUTImpl * time_zone, bool)
-{
-    time_t time = 0;
-    if (!tryReadDateTimeText(time, rb, *time_zone))
-        return false;
-    convertFromTime<DataTypeDateTime>(x, time);
-    return true;
-}
-
-template <>
-inline bool tryParseImpl<DataTypeUUID>(DataTypeUUID::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool)
-{
-    UUID tmp;
-    if (!tryReadUUIDText(tmp, rb))
-        return false;
-
-    x = tmp.toUnderType();
-    return true;
-}
-
-template <>
-inline bool tryParseImpl<DataTypeIPv4>(DataTypeIPv4::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool)
-{
-    IPv4 tmp;
-    if (!tryReadIPv4Text(tmp, rb))
-        return false;
-
-    x = tmp.toUnderType();
-    return true;
-}
-
-template <>
-inline bool tryParseImpl<DataTypeIPv6>(DataTypeIPv6::FieldType & x, ReadBuffer & rb, const DateLUTImpl *, bool)
-{
-    IPv6 tmp;
-    if (!tryReadIPv6Text(tmp, rb))
-        return false;
-
-    x = tmp;
-    return true;
-}
-
-
-/** Throw exception with verbose message when string value is not parsed completely.
-  */
-[[noreturn]] inline void throwExceptionForIncompletelyParsedValue(ReadBuffer & read_buffer, const IDataType & result_type)
-{
-    WriteBufferFromOwnString message_buf;
-    message_buf << "Cannot parse string " << quote << String(read_buffer.buffer().begin(), read_buffer.buffer().size())
-                << " as " << result_type.getName()
-                << ": syntax error";
-
-    if (read_buffer.offset())
-        message_buf << " at position " << read_buffer.offset()
-                    << " (parsed just " << quote << String(read_buffer.buffer().begin(), read_buffer.offset()) << ")";
-    else
-        message_buf << " at begin of string";
-
-    // Currently there are no functions toIPv{4,6}Or{Null,Zero}
-    if (isNativeNumber(result_type) && !(result_type.getName() == "IPv4" || result_type.getName() == "IPv6"))
-        message_buf << ". Note: there are to" << result_type.getName() << "OrZero and to" << result_type.getName() << "OrNull functions, which returns zero/NULL instead of throwing exception.";
-
-    throw Exception(PreformattedMessage{message_buf.str(), "Cannot parse string {} as {}: syntax error {}"}, ErrorCodes::CANNOT_PARSE_TEXT);
-}
-
-
-enum class ConvertFromStringExceptionMode
-{
-    Throw,  /// Throw exception if value cannot be parsed.
-    Zero,   /// Fill with zero or default if value cannot be parsed.
-    Null    /// Return ColumnNullable with NULLs when value cannot be parsed.
-};
-
-enum class ConvertFromStringParsingMode
-{
-    Normal,
-    BestEffort,  /// Only applicable for DateTime. Will use sophisticated method, that is slower.
-    BestEffortUS
-};
-
-template <typename FromDataType, typename ToDataType, typename Name,
-    ConvertFromStringExceptionMode exception_mode, ConvertFromStringParsingMode parsing_mode>
-struct ConvertThroughParsing
-{
-    static_assert(std::is_same_v<FromDataType, DataTypeString> || std::is_same_v<FromDataType, DataTypeFixedString>,
-        "ConvertThroughParsing is only applicable for String or FixedString data types");
-
-    static constexpr bool to_datetime64 = std::is_same_v<ToDataType, DataTypeDateTime64>;
-
-    static bool isAllRead(ReadBuffer & in)
-    {
-        /// In case of FixedString, skip zero bytes at end.
-        if constexpr (std::is_same_v<FromDataType, DataTypeFixedString>)
-            while (!in.eof() && *in.position() == 0)
-                ++in.position();
-
-        if (in.eof())
-            return true;
-
-        /// Special case, that allows to parse string with DateTime or DateTime64 as Date or Date32.
-        if constexpr (std::is_same_v<ToDataType, DataTypeDate> || std::is_same_v<ToDataType, DataTypeDate32>)
-        {
-            if (!in.eof() && (*in.position() == ' ' || *in.position() == 'T'))
-            {
-                if (in.buffer().size() == strlen("YYYY-MM-DD hh:mm:ss"))
-                    return true;
-
-                if (in.buffer().size() >= strlen("YYYY-MM-DD hh:mm:ss.x")
-                    && in.buffer().begin()[19] == '.')
-                {
-                    in.position() = in.buffer().begin() + 20;
-
-                    while (!in.eof() && isNumericASCII(*in.position()))
-                        ++in.position();
-
-                    if (in.eof())
-                        return true;
-                }
-            }
-        }
-
-        return false;
-    }
-
-    template <typename Additions = void *>
-    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr & res_type, size_t input_rows_count,
-                        Additions additions [[maybe_unused]] = Additions())
-    {
-        using ColVecTo = typename ToDataType::ColumnType;
-
-        const DateLUTImpl * local_time_zone [[maybe_unused]] = nullptr;
-        const DateLUTImpl * utc_time_zone [[maybe_unused]] = nullptr;
-
-        /// For conversion to Date or DateTime type, second argument with time zone could be specified.
-        if constexpr (std::is_same_v<ToDataType, DataTypeDateTime> || to_datetime64)
-        {
-            const auto result_type = removeNullable(res_type);
-            // Time zone is already figured out during result type resolution, no need to do it here.
-            if (const auto dt_col = checkAndGetDataType<ToDataType>(result_type.get()))
-                local_time_zone = &dt_col->getTimeZone();
-            else
-                local_time_zone = &extractTimeZoneFromFunctionArguments(arguments, 1, 0);
-
-            if constexpr (parsing_mode == ConvertFromStringParsingMode::BestEffort || parsing_mode == ConvertFromStringParsingMode::BestEffortUS)
-                utc_time_zone = &DateLUT::instance("UTC");
-        }
-        else if constexpr (std::is_same_v<ToDataType, DataTypeDate> || std::is_same_v<ToDataType, DataTypeDate32>)
-        {
-            // Timezone is more or less dummy when parsing Date/Date32 from string.
-            local_time_zone = &DateLUT::instance();
-            utc_time_zone = &DateLUT::instance("UTC");
-        }
-
-        const IColumn * col_from = arguments[0].column.get();
-        const ColumnString * col_from_string = checkAndGetColumn<ColumnString>(col_from);
-        const ColumnFixedString * col_from_fixed_string = checkAndGetColumn<ColumnFixedString>(col_from);
-
-        if (std::is_same_v<FromDataType, DataTypeString> && !col_from_string)
-            throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                col_from->getName(), Name::name);
-
-        if (std::is_same_v<FromDataType, DataTypeFixedString> && !col_from_fixed_string)
-            throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                col_from->getName(), Name::name);
-
-        size_t size = input_rows_count;
-        typename ColVecTo::MutablePtr col_to = nullptr;
-
-        if constexpr (IsDataTypeDecimal<ToDataType>)
-        {
-            UInt32 scale = additions;
-            if constexpr (to_datetime64)
-            {
-                ToDataType check_bounds_in_ctor(scale, local_time_zone ? local_time_zone->getTimeZone() : String{});
-            }
-            else
-            {
-                ToDataType check_bounds_in_ctor(ToDataType::maxPrecision(), scale);
-            }
-            col_to = ColVecTo::create(size, scale);
-        }
-        else
-            col_to = ColVecTo::create(size);
-
-        typename ColVecTo::Container & vec_to = col_to->getData();
-
-        ColumnUInt8::MutablePtr col_null_map_to;
-        ColumnUInt8::Container * vec_null_map_to [[maybe_unused]] = nullptr;
-        if constexpr (exception_mode == ConvertFromStringExceptionMode::Null)
-        {
-            col_null_map_to = ColumnUInt8::create(size);
-            vec_null_map_to = &col_null_map_to->getData();
-        }
-
-        const ColumnString::Chars * chars = nullptr;
-        const IColumn::Offsets * offsets = nullptr;
-        size_t fixed_string_size = 0;
-
-        if constexpr (std::is_same_v<FromDataType, DataTypeString>)
-        {
-            chars = &col_from_string->getChars();
-            offsets = &col_from_string->getOffsets();
-        }
-        else
-        {
-            chars = &col_from_fixed_string->getChars();
-            fixed_string_size = col_from_fixed_string->getN();
-        }
-
-        size_t current_offset = 0;
-
-        bool precise_float_parsing = false;
-
-        if (DB::CurrentThread::isInitialized())
-        {
-            const DB::ContextPtr query_context = DB::CurrentThread::get().getQueryContext();
-
-            if (query_context)
-                precise_float_parsing = query_context->getSettingsRef().precise_float_parsing;
-        }
-
-        for (size_t i = 0; i < size; ++i)
-        {
-            size_t next_offset = std::is_same_v<FromDataType, DataTypeString> ? (*offsets)[i] : (current_offset + fixed_string_size);
-            size_t string_size = std::is_same_v<FromDataType, DataTypeString> ? next_offset - current_offset - 1 : fixed_string_size;
-
-            ReadBufferFromMemory read_buffer(&(*chars)[current_offset], string_size);
-
-            if constexpr (exception_mode == ConvertFromStringExceptionMode::Throw)
-            {
-                if constexpr (parsing_mode == ConvertFromStringParsingMode::BestEffort)
-                {
-                    if constexpr (to_datetime64)
-                    {
-                        DateTime64 res = 0;
-                        parseDateTime64BestEffort(res, col_to->getScale(), read_buffer, *local_time_zone, *utc_time_zone);
-                        vec_to[i] = res;
-                    }
-                    else
-                    {
-                        time_t res;
-                        parseDateTimeBestEffort(res, read_buffer, *local_time_zone, *utc_time_zone);
-                        convertFromTime<ToDataType>(vec_to[i], res);
-                    }
-                }
-                else if constexpr (parsing_mode == ConvertFromStringParsingMode::BestEffortUS)
-                {
-                    if constexpr (to_datetime64)
-                    {
-                        DateTime64 res = 0;
-                        parseDateTime64BestEffortUS(res, col_to->getScale(), read_buffer, *local_time_zone, *utc_time_zone);
-                        vec_to[i] = res;
-                    }
-                    else
-                    {
-                        time_t res;
-                        parseDateTimeBestEffortUS(res, read_buffer, *local_time_zone, *utc_time_zone);
-                        convertFromTime<ToDataType>(vec_to[i], res);
-                    }
-                }
-                else
-                {
-                    if constexpr (to_datetime64)
-                    {
-                        DateTime64 value = 0;
-                        readDateTime64Text(value, col_to->getScale(), read_buffer, *local_time_zone);
-                        vec_to[i] = value;
-                    }
-                    else if constexpr (IsDataTypeDecimal<ToDataType>)
-                    {
-                        SerializationDecimal<typename ToDataType::FieldType>::readText(
-                            vec_to[i], read_buffer, ToDataType::maxPrecision(), col_to->getScale());
-                    }
-                    else
-                    {
-                        /// we want to utilize constexpr condition here, which is not mixable with value comparison
-                        do
-                        {
-                            if constexpr (std::is_same_v<FromDataType, DataTypeFixedString> && std::is_same_v<ToDataType, DataTypeIPv6>)
-                            {
-                                if (fixed_string_size == IPV6_BINARY_LENGTH)
-                                {
-                                    readBinary(vec_to[i], read_buffer);
-                                    break;
-                                }
-                            }
-                            parseImpl<ToDataType>(vec_to[i], read_buffer, local_time_zone, precise_float_parsing);
-                        } while (false);
-                    }
-                }
-
-                if (!isAllRead(read_buffer))
-                    throwExceptionForIncompletelyParsedValue(read_buffer, *res_type);
-            }
-            else
-            {
-                bool parsed;
-
-                if constexpr (parsing_mode == ConvertFromStringParsingMode::BestEffort)
-                {
-                    if constexpr (to_datetime64)
-                    {
-                        DateTime64 res = 0;
-                        parsed = tryParseDateTime64BestEffort(res, col_to->getScale(), read_buffer, *local_time_zone, *utc_time_zone);
-                        vec_to[i] = res;
-                    }
-                    else
-                    {
-                        time_t res;
-                        parsed = tryParseDateTimeBestEffort(res, read_buffer, *local_time_zone, *utc_time_zone);
-                        convertFromTime<ToDataType>(vec_to[i],res);
-                    }
-                }
-                else if constexpr (parsing_mode == ConvertFromStringParsingMode::BestEffortUS)
-                {
-                    if constexpr (to_datetime64)
-                    {
-                        DateTime64 res = 0;
-                        parsed = tryParseDateTime64BestEffortUS(res, col_to->getScale(), read_buffer, *local_time_zone, *utc_time_zone);
-                        vec_to[i] = res;
-                    }
-                    else
-                    {
-                        time_t res;
-                        parsed = tryParseDateTimeBestEffortUS(res, read_buffer, *local_time_zone, *utc_time_zone);
-                        convertFromTime<ToDataType>(vec_to[i],res);
-                    }
-                }
-                else
-                {
-                    if constexpr (to_datetime64)
-                    {
-                        DateTime64 value = 0;
-                        parsed = tryReadDateTime64Text(value, col_to->getScale(), read_buffer, *local_time_zone);
-                        vec_to[i] = value;
-                    }
-                    else if constexpr (IsDataTypeDecimal<ToDataType>)
-                    {
-                        parsed = SerializationDecimal<typename ToDataType::FieldType>::tryReadText(
-                            vec_to[i], read_buffer, ToDataType::maxPrecision(), col_to->getScale());
-                    }
-                    else
-                    {
-                        /// we want to utilize constexpr condition here, which is not mixable with value comparison
-                        do
-                        {
-                            if constexpr (std::is_same_v<FromDataType, DataTypeFixedString> && std::is_same_v<ToDataType, DataTypeIPv6>)
-                            {
-                                if (fixed_string_size == IPV6_BINARY_LENGTH)
-                                {
-                                    readBinary(vec_to[i], read_buffer);
-                                    parsed = true;
-                                    break;
-                                }
-                            }
-
-                            parsed = tryParseImpl<ToDataType>(vec_to[i], read_buffer, local_time_zone, precise_float_parsing);
-                        } while (false);
-                    }
-                }
-
-                if (!isAllRead(read_buffer))
-                    parsed = false;
-
-                if (!parsed)
-                {
-                    if constexpr (std::is_same_v<ToDataType, DataTypeDate32>)
-                    {
-                        vec_to[i] = -static_cast<Int32>(DateLUT::instance().getDayNumOffsetEpoch());
-                    }
-                    else
-                    {
-                        vec_to[i] = static_cast<typename ToDataType::FieldType>(0);
-                    }
-                }
-
-                if constexpr (exception_mode == ConvertFromStringExceptionMode::Null)
-                    (*vec_null_map_to)[i] = !parsed;
-            }
-
-            current_offset = next_offset;
-        }
-
-        if constexpr (exception_mode == ConvertFromStringExceptionMode::Null)
-            return ColumnNullable::create(std::move(col_to), std::move(col_null_map_to));
-        else
-            return col_to;
-    }
-};
-
-
-template <typename ToDataType, typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-requires (!std::is_same_v<ToDataType, DataTypeString>)
-struct ConvertImpl<DataTypeString, ToDataType, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : ConvertThroughParsing<DataTypeString, ToDataType, Name, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::Normal> {};
-
-template <typename ToDataType, typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-requires (!std::is_same_v<ToDataType, DataTypeFixedString>)
-struct ConvertImpl<DataTypeFixedString, ToDataType, Name, ConvertDefaultBehaviorTag, date_time_overflow_behavior>
-    : ConvertThroughParsing<DataTypeFixedString, ToDataType, Name, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::Normal> {};
-
-template <typename ToDataType, typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-requires (!std::is_same_v<ToDataType, DataTypeString>)
-struct ConvertImpl<DataTypeString, ToDataType, Name, ConvertReturnNullOnErrorTag, date_time_overflow_behavior>
-    : ConvertThroughParsing<DataTypeString, ToDataType, Name, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::Normal> {};
-
-template <typename ToDataType, typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-requires (!std::is_same_v<ToDataType, DataTypeFixedString>)
-struct ConvertImpl<DataTypeFixedString, ToDataType, Name, ConvertReturnNullOnErrorTag, date_time_overflow_behavior>
-    : ConvertThroughParsing<DataTypeFixedString, ToDataType, Name, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::Normal> {};
-
-template <typename FromDataType, typename ToDataType, typename Name, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-requires (is_any_of<FromDataType, DataTypeString, DataTypeFixedString> && is_any_of<ToDataType, DataTypeIPv4, DataTypeIPv6>)
-struct ConvertImpl<FromDataType, ToDataType, Name, ConvertReturnZeroOnErrorTag, date_time_overflow_behavior>
-    : ConvertThroughParsing<FromDataType, ToDataType, Name, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::Normal> {};
-
-/// Generic conversion of any type from String. Used for complex types: Array and Tuple or types with custom serialization.
-template <typename StringColumnType>
-struct ConvertImplGenericFromString
-{
-    static ColumnPtr execute(ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable * column_nullable, size_t input_rows_count)
-    {
-        static_assert(std::is_same_v<StringColumnType, ColumnString> || std::is_same_v<StringColumnType, ColumnFixedString>,
-                "Can be used only to parse from ColumnString or ColumnFixedString");
-
-        const IColumn & column_from = *arguments[0].column;
-        const IDataType & data_type_to = *result_type;
-        auto res = data_type_to.createColumn();
-        auto serialization = data_type_to.getDefaultSerialization();
-        const auto * null_map = column_nullable ? &column_nullable->getNullMapData() : nullptr;
-
-        executeImpl(column_from, *res, *serialization, input_rows_count, null_map, result_type.get());
-        return res;
-    }
-
-    static void executeImpl(
-        const IColumn & column_from,
-        IColumn & column_to,
-        const ISerialization & serialization_from,
-        size_t input_rows_count,
-        const PaddedPODArray<UInt8> * null_map = nullptr,
-        const IDataType * result_type = nullptr)
-    {
-        static_assert(std::is_same_v<StringColumnType, ColumnString> || std::is_same_v<StringColumnType, ColumnFixedString>,
-                "Can be used only to parse from ColumnString or ColumnFixedString");
-
-        if (const StringColumnType * col_from_string = checkAndGetColumn<StringColumnType>(&column_from))
-        {
-            column_to.reserve(input_rows_count);
-
-            FormatSettings format_settings;
-            for (size_t i = 0; i < input_rows_count; ++i)
-            {
-                if (null_map && (*null_map)[i])
-                {
-                    column_to.insertDefault();
-                    continue;
-                }
-
-                const auto & val = col_from_string->getDataAt(i);
-                ReadBufferFromMemory read_buffer(val.data, val.size);
-                try
-                {
-                    serialization_from.deserializeWholeText(column_to, read_buffer, format_settings);
-                }
-                catch (const Exception & e)
-                {
-                    auto * nullable_column = typeid_cast<ColumnNullable *>(&column_to);
-                    if (e.code() == ErrorCodes::CANNOT_PARSE_BOOL && nullable_column)
-                    {
-                        auto & col_nullmap = nullable_column->getNullMapData();
-                        if (col_nullmap.size() != nullable_column->size())
-                            col_nullmap.resize_fill(nullable_column->size());
-                        if (nullable_column->size() == (i + 1))
-                            nullable_column->popBack(1);
-                        nullable_column->insertDefault();
-                        continue;
-                    }
-                    throw;
-                }
-
-                if (!read_buffer.eof())
-                {
-                    if (result_type)
-                        throwExceptionForIncompletelyParsedValue(read_buffer, *result_type);
-                    else
-                        throw Exception(ErrorCodes::CANNOT_PARSE_TEXT,
-                            "Cannot parse string to column {}. Expected eof", column_to.getName());
-                }
-            }
-        }
-        else
-            throw Exception(ErrorCodes::ILLEGAL_COLUMN,
-                "Illegal column {} of first argument of conversion function from string",
-                column_from.getName());
-    }
-
-};
-
-
-template <>
-struct ConvertImpl<DataTypeString, DataTypeUInt32, NameToUnixTimestamp, ConvertDefaultBehaviorTag>
-    : ConvertImpl<DataTypeString, DataTypeDateTime, NameToUnixTimestamp, ConvertDefaultBehaviorTag> {};
-
-template <>
-struct ConvertImpl<DataTypeString, DataTypeUInt32, NameToUnixTimestamp, ConvertReturnNullOnErrorTag>
-    : ConvertImpl<DataTypeString, DataTypeDateTime, NameToUnixTimestamp, ConvertReturnNullOnErrorTag> {};
-
-/** If types are identical, just take reference to column.
-  */
-template <typename T, typename Name>
-requires (!T::is_parametric)
-struct ConvertImpl<T, T, Name, ConvertDefaultBehaviorTag>
-{
-    template <typename Additions = void *>
-    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/,
-        Additions additions [[maybe_unused]] = Additions())
-    {
-        return arguments[0].column;
-    }
-};
-
-template <typename Name>
-struct ConvertImpl<DataTypeUInt8, DataTypeUInt8, Name, ConvertDefaultBehaviorTag>
-{
-    template <typename Additions = void *>
-    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/,
-                             Additions additions [[maybe_unused]] = Additions())
-    {
-
-        return arguments[0].column;
-    }
-};
-
-
-/** Conversion from FixedString to String.
-  * Cutting sequences of zero bytes from end of strings.
-  */
-template <typename Name>
-struct ConvertImpl<DataTypeFixedString, DataTypeString, Name, ConvertDefaultBehaviorTag>
-{
-    static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr & return_type, size_t /*input_rows_count*/)
-    {
-        ColumnUInt8::MutablePtr null_map = copyNullMap(arguments[0].column);
-        const auto & nested =  columnGetNested(arguments[0]);
-        if (const ColumnFixedString * col_from = checkAndGetColumn<ColumnFixedString>(nested.column.get()))
-        {
-            auto col_to = ColumnString::create();
-
-            const ColumnFixedString::Chars & data_from = col_from->getChars();
-            ColumnString::Chars & data_to = col_to->getChars();
-            ColumnString::Offsets & offsets_to = col_to->getOffsets();
-            size_t size = col_from->size();
-            size_t n = col_from->getN();
-            data_to.resize(size * (n + 1)); /// + 1 - zero terminator
-            offsets_to.resize(size);
-
-            size_t offset_from = 0;
-            size_t offset_to = 0;
-            for (size_t i = 0; i < size; ++i)
-            {
-                if (!null_map || !null_map->getData()[i])
-                {
-                    size_t bytes_to_copy = n;
-                    while (bytes_to_copy > 0 && data_from[offset_from + bytes_to_copy - 1] == 0)
-                        --bytes_to_copy;
-
-                    memcpy(&data_to[offset_to], &data_from[offset_from], bytes_to_copy);
-                    offset_to += bytes_to_copy;
-                }
-                data_to[offset_to] = 0;
-                ++offset_to;
-                offsets_to[i] = offset_to;
-                offset_from += n;
-            }
-
-            data_to.resize(offset_to);
-            if (return_type->isNullable() && null_map)
-                return ColumnNullable::create(std::move(col_to), std::move(null_map));
-            return col_to;
-        }
-        else
-            throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}",
-                    arguments[0].column->getName(), Name::name);
-    }
-};
-
-
-/// Declared early because used below.
-struct NameToDate { static constexpr auto name = "toDate"; };
-struct NameToDate32 { static constexpr auto name = "toDate32"; };
-struct NameToDateTime { static constexpr auto name = "toDateTime"; };
-struct NameToDateTime32 { static constexpr auto name = "toDateTime32"; };
-struct NameToDateTime64 { static constexpr auto name = "toDateTime64"; };
-struct NameToString { static constexpr auto name = "toString"; };
-struct NameToDecimal32 { static constexpr auto name = "toDecimal32"; };
-struct NameToDecimal64 { static constexpr auto name = "toDecimal64"; };
-struct NameToDecimal128 { static constexpr auto name = "toDecimal128"; };
-struct NameToDecimal256 { static constexpr auto name = "toDecimal256"; };
-
-
-#define DEFINE_NAME_TO_INTERVAL(INTERVAL_KIND) \
-    struct NameToInterval ## INTERVAL_KIND \
-    { \
-        static constexpr auto name = "toInterval" #INTERVAL_KIND; \
-        static constexpr auto kind = IntervalKind::Kind::INTERVAL_KIND; \
-    };
-
-DEFINE_NAME_TO_INTERVAL(Nanosecond)
-DEFINE_NAME_TO_INTERVAL(Microsecond)
-DEFINE_NAME_TO_INTERVAL(Millisecond)
-DEFINE_NAME_TO_INTERVAL(Second)
-DEFINE_NAME_TO_INTERVAL(Minute)
-DEFINE_NAME_TO_INTERVAL(Hour)
-DEFINE_NAME_TO_INTERVAL(Day)
-DEFINE_NAME_TO_INTERVAL(Week)
-DEFINE_NAME_TO_INTERVAL(Month)
-DEFINE_NAME_TO_INTERVAL(Quarter)
-DEFINE_NAME_TO_INTERVAL(Year)
-
-#undef DEFINE_NAME_TO_INTERVAL
-
-struct NameParseDateTimeBestEffort;
-struct NameParseDateTimeBestEffortOrZero;
-struct NameParseDateTimeBestEffortOrNull;
-
-template<typename Name, typename ToDataType>
-static inline bool isDateTime64(const ColumnsWithTypeAndName & arguments)
-{
-    if constexpr (std::is_same_v<ToDataType, DataTypeDateTime64>)
-        return true;
-    else if constexpr (std::is_same_v<Name, NameToDateTime> || std::is_same_v<Name, NameParseDateTimeBestEffort>
-        || std::is_same_v<Name, NameParseDateTimeBestEffortOrZero> || std::is_same_v<Name, NameParseDateTimeBestEffortOrNull>)
-    {
-        return (arguments.size() == 2 && isUInt(arguments[1].type)) || arguments.size() == 3;
-    }
-
-    return false;
-}
-
-template <typename ToDataType, typename Name, typename MonotonicityImpl>
-class FunctionConvert : public IFunction
-{
-public:
-    using Monotonic = MonotonicityImpl;
-
-    static constexpr auto name = Name::name;
-    static constexpr bool to_decimal =
-        std::is_same_v<Name, NameToDecimal32> || std::is_same_v<Name, NameToDecimal64>
-         || std::is_same_v<Name, NameToDecimal128> || std::is_same_v<Name, NameToDecimal256>;
-
-    static constexpr bool to_datetime64 = std::is_same_v<ToDataType, DataTypeDateTime64>;
-
-    static constexpr bool to_string_or_fixed_string = std::is_same_v<ToDataType, DataTypeFixedString> ||
-                                                      std::is_same_v<ToDataType, DataTypeString>;
-
-    static constexpr bool to_date_or_datetime = std::is_same_v<ToDataType, DataTypeDate> ||
-                                                std::is_same_v<ToDataType, DataTypeDate32> ||
-                                                std::is_same_v<ToDataType, DataTypeDateTime>;
-
-    static FunctionPtr create(ContextPtr context) { return std::make_shared<FunctionConvert>(context); }
-    static FunctionPtr create() { return std::make_shared<FunctionConvert>(); }
-
-    FunctionConvert() = default;
-    explicit FunctionConvert(ContextPtr context_) : context(context_) {}
-
-    String getName() const override
-    {
-        return name;
-    }
-
-    bool isVariadic() const override { return true; }
-    size_t getNumberOfArguments() const override { return 0; }
-    bool isInjective(const ColumnsWithTypeAndName &) const override { return std::is_same_v<Name, NameToString>; }
-    bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & arguments) const override
-    {
-        /// TODO: We can make more optimizations here.
-        return !(to_date_or_datetime && isNumber(*arguments[0].type));
-    }
-
-    using DefaultReturnTypeGetter = std::function<DataTypePtr(const ColumnsWithTypeAndName &)>;
-    static DataTypePtr getReturnTypeDefaultImplementationForNulls(const ColumnsWithTypeAndName & arguments, const DefaultReturnTypeGetter & getter)
-    {
-        NullPresence null_presence = getNullPresense(arguments);
-
-        if (null_presence.has_null_constant)
-        {
-            return makeNullable(std::make_shared<DataTypeNothing>());
-        }
-        if (null_presence.has_nullable)
-        {
-            auto nested_columns = Block(createBlockWithNestedColumns(arguments));
-            auto return_type = getter(ColumnsWithTypeAndName(nested_columns.begin(), nested_columns.end()));
-            return makeNullable(return_type);
-        }
-
-        return getter(arguments);
-    }
-
-    DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override
-    {
-        auto getter = [&] (const auto & args) { return getReturnTypeImplRemovedNullable(args); };
-        auto res = getReturnTypeDefaultImplementationForNulls(arguments, getter);
-        to_nullable = res->isNullable();
-        checked_return_type = true;
-        return res;
-    }
-
-    DataTypePtr getReturnTypeImplRemovedNullable(const ColumnsWithTypeAndName & arguments) const
-    {
-        FunctionArgumentDescriptors mandatory_args = {{"Value", nullptr, nullptr, nullptr}};
-        FunctionArgumentDescriptors optional_args;
-
-        if constexpr (to_decimal)
-        {
-            mandatory_args.push_back({"scale", static_cast<FunctionArgumentDescriptor::TypeValidator>(&isNativeInteger), &isColumnConst, "const Integer"});
-        }
-
-        if (!to_decimal && isDateTime64<Name, ToDataType>(arguments))
-        {
-            mandatory_args.push_back({"scale", static_cast<FunctionArgumentDescriptor::TypeValidator>(&isNativeInteger), &isColumnConst, "const Integer"});
-        }
-
-        // toString(DateTime or DateTime64, [timezone: String])
-        if ((std::is_same_v<Name, NameToString> && !arguments.empty() && (isDateTime64(arguments[0].type) || isDateTime(arguments[0].type)))
-            // toUnixTimestamp(value[, timezone : String])
-            || std::is_same_v<Name, NameToUnixTimestamp>
-            // toDate(value[, timezone : String])
-            || std::is_same_v<ToDataType, DataTypeDate> // TODO: shall we allow timestamp argument for toDate? DateTime knows nothing about timezones and this argument is ignored below.
-            // toDate32(value[, timezone : String])
-            || std::is_same_v<ToDataType, DataTypeDate32>
-            // toDateTime(value[, timezone: String])
-            || std::is_same_v<ToDataType, DataTypeDateTime>
-            // toDateTime64(value, scale : Integer[, timezone: String])
-            || std::is_same_v<ToDataType, DataTypeDateTime64>)
-        {
-            optional_args.push_back({"timezone", static_cast<FunctionArgumentDescriptor::TypeValidator>(&isString), nullptr, "String"});
-        }
-
-        validateFunctionArgumentTypes(*this, arguments, mandatory_args, optional_args);
-
-        if constexpr (std::is_same_v<ToDataType, DataTypeInterval>)
-        {
-            return std::make_shared<DataTypeInterval>(Name::kind);
-        }
-        else if constexpr (to_decimal)
-        {
-            UInt64 scale = extractToDecimalScale(arguments[1]);
-
-            if constexpr (std::is_same_v<Name, NameToDecimal32>)
-                return createDecimalMaxPrecision<Decimal32>(scale);
-            else if constexpr (std::is_same_v<Name, NameToDecimal64>)
-                return createDecimalMaxPrecision<Decimal64>(scale);
-            else if constexpr (std::is_same_v<Name, NameToDecimal128>)
-                return createDecimalMaxPrecision<Decimal128>(scale);
-            else if constexpr (std::is_same_v<Name, NameToDecimal256>)
-                return createDecimalMaxPrecision<Decimal256>(scale);
-
-            throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected branch in code of conversion function: it is a bug.");
-        }
-        else
-        {
-            // Optional second argument with time zone for DateTime.
-            UInt8 timezone_arg_position = 1;
-            UInt32 scale [[maybe_unused]] = DataTypeDateTime64::default_scale;
-
-            // DateTime64 requires more arguments: scale and timezone. Since timezone is optional, scale should be first.
-            if (isDateTime64<Name, ToDataType>(arguments))
-            {
-                timezone_arg_position += 1;
-                scale = static_cast<UInt32>(arguments[1].column->get64(0));
-
-                if (to_datetime64 || scale != 0) /// toDateTime('xxxx-xx-xx xx:xx:xx', 0) return DateTime
-                    return std::make_shared<DataTypeDateTime64>(scale,
-                        extractTimeZoneNameFromFunctionArguments(arguments, timezone_arg_position, 0, false));
-
-                return std::make_shared<DataTypeDateTime>(extractTimeZoneNameFromFunctionArguments(arguments, timezone_arg_position, 0, false));
-            }
-
-            if constexpr (std::is_same_v<ToDataType, DataTypeDateTime>)
-                return std::make_shared<DataTypeDateTime>(extractTimeZoneNameFromFunctionArguments(arguments, timezone_arg_position, 0, false));
-            else if constexpr (std::is_same_v<ToDataType, DataTypeDateTime64>)
-                throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected branch in code of conversion function: it is a bug.");
-            else
-                return std::make_shared<ToDataType>();
-        }
-    }
-
-    /// Function actually uses default implementation for nulls,
-    /// but we need to know if return type is Nullable or not,
-    /// so we use checked_return_type only to intercept the first call to getReturnTypeImpl(...).
-    bool useDefaultImplementationForNulls() const override
-    {
-        bool to_nullable_string = to_nullable && std::is_same_v<ToDataType, DataTypeString>;
-        return checked_return_type && !to_nullable_string;
-    }
-
-    bool useDefaultImplementationForConstants() const override { return true; }
-    ColumnNumbers getArgumentsThatAreAlwaysConstant() const override
-    {
-        if constexpr (std::is_same_v<ToDataType, DataTypeString>)
-            return {};
-        else if constexpr (std::is_same_v<ToDataType, DataTypeDateTime64>)
-            return {2};
-        return {1};
-    }
-    bool canBeExecutedOnDefaultArguments() const override { return false; }
-
-    ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const override
-    {
-        try
-        {
-            return executeInternal(arguments, result_type, input_rows_count);
-        }
-        catch (Exception & e)
-        {
-            /// More convenient error message.
-            if (e.code() == ErrorCodes::ATTEMPT_TO_READ_AFTER_EOF)
-            {
-                e.addMessage("Cannot parse "
-                    + result_type->getName() + " from "
-                    + arguments[0].type->getName()
-                    + ", because value is too short");
-            }
-            else if (e.code() == ErrorCodes::CANNOT_PARSE_NUMBER
-                || e.code() == ErrorCodes::CANNOT_READ_ARRAY_FROM_TEXT
-                || e.code() == ErrorCodes::CANNOT_PARSE_INPUT_ASSERTION_FAILED
-                || e.code() == ErrorCodes::CANNOT_PARSE_QUOTED_STRING
-                || e.code() == ErrorCodes::CANNOT_PARSE_ESCAPE_SEQUENCE
-                || e.code() == ErrorCodes::CANNOT_PARSE_DATE
-                || e.code() == ErrorCodes::CANNOT_PARSE_DATETIME
-                || e.code() == ErrorCodes::CANNOT_PARSE_UUID
-                || e.code() == ErrorCodes::CANNOT_PARSE_IPV4
-                || e.code() == ErrorCodes::CANNOT_PARSE_IPV6)
-            {
-                e.addMessage("Cannot parse "
-                    + result_type->getName() + " from "
-                    + arguments[0].type->getName());
-            }
-
-            throw;
-        }
-    }
-
-    bool hasInformationAboutMonotonicity() const override
-    {
-        return Monotonic::has();
-    }
-
-    Monotonicity getMonotonicityForRange(const IDataType & type, const Field & left, const Field & right) const override
-    {
-        return Monotonic::get(type, left, right);
-    }
-
-private:
-    ContextPtr context;
-    mutable bool checked_return_type = false;
-    mutable bool to_nullable = false;
-
-    ColumnPtr executeInternal(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const
-    {
-        if (arguments.empty())
-            throw Exception(ErrorCodes::TOO_FEW_ARGUMENTS_FOR_FUNCTION, "Function {} expects at least 1 argument", getName());
-
-        if (result_type->onlyNull())
-            return result_type->createColumnConstWithDefaultValue(input_rows_count);
-
-        const DataTypePtr from_type = removeNullable(arguments[0].type);
-        ColumnPtr result_column;
-
-        [[maybe_unused]] FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior = default_date_time_overflow_behavior;
-
-        if (context)
-            date_time_overflow_behavior = context->getSettingsRef().date_time_overflow_behavior.value;
-
-        auto call = [&](const auto & types, const auto & tag) -> bool
-        {
-            using Types = std::decay_t<decltype(types)>;
-            using LeftDataType = typename Types::LeftType;
-            using RightDataType = typename Types::RightType;
-            using SpecialTag = std::decay_t<decltype(tag)>;
-
-            if constexpr (IsDataTypeDecimal<RightDataType>)
-            {
-                if constexpr (std::is_same_v<RightDataType, DataTypeDateTime64>)
-                {
-                    /// Account for optional timezone argument.
-                    if (arguments.size() != 2 && arguments.size() != 3)
-                        throw Exception(ErrorCodes::TOO_FEW_ARGUMENTS_FOR_FUNCTION, "Function {} expects 2 or 3 arguments for DataTypeDateTime64.", getName());
-                }
-                else if (arguments.size() != 2)
-                {
-                    throw Exception(ErrorCodes::TOO_FEW_ARGUMENTS_FOR_FUNCTION, "Function {} expects 2 arguments for Decimal.", getName());
-                }
-
-                const ColumnWithTypeAndName & scale_column = arguments[1];
-                UInt32 scale = extractToDecimalScale(scale_column);
-
-                switch (date_time_overflow_behavior)
-                {
-                    case FormatSettings::DateTimeOverflowBehavior::Throw:
-                        result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag, FormatSettings::DateTimeOverflowBehavior::Throw>::execute(arguments, result_type, input_rows_count, scale);
-                        break;
-                    case FormatSettings::DateTimeOverflowBehavior::Ignore:
-                        result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag, FormatSettings::DateTimeOverflowBehavior::Ignore>::execute(arguments, result_type, input_rows_count, scale);
-                        break;
-                    case FormatSettings::DateTimeOverflowBehavior::Saturate:
-                        result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag, FormatSettings::DateTimeOverflowBehavior::Saturate>::execute(arguments, result_type, input_rows_count, scale);
-                        break;
-                }
-
-            }
-            else if constexpr (IsDataTypeDateOrDateTime<RightDataType> && std::is_same_v<LeftDataType, DataTypeDateTime64>)
-            {
-                const auto * dt64 = assert_cast<const DataTypeDateTime64 *>(arguments[0].type.get());
-                switch (date_time_overflow_behavior)
-                {
-                    case FormatSettings::DateTimeOverflowBehavior::Throw:
-                        result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag, FormatSettings::DateTimeOverflowBehavior::Throw>::execute(arguments, result_type, input_rows_count, dt64->getScale());
-                        break;
-                    case FormatSettings::DateTimeOverflowBehavior::Ignore:
-                        result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag, FormatSettings::DateTimeOverflowBehavior::Ignore>::execute(arguments, result_type, input_rows_count, dt64->getScale());
-                        break;
-                    case FormatSettings::DateTimeOverflowBehavior::Saturate:
-                        result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag, FormatSettings::DateTimeOverflowBehavior::Saturate>::execute(arguments, result_type, input_rows_count, dt64->getScale());
-                        break;
-                }
-            }
-#define GENERATE_OVERFLOW_MODE_CASE(OVERFLOW_MODE) \
-            case FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE: \
-                result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag, FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE>::execute( \
-                arguments, result_type, input_rows_count); \
-                break;
-
-            else if constexpr (IsDataTypeDecimalOrNumber<LeftDataType> && IsDataTypeDecimalOrNumber<RightDataType>)
-            {
-                using LeftT = typename LeftDataType::FieldType;
-                using RightT = typename RightDataType::FieldType;
-
-                static constexpr bool bad_left =
-                    is_decimal<LeftT> || std::is_floating_point_v<LeftT> || is_big_int_v<LeftT> || is_signed_v<LeftT>;
-                static constexpr bool bad_right =
-                    is_decimal<RightT> || std::is_floating_point_v<RightT> || is_big_int_v<RightT> || is_signed_v<RightT>;
-
-                /// Disallow int vs UUID conversion (but support int vs UInt128 conversion)
-                if constexpr ((bad_left && std::is_same_v<RightDataType, DataTypeUUID>) ||
-                              (bad_right && std::is_same_v<LeftDataType, DataTypeUUID>))
-                {
-                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Wrong UUID conversion");
-                }
-                else
-                {
-                    switch (date_time_overflow_behavior)
-                    {
-                        GENERATE_OVERFLOW_MODE_CASE(Throw)
-                        GENERATE_OVERFLOW_MODE_CASE(Ignore)
-                        GENERATE_OVERFLOW_MODE_CASE(Saturate)
-                    }
-                }
-            }
-            else if constexpr ((IsDataTypeNumber<LeftDataType> || IsDataTypeDateOrDateTime<LeftDataType>)
-                               && IsDataTypeDateOrDateTime<RightDataType>)
-            {
-                switch (date_time_overflow_behavior)
-                {
-                    GENERATE_OVERFLOW_MODE_CASE(Throw)
-                    GENERATE_OVERFLOW_MODE_CASE(Ignore)
-                    GENERATE_OVERFLOW_MODE_CASE(Saturate)
-                }
-            }
-#undef GENERATE_OVERFLOW_MODE_CASE
-            else
-                  result_column = ConvertImpl<LeftDataType, RightDataType, Name, SpecialTag>::execute(arguments, result_type, input_rows_count);
-
-            return true;
-        };
-
-        if (isDateTime64<Name, ToDataType>(arguments))
-        {
-            /// For toDateTime('xxxx-xx-xx xx:xx:xx.00', 2[, 'timezone']) we need to it convert to DateTime64
-            const ColumnWithTypeAndName & scale_column = arguments[1];
-            UInt32 scale = extractToDecimalScale(scale_column);
-
-            if (to_datetime64 || scale != 0) /// When scale = 0, the data type is DateTime otherwise the data type is DateTime64
-            {
-                if (!callOnIndexAndDataType<DataTypeDateTime64>(from_type->getTypeId(), call, ConvertDefaultBehaviorTag{}))
-                    throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of argument of function {}",
-                                    arguments[0].type->getName(), getName());
-
-                return result_column;
-            }
-        }
-
-        if constexpr (std::is_same_v<ToDataType, DataTypeString>)
-        {
-            if (from_type->getCustomSerialization())
-                return ConvertImplGenericToString<ColumnString>::execute(arguments, result_type, input_rows_count);
-        }
-
-        bool done = false;
-        if constexpr (to_string_or_fixed_string)
-        {
-            done = callOnIndexAndDataType<ToDataType>(from_type->getTypeId(), call, ConvertDefaultBehaviorTag{});
-        }
-        else
-        {
-            bool cast_ipv4_ipv6_default_on_conversion_error = false;
-            if constexpr (is_any_of<ToDataType, DataTypeIPv4, DataTypeIPv6>)
-                if (context && (cast_ipv4_ipv6_default_on_conversion_error = context->getSettingsRef().cast_ipv4_ipv6_default_on_conversion_error))
-                    done = callOnIndexAndDataType<ToDataType>(from_type->getTypeId(), call, ConvertReturnZeroOnErrorTag{});
-
-            if (!cast_ipv4_ipv6_default_on_conversion_error)
-            {
-                /// We should use ConvertFromStringExceptionMode::Null mode when converting from String (or FixedString)
-                /// to Nullable type, to avoid 'value is too short' error on attempt to parse empty string from NULL values.
-                if (to_nullable && WhichDataType(from_type).isStringOrFixedString())
-                    done = callOnIndexAndDataType<ToDataType>(from_type->getTypeId(), call, ConvertReturnNullOnErrorTag{});
-                else
-                    done = callOnIndexAndDataType<ToDataType>(from_type->getTypeId(), call, ConvertDefaultBehaviorTag{});
-            }
-        }
-
-        if (!done)
-        {
-            /// Generic conversion of any type to String.
-            if (std::is_same_v<ToDataType, DataTypeString>)
-            {
-                return ConvertImplGenericToString<ColumnString>::execute(arguments, result_type, input_rows_count);
-            }
-            else
-                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of argument of function {}",
-                    arguments[0].type->getName(), getName());
-        }
-
-        return result_column;
-    }
-};
-
-
-/** Function toTOrZero (where T is number of date or datetime type):
-  *  try to convert from String to type T through parsing,
-  *  if cannot parse, return default value instead of throwing exception.
-  * Function toTOrNull will return Nullable type with NULL when cannot parse.
-  * NOTE Also need to implement tryToUnixTimestamp with timezone.
-  */
-template <typename ToDataType, typename Name,
-    ConvertFromStringExceptionMode exception_mode,
-    ConvertFromStringParsingMode parsing_mode = ConvertFromStringParsingMode::Normal>
-class FunctionConvertFromString : public IFunction
-{
-public:
-    static constexpr auto name = Name::name;
-    static constexpr bool to_decimal =
-        std::is_same_v<ToDataType, DataTypeDecimal<Decimal32>> ||
-        std::is_same_v<ToDataType, DataTypeDecimal<Decimal64>> ||
-        std::is_same_v<ToDataType, DataTypeDecimal<Decimal128>> ||
-        std::is_same_v<ToDataType, DataTypeDecimal<Decimal256>>;
-
-    static constexpr bool to_datetime64 = std::is_same_v<ToDataType, DataTypeDateTime64>;
-
-    static FunctionPtr create(ContextPtr) { return std::make_shared<FunctionConvertFromString>(); }
-    static FunctionPtr create() { return std::make_shared<FunctionConvertFromString>(); }
-
-    String getName() const override
-    {
-        return name;
-    }
-
-    bool isVariadic() const override { return true; }
-    bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return true; }
-    size_t getNumberOfArguments() const override { return 0; }
-
-    bool useDefaultImplementationForConstants() const override { return true; }
-    bool canBeExecutedOnDefaultArguments() const override { return false; }
-
-    ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {1}; }
-
-    DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override
-    {
-        DataTypePtr res;
-
-        if (isDateTime64<Name, ToDataType>(arguments))
-        {
-            validateFunctionArgumentTypes(*this, arguments,
-                FunctionArgumentDescriptors{{"string", static_cast<FunctionArgumentDescriptor::TypeValidator>(&isStringOrFixedString), nullptr, "String or FixedString"}},
-                // optional
-                FunctionArgumentDescriptors{
-                    {"precision", static_cast<FunctionArgumentDescriptor::TypeValidator>(&isUInt8), isColumnConst, "const UInt8"},
-                    {"timezone", static_cast<FunctionArgumentDescriptor::TypeValidator>(&isStringOrFixedString), isColumnConst, "const String or FixedString"},
-                });
-
-            UInt64 scale = to_datetime64 ? DataTypeDateTime64::default_scale : 0;
-            if (arguments.size() > 1)
-                scale = extractToDecimalScale(arguments[1]);
-            const auto timezone = extractTimeZoneNameFromFunctionArguments(arguments, 2, 0, false);
-
-            res = scale == 0 ? res = std::make_shared<DataTypeDateTime>(timezone) : std::make_shared<DataTypeDateTime64>(scale, timezone);
-        }
-        else
-        {
-            if ((arguments.size() != 1 && arguments.size() != 2) || (to_decimal && arguments.size() != 2))
-                throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH,
-                    "Number of arguments for function {} doesn't match: passed {}, should be 1 or 2. "
-                    "Second argument only make sense for DateTime (time zone, optional) and Decimal (scale).",
-                    getName(), arguments.size());
-
-            if (!isStringOrFixedString(arguments[0].type))
-            {
-                if (this->getName().find("OrZero") != std::string::npos ||
-                    this->getName().find("OrNull") != std::string::npos)
-                    throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of first argument of function {}. "
-                            "Conversion functions with postfix 'OrZero' or 'OrNull' should take String argument",
-                            arguments[0].type->getName(), getName());
-                else
-                    throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of first argument of function {}",
-                            arguments[0].type->getName(), getName());
-            }
-
-            if (arguments.size() == 2)
-            {
-                if constexpr (std::is_same_v<ToDataType, DataTypeDateTime>)
-                {
-                    if (!isString(arguments[1].type))
-                        throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of 2nd argument of function {}",
-                            arguments[1].type->getName(), getName());
-                }
-                else if constexpr (to_decimal)
-                {
-                    if (!isInteger(arguments[1].type))
-                        throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of 2nd argument of function {}",
-                            arguments[1].type->getName(), getName());
-                    if (!arguments[1].column)
-                        throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Second argument for function {} must be constant", getName());
-                }
-                else
-                {
-                    throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH,
-                        "Number of arguments for function {} doesn't match: passed {}, should be 1. "
-                        "Second argument makes sense only for DateTime and Decimal.",
-                        getName(), arguments.size());
-                }
-            }
-
-            if constexpr (std::is_same_v<ToDataType, DataTypeDateTime>)
-                res = std::make_shared<DataTypeDateTime>(extractTimeZoneNameFromFunctionArguments(arguments, 1, 0, false));
-            else if constexpr (std::is_same_v<ToDataType, DataTypeDateTime64>)
-                throw Exception(ErrorCodes::LOGICAL_ERROR, "MaterializedMySQL is a bug.");
-            else if constexpr (to_decimal)
-            {
-                UInt64 scale = extractToDecimalScale(arguments[1]);
-                res = createDecimalMaxPrecision<typename ToDataType::FieldType>(scale);
-                if (!res)
-                    throw Exception(ErrorCodes::LOGICAL_ERROR, "Something wrong with toDecimalNNOrZero() or toDecimalNNOrNull()");
-            }
-            else
-                res = std::make_shared<ToDataType>();
-        }
-
-        if constexpr (exception_mode == ConvertFromStringExceptionMode::Null)
-            res = std::make_shared<DataTypeNullable>(res);
-
-        return res;
-    }
-
-    template <typename ConvertToDataType>
-    ColumnPtr executeInternal(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count, UInt32 scale = 0) const
-    {
-        const IDataType * from_type = arguments[0].type.get();
-
-        if (checkAndGetDataType<DataTypeString>(from_type))
-        {
-            return ConvertThroughParsing<DataTypeString, ConvertToDataType, Name, exception_mode, parsing_mode>::execute(
-                arguments, result_type, input_rows_count, scale);
-        }
-        else if (checkAndGetDataType<DataTypeFixedString>(from_type))
-        {
-            return ConvertThroughParsing<DataTypeFixedString, ConvertToDataType, Name, exception_mode, parsing_mode>::execute(
-                arguments, result_type, input_rows_count, scale);
-        }
-
-        return nullptr;
-    }
-
-    ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const override
-    {
-        ColumnPtr result_column;
-
-        if constexpr (to_decimal)
-            result_column = executeInternal<ToDataType>(arguments, result_type, input_rows_count,
-                assert_cast<const ToDataType &>(*removeNullable(result_type)).getScale());
-        else
-        {
-            if (isDateTime64<Name, ToDataType>(arguments))
-            {
-                UInt64 scale = to_datetime64 ? DataTypeDateTime64::default_scale : 0;
-                if (arguments.size() > 1)
-                    scale = extractToDecimalScale(arguments[1]);
-
-                if (scale == 0)
-                    result_column = executeInternal<DataTypeDateTime>(arguments, result_type, input_rows_count);
-                else
-                {
-                    result_column = executeInternal<DataTypeDateTime64>(arguments, result_type, input_rows_count, static_cast<UInt32>(scale));
-                }
-            }
-            else
-            {
-                result_column = executeInternal<ToDataType>(arguments, result_type, input_rows_count);
-            }
-        }
-
-        if (!result_column)
-            throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of argument of function {}. "
-                "Only String or FixedString argument is accepted for try-conversion function. For other arguments, "
-                "use function without 'orZero' or 'orNull'.", arguments[0].type->getName(), getName());
-
-        return result_column;
-    }
-};
-
-
-/// Monotonicity.
-
-struct PositiveMonotonicity
-{
-    static bool has() { return true; }
-    static IFunction::Monotonicity get(const IDataType &, const Field &, const Field &)
-    {
-        return { .is_monotonic = true };
-    }
-};
-
-struct UnknownMonotonicity
-{
-    static bool has() { return false; }
-    static IFunction::Monotonicity get(const IDataType &, const Field &, const Field &)
-    {
-        return { };
-    }
-};
-
-template <typename T>
-struct ToNumberMonotonicity
-{
-    static bool has() { return true; }
-
-    static UInt64 divideByRangeOfType(UInt64 x)
-    {
-        if constexpr (sizeof(T) < sizeof(UInt64))
-            return x >> (sizeof(T) * 8);
-        else
-            return 0;
-    }
-
-    static IFunction::Monotonicity get(const IDataType & type, const Field & left, const Field & right)
-    {
-        if (!type.isValueRepresentedByNumber())
-            return {};
-
-        /// If type is same, the conversion is always monotonic.
-        /// (Enum has separate case, because it is different data type)
-        if (checkAndGetDataType<DataTypeNumber<T>>(&type) ||
-            checkAndGetDataType<DataTypeEnum<T>>(&type))
-            return { .is_monotonic = true, .is_always_monotonic = true };
-
-        /// Float cases.
-
-        /// When converting to Float, the conversion is always monotonic.
-        if constexpr (std::is_floating_point_v<T>)
-            return { .is_monotonic = true, .is_always_monotonic = true };
-
-        const auto * low_cardinality = typeid_cast<const DataTypeLowCardinality *>(&type);
-        const IDataType * low_cardinality_dictionary_type = nullptr;
-        if (low_cardinality)
-            low_cardinality_dictionary_type = low_cardinality->getDictionaryType().get();
-
-        WhichDataType which_type(type);
-        WhichDataType which_inner_type = low_cardinality
-            ? WhichDataType(low_cardinality_dictionary_type)
-            : WhichDataType(type);
-
-        /// If converting from Float, for monotonicity, arguments must fit in range of result type.
-        if (which_inner_type.isFloat())
-        {
-            if (left.isNull() || right.isNull())
-                return {};
-
-            Float64 left_float = left.get<Float64>();
-            Float64 right_float = right.get<Float64>();
-
-            if (left_float >= static_cast<Float64>(std::numeric_limits<T>::min())
-                && left_float <= static_cast<Float64>(std::numeric_limits<T>::max())
-                && right_float >= static_cast<Float64>(std::numeric_limits<T>::min())
-                && right_float <= static_cast<Float64>(std::numeric_limits<T>::max()))
-                return { .is_monotonic = true };
-
-            return {};
-        }
-
-        /// Integer cases.
-
-        /// Only support types represented by native integers.
-        /// It can be extended to big integers, decimals and DateTime64 later.
-        /// By the way, NULLs are representing unbounded ranges.
-        if (!((left.isNull() || left.getType() == Field::Types::UInt64 || left.getType() == Field::Types::Int64)
-            && (right.isNull() || right.getType() == Field::Types::UInt64 || right.getType() == Field::Types::Int64)))
-            return {};
-
-        const bool from_is_unsigned = type.isValueRepresentedByUnsignedInteger();
-        const bool to_is_unsigned = is_unsigned_v<T>;
-
-        const size_t size_of_from = type.getSizeOfValueInMemory();
-        const size_t size_of_to = sizeof(T);
-
-        const bool left_in_first_half = left.isNull()
-            ? from_is_unsigned
-            : (left.get<Int64>() >= 0);
-
-        const bool right_in_first_half = right.isNull()
-            ? !from_is_unsigned
-            : (right.get<Int64>() >= 0);
-
-        /// Size of type is the same.
-        if (size_of_from == size_of_to)
-        {
-            if (from_is_unsigned == to_is_unsigned)
-                return { .is_monotonic = true, .is_always_monotonic = true };
-
-            if (left_in_first_half == right_in_first_half)
-                return { .is_monotonic = true };
-
-            return {};
-        }
-
-        /// Size of type is expanded.
-        if (size_of_from < size_of_to)
-        {
-            if (from_is_unsigned == to_is_unsigned)
-                return { .is_monotonic = true, .is_always_monotonic = true };
-
-            if (!to_is_unsigned)
-                return { .is_monotonic = true, .is_always_monotonic = true };
-
-            /// signed -> unsigned. If arguments from the same half, then function is monotonic.
-            if (left_in_first_half == right_in_first_half)
-                return { .is_monotonic = true };
-
-            return {};
-        }
-
-        /// Size of type is shrunk.
-        if (size_of_from > size_of_to)
-        {
-            /// Function cannot be monotonic on unbounded ranges.
-            if (left.isNull() || right.isNull())
-                return {};
-
-            /// Function cannot be monotonic when left and right are not on the same ranges.
-            if (divideByRangeOfType(left.get<UInt64>()) != divideByRangeOfType(right.get<UInt64>()))
-                return {};
-
-            if (to_is_unsigned)
-                return { .is_monotonic = true };
-            else
-            {
-                // If To is signed, it's possible that the signedness is different after conversion. So we check it explicitly.
-                const bool is_monotonic = (T(left.get<UInt64>()) >= 0) == (T(right.get<UInt64>()) >= 0);
-
-                return { .is_monotonic = is_monotonic };
-            }
-        }
-
-        UNREACHABLE();
-    }
-};
-
-struct ToDateMonotonicity
-{
-    static bool has() { return true; }
-
-    static IFunction::Monotonicity get(const IDataType & type, const Field & left, const Field & right)
-    {
-        auto which = WhichDataType(type);
-        if (which.isDateOrDate32() || which.isDateTime() || which.isDateTime64() || which.isInt8() || which.isInt16() || which.isUInt8()
-            || which.isUInt16())
-        {
-            return {.is_monotonic = true, .is_always_monotonic = true};
-        }
-        else if (
-            ((left.getType() == Field::Types::UInt64 || left.isNull()) && (right.getType() == Field::Types::UInt64 || right.isNull())
-             && ((left.isNull() || left.get<UInt64>() < 0xFFFF) && (right.isNull() || right.get<UInt64>() >= 0xFFFF)))
-            || ((left.getType() == Field::Types::Int64 || left.isNull()) && (right.getType() == Field::Types::Int64 || right.isNull())
-                && ((left.isNull() || left.get<Int64>() < 0xFFFF) && (right.isNull() || right.get<Int64>() >= 0xFFFF)))
-            || ((
-                (left.getType() == Field::Types::Float64 || left.isNull())
-                && (right.getType() == Field::Types::Float64 || right.isNull())
-                && ((left.isNull() || left.get<Float64>() < 0xFFFF) && (right.isNull() || right.get<Float64>() >= 0xFFFF))))
-            || !isNativeNumber(type))
-        {
-            return {};
-        }
-        else
-        {
-            return {.is_monotonic = true, .is_always_monotonic = true};
-        }
-    }
-};
-
-struct ToDateTimeMonotonicity
-{
-    static bool has() { return true; }
-
-    static IFunction::Monotonicity get(const IDataType & type, const Field &, const Field &)
-    {
-        if (type.isValueRepresentedByNumber())
-            return {.is_monotonic = true, .is_always_monotonic = true};
-        else
-            return {};
-    }
-};
-
-/** The monotonicity for the `toString` function is mainly determined for test purposes.
-  * It is doubtful that anyone is looking to optimize queries with conditions `toString(CounterID) = 34`.
-  */
-struct ToStringMonotonicity
-{
-    static bool has() { return true; }
-
-    static IFunction::Monotonicity get(const IDataType & type, const Field & left, const Field & right)
-    {
-        IFunction::Monotonicity positive{ .is_monotonic = true };
-        IFunction::Monotonicity not_monotonic;
-
-        const auto * type_ptr = &type;
-        if (const auto * low_cardinality_type = checkAndGetDataType<DataTypeLowCardinality>(type_ptr))
-            type_ptr = low_cardinality_type->getDictionaryType().get();
-
-        /// Order on enum values (which is the order on integers) is completely arbitrary in respect to the order on strings.
-        if (WhichDataType(type).isEnum())
-            return not_monotonic;
-
-        /// `toString` function is monotonous if the argument is Date or Date32 or DateTime or String, or non-negative numbers with the same number of symbols.
-        if (checkDataTypes<DataTypeDate, DataTypeDate32, DataTypeDateTime, DataTypeString>(type_ptr))
-            return positive;
-
-        if (left.isNull() || right.isNull())
-            return {};
-
-        if (left.getType() == Field::Types::UInt64
-            && right.getType() == Field::Types::UInt64)
-        {
-            return (left.get<Int64>() == 0 && right.get<Int64>() == 0)
-                || (floor(log10(left.get<UInt64>())) == floor(log10(right.get<UInt64>())))
-                ? positive : not_monotonic;
-        }
-
-        if (left.getType() == Field::Types::Int64
-            && right.getType() == Field::Types::Int64)
-        {
-            return (left.get<Int64>() == 0 && right.get<Int64>() == 0)
-                || (left.get<Int64>() > 0 && right.get<Int64>() > 0 && floor(log10(left.get<Int64>())) == floor(log10(right.get<Int64>())))
-                ? positive : not_monotonic;
-        }
-
-        return not_monotonic;
-    }
-};
-
-
-struct NameToUInt8 { static constexpr auto name = "toUInt8"; };
-struct NameToUInt16 { static constexpr auto name = "toUInt16"; };
-struct NameToUInt32 { static constexpr auto name = "toUInt32"; };
-struct NameToUInt64 { static constexpr auto name = "toUInt64"; };
-struct NameToUInt128 { static constexpr auto name = "toUInt128"; };
-struct NameToUInt256 { static constexpr auto name = "toUInt256"; };
-struct NameToInt8 { static constexpr auto name = "toInt8"; };
-struct NameToInt16 { static constexpr auto name = "toInt16"; };
-struct NameToInt32 { static constexpr auto name = "toInt32"; };
-struct NameToInt64 { static constexpr auto name = "toInt64"; };
-struct NameToInt128 { static constexpr auto name = "toInt128"; };
-struct NameToInt256 { static constexpr auto name = "toInt256"; };
-struct NameToFloat32 { static constexpr auto name = "toFloat32"; };
-struct NameToFloat64 { static constexpr auto name = "toFloat64"; };
-struct NameToUUID { static constexpr auto name = "toUUID"; };
-struct NameToIPv4 { static constexpr auto name = "toIPv4"; };
-struct NameToIPv6 { static constexpr auto name = "toIPv6"; };
-
-using FunctionToUInt8 = FunctionConvert<DataTypeUInt8, NameToUInt8, ToNumberMonotonicity<UInt8>>;
-using FunctionToUInt16 = FunctionConvert<DataTypeUInt16, NameToUInt16, ToNumberMonotonicity<UInt16>>;
-using FunctionToUInt32 = FunctionConvert<DataTypeUInt32, NameToUInt32, ToNumberMonotonicity<UInt32>>;
-using FunctionToUInt64 = FunctionConvert<DataTypeUInt64, NameToUInt64, ToNumberMonotonicity<UInt64>>;
-using FunctionToUInt128 = FunctionConvert<DataTypeUInt128, NameToUInt128, ToNumberMonotonicity<UInt128>>;
-using FunctionToUInt256 = FunctionConvert<DataTypeUInt256, NameToUInt256, ToNumberMonotonicity<UInt256>>;
-using FunctionToInt8 = FunctionConvert<DataTypeInt8, NameToInt8, ToNumberMonotonicity<Int8>>;
-using FunctionToInt16 = FunctionConvert<DataTypeInt16, NameToInt16, ToNumberMonotonicity<Int16>>;
-using FunctionToInt32 = FunctionConvert<DataTypeInt32, NameToInt32, ToNumberMonotonicity<Int32>>;
-using FunctionToInt64 = FunctionConvert<DataTypeInt64, NameToInt64, ToNumberMonotonicity<Int64>>;
-using FunctionToInt128 = FunctionConvert<DataTypeInt128, NameToInt128, ToNumberMonotonicity<Int128>>;
-using FunctionToInt256 = FunctionConvert<DataTypeInt256, NameToInt256, ToNumberMonotonicity<Int256>>;
-using FunctionToFloat32 = FunctionConvert<DataTypeFloat32, NameToFloat32, ToNumberMonotonicity<Float32>>;
-using FunctionToFloat64 = FunctionConvert<DataTypeFloat64, NameToFloat64, ToNumberMonotonicity<Float64>>;
-
-using FunctionToDate = FunctionConvert<DataTypeDate, NameToDate, ToDateMonotonicity>;
-
-using FunctionToDate32 = FunctionConvert<DataTypeDate32, NameToDate32, ToDateMonotonicity>;
-
-using FunctionToDateTime = FunctionConvert<DataTypeDateTime, NameToDateTime, ToDateTimeMonotonicity>;
-
-using FunctionToDateTime32 = FunctionConvert<DataTypeDateTime, NameToDateTime32, ToDateTimeMonotonicity>;
-
-using FunctionToDateTime64 = FunctionConvert<DataTypeDateTime64, NameToDateTime64, ToDateTimeMonotonicity>;
-
-using FunctionToUUID = FunctionConvert<DataTypeUUID, NameToUUID, ToNumberMonotonicity<UInt128>>;
-using FunctionToIPv4 = FunctionConvert<DataTypeIPv4, NameToIPv4, ToNumberMonotonicity<UInt32>>;
-using FunctionToIPv6 = FunctionConvert<DataTypeIPv6, NameToIPv6, ToNumberMonotonicity<UInt128>>;
-using FunctionToString = FunctionConvert<DataTypeString, NameToString, ToStringMonotonicity>;
-using FunctionToUnixTimestamp = FunctionConvert<DataTypeUInt32, NameToUnixTimestamp, ToNumberMonotonicity<UInt32>>;
-using FunctionToDecimal32 = FunctionConvert<DataTypeDecimal<Decimal32>, NameToDecimal32, UnknownMonotonicity>;
-using FunctionToDecimal64 = FunctionConvert<DataTypeDecimal<Decimal64>, NameToDecimal64, UnknownMonotonicity>;
-using FunctionToDecimal128 = FunctionConvert<DataTypeDecimal<Decimal128>, NameToDecimal128, UnknownMonotonicity>;
-using FunctionToDecimal256 = FunctionConvert<DataTypeDecimal<Decimal256>, NameToDecimal256, UnknownMonotonicity>;
-
-template <typename DataType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior = default_date_time_overflow_behavior> struct FunctionTo;
-
-template <> struct FunctionTo<DataTypeUInt8> { using Type = FunctionToUInt8; };
-template <> struct FunctionTo<DataTypeUInt16> { using Type = FunctionToUInt16; };
-template <> struct FunctionTo<DataTypeUInt32> { using Type = FunctionToUInt32; };
-template <> struct FunctionTo<DataTypeUInt64> { using Type = FunctionToUInt64; };
-template <> struct FunctionTo<DataTypeUInt128> { using Type = FunctionToUInt128; };
-template <> struct FunctionTo<DataTypeUInt256> { using Type = FunctionToUInt256; };
-template <> struct FunctionTo<DataTypeInt8> { using Type = FunctionToInt8; };
-template <> struct FunctionTo<DataTypeInt16> { using Type = FunctionToInt16; };
-template <> struct FunctionTo<DataTypeInt32> { using Type = FunctionToInt32; };
-template <> struct FunctionTo<DataTypeInt64> { using Type = FunctionToInt64; };
-template <> struct FunctionTo<DataTypeInt128> { using Type = FunctionToInt128; };
-template <> struct FunctionTo<DataTypeInt256> { using Type = FunctionToInt256; };
-template <> struct FunctionTo<DataTypeFloat32> { using Type = FunctionToFloat32; };
-template <> struct FunctionTo<DataTypeFloat64> { using Type = FunctionToFloat64; };
-
-template <FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct FunctionTo<DataTypeDate, date_time_overflow_behavior> { using Type = FunctionToDate; };
-
-template <FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct FunctionTo<DataTypeDate32, date_time_overflow_behavior> { using Type = FunctionToDate32; };
-
-template <FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct FunctionTo<DataTypeDateTime, date_time_overflow_behavior> { using Type = FunctionToDateTime; };
-
-template <FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
-struct FunctionTo<DataTypeDateTime64, date_time_overflow_behavior> { using Type = FunctionToDateTime64; };
-
-template <> struct FunctionTo<DataTypeUUID> { using Type = FunctionToUUID; };
-template <> struct FunctionTo<DataTypeIPv4> { using Type = FunctionToIPv4; };
-template <> struct FunctionTo<DataTypeIPv6> { using Type = FunctionToIPv6; };
-template <> struct FunctionTo<DataTypeString> { using Type = FunctionToString; };
-template <> struct FunctionTo<DataTypeFixedString> { using Type = FunctionToFixedString; };
-template <> struct FunctionTo<DataTypeDecimal<Decimal32>> { using Type = FunctionToDecimal32; };
-template <> struct FunctionTo<DataTypeDecimal<Decimal64>> { using Type = FunctionToDecimal64; };
-template <> struct FunctionTo<DataTypeDecimal<Decimal128>> { using Type = FunctionToDecimal128; };
-template <> struct FunctionTo<DataTypeDecimal<Decimal256>> { using Type = FunctionToDecimal256; };
-
-template <typename FieldType> struct FunctionTo<DataTypeEnum<FieldType>>
-    : FunctionTo<DataTypeNumber<FieldType>>
-{
-};
-
-struct NameToUInt8OrZero { static constexpr auto name = "toUInt8OrZero"; };
-struct NameToUInt16OrZero { static constexpr auto name = "toUInt16OrZero"; };
-struct NameToUInt32OrZero { static constexpr auto name = "toUInt32OrZero"; };
-struct NameToUInt64OrZero { static constexpr auto name = "toUInt64OrZero"; };
-struct NameToUInt128OrZero { static constexpr auto name = "toUInt128OrZero"; };
-struct NameToUInt256OrZero { static constexpr auto name = "toUInt256OrZero"; };
-struct NameToInt8OrZero { static constexpr auto name = "toInt8OrZero"; };
-struct NameToInt16OrZero { static constexpr auto name = "toInt16OrZero"; };
-struct NameToInt32OrZero { static constexpr auto name = "toInt32OrZero"; };
-struct NameToInt64OrZero { static constexpr auto name = "toInt64OrZero"; };
-struct NameToInt128OrZero { static constexpr auto name = "toInt128OrZero"; };
-struct NameToInt256OrZero { static constexpr auto name = "toInt256OrZero"; };
-struct NameToFloat32OrZero { static constexpr auto name = "toFloat32OrZero"; };
-struct NameToFloat64OrZero { static constexpr auto name = "toFloat64OrZero"; };
-struct NameToDateOrZero { static constexpr auto name = "toDateOrZero"; };
-struct NameToDate32OrZero { static constexpr auto name = "toDate32OrZero"; };
-struct NameToDateTimeOrZero { static constexpr auto name = "toDateTimeOrZero"; };
-struct NameToDateTime64OrZero { static constexpr auto name = "toDateTime64OrZero"; };
-struct NameToDecimal32OrZero { static constexpr auto name = "toDecimal32OrZero"; };
-struct NameToDecimal64OrZero { static constexpr auto name = "toDecimal64OrZero"; };
-struct NameToDecimal128OrZero { static constexpr auto name = "toDecimal128OrZero"; };
-struct NameToDecimal256OrZero { static constexpr auto name = "toDecimal256OrZero"; };
-struct NameToUUIDOrZero { static constexpr auto name = "toUUIDOrZero"; };
-struct NameToIPv4OrZero { static constexpr auto name = "toIPv4OrZero"; };
-struct NameToIPv6OrZero { static constexpr auto name = "toIPv6OrZero"; };
-
-using FunctionToUInt8OrZero = FunctionConvertFromString<DataTypeUInt8, NameToUInt8OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToUInt16OrZero = FunctionConvertFromString<DataTypeUInt16, NameToUInt16OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToUInt32OrZero = FunctionConvertFromString<DataTypeUInt32, NameToUInt32OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToUInt64OrZero = FunctionConvertFromString<DataTypeUInt64, NameToUInt64OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToUInt128OrZero = FunctionConvertFromString<DataTypeUInt128, NameToUInt128OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToUInt256OrZero = FunctionConvertFromString<DataTypeUInt256, NameToUInt256OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToInt8OrZero = FunctionConvertFromString<DataTypeInt8, NameToInt8OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToInt16OrZero = FunctionConvertFromString<DataTypeInt16, NameToInt16OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToInt32OrZero = FunctionConvertFromString<DataTypeInt32, NameToInt32OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToInt64OrZero = FunctionConvertFromString<DataTypeInt64, NameToInt64OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToInt128OrZero = FunctionConvertFromString<DataTypeInt128, NameToInt128OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToInt256OrZero = FunctionConvertFromString<DataTypeInt256, NameToInt256OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToFloat32OrZero = FunctionConvertFromString<DataTypeFloat32, NameToFloat32OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToFloat64OrZero = FunctionConvertFromString<DataTypeFloat64, NameToFloat64OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToDateOrZero = FunctionConvertFromString<DataTypeDate, NameToDateOrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToDate32OrZero = FunctionConvertFromString<DataTypeDate32, NameToDate32OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToDateTimeOrZero = FunctionConvertFromString<DataTypeDateTime, NameToDateTimeOrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToDateTime64OrZero = FunctionConvertFromString<DataTypeDateTime64, NameToDateTime64OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToDecimal32OrZero = FunctionConvertFromString<DataTypeDecimal<Decimal32>, NameToDecimal32OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToDecimal64OrZero = FunctionConvertFromString<DataTypeDecimal<Decimal64>, NameToDecimal64OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToDecimal128OrZero = FunctionConvertFromString<DataTypeDecimal<Decimal128>, NameToDecimal128OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToDecimal256OrZero = FunctionConvertFromString<DataTypeDecimal<Decimal256>, NameToDecimal256OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToUUIDOrZero = FunctionConvertFromString<DataTypeUUID, NameToUUIDOrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToIPv4OrZero = FunctionConvertFromString<DataTypeIPv4, NameToIPv4OrZero, ConvertFromStringExceptionMode::Zero>;
-using FunctionToIPv6OrZero = FunctionConvertFromString<DataTypeIPv6, NameToIPv6OrZero, ConvertFromStringExceptionMode::Zero>;
-
-struct NameToUInt8OrNull { static constexpr auto name = "toUInt8OrNull"; };
-struct NameToUInt16OrNull { static constexpr auto name = "toUInt16OrNull"; };
-struct NameToUInt32OrNull { static constexpr auto name = "toUInt32OrNull"; };
-struct NameToUInt64OrNull { static constexpr auto name = "toUInt64OrNull"; };
-struct NameToUInt128OrNull { static constexpr auto name = "toUInt128OrNull"; };
-struct NameToUInt256OrNull { static constexpr auto name = "toUInt256OrNull"; };
-struct NameToInt8OrNull { static constexpr auto name = "toInt8OrNull"; };
-struct NameToInt16OrNull { static constexpr auto name = "toInt16OrNull"; };
-struct NameToInt32OrNull { static constexpr auto name = "toInt32OrNull"; };
-struct NameToInt64OrNull { static constexpr auto name = "toInt64OrNull"; };
-struct NameToInt128OrNull { static constexpr auto name = "toInt128OrNull"; };
-struct NameToInt256OrNull { static constexpr auto name = "toInt256OrNull"; };
-struct NameToFloat32OrNull { static constexpr auto name = "toFloat32OrNull"; };
-struct NameToFloat64OrNull { static constexpr auto name = "toFloat64OrNull"; };
-struct NameToDateOrNull { static constexpr auto name = "toDateOrNull"; };
-struct NameToDate32OrNull { static constexpr auto name = "toDate32OrNull"; };
-struct NameToDateTimeOrNull { static constexpr auto name = "toDateTimeOrNull"; };
-struct NameToDateTime64OrNull { static constexpr auto name = "toDateTime64OrNull"; };
-struct NameToDecimal32OrNull { static constexpr auto name = "toDecimal32OrNull"; };
-struct NameToDecimal64OrNull { static constexpr auto name = "toDecimal64OrNull"; };
-struct NameToDecimal128OrNull { static constexpr auto name = "toDecimal128OrNull"; };
-struct NameToDecimal256OrNull { static constexpr auto name = "toDecimal256OrNull"; };
-struct NameToUUIDOrNull { static constexpr auto name = "toUUIDOrNull"; };
-struct NameToIPv4OrNull { static constexpr auto name = "toIPv4OrNull"; };
-struct NameToIPv6OrNull { static constexpr auto name = "toIPv6OrNull"; };
-
-using FunctionToUInt8OrNull = FunctionConvertFromString<DataTypeUInt8, NameToUInt8OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToUInt16OrNull = FunctionConvertFromString<DataTypeUInt16, NameToUInt16OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToUInt32OrNull = FunctionConvertFromString<DataTypeUInt32, NameToUInt32OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToUInt64OrNull = FunctionConvertFromString<DataTypeUInt64, NameToUInt64OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToUInt128OrNull = FunctionConvertFromString<DataTypeUInt128, NameToUInt128OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToUInt256OrNull = FunctionConvertFromString<DataTypeUInt256, NameToUInt256OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToInt8OrNull = FunctionConvertFromString<DataTypeInt8, NameToInt8OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToInt16OrNull = FunctionConvertFromString<DataTypeInt16, NameToInt16OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToInt32OrNull = FunctionConvertFromString<DataTypeInt32, NameToInt32OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToInt64OrNull = FunctionConvertFromString<DataTypeInt64, NameToInt64OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToInt128OrNull = FunctionConvertFromString<DataTypeInt128, NameToInt128OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToInt256OrNull = FunctionConvertFromString<DataTypeInt256, NameToInt256OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToFloat32OrNull = FunctionConvertFromString<DataTypeFloat32, NameToFloat32OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToFloat64OrNull = FunctionConvertFromString<DataTypeFloat64, NameToFloat64OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToDateOrNull = FunctionConvertFromString<DataTypeDate, NameToDateOrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToDate32OrNull = FunctionConvertFromString<DataTypeDate32, NameToDate32OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToDateTimeOrNull = FunctionConvertFromString<DataTypeDateTime, NameToDateTimeOrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToDateTime64OrNull = FunctionConvertFromString<DataTypeDateTime64, NameToDateTime64OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToDecimal32OrNull = FunctionConvertFromString<DataTypeDecimal<Decimal32>, NameToDecimal32OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToDecimal64OrNull = FunctionConvertFromString<DataTypeDecimal<Decimal64>, NameToDecimal64OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToDecimal128OrNull = FunctionConvertFromString<DataTypeDecimal<Decimal128>, NameToDecimal128OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToDecimal256OrNull = FunctionConvertFromString<DataTypeDecimal<Decimal256>, NameToDecimal256OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToUUIDOrNull = FunctionConvertFromString<DataTypeUUID, NameToUUIDOrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToIPv4OrNull = FunctionConvertFromString<DataTypeIPv4, NameToIPv4OrNull, ConvertFromStringExceptionMode::Null>;
-using FunctionToIPv6OrNull = FunctionConvertFromString<DataTypeIPv6, NameToIPv6OrNull, ConvertFromStringExceptionMode::Null>;
-
-struct NameParseDateTimeBestEffort { static constexpr auto name = "parseDateTimeBestEffort"; };
-struct NameParseDateTimeBestEffortOrZero { static constexpr auto name = "parseDateTimeBestEffortOrZero"; };
-struct NameParseDateTimeBestEffortOrNull { static constexpr auto name = "parseDateTimeBestEffortOrNull"; };
-struct NameParseDateTimeBestEffortUS { static constexpr auto name = "parseDateTimeBestEffortUS"; };
-struct NameParseDateTimeBestEffortUSOrZero { static constexpr auto name = "parseDateTimeBestEffortUSOrZero"; };
-struct NameParseDateTimeBestEffortUSOrNull { static constexpr auto name = "parseDateTimeBestEffortUSOrNull"; };
-struct NameParseDateTime32BestEffort { static constexpr auto name = "parseDateTime32BestEffort"; };
-struct NameParseDateTime32BestEffortOrZero { static constexpr auto name = "parseDateTime32BestEffortOrZero"; };
-struct NameParseDateTime32BestEffortOrNull { static constexpr auto name = "parseDateTime32BestEffortOrNull"; };
-struct NameParseDateTime64BestEffort { static constexpr auto name = "parseDateTime64BestEffort"; };
-struct NameParseDateTime64BestEffortOrZero { static constexpr auto name = "parseDateTime64BestEffortOrZero"; };
-struct NameParseDateTime64BestEffortOrNull { static constexpr auto name = "parseDateTime64BestEffortOrNull"; };
-struct NameParseDateTime64BestEffortUS { static constexpr auto name = "parseDateTime64BestEffortUS"; };
-struct NameParseDateTime64BestEffortUSOrZero { static constexpr auto name = "parseDateTime64BestEffortUSOrZero"; };
-struct NameParseDateTime64BestEffortUSOrNull { static constexpr auto name = "parseDateTime64BestEffortUSOrNull"; };
-
-
-using FunctionParseDateTimeBestEffort = FunctionConvertFromString<
-    DataTypeDateTime, NameParseDateTimeBestEffort, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::BestEffort>;
-using FunctionParseDateTimeBestEffortOrZero = FunctionConvertFromString<
-    DataTypeDateTime, NameParseDateTimeBestEffortOrZero, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::BestEffort>;
-using FunctionParseDateTimeBestEffortOrNull = FunctionConvertFromString<
-    DataTypeDateTime, NameParseDateTimeBestEffortOrNull, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::BestEffort>;
-
-using FunctionParseDateTimeBestEffortUS = FunctionConvertFromString<
-    DataTypeDateTime, NameParseDateTimeBestEffortUS, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::BestEffortUS>;
-using FunctionParseDateTimeBestEffortUSOrZero = FunctionConvertFromString<
-    DataTypeDateTime, NameParseDateTimeBestEffortUSOrZero, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::BestEffortUS>;
-using FunctionParseDateTimeBestEffortUSOrNull = FunctionConvertFromString<
-    DataTypeDateTime, NameParseDateTimeBestEffortUSOrNull, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::BestEffortUS>;
-
-using FunctionParseDateTime32BestEffort = FunctionConvertFromString<
-    DataTypeDateTime, NameParseDateTime32BestEffort, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::BestEffort>;
-using FunctionParseDateTime32BestEffortOrZero = FunctionConvertFromString<
-    DataTypeDateTime, NameParseDateTime32BestEffortOrZero, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::BestEffort>;
-using FunctionParseDateTime32BestEffortOrNull = FunctionConvertFromString<
-    DataTypeDateTime, NameParseDateTime32BestEffortOrNull, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::BestEffort>;
-
-using FunctionParseDateTime64BestEffort = FunctionConvertFromString<
-    DataTypeDateTime64, NameParseDateTime64BestEffort, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::BestEffort>;
-using FunctionParseDateTime64BestEffortOrZero = FunctionConvertFromString<
-    DataTypeDateTime64, NameParseDateTime64BestEffortOrZero, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::BestEffort>;
-using FunctionParseDateTime64BestEffortOrNull = FunctionConvertFromString<
-    DataTypeDateTime64, NameParseDateTime64BestEffortOrNull, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::BestEffort>;
-
-using FunctionParseDateTime64BestEffortUS = FunctionConvertFromString<
-    DataTypeDateTime64, NameParseDateTime64BestEffortUS, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::BestEffortUS>;
-using FunctionParseDateTime64BestEffortUSOrZero = FunctionConvertFromString<
-    DataTypeDateTime64, NameParseDateTime64BestEffortUSOrZero, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::BestEffortUS>;
-using FunctionParseDateTime64BestEffortUSOrNull = FunctionConvertFromString<
-    DataTypeDateTime64, NameParseDateTime64BestEffortUSOrNull, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::BestEffortUS>;
-
-
-class ExecutableFunctionCast : public IExecutableFunction
-{
-public:
-    using WrapperType = std::function<ColumnPtr(ColumnsWithTypeAndName &, const DataTypePtr &, const ColumnNullable *, size_t)>;
-
-    explicit ExecutableFunctionCast(
-            WrapperType && wrapper_function_, const char * name_, std::optional<CastDiagnostic> diagnostic_)
-            : wrapper_function(std::move(wrapper_function_)), name(name_), diagnostic(std::move(diagnostic_)) {}
-
-    String getName() const override { return name; }
-
-protected:
-    ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const override
-    {
-        /// drop second argument, pass others
-        ColumnsWithTypeAndName new_arguments{arguments.front()};
-        if (arguments.size() > 2)
-            new_arguments.insert(std::end(new_arguments), std::next(std::begin(arguments), 2), std::end(arguments));
-
-        try
-        {
-            return wrapper_function(new_arguments, result_type, nullptr, input_rows_count);
-        }
-        catch (Exception & e)
-        {
-            if (diagnostic)
-                e.addMessage("while converting source column " + backQuoteIfNeed(diagnostic->column_from) +
-                             " to destination column " + backQuoteIfNeed(diagnostic->column_to));
-            throw;
-        }
-    }
-
-    bool useDefaultImplementationForNulls() const override { return false; }
-    /// CAST(Nothing, T) -> T
-    bool useDefaultImplementationForNothing() const override { return false; }
-    bool useDefaultImplementationForConstants() const override { return true; }
-    bool useDefaultImplementationForLowCardinalityColumns() const override { return false; }
-    ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {1}; }
-
-private:
-    WrapperType wrapper_function;
-    const char * name;
-    std::optional<CastDiagnostic> diagnostic;
-};
-
-struct CastName { static constexpr auto name = "CAST"; };
-struct CastInternalName { static constexpr auto name = "_CAST"; };
-
-class FunctionCastBase : public IFunctionBase
-{
-public:
-    using MonotonicityForRange = std::function<Monotonicity(const IDataType &, const Field &, const Field &)>;
-};
-
-template <typename FunctionName>
-class FunctionCast final : public FunctionCastBase
-{
-public:
-    using WrapperType = std::function<ColumnPtr(ColumnsWithTypeAndName &, const DataTypePtr &, const ColumnNullable *, size_t)>;
-
-    FunctionCast(ContextPtr context_
-            , const char * cast_name_
-            , MonotonicityForRange && monotonicity_for_range_
-            , const DataTypes & argument_types_
-            , const DataTypePtr & return_type_
-            , std::optional<CastDiagnostic> diagnostic_
-            , CastType cast_type_)
-        : cast_name(cast_name_), monotonicity_for_range(std::move(monotonicity_for_range_))
-        , argument_types(argument_types_), return_type(return_type_), diagnostic(std::move(diagnostic_))
-        , cast_type(cast_type_)
-        , context(context_)
-    {
-    }
-
-    const DataTypes & getArgumentTypes() const override { return argument_types; }
-    const DataTypePtr & getResultType() const override { return return_type; }
-
-    ExecutableFunctionPtr prepare(const ColumnsWithTypeAndName & /*sample_columns*/) const override
-    {
-        try
-        {
-            return std::make_unique<ExecutableFunctionCast>(
-                    prepareUnpackDictionaries(getArgumentTypes()[0], getResultType()), cast_name, diagnostic);
-        }
-        catch (Exception & e)
-        {
-            if (diagnostic)
-                e.addMessage("while converting source column " + backQuoteIfNeed(diagnostic->column_from) +
-                             " to destination column " + backQuoteIfNeed(diagnostic->column_to));
-            throw;
-        }
-    }
-
-    String getName() const override { return cast_name; }
-
-    bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return true; }
-
-    bool hasInformationAboutMonotonicity() const override
-    {
-        return static_cast<bool>(monotonicity_for_range);
-    }
-
-    Monotonicity getMonotonicityForRange(const IDataType & type, const Field & left, const Field & right) const override
-    {
-        return monotonicity_for_range(type, left, right);
-    }
-
-private:
-
-    const char * cast_name;
-    MonotonicityForRange monotonicity_for_range;
-
-    DataTypes argument_types;
-    DataTypePtr return_type;
-
-    std::optional<CastDiagnostic> diagnostic;
-    CastType cast_type;
-    ContextPtr context;
-
-    static WrapperType createFunctionAdaptor(FunctionPtr function, const DataTypePtr & from_type)
-    {
-        auto function_adaptor = std::make_unique<FunctionToOverloadResolverAdaptor>(function)->build({ColumnWithTypeAndName{nullptr, from_type, ""}});
-
-        return [function_adaptor]
-            (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count)
-        {
-            return function_adaptor->execute(arguments, result_type, input_rows_count);
-        };
-    }
-
-    static WrapperType createToNullableColumnWrapper()
-    {
-        return [] (ColumnsWithTypeAndName &, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count)
-        {
-            ColumnPtr res = result_type->createColumn();
-            ColumnUInt8::Ptr col_null_map_to = ColumnUInt8::create(input_rows_count, true);
-            return ColumnNullable::create(res->cloneResized(input_rows_count), std::move(col_null_map_to));
-        };
-    }
-
-    template <typename ToDataType>
-    WrapperType createWrapper(const DataTypePtr & from_type, const ToDataType * const to_type, bool requested_result_is_nullable) const
-    {
-        TypeIndex from_type_index = from_type->getTypeId();
-        WhichDataType which(from_type_index);
-        bool can_apply_accurate_cast = (cast_type == CastType::accurate || cast_type == CastType::accurateOrNull)
-            && (which.isInt() || which.isUInt() || which.isFloat());
-
-        FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior = default_date_time_overflow_behavior;
-        if (context)
-            date_time_overflow_behavior = context->getSettingsRef().date_time_overflow_behavior;
-
-        if (requested_result_is_nullable && checkAndGetDataType<DataTypeString>(from_type.get()))
-        {
-            /// In case when converting to Nullable type, we apply different parsing rule,
-            /// that will not throw an exception but return NULL in case of malformed input.
-            FunctionPtr function = FunctionConvertFromString<ToDataType, FunctionName, ConvertFromStringExceptionMode::Null>::create();
-            return createFunctionAdaptor(function, from_type);
-        }
-        else if (!can_apply_accurate_cast)
-        {
-            FunctionPtr function = FunctionTo<ToDataType>::Type::create(context);
-            return createFunctionAdaptor(function, from_type);
-        }
-
-        auto wrapper_cast_type = cast_type;
-
-        return [wrapper_cast_type, from_type_index, to_type, date_time_overflow_behavior]
-            (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *column_nullable, size_t input_rows_count)
-        {
-            ColumnPtr result_column;
-            auto res = callOnIndexAndDataType<ToDataType>(from_type_index, [&](const auto & types) -> bool {
-                using Types = std::decay_t<decltype(types)>;
-                using LeftDataType = typename Types::LeftType;
-                using RightDataType = typename Types::RightType;
-
-                if constexpr (IsDataTypeNumber<LeftDataType>)
-                {
-                    if constexpr (IsDataTypeNumber<RightDataType>)
-                    {
-#define GENERATE_OVERFLOW_MODE_CASE(OVERFLOW_MODE, ADDITIONS) \
-            case FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE: \
-                result_column = ConvertImpl<LeftDataType, RightDataType, FunctionName, ConvertDefaultBehaviorTag, FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE>::execute( \
-                arguments, result_type, input_rows_count, ADDITIONS()); \
-                break;
-                        if (wrapper_cast_type == CastType::accurate)
-                        {
-                            switch (date_time_overflow_behavior)
-                            {
-                                GENERATE_OVERFLOW_MODE_CASE(Throw, AccurateConvertStrategyAdditions)
-                                GENERATE_OVERFLOW_MODE_CASE(Ignore, AccurateConvertStrategyAdditions)
-                                GENERATE_OVERFLOW_MODE_CASE(Saturate, AccurateConvertStrategyAdditions)
-                            }
-                        }
-                        else
-                        {
-                            switch (date_time_overflow_behavior)
-                            {
-                                GENERATE_OVERFLOW_MODE_CASE(Throw, AccurateOrNullConvertStrategyAdditions)
-                                GENERATE_OVERFLOW_MODE_CASE(Ignore, AccurateOrNullConvertStrategyAdditions)
-                                GENERATE_OVERFLOW_MODE_CASE(Saturate, AccurateOrNullConvertStrategyAdditions)
-                            }
-                        }
-#undef GENERATE_OVERFLOW_MODE_CASE
-
-                        return true;
-                    }
-
-                    if constexpr (std::is_same_v<RightDataType, DataTypeDate> || std::is_same_v<RightDataType, DataTypeDateTime>)
-                    {
-#define GENERATE_OVERFLOW_MODE_CASE(OVERFLOW_MODE, ADDITIONS) \
-            case FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE: \
-            result_column = ConvertImpl<LeftDataType, RightDataType, FunctionName, ConvertDefaultBehaviorTag, FormatSettings::DateTimeOverflowBehavior::OVERFLOW_MODE>::template execute<ADDITIONS>( \
-arguments, result_type, input_rows_count); \
-                break;
-                        if (wrapper_cast_type == CastType::accurate)
-                        {
-                            switch (date_time_overflow_behavior)
-                            {
-                                GENERATE_OVERFLOW_MODE_CASE(Throw, DateTimeAccurateConvertStrategyAdditions)
-                                GENERATE_OVERFLOW_MODE_CASE(Ignore, DateTimeAccurateConvertStrategyAdditions)
-                                GENERATE_OVERFLOW_MODE_CASE(Saturate, DateTimeAccurateConvertStrategyAdditions)
-                            }
-                        }
-                        else
-                        {
-                            switch (date_time_overflow_behavior)
-                            {
-                                GENERATE_OVERFLOW_MODE_CASE(Throw, DateTimeAccurateOrNullConvertStrategyAdditions)
-                                GENERATE_OVERFLOW_MODE_CASE(Ignore, DateTimeAccurateOrNullConvertStrategyAdditions)
-                                GENERATE_OVERFLOW_MODE_CASE(Saturate, DateTimeAccurateOrNullConvertStrategyAdditions)
-                            }
-                        }
-#undef GENERATE_OVERFLOW_MODE_CASE
-                        return true;
-                    }
-                }
-
-                return false;
-            });
-
-            /// Additionally check if callOnIndexAndDataType wasn't called at all.
-            if (!res)
-            {
-                if (wrapper_cast_type == CastType::accurateOrNull)
-                {
-                    auto nullable_column_wrapper = FunctionCast<FunctionName>::createToNullableColumnWrapper();
-                    return nullable_column_wrapper(arguments, result_type, column_nullable, input_rows_count);
-                }
-                else
-                {
-                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE,
-                        "Conversion from {} to {} is not supported",
-                        from_type_index, to_type->getName());
-                }
-            }
-
-            return result_column;
-        };
-    }
-
-    template <typename ToDataType>
-    WrapperType createBoolWrapper(const DataTypePtr & from_type, const ToDataType * const to_type, bool requested_result_is_nullable) const
-    {
-        if (checkAndGetDataType<DataTypeString>(from_type.get()))
-        {
-            return &ConvertImplGenericFromString<ColumnString>::execute;
-        }
-
-        return createWrapper<ToDataType>(from_type, to_type, requested_result_is_nullable);
-    }
-
-    WrapperType createUInt8ToBoolWrapper(const DataTypePtr from_type, const DataTypePtr to_type) const
-    {
-        return [from_type, to_type] (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable *, size_t /*input_rows_count*/) -> ColumnPtr
-        {
-            /// Special case when we convert UInt8 column to Bool column.
-            /// both columns have type UInt8, but we shouldn't use identity wrapper,
-            /// because Bool column can contain only 0 and 1.
-            auto res_column = to_type->createColumn();
-            const auto & data_from = checkAndGetColumn<ColumnUInt8>(arguments[0].column.get())->getData();
-            auto & data_to = assert_cast<ColumnUInt8 *>(res_column.get())->getData();
-            data_to.resize(data_from.size());
-            for (size_t i = 0; i != data_from.size(); ++i)
-                data_to[i] = static_cast<bool>(data_from[i]);
-            return res_column;
-        };
-    }
-
-    static WrapperType createStringWrapper(const DataTypePtr & from_type)
-    {
-        FunctionPtr function = FunctionToString::create();
-        return createFunctionAdaptor(function, from_type);
-    }
-
-    WrapperType createFixedStringWrapper(const DataTypePtr & from_type, const size_t N) const
-    {
-        if (!isStringOrFixedString(from_type))
-            throw Exception(ErrorCodes::NOT_IMPLEMENTED, "CAST AS FixedString is only implemented for types String and FixedString");
-
-        bool exception_mode_null = cast_type == CastType::accurateOrNull;
-        return [exception_mode_null, N] (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable *, size_t /*input_rows_count*/)
-        {
-            if (exception_mode_null)
-                return FunctionToFixedString::executeForN<ConvertToFixedStringExceptionMode::Null>(arguments, N);
-            else
-                return FunctionToFixedString::executeForN<ConvertToFixedStringExceptionMode::Throw>(arguments, N);
-        };
-    }
-
-#define GENERATE_INTERVAL_CASE(INTERVAL_KIND) \
-            case IntervalKind::Kind::INTERVAL_KIND: \
-                return createFunctionAdaptor(FunctionConvert<DataTypeInterval, NameToInterval##INTERVAL_KIND, PositiveMonotonicity>::create(), from_type);
-
-    static WrapperType createIntervalWrapper(const DataTypePtr & from_type, IntervalKind kind)
-    {
-        switch (kind)
-        {
-            GENERATE_INTERVAL_CASE(Nanosecond)
-            GENERATE_INTERVAL_CASE(Microsecond)
-            GENERATE_INTERVAL_CASE(Millisecond)
-            GENERATE_INTERVAL_CASE(Second)
-            GENERATE_INTERVAL_CASE(Minute)
-            GENERATE_INTERVAL_CASE(Hour)
-            GENERATE_INTERVAL_CASE(Day)
-            GENERATE_INTERVAL_CASE(Week)
-            GENERATE_INTERVAL_CASE(Month)
-            GENERATE_INTERVAL_CASE(Quarter)
-            GENERATE_INTERVAL_CASE(Year)
-        }
-        throw Exception{ErrorCodes::CANNOT_CONVERT_TYPE, "Conversion to unexpected IntervalKind: {}", kind.toString()};
-    }
-
-#undef GENERATE_INTERVAL_CASE
-
-    template <typename ToDataType>
-    requires IsDataTypeDecimal<ToDataType>
-    WrapperType createDecimalWrapper(const DataTypePtr & from_type, const ToDataType * to_type, bool requested_result_is_nullable) const
-    {
-        TypeIndex type_index = from_type->getTypeId();
-        UInt32 scale = to_type->getScale();
-
-        WhichDataType which(type_index);
-        bool ok = which.isNativeInt() || which.isNativeUInt() || which.isDecimal() || which.isFloat() || which.isDateOrDate32() || which.isDateTime() || which.isDateTime64()
-            || which.isStringOrFixedString();
-        if (!ok)
-        {
-            if (cast_type == CastType::accurateOrNull)
-                return createToNullableColumnWrapper();
-            else
-                throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Conversion from {} to {} is not supported",
-                    from_type->getName(), to_type->getName());
-        }
-
-        auto wrapper_cast_type = cast_type;
-
-        return [wrapper_cast_type, type_index, scale, to_type, requested_result_is_nullable]
-            (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *column_nullable, size_t input_rows_count)
-        {
-            ColumnPtr result_column;
-            auto res = callOnIndexAndDataType<ToDataType>(type_index, [&](const auto & types) -> bool
-            {
-                using Types = std::decay_t<decltype(types)>;
-                using LeftDataType = typename Types::LeftType;
-                using RightDataType = typename Types::RightType;
-
-                if constexpr (IsDataTypeDecimalOrNumber<LeftDataType> && IsDataTypeDecimalOrNumber<RightDataType> && !std::is_same_v<DataTypeDateTime64, RightDataType>)
-                {
-                    if (wrapper_cast_type == CastType::accurate)
-                    {
-                        AccurateConvertStrategyAdditions additions;
-                        additions.scale = scale;
-                        result_column = ConvertImpl<LeftDataType, RightDataType, FunctionName>::execute(
-                            arguments, result_type, input_rows_count, additions);
-
-                        return true;
-                    }
-                    else if (wrapper_cast_type == CastType::accurateOrNull)
-                    {
-                        AccurateOrNullConvertStrategyAdditions additions;
-                        additions.scale = scale;
-                        result_column = ConvertImpl<LeftDataType, RightDataType, FunctionName>::execute(
-                            arguments, result_type, input_rows_count, additions);
-
-                        return true;
-                    }
-                }
-                else if constexpr (std::is_same_v<LeftDataType, DataTypeString>)
-                {
-                    if (requested_result_is_nullable)
-                    {
-                        /// Consistent with CAST(Nullable(String) AS Nullable(Numbers))
-                        /// In case when converting to Nullable type, we apply different parsing rule,
-                        /// that will not throw an exception but return NULL in case of malformed input.
-                        result_column = ConvertImpl<LeftDataType, RightDataType, FunctionName, ConvertReturnNullOnErrorTag>::execute(
-                            arguments, result_type, input_rows_count, scale);
-
-                        return true;
-                    }
-                }
-
-                result_column = ConvertImpl<LeftDataType, RightDataType, FunctionName>::execute(arguments, result_type, input_rows_count, scale);
-
-                return true;
-            });
-
-            /// Additionally check if callOnIndexAndDataType wasn't called at all.
-            if (!res)
-            {
-                if (wrapper_cast_type == CastType::accurateOrNull)
-                {
-                    auto nullable_column_wrapper = FunctionCast<FunctionName>::createToNullableColumnWrapper();
-                    return nullable_column_wrapper(arguments, result_type, column_nullable, input_rows_count);
-                }
-                else
-                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE,
-                        "Conversion from {} to {} is not supported",
-                        type_index, to_type->getName());
-            }
-
-            return result_column;
-        };
-    }
-
-    WrapperType createAggregateFunctionWrapper(const DataTypePtr & from_type_untyped, const DataTypeAggregateFunction * to_type) const
-    {
-        /// Conversion from String through parsing.
-        if (checkAndGetDataType<DataTypeString>(from_type_untyped.get()))
-        {
-            return &ConvertImplGenericFromString<ColumnString>::execute;
-        }
-        else if (const auto * agg_type = checkAndGetDataType<DataTypeAggregateFunction>(from_type_untyped.get()))
-        {
-            if (agg_type->getFunction()->haveSameStateRepresentation(*to_type->getFunction()))
-            {
-                return [function = to_type->getFunction()](
-                           ColumnsWithTypeAndName & arguments,
-                           const DataTypePtr & /* result_type */,
-                           const ColumnNullable * /* nullable_source */,
-                           size_t /*input_rows_count*/) -> ColumnPtr
-                {
-                    const auto & argument_column = arguments.front();
-                    const auto * col_agg = checkAndGetColumn<ColumnAggregateFunction>(argument_column.column.get());
-                    if (col_agg)
-                    {
-                        auto new_col_agg = ColumnAggregateFunction::create(*col_agg);
-                        new_col_agg->set(function);
-                        return new_col_agg;
-                    }
-                    else
-                    {
-                        throw Exception(
-                            ErrorCodes::LOGICAL_ERROR,
-                            "Illegal column {} for function CAST AS AggregateFunction",
-                            argument_column.column->getName());
-                    }
-                };
-            }
-        }
-
-        if (cast_type == CastType::accurateOrNull)
-            return createToNullableColumnWrapper();
-        else
-            throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Conversion from {} to {} is not supported",
-                from_type_untyped->getName(), to_type->getName());
-    }
-
-    WrapperType createArrayWrapper(const DataTypePtr & from_type_untyped, const DataTypeArray & to_type) const
-    {
-        /// Conversion from String through parsing.
-        if (checkAndGetDataType<DataTypeString>(from_type_untyped.get()))
-        {
-            return &ConvertImplGenericFromString<ColumnString>::execute;
-        }
-
-        DataTypePtr from_type_holder;
-        const auto * from_type = checkAndGetDataType<DataTypeArray>(from_type_untyped.get());
-        const auto * from_type_map = checkAndGetDataType<DataTypeMap>(from_type_untyped.get());
-
-        /// Convert from Map
-        if (from_type_map)
-        {
-            /// Recreate array of unnamed tuples because otherwise it may work
-            /// unexpectedly while converting to array of named tuples.
-            from_type_holder = from_type_map->getNestedTypeWithUnnamedTuple();
-            from_type = assert_cast<const DataTypeArray *>(from_type_holder.get());
-        }
-
-        if (!from_type)
-        {
-            throw Exception(ErrorCodes::TYPE_MISMATCH,
-                "CAST AS Array can only be performed between same-dimensional Array, Map or String types");
-        }
-
-        DataTypePtr from_nested_type = from_type->getNestedType();
-
-        /// In query SELECT CAST([] AS Array(Array(String))) from type is Array(Nothing)
-        bool from_empty_array = isNothing(from_nested_type);
-
-        if (from_type->getNumberOfDimensions() != to_type.getNumberOfDimensions() && !from_empty_array)
-            throw Exception(ErrorCodes::TYPE_MISMATCH,
-                "CAST AS Array can only be performed between same-dimensional array types");
-
-        const DataTypePtr & to_nested_type = to_type.getNestedType();
-
-        /// Prepare nested type conversion
-        const auto nested_function = prepareUnpackDictionaries(from_nested_type, to_nested_type);
-
-        return [nested_function, from_nested_type, to_nested_type](
-                ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * nullable_source, size_t /*input_rows_count*/) -> ColumnPtr
-        {
-            const auto & argument_column = arguments.front();
-
-            const ColumnArray * col_array = nullptr;
-
-            if (const ColumnMap * col_map = checkAndGetColumn<ColumnMap>(argument_column.column.get()))
-                col_array = &col_map->getNestedColumn();
-            else
-                col_array = checkAndGetColumn<ColumnArray>(argument_column.column.get());
-
-            if (col_array)
-            {
-                /// create columns for converting nested column containing original and result columns
-                ColumnsWithTypeAndName nested_columns{{ col_array->getDataPtr(), from_nested_type, "" }};
-
-                /// convert nested column
-                auto result_column = nested_function(nested_columns, to_nested_type, nullable_source, nested_columns.front().column->size());
-
-                /// set converted nested column to result
-                return ColumnArray::create(result_column, col_array->getOffsetsPtr());
-            }
-            else
-            {
-                throw Exception(ErrorCodes::LOGICAL_ERROR,
-                    "Illegal column {} for function CAST AS Array",
-                    argument_column.column->getName());
-            }
-        };
-    }
-
-    using ElementWrappers = std::vector<WrapperType>;
-
-    ElementWrappers getElementWrappers(const DataTypes & from_element_types, const DataTypes & to_element_types) const
-    {
-        ElementWrappers element_wrappers;
-        element_wrappers.reserve(from_element_types.size());
-
-        /// Create conversion wrapper for each element in tuple
-        for (size_t i = 0; i < from_element_types.size(); ++i)
-        {
-            const DataTypePtr & from_element_type = from_element_types[i];
-            const DataTypePtr & to_element_type = to_element_types[i];
-            element_wrappers.push_back(prepareUnpackDictionaries(from_element_type, to_element_type));
-        }
-
-        return element_wrappers;
-    }
-
-    WrapperType createTupleWrapper(const DataTypePtr & from_type_untyped, const DataTypeTuple * to_type) const
-    {
-        /// Conversion from String through parsing.
-        if (checkAndGetDataType<DataTypeString>(from_type_untyped.get()))
-        {
-            return &ConvertImplGenericFromString<ColumnString>::execute;
-        }
-
-        const auto * from_type = checkAndGetDataType<DataTypeTuple>(from_type_untyped.get());
-        if (!from_type)
-            throw Exception(ErrorCodes::TYPE_MISMATCH, "CAST AS Tuple can only be performed between tuple types or from String.\n"
-                            "Left type: {}, right type: {}", from_type_untyped->getName(), to_type->getName());
-
-        const auto & from_element_types = from_type->getElements();
-        const auto & to_element_types = to_type->getElements();
-
-        std::vector<WrapperType> element_wrappers;
-        std::vector<std::optional<size_t>> to_reverse_index;
-
-        /// For named tuples allow conversions for tuples with
-        /// different sets of elements. If element exists in @to_type
-        /// and doesn't exist in @to_type it will be filled by default values.
-        if (from_type->haveExplicitNames() && to_type->haveExplicitNames())
-        {
-            const auto & from_names = from_type->getElementNames();
-            std::unordered_map<String, size_t> from_positions;
-            from_positions.reserve(from_names.size());
-            for (size_t i = 0; i < from_names.size(); ++i)
-                from_positions[from_names[i]] = i;
-
-            const auto & to_names = to_type->getElementNames();
-            element_wrappers.reserve(to_names.size());
-            to_reverse_index.reserve(from_names.size());
-
-            for (size_t i = 0; i < to_names.size(); ++i)
-            {
-                auto it = from_positions.find(to_names[i]);
-                if (it != from_positions.end())
-                {
-                    element_wrappers.emplace_back(prepareUnpackDictionaries(from_element_types[it->second], to_element_types[i]));
-                    to_reverse_index.emplace_back(it->second);
-                }
-                else
-                {
-                    element_wrappers.emplace_back();
-                    to_reverse_index.emplace_back();
-                }
-            }
-        }
-        else
-        {
-            if (from_element_types.size() != to_element_types.size())
-                throw Exception(ErrorCodes::TYPE_MISMATCH, "CAST AS Tuple can only be performed between tuple types "
-                                "with the same number of elements or from String.\nLeft type: {}, right type: {}",
-                                from_type->getName(), to_type->getName());
-
-            element_wrappers = getElementWrappers(from_element_types, to_element_types);
-            to_reverse_index.reserve(to_element_types.size());
-            for (size_t i = 0; i < to_element_types.size(); ++i)
-                to_reverse_index.emplace_back(i);
-        }
-
-        return [element_wrappers, from_element_types, to_element_types, to_reverse_index]
-            (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * nullable_source, size_t input_rows_count) -> ColumnPtr
-        {
-            const auto * col = arguments.front().column.get();
-
-            size_t tuple_size = to_element_types.size();
-            const ColumnTuple & column_tuple = typeid_cast<const ColumnTuple &>(*col);
-
-            Columns converted_columns(tuple_size);
-
-            /// invoke conversion for each element
-            for (size_t i = 0; i < tuple_size; ++i)
-            {
-                if (to_reverse_index[i])
-                {
-                    size_t from_idx = *to_reverse_index[i];
-                    ColumnsWithTypeAndName element = {{column_tuple.getColumns()[from_idx], from_element_types[from_idx], "" }};
-                    converted_columns[i] = element_wrappers[i](element, to_element_types[i], nullable_source, input_rows_count);
-                }
-                else
-                {
-                    converted_columns[i] = to_element_types[i]->createColumn()->cloneResized(input_rows_count);
-                }
-            }
-
-            return ColumnTuple::create(converted_columns);
-        };
-    }
-
-    /// The case of: tuple([key1, key2, ..., key_n], [value1, value2, ..., value_n])
-    WrapperType createTupleToMapWrapper(const DataTypes & from_kv_types, const DataTypes & to_kv_types) const
-    {
-        return [element_wrappers = getElementWrappers(from_kv_types, to_kv_types), from_kv_types, to_kv_types]
-            (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * nullable_source, size_t /*input_rows_count*/) -> ColumnPtr
-        {
-            const auto * col = arguments.front().column.get();
-            const auto & column_tuple = assert_cast<const ColumnTuple &>(*col);
-
-            Columns offsets(2);
-            Columns converted_columns(2);
-            for (size_t i = 0; i < 2; ++i)
-            {
-                const auto & column_array = assert_cast<const ColumnArray &>(column_tuple.getColumn(i));
-                ColumnsWithTypeAndName element = {{column_array.getDataPtr(), from_kv_types[i], ""}};
-                converted_columns[i] = element_wrappers[i](element, to_kv_types[i], nullable_source, (element[0].column)->size());
-                offsets[i] = column_array.getOffsetsPtr();
-            }
-
-            const auto & keys_offsets = assert_cast<const ColumnArray::ColumnOffsets &>(*offsets[0]).getData();
-            const auto & values_offsets = assert_cast<const ColumnArray::ColumnOffsets &>(*offsets[1]).getData();
-            if (keys_offsets != values_offsets)
-                throw Exception(ErrorCodes::TYPE_MISMATCH,
-                    "CAST AS Map can only be performed from tuple of arrays with equal sizes.");
-
-            return ColumnMap::create(converted_columns[0], converted_columns[1], offsets[0]);
-        };
-    }
-
-    WrapperType createMapToMapWrapper(const DataTypes & from_kv_types, const DataTypes & to_kv_types) const
-    {
-        return [element_wrappers = getElementWrappers(from_kv_types, to_kv_types), from_kv_types, to_kv_types]
-            (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * nullable_source, size_t /*input_rows_count*/) -> ColumnPtr
-        {
-            const auto * col = arguments.front().column.get();
-            const auto & column_map = typeid_cast<const ColumnMap &>(*col);
-            const auto & nested_data = column_map.getNestedData();
-
-            Columns converted_columns(2);
-            for (size_t i = 0; i < 2; ++i)
-            {
-                ColumnsWithTypeAndName element = {{nested_data.getColumnPtr(i), from_kv_types[i], ""}};
-                converted_columns[i] = element_wrappers[i](element, to_kv_types[i], nullable_source, (element[0].column)->size());
-            }
-
-            return ColumnMap::create(converted_columns[0], converted_columns[1], column_map.getNestedColumn().getOffsetsPtr());
-        };
-    }
-
-    /// The case of: [(key1, value1), (key2, value2), ...]
-    WrapperType createArrayToMapWrapper(const DataTypes & from_kv_types, const DataTypes & to_kv_types) const
-    {
-        return [element_wrappers = getElementWrappers(from_kv_types, to_kv_types), from_kv_types, to_kv_types]
-            (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * nullable_source, size_t /*input_rows_count*/) -> ColumnPtr
-        {
-            const auto * col = arguments.front().column.get();
-            const auto & column_array = typeid_cast<const ColumnArray &>(*col);
-            const auto & nested_data = typeid_cast<const ColumnTuple &>(column_array.getData());
-
-            Columns converted_columns(2);
-            for (size_t i = 0; i < 2; ++i)
-            {
-                ColumnsWithTypeAndName element = {{nested_data.getColumnPtr(i), from_kv_types[i], ""}};
-                converted_columns[i] = element_wrappers[i](element, to_kv_types[i], nullable_source, (element[0].column)->size());
-            }
-
-            return ColumnMap::create(converted_columns[0], converted_columns[1], column_array.getOffsetsPtr());
-        };
-    }
-
-
-    WrapperType createMapWrapper(const DataTypePtr & from_type_untyped, const DataTypeMap * to_type) const
-    {
-        if (const auto * from_tuple = checkAndGetDataType<DataTypeTuple>(from_type_untyped.get()))
-        {
-            if (from_tuple->getElements().size() != 2)
-                throw Exception(
-                    ErrorCodes::TYPE_MISMATCH,
-                    "CAST AS Map from tuple requires 2 elements. "
-                    "Left type: {}, right type: {}",
-                    from_tuple->getName(),
-                    to_type->getName());
-
-            DataTypes from_kv_types;
-            const auto & to_kv_types = to_type->getKeyValueTypes();
-
-            for (const auto & elem : from_tuple->getElements())
-            {
-                const auto * type_array = checkAndGetDataType<DataTypeArray>(elem.get());
-                if (!type_array)
-                    throw Exception(ErrorCodes::TYPE_MISMATCH,
-                        "CAST AS Map can only be performed from tuples of array. Got: {}", from_tuple->getName());
-
-                from_kv_types.push_back(type_array->getNestedType());
-            }
-
-            return createTupleToMapWrapper(from_kv_types, to_kv_types);
-        }
-        else if (const auto * from_array = typeid_cast<const DataTypeArray *>(from_type_untyped.get()))
-        {
-            const auto * nested_tuple = typeid_cast<const DataTypeTuple *>(from_array->getNestedType().get());
-            if (!nested_tuple || nested_tuple->getElements().size() != 2)
-                throw Exception(
-                    ErrorCodes::TYPE_MISMATCH,
-                    "CAST AS Map from array requires nested tuple of 2 elements. "
-                    "Left type: {}, right type: {}",
-                    from_array->getName(),
-                    to_type->getName());
-
-            return createArrayToMapWrapper(nested_tuple->getElements(), to_type->getKeyValueTypes());
-        }
-        else if (const auto * from_type = checkAndGetDataType<DataTypeMap>(from_type_untyped.get()))
-        {
-            return createMapToMapWrapper(from_type->getKeyValueTypes(), to_type->getKeyValueTypes());
-        }
-        else
-        {
-            throw Exception(ErrorCodes::TYPE_MISMATCH, "Unsupported types to CAST AS Map. "
-                "Left type: {}, right type: {}", from_type_untyped->getName(), to_type->getName());
-        }
-    }
-
-    WrapperType createTupleToObjectWrapper(const DataTypeTuple & from_tuple, bool has_nullable_subcolumns) const
-    {
-        if (!from_tuple.haveExplicitNames())
-            throw Exception(ErrorCodes::TYPE_MISMATCH,
-            "Cast to Object can be performed only from flatten Named Tuple. Got: {}", from_tuple.getName());
-
-        PathsInData paths;
-        DataTypes from_types;
-
-        std::tie(paths, from_types) = flattenTuple(from_tuple.getPtr());
-        auto to_types = from_types;
-
-        for (auto & type : to_types)
-        {
-            if (isTuple(type) || isNested(type))
-                throw Exception(ErrorCodes::TYPE_MISMATCH,
-                    "Cast to Object can be performed only from flatten Named Tuple. Got: {}",
-                    from_tuple.getName());
-
-            type = recursiveRemoveLowCardinality(type);
-        }
-
-        return [element_wrappers = getElementWrappers(from_types, to_types),
-            has_nullable_subcolumns, from_types, to_types, paths]
-            (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * nullable_source, size_t input_rows_count)
-        {
-            size_t tuple_size = to_types.size();
-            auto flattened_column = flattenTuple(arguments.front().column);
-            const auto & column_tuple = assert_cast<const ColumnTuple &>(*flattened_column);
-
-            if (tuple_size != column_tuple.getColumns().size())
-                throw Exception(ErrorCodes::TYPE_MISMATCH,
-                    "Expected tuple with {} subcolumn, but got {} subcolumns",
-                    tuple_size, column_tuple.getColumns().size());
-
-            auto res = ColumnObject::create(has_nullable_subcolumns);
-            for (size_t i = 0; i < tuple_size; ++i)
-            {
-                ColumnsWithTypeAndName element = {{column_tuple.getColumns()[i], from_types[i], "" }};
-                auto converted_column = element_wrappers[i](element, to_types[i], nullable_source, input_rows_count);
-                res->addSubcolumn(paths[i], converted_column->assumeMutable());
-            }
-
-            return res;
-        };
-    }
-
-    WrapperType createMapToObjectWrapper(const DataTypeMap & from_map, bool has_nullable_subcolumns) const
-    {
-        auto key_value_types = from_map.getKeyValueTypes();
-
-        if (!isStringOrFixedString(key_value_types[0]))
-            throw Exception(ErrorCodes::TYPE_MISMATCH,
-                "Cast to Object from Map can be performed only from Map "
-                "with String or FixedString key. Got: {}", from_map.getName());
-
-        const auto & value_type = key_value_types[1];
-        auto to_value_type = value_type;
-
-        if (!has_nullable_subcolumns && value_type->isNullable())
-            to_value_type = removeNullable(value_type);
-
-        if (has_nullable_subcolumns && !value_type->isNullable())
-            to_value_type = makeNullable(value_type);
-
-        DataTypes to_key_value_types{std::make_shared<DataTypeString>(), std::move(to_value_type)};
-        auto element_wrappers = getElementWrappers(key_value_types, to_key_value_types);
-
-        return [has_nullable_subcolumns, element_wrappers, key_value_types, to_key_value_types]
-            (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * nullable_source, size_t) -> ColumnPtr
-        {
-            const auto & column_map = assert_cast<const ColumnMap &>(*arguments.front().column);
-            const auto & offsets = column_map.getNestedColumn().getOffsets();
-            auto key_value_columns = column_map.getNestedData().getColumnsCopy();
-
-            for (size_t i = 0; i < 2; ++i)
-            {
-                ColumnsWithTypeAndName element{{key_value_columns[i], key_value_types[i], ""}};
-                key_value_columns[i] = element_wrappers[i](element, to_key_value_types[i], nullable_source, key_value_columns[i]->size());
-            }
-
-            const auto & key_column_str = assert_cast<const ColumnString &>(*key_value_columns[0]);
-            const auto & value_column = *key_value_columns[1];
-
-            using SubcolumnsMap = HashMap<StringRef, MutableColumnPtr, StringRefHash>;
-            SubcolumnsMap subcolumns;
-
-            for (size_t row = 0; row < offsets.size(); ++row)
-            {
-                for (size_t i = offsets[static_cast<ssize_t>(row) - 1]; i < offsets[row]; ++i)
-                {
-                    auto ref = key_column_str.getDataAt(i);
-
-                    bool inserted;
-                    SubcolumnsMap::LookupResult it;
-                    subcolumns.emplace(ref, it, inserted);
-                    auto & subcolumn = it->getMapped();
-
-                    if (inserted)
-                        subcolumn = value_column.cloneEmpty()->cloneResized(row);
-
-                    /// Map can have duplicated keys. We insert only first one.
-                    if (subcolumn->size() == row)
-                        subcolumn->insertFrom(value_column, i);
-                }
-
-                /// Insert default values for keys missed in current row.
-                for (const auto & [_, subcolumn] : subcolumns)
-                    if (subcolumn->size() == row)
-                        subcolumn->insertDefault();
-            }
-
-            auto column_object = ColumnObject::create(has_nullable_subcolumns);
-            for (auto && [key, subcolumn] : subcolumns)
-            {
-                PathInData path(key.toView());
-                column_object->addSubcolumn(path, std::move(subcolumn));
-            }
-
-            return column_object;
-        };
-    }
-
-    WrapperType createObjectWrapper(const DataTypePtr & from_type, const DataTypeObject * to_type) const
-    {
-        if (const auto * from_tuple = checkAndGetDataType<DataTypeTuple>(from_type.get()))
-        {
-            return createTupleToObjectWrapper(*from_tuple, to_type->hasNullableSubcolumns());
-        }
-        else if (const auto * from_map = checkAndGetDataType<DataTypeMap>(from_type.get()))
-        {
-            return createMapToObjectWrapper(*from_map, to_type->hasNullableSubcolumns());
-        }
-        else if (checkAndGetDataType<DataTypeString>(from_type.get()))
-        {
-            return [] (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable * nullable_source, size_t input_rows_count)
-            {
-                auto res = ConvertImplGenericFromString<ColumnString>::execute(arguments, result_type, nullable_source, input_rows_count)->assumeMutable();
-                res->finalize();
-                return res;
-            };
-        }
-        else if (checkAndGetDataType<DataTypeObject>(from_type.get()))
-        {
-            return [is_nullable = to_type->hasNullableSubcolumns()] (ColumnsWithTypeAndName & arguments, const DataTypePtr & , const ColumnNullable * , size_t) -> ColumnPtr
-            {
-                auto & column_object = assert_cast<const ColumnObject &>(*arguments.front().column);
-                auto res = ColumnObject::create(is_nullable);
-                for (size_t i = 0; i < column_object.size(); i++)
-                    res->insert(column_object[i]);
-
-                res->finalize();
-                return res;
-            };
-        }
-
-        throw Exception(ErrorCodes::TYPE_MISMATCH,
-            "Cast to Object can be performed only from flatten named Tuple, Map or String. Got: {}", from_type->getName());
-    }
-
-    WrapperType createVariantToVariantWrapper(const DataTypeVariant & from_variant, const DataTypeVariant & to_variant) const
-    {
-        /// We support only extension of variant type, so, only new types can be added.
-        /// For example: Variant(T1, T2) -> Variant(T1, T2, T3) is supported, but Variant(T1, T2) -> Variant(T1, T3) is not supported.
-        /// We want to extend Variant type for free without rewriting the data, but we sort data types inside Variant during type creation
-        /// (we do it because we want Variant(T1, T2) to be the same as Variant(T2, T1)), but after extension the order of variant types
-        /// (and so their discriminators) can be different. For example: Variant(T1, T3) -> Variant(T1, T2, T3).
-        /// To avoid full rewrite of discriminators column, ColumnVariant supports it's local order of variant columns (and so local
-        /// discriminators) and stores mapping global order -> local order.
-        /// So, to extend Variant with new types for free, we should keep old local order for old variants, append new variants and change
-        /// mapping global order -> local order according to the new global order.
-
-        /// Create map (new variant type) -> (it's global discriminator in new order).
-        const auto & new_variants = to_variant.getVariants();
-        std::unordered_map<String, ColumnVariant::Discriminator> new_variant_types_to_new_global_discriminator;
-        new_variant_types_to_new_global_discriminator.reserve(new_variants.size());
-        for (size_t i = 0; i != new_variants.size(); ++i)
-            new_variant_types_to_new_global_discriminator[new_variants[i]->getName()] = i;
-
-        /// Create set of old variant types.
-        const auto & old_variants = from_variant.getVariants();
-        std::unordered_map<String, ColumnVariant::Discriminator> old_variant_types_to_old_global_discriminator;
-        old_variant_types_to_old_global_discriminator.reserve(old_variants.size());
-        for (size_t i = 0; i != old_variants.size(); ++i)
-            old_variant_types_to_old_global_discriminator[old_variants[i]->getName()] = i;
-
-        /// Check that the set of old variants types is a subset of new variant types and collect new global discriminator for each old global discriminator.
-        std::unordered_map<ColumnVariant::Discriminator, ColumnVariant::Discriminator> old_global_discriminator_to_new;
-        old_global_discriminator_to_new.reserve(old_variants.size());
-        for (const auto & [old_variant_type, old_discriminator] : old_variant_types_to_old_global_discriminator)
-        {
-            auto it = new_variant_types_to_new_global_discriminator.find(old_variant_type);
-            if (it == new_variant_types_to_new_global_discriminator.end())
-                throw Exception(
-                    ErrorCodes::CANNOT_CONVERT_TYPE,
-                    "Cannot convert type {} to {}. Conversion between Variant types is allowed only when new Variant type is an extension "
-                    "of an initial one", from_variant.getName(), to_variant.getName());
-            old_global_discriminator_to_new[old_discriminator] = it->second;
-        }
-
-        /// Collect variant types and their global discriminators that should be added to the old Variant to get the new Variant.
-        std::vector<std::pair<DataTypePtr, ColumnVariant::Discriminator>> variant_types_and_discriminators_to_add;
-        variant_types_and_discriminators_to_add.reserve(new_variants.size() - old_variants.size());
-        for (size_t i = 0; i != new_variants.size(); ++i)
-        {
-            if (!old_variant_types_to_old_global_discriminator.contains(new_variants[i]->getName()))
-                variant_types_and_discriminators_to_add.emplace_back(new_variants[i], i);
-        }
-
-        return [old_global_discriminator_to_new, variant_types_and_discriminators_to_add]
-               (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable *, size_t) -> ColumnPtr
-        {
-            const auto & column_variant = assert_cast<const ColumnVariant &>(*arguments.front().column.get());
-            size_t num_old_variants = column_variant.getNumVariants();
-            Columns new_variant_columns;
-            new_variant_columns.reserve(num_old_variants + variant_types_and_discriminators_to_add.size());
-            std::vector<ColumnVariant::Discriminator> new_local_to_global_discriminators;
-            new_local_to_global_discriminators.reserve(num_old_variants + variant_types_and_discriminators_to_add.size());
-            for (size_t i = 0; i != num_old_variants; ++i)
-            {
-                new_variant_columns.push_back(column_variant.getVariantPtrByLocalDiscriminator(i));
-                new_local_to_global_discriminators.push_back(old_global_discriminator_to_new.at(column_variant.globalDiscriminatorByLocal(i)));
-            }
-
-            for (const auto & [new_variant_type, new_global_discriminator] : variant_types_and_discriminators_to_add)
-            {
-                new_variant_columns.push_back(new_variant_type->createColumn());
-                new_local_to_global_discriminators.push_back(new_global_discriminator);
-            }
-
-            return ColumnVariant::create(column_variant.getLocalDiscriminatorsPtr(), column_variant.getOffsetsPtr(), new_variant_columns, new_local_to_global_discriminators);
-        };
-    }
-
-    WrapperType createVariantToColumnWrapper(const DataTypeVariant & from_variant, const DataTypePtr & to_type) const
-    {
-        const auto & variant_types = from_variant.getVariants();
-        std::vector<WrapperType> variant_wrappers;
-        variant_wrappers.reserve(variant_types.size());
-
-        /// Create conversion wrapper for each variant.
-        for (const auto & variant_type : variant_types)
-            variant_wrappers.push_back(prepareUnpackDictionaries(variant_type, to_type));
-
-        return [variant_wrappers, variant_types, to_type]
-               (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count) -> ColumnPtr
-        {
-            const auto & column_variant = assert_cast<const ColumnVariant &>(*arguments.front().column.get());
-
-            /// First, cast each variant to the result type.
-            std::vector<ColumnPtr> casted_variant_columns;
-            casted_variant_columns.reserve(variant_types.size());
-            for (size_t i = 0; i != variant_types.size(); ++i)
-            {
-                auto variant_col = column_variant.getVariantPtrByLocalDiscriminator(i);
-                ColumnsWithTypeAndName variant = {{variant_col, variant_types[i], "" }};
-                const auto & variant_wrapper = variant_wrappers[column_variant.globalDiscriminatorByLocal(i)];
-                casted_variant_columns.push_back(variant_wrapper(variant, result_type, nullptr, variant_col->size()));
-            }
-
-            /// Second, construct resulting column from casted variant columns according to discriminators.
-            const auto & local_discriminators = column_variant.getLocalDiscriminators();
-            auto res = result_type->createColumn();
-            res->reserve(input_rows_count);
-            for (size_t i = 0; i != input_rows_count; ++i)
-            {
-                auto local_discr = local_discriminators[i];
-                if (local_discr == ColumnVariant::NULL_DISCRIMINATOR)
-                    res->insertDefault();
-                else
-                    res->insertFrom(*casted_variant_columns[local_discr], column_variant.offsetAt(i));
-            }
-
-            return res;
-        };
-    }
-
-    static ColumnPtr createVariantFromDescriptorsAndOneNonEmptyVariant(const DataTypes & variant_types, const ColumnPtr & discriminators, const ColumnPtr & variant, ColumnVariant::Discriminator variant_discr)
-    {
-        Columns variants;
-        variants.reserve(variant_types.size());
-        for (size_t i = 0; i != variant_types.size(); ++i)
-        {
-            if (i == variant_discr)
-                variants.emplace_back(variant);
-            else
-                variants.push_back(variant_types[i]->createColumn());
-        }
-
-        return ColumnVariant::create(discriminators, variants);
-    }
-
-    WrapperType createColumnToVariantWrapper(const DataTypePtr & from_type, const DataTypeVariant & to_variant) const
-    {
-        /// We allow converting NULL to Variant(...) as Variant can store NULLs.
-        if (from_type->onlyNull())
-        {
-            return [](ColumnsWithTypeAndName &, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count) -> ColumnPtr
-            {
-                auto result_column = result_type->createColumn();
-                result_column->insertManyDefaults(input_rows_count);
-                return result_column;
-            };
-        }
-
-        auto variant_discr_opt = to_variant.tryGetVariantDiscriminator(*removeNullableOrLowCardinalityNullable(from_type));
-        if (!variant_discr_opt)
-            throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Cannot convert type {} to {}. Conversion to Variant allowed only for types from this Variant", from_type->getName(), to_variant.getName());
-
-        return [variant_discr = *variant_discr_opt]
-               (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *, size_t) -> ColumnPtr
-        {
-            const auto & result_variant_type = assert_cast<const DataTypeVariant &>(*result_type);
-            const auto & variant_types = result_variant_type.getVariants();
-            if (const ColumnNullable * col_nullable = typeid_cast<const ColumnNullable *>(arguments.front().column.get()))
-            {
-                const auto & column = col_nullable->getNestedColumnPtr();
-                const auto & null_map = col_nullable->getNullMapData();
-                IColumn::Filter filter;
-                filter.reserve(column->size());
-                auto discriminators = ColumnVariant::ColumnDiscriminators::create();
-                auto & discriminators_data = discriminators->getData();
-                discriminators_data.reserve(column->size());
-                size_t variant_size_hint = 0;
-                for (size_t i = 0; i != column->size(); ++i)
-                {
-                    if (null_map[i])
-                    {
-                        discriminators_data.push_back(ColumnVariant::NULL_DISCRIMINATOR);
-                        filter.push_back(0);
-                    }
-                    else
-                    {
-                        discriminators_data.push_back(variant_discr);
-                        filter.push_back(1);
-                        ++variant_size_hint;
-                    }
-                }
-
-                ColumnPtr variant_column;
-                /// If there were no NULLs, just use the column.
-                if (variant_size_hint == column->size())
-                    variant_column = column;
-                /// Otherwise we should use filtered column.
-                else
-                    variant_column = column->filter(filter, variant_size_hint);
-                return createVariantFromDescriptorsAndOneNonEmptyVariant(variant_types, std::move(discriminators), variant_column, variant_discr);
-            }
-            else if (isColumnLowCardinalityNullable(*arguments.front().column))
-            {
-                const auto & column = arguments.front().column;
-
-                /// Variant column cannot have LowCardinality(Nullable(...)) variant, as Variant column stores NULLs itself.
-                /// We should create a null-map, insert NULL_DISCRIMINATOR on NULL values and filter initial column.
-                const auto & col_lc = assert_cast<const ColumnLowCardinality &>(*column);
-                const auto & indexes = col_lc.getIndexes();
-                auto null_index = col_lc.getDictionary().getNullValueIndex();
-                IColumn::Filter filter;
-                filter.reserve(col_lc.size());
-                auto discriminators = ColumnVariant::ColumnDiscriminators::create();
-                auto & discriminators_data = discriminators->getData();
-                discriminators_data.reserve(col_lc.size());
-                size_t variant_size_hint = 0;
-                for (size_t i = 0; i != col_lc.size(); ++i)
-                {
-                    if (indexes.getUInt(i) == null_index)
-                    {
-                        discriminators_data.push_back(ColumnVariant::NULL_DISCRIMINATOR);
-                        filter.push_back(0);
-                    }
-                    else
-                    {
-                        discriminators_data.push_back(variant_discr);
-                        filter.push_back(1);
-                        ++variant_size_hint;
-                    }
-                }
-
-                MutableColumnPtr variant_column;
-                /// If there were no NULLs, we can just clone the column.
-                if (variant_size_hint == col_lc.size())
-                    variant_column = IColumn::mutate(column);
-                /// Otherwise we should filter column.
-                else
-                    variant_column = column->filter(filter, variant_size_hint)->assumeMutable();
-
-                assert_cast<ColumnLowCardinality &>(*variant_column).nestedRemoveNullable();
-                return createVariantFromDescriptorsAndOneNonEmptyVariant(variant_types, std::move(discriminators), std::move(variant_column), variant_discr);
-            }
-            else
-            {
-                const auto & column = arguments.front().column;
-                auto discriminators = ColumnVariant::ColumnDiscriminators::create();
-                discriminators->getData().resize_fill(column->size(), variant_discr);
-                return createVariantFromDescriptorsAndOneNonEmptyVariant(variant_types, std::move(discriminators), column, variant_discr);
-            }
-        };
-    }
-
-    /// Wrapper for conversion to/from Variant type
-    WrapperType createVariantWrapper(const DataTypePtr & from_type, const DataTypePtr & to_type) const
-    {
-        if (const auto * from_variant = checkAndGetDataType<DataTypeVariant>(from_type.get()))
-        {
-            if (const auto * to_variant = checkAndGetDataType<DataTypeVariant>(to_type.get()))
-                return createVariantToVariantWrapper(*from_variant, *to_variant);
-
-            return createVariantToColumnWrapper(*from_variant, to_type);
-        }
-
-        return createColumnToVariantWrapper(from_type, assert_cast<const DataTypeVariant &>(*to_type));
-    }
-
-    template <typename FieldType>
-    WrapperType createEnumWrapper(const DataTypePtr & from_type, const DataTypeEnum<FieldType> * to_type) const
-    {
-        using EnumType = DataTypeEnum<FieldType>;
-        using Function = typename FunctionTo<EnumType>::Type;
-
-        if (const auto * from_enum8 = checkAndGetDataType<DataTypeEnum8>(from_type.get()))
-            checkEnumToEnumConversion(from_enum8, to_type);
-        else if (const auto * from_enum16 = checkAndGetDataType<DataTypeEnum16>(from_type.get()))
-            checkEnumToEnumConversion(from_enum16, to_type);
-
-        if (checkAndGetDataType<DataTypeString>(from_type.get()))
-            return createStringToEnumWrapper<ColumnString, EnumType>();
-        else if (checkAndGetDataType<DataTypeFixedString>(from_type.get()))
-            return createStringToEnumWrapper<ColumnFixedString, EnumType>();
-        else if (isNativeNumber(from_type) || isEnum(from_type))
-        {
-            auto function = Function::create();
-            return createFunctionAdaptor(function, from_type);
-        }
-        else
-        {
-            if (cast_type == CastType::accurateOrNull)
-                return createToNullableColumnWrapper();
-            else
-                throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Conversion from {} to {} is not supported",
-                    from_type->getName(), to_type->getName());
-        }
-    }
-
-    template <typename EnumTypeFrom, typename EnumTypeTo>
-    void checkEnumToEnumConversion(const EnumTypeFrom * from_type, const EnumTypeTo * to_type) const
-    {
-        const auto & from_values = from_type->getValues();
-        const auto & to_values = to_type->getValues();
-
-        using ValueType = std::common_type_t<typename EnumTypeFrom::FieldType, typename EnumTypeTo::FieldType>;
-        using NameValuePair = std::pair<std::string, ValueType>;
-        using EnumValues = std::vector<NameValuePair>;
-
-        EnumValues name_intersection;
-        std::set_intersection(std::begin(from_values), std::end(from_values),
-            std::begin(to_values), std::end(to_values), std::back_inserter(name_intersection),
-            [] (auto && from, auto && to) { return from.first < to.first; });
-
-        for (const auto & name_value : name_intersection)
-        {
-            const auto & old_value = name_value.second;
-            const auto & new_value = to_type->getValue(name_value.first);
-            if (old_value != new_value)
-                throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Enum conversion changes value for element '{}' from {} to {}",
-                    name_value.first, toString(old_value), toString(new_value));
-        }
-    }
-
-    template <typename ColumnStringType, typename EnumType>
-    WrapperType createStringToEnumWrapper() const
-    {
-        const char * function_name = cast_name;
-        return [function_name] (
-            ColumnsWithTypeAndName & arguments, const DataTypePtr & res_type, const ColumnNullable * nullable_col, size_t /*input_rows_count*/)
-        {
-            const auto & first_col = arguments.front().column.get();
-            const auto & result_type = typeid_cast<const EnumType &>(*res_type);
-
-            const ColumnStringType * col = typeid_cast<const ColumnStringType *>(first_col);
-
-            if (col && nullable_col && nullable_col->size() != col->size())
-                throw Exception(ErrorCodes::LOGICAL_ERROR, "ColumnNullable is not compatible with original");
-
-            if (col)
-            {
-                const auto size = col->size();
-
-                auto res = result_type.createColumn();
-                auto & out_data = static_cast<typename EnumType::ColumnType &>(*res).getData();
-                out_data.resize(size);
-
-                auto default_enum_value = result_type.getValues().front().second;
-
-                if (nullable_col)
-                {
-                    for (size_t i = 0; i < size; ++i)
-                    {
-                        if (!nullable_col->isNullAt(i))
-                            out_data[i] = result_type.getValue(col->getDataAt(i));
-                        else
-                            out_data[i] = default_enum_value;
-                    }
-                }
-                else
-                {
-                    for (size_t i = 0; i < size; ++i)
-                        out_data[i] = result_type.getValue(col->getDataAt(i));
-                }
-
-                return res;
-            }
-            else
-                throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected column {} as first argument of function {}",
-                    first_col->getName(), function_name);
-        };
-    }
-
-    template <typename EnumType>
-    WrapperType createEnumToStringWrapper() const
-    {
-        const char * function_name = cast_name;
-        return [function_name] (
-            ColumnsWithTypeAndName & arguments, const DataTypePtr & res_type, const ColumnNullable * nullable_col, size_t /*input_rows_count*/)
-        {
-            using ColumnEnumType = EnumType::ColumnType;
-
-            const auto & first_col = arguments.front().column.get();
-            const auto & first_type = arguments.front().type.get();
-
-            const ColumnEnumType * enum_col = typeid_cast<const ColumnEnumType *>(first_col);
-            const EnumType * enum_type = typeid_cast<const EnumType *>(first_type);
-
-            if (enum_col && nullable_col && nullable_col->size() != enum_col->size())
-                throw Exception(ErrorCodes::LOGICAL_ERROR, "ColumnNullable is not compatible with original");
-
-            if (enum_col && enum_type)
-            {
-                const auto size = enum_col->size();
-                const auto & enum_data = enum_col->getData();
-
-                auto res = res_type->createColumn();
-
-                if (nullable_col)
-                {
-                    for (size_t i = 0; i < size; ++i)
-                    {
-                        if (!nullable_col->isNullAt(i))
-                        {
-                            const auto & value = enum_type->getNameForValue(enum_data[i]);
-                            res->insertData(value.data, value.size);
-                        }
-                        else
-                            res->insertDefault();
-                    }
-                }
-                else
-                {
-                    for (size_t i = 0; i < size; ++i)
-                    {
-                        const auto & value = enum_type->getNameForValue(enum_data[i]);
-                        res->insertData(value.data, value.size);
-                    }
-                }
-
-                return res;
-            }
-            else
-                throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected column {} as first argument of function {}",
-                    first_col->getName(), function_name);
-        };
-    }
-
-    static WrapperType createIdentityWrapper(const DataTypePtr &)
-    {
-        return [] (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable *, size_t /*input_rows_count*/)
-        {
-            return arguments.front().column;
-        };
-    }
-
-    static WrapperType createNothingWrapper(const IDataType * to_type)
-    {
-        ColumnPtr res = to_type->createColumnConstWithDefaultValue(1);
-        return [res] (ColumnsWithTypeAndName &, const DataTypePtr &, const ColumnNullable *, size_t input_rows_count)
-        {
-            /// Column of Nothing type is trivially convertible to any other column
-            return res->cloneResized(input_rows_count)->convertToFullColumnIfConst();
-        };
-    }
-
-    WrapperType prepareUnpackDictionaries(const DataTypePtr & from_type, const DataTypePtr & to_type) const
-    {
-        /// Conversion from/to Variant data type is processed in a special way.
-        /// We don't need to remove LowCardinality/Nullable.
-        if (isVariant(to_type) || isVariant(from_type))
-            return createVariantWrapper(from_type, to_type);
-
-        const auto * from_low_cardinality = typeid_cast<const DataTypeLowCardinality *>(from_type.get());
-        const auto * to_low_cardinality = typeid_cast<const DataTypeLowCardinality *>(to_type.get());
-        const auto & from_nested = from_low_cardinality ? from_low_cardinality->getDictionaryType() : from_type;
-        const auto & to_nested = to_low_cardinality ? to_low_cardinality->getDictionaryType() : to_type;
-
-        if (from_type->onlyNull())
-        {
-            if (!to_nested->isNullable() && !isVariant(to_type))
-            {
-                if (cast_type == CastType::accurateOrNull)
-                {
-                    return createToNullableColumnWrapper();
-                }
-                else
-                {
-                    throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Cannot convert NULL to a non-nullable type");
-                }
-            }
-
-            return [](ColumnsWithTypeAndName &, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count)
-            {
-                return result_type->createColumnConstWithDefaultValue(input_rows_count)->convertToFullColumnIfConst();
-            };
-        }
-
-        bool skip_not_null_check = false;
-
-        if (from_low_cardinality && from_nested->isNullable() && !to_nested->isNullable())
-            /// Disable check for dictionary. Will check that column doesn't contain NULL in wrapper below.
-            skip_not_null_check = true;
-
-        auto wrapper = prepareRemoveNullable(from_nested, to_nested, skip_not_null_check);
-        if (!from_low_cardinality && !to_low_cardinality)
-            return wrapper;
-
-        return [wrapper, from_low_cardinality, to_low_cardinality, skip_not_null_check]
-                (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable * nullable_source, size_t input_rows_count) -> ColumnPtr
-        {
-            ColumnsWithTypeAndName args = {arguments[0]};
-            auto & arg = args.front();
-            auto res_type = result_type;
-
-            ColumnPtr converted_column;
-
-            ColumnPtr res_indexes;
-            /// For some types default can't be casted (for example, String to Int). In that case convert column to full.
-            bool src_converted_to_full_column = false;
-
-            {
-                auto tmp_rows_count = input_rows_count;
-
-                if (to_low_cardinality)
-                    res_type = to_low_cardinality->getDictionaryType();
-
-                if (from_low_cardinality)
-                {
-                    const auto * col_low_cardinality = assert_cast<const ColumnLowCardinality *>(arguments[0].column.get());
-
-                    if (skip_not_null_check && col_low_cardinality->containsNull())
-                        throw Exception(ErrorCodes::CANNOT_INSERT_NULL_IN_ORDINARY_COLUMN, "Cannot convert NULL value to non-Nullable type");
-
-                    arg.column = col_low_cardinality->getDictionary().getNestedColumn();
-                    arg.type = from_low_cardinality->getDictionaryType();
-
-                    /// TODO: Make map with defaults conversion.
-                    src_converted_to_full_column = !removeNullable(arg.type)->equals(*removeNullable(res_type));
-                    if (src_converted_to_full_column)
-                        arg.column = arg.column->index(col_low_cardinality->getIndexes(), 0);
-                    else
-                        res_indexes = col_low_cardinality->getIndexesPtr();
-
-                    tmp_rows_count = arg.column->size();
-                }
-
-                /// Perform the requested conversion.
-                converted_column = wrapper(args, res_type, nullable_source, tmp_rows_count);
-            }
-
-            if (to_low_cardinality)
-            {
-                auto res_column = to_low_cardinality->createColumn();
-                auto * col_low_cardinality = assert_cast<ColumnLowCardinality *>(res_column.get());
-
-                if (from_low_cardinality && !src_converted_to_full_column)
-                {
-                    col_low_cardinality->insertRangeFromDictionaryEncodedColumn(*converted_column, *res_indexes);
-                }
-                else
-                    col_low_cardinality->insertRangeFromFullColumn(*converted_column, 0, converted_column->size());
-
-                return res_column;
-            }
-            else if (!src_converted_to_full_column)
-                return converted_column->index(*res_indexes, 0);
-            else
-                return converted_column;
-        };
-    }
-
-    WrapperType prepareRemoveNullable(const DataTypePtr & from_type, const DataTypePtr & to_type, bool skip_not_null_check) const
-    {
-        /// Determine whether pre-processing and/or post-processing must take place during conversion.
-
-        bool source_is_nullable = from_type->isNullable();
-        bool result_is_nullable = to_type->isNullable();
-
-        auto wrapper = prepareImpl(removeNullable(from_type), removeNullable(to_type), result_is_nullable);
-
-        if (result_is_nullable)
-        {
-            return [wrapper, source_is_nullable]
-                (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count) -> ColumnPtr
-            {
-                /// Create a temporary columns on which to perform the operation.
-                const auto & nullable_type = static_cast<const DataTypeNullable &>(*result_type);
-                const auto & nested_type = nullable_type.getNestedType();
-
-                ColumnsWithTypeAndName tmp_args;
-                if (source_is_nullable)
-                    tmp_args = createBlockWithNestedColumns(arguments);
-                else
-                    tmp_args = arguments;
-
-                const ColumnNullable * nullable_source = nullptr;
-
-                /// Add original ColumnNullable for createStringToEnumWrapper()
-                if (source_is_nullable)
-                {
-                    if (arguments.size() != 1)
-                        throw Exception(ErrorCodes::LOGICAL_ERROR, "Invalid number of arguments");
-                    nullable_source = typeid_cast<const ColumnNullable *>(arguments.front().column.get());
-                }
-
-                /// Perform the requested conversion.
-                auto tmp_res = wrapper(tmp_args, nested_type, nullable_source, input_rows_count);
-
-                /// May happen in fuzzy tests. For debug purpose.
-                if (!tmp_res)
-                    throw Exception(ErrorCodes::LOGICAL_ERROR, "Couldn't convert {} to {} in prepareRemoveNullable wrapper.",
-                                    arguments[0].type->getName(), nested_type->getName());
-
-                return wrapInNullable(tmp_res, arguments, nested_type, input_rows_count);
-            };
-        }
-        else if (source_is_nullable)
-        {
-            /// Conversion from Nullable to non-Nullable.
-
-            return [wrapper, skip_not_null_check]
-                (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count) -> ColumnPtr
-            {
-                auto tmp_args = createBlockWithNestedColumns(arguments);
-                auto nested_type = removeNullable(result_type);
-
-                /// Check that all values are not-NULL.
-                /// Check can be skipped in case if LowCardinality dictionary is transformed.
-                /// In that case, correctness will be checked beforehand.
-                if (!skip_not_null_check)
-                {
-                    const auto & col = arguments[0].column;
-                    const auto & nullable_col = assert_cast<const ColumnNullable &>(*col);
-                    const auto & null_map = nullable_col.getNullMapData();
-
-                    if (!memoryIsZero(null_map.data(), 0, null_map.size()))
-                        throw Exception(ErrorCodes::CANNOT_INSERT_NULL_IN_ORDINARY_COLUMN, "Cannot convert NULL value to non-Nullable type");
-                }
-                const ColumnNullable * nullable_source = typeid_cast<const ColumnNullable *>(arguments.front().column.get());
-                return wrapper(tmp_args, nested_type, nullable_source, input_rows_count);
-            };
-        }
-        else
-            return wrapper;
-    }
-
-    /// 'from_type' and 'to_type' are nested types in case of Nullable.
-    /// 'requested_result_is_nullable' is true if CAST to Nullable type is requested.
-    WrapperType prepareImpl(const DataTypePtr & from_type, const DataTypePtr & to_type, bool requested_result_is_nullable) const
-    {
-        if (isUInt8(from_type) && isBool(to_type))
-            return createUInt8ToBoolWrapper(from_type, to_type);
-
-        /// We can cast IPv6 into IPv6, IPv4 into IPv4, but we should not allow to cast FixedString(16) into IPv6 as part of identity cast
-        bool safe_convert_custom_types = true;
-
-        if (const auto * to_type_custom_name = to_type->getCustomName())
-            safe_convert_custom_types = from_type->getCustomName() && from_type->getCustomName()->getName() == to_type_custom_name->getName();
-        else if (const auto * from_type_custom_name = from_type->getCustomName())
-            safe_convert_custom_types = to_type->getCustomName() && from_type_custom_name->getName() == to_type->getCustomName()->getName();
-
-        if (from_type->equals(*to_type) && safe_convert_custom_types)
-        {
-            /// We can only use identity conversion for DataTypeAggregateFunction when they are strictly equivalent.
-            if (typeid_cast<const DataTypeAggregateFunction *>(from_type.get()))
-            {
-                if (DataTypeAggregateFunction::strictEquals(from_type, to_type))
-                    return createIdentityWrapper(from_type);
-            }
-            else
-                return createIdentityWrapper(from_type);
-        }
-        else if (WhichDataType(from_type).isNothing())
-            return createNothingWrapper(to_type.get());
-
-        WrapperType ret;
-
-        auto make_default_wrapper = [&](const auto & types) -> bool
-        {
-            using Types = std::decay_t<decltype(types)>;
-            using ToDataType = typename Types::LeftType;
-
-            if constexpr (
-                std::is_same_v<ToDataType, DataTypeUInt16> ||
-                std::is_same_v<ToDataType, DataTypeUInt32> ||
-                std::is_same_v<ToDataType, DataTypeUInt64> ||
-                std::is_same_v<ToDataType, DataTypeUInt128> ||
-                std::is_same_v<ToDataType, DataTypeUInt256> ||
-                std::is_same_v<ToDataType, DataTypeInt8> ||
-                std::is_same_v<ToDataType, DataTypeInt16> ||
-                std::is_same_v<ToDataType, DataTypeInt32> ||
-                std::is_same_v<ToDataType, DataTypeInt64> ||
-                std::is_same_v<ToDataType, DataTypeInt128> ||
-                std::is_same_v<ToDataType, DataTypeInt256> ||
-                std::is_same_v<ToDataType, DataTypeFloat32> ||
-                std::is_same_v<ToDataType, DataTypeFloat64> ||
-                std::is_same_v<ToDataType, DataTypeDate> ||
-                std::is_same_v<ToDataType, DataTypeDate32> ||
-                std::is_same_v<ToDataType, DataTypeDateTime> ||
-                std::is_same_v<ToDataType, DataTypeUUID> ||
-                std::is_same_v<ToDataType, DataTypeIPv4> ||
-                std::is_same_v<ToDataType, DataTypeIPv6>)
-            {
-                ret = createWrapper(from_type, checkAndGetDataType<ToDataType>(to_type.get()), requested_result_is_nullable);
-                return true;
-            }
-            if constexpr (std::is_same_v<ToDataType, DataTypeUInt8>)
-            {
-                if (isBool(to_type))
-                    ret = createBoolWrapper<ToDataType>(from_type, checkAndGetDataType<ToDataType>(to_type.get()), requested_result_is_nullable);
-                else
-                    ret = createWrapper(from_type, checkAndGetDataType<ToDataType>(to_type.get()), requested_result_is_nullable);
-                return true;
-            }
-            if constexpr (
-                std::is_same_v<ToDataType, DataTypeEnum8> ||
-                std::is_same_v<ToDataType, DataTypeEnum16>)
-            {
-                ret = createEnumWrapper(from_type, checkAndGetDataType<ToDataType>(to_type.get()));
-                return true;
-            }
-            if constexpr (
-                std::is_same_v<ToDataType, DataTypeDecimal<Decimal32>> ||
-                std::is_same_v<ToDataType, DataTypeDecimal<Decimal64>> ||
-                std::is_same_v<ToDataType, DataTypeDecimal<Decimal128>> ||
-                std::is_same_v<ToDataType, DataTypeDecimal<Decimal256>> ||
-                std::is_same_v<ToDataType, DataTypeDateTime64>)
-            {
-                ret = createDecimalWrapper(from_type, checkAndGetDataType<ToDataType>(to_type.get()), requested_result_is_nullable);
-                return true;
-            }
-
-            return false;
-        };
-
-        bool cast_ipv4_ipv6_default_on_conversion_error_value = context && context->getSettingsRef().cast_ipv4_ipv6_default_on_conversion_error;
-        bool input_format_ipv4_default_on_conversion_error_value = context && context->getSettingsRef().input_format_ipv4_default_on_conversion_error;
-        bool input_format_ipv6_default_on_conversion_error_value = context && context->getSettingsRef().input_format_ipv6_default_on_conversion_error;
-
-        auto make_custom_serialization_wrapper = [&, cast_ipv4_ipv6_default_on_conversion_error_value, input_format_ipv4_default_on_conversion_error_value, input_format_ipv6_default_on_conversion_error_value](const auto & types) -> bool
-        {
-            using Types = std::decay_t<decltype(types)>;
-            using ToDataType = typename Types::RightType;
-            using FromDataType = typename Types::LeftType;
-
-            if constexpr (WhichDataType(FromDataType::type_id).isStringOrFixedString())
-            {
-                if constexpr (std::is_same_v<ToDataType, DataTypeIPv4>)
-                {
-                    ret = [cast_ipv4_ipv6_default_on_conversion_error_value,
-                           input_format_ipv4_default_on_conversion_error_value,
-                           requested_result_is_nullable](
-                              ColumnsWithTypeAndName & arguments,
-                              const DataTypePtr & result_type,
-                              const ColumnNullable * column_nullable,
-                              size_t) -> ColumnPtr
-                    {
-                        if (!WhichDataType(result_type).isIPv4())
-                            throw Exception(ErrorCodes::TYPE_MISMATCH, "Wrong result type {}. Expected IPv4", result_type->getName());
-
-                        const auto * null_map = column_nullable ? &column_nullable->getNullMapData() : nullptr;
-                        if (requested_result_is_nullable)
-                            return convertToIPv4<IPStringToNumExceptionMode::Null>(arguments[0].column, null_map);
-                        else if (cast_ipv4_ipv6_default_on_conversion_error_value || input_format_ipv4_default_on_conversion_error_value)
-                            return convertToIPv4<IPStringToNumExceptionMode::Default>(arguments[0].column, null_map);
-                        else
-                            return convertToIPv4<IPStringToNumExceptionMode::Throw>(arguments[0].column, null_map);
-                    };
-
-                    return true;
-                }
-
-                if constexpr (std::is_same_v<ToDataType, DataTypeIPv6>)
-                {
-                    ret = [cast_ipv4_ipv6_default_on_conversion_error_value,
-                           input_format_ipv6_default_on_conversion_error_value,
-                           requested_result_is_nullable](
-                              ColumnsWithTypeAndName & arguments,
-                              const DataTypePtr & result_type,
-                              const ColumnNullable * column_nullable,
-                              size_t) -> ColumnPtr
-                    {
-                        if (!WhichDataType(result_type).isIPv6())
-                            throw Exception(
-                                ErrorCodes::TYPE_MISMATCH, "Wrong result type {}. Expected IPv6", result_type->getName());
-
-                        const auto * null_map = column_nullable ? &column_nullable->getNullMapData() : nullptr;
-                        if (requested_result_is_nullable)
-                            return convertToIPv6<IPStringToNumExceptionMode::Null>(arguments[0].column, null_map);
-                        else if (cast_ipv4_ipv6_default_on_conversion_error_value || input_format_ipv6_default_on_conversion_error_value)
-                            return convertToIPv6<IPStringToNumExceptionMode::Default>(arguments[0].column, null_map);
-                        else
-                            return convertToIPv6<IPStringToNumExceptionMode::Throw>(arguments[0].column, null_map);
-                    };
-
-                    return true;
-                }
-
-                if (to_type->getCustomSerialization() && to_type->getCustomName())
-                {
-                    ret = [requested_result_is_nullable](
-                              ColumnsWithTypeAndName & arguments,
-                              const DataTypePtr & result_type,
-                              const ColumnNullable * column_nullable,
-                              size_t input_rows_count) -> ColumnPtr
-                    {
-                        auto wrapped_result_type = result_type;
-                        if (requested_result_is_nullable)
-                            wrapped_result_type = makeNullable(result_type);
-                        return ConvertImplGenericFromString<typename FromDataType::ColumnType>::execute(
-                            arguments, wrapped_result_type, column_nullable, input_rows_count);
-                    };
-                    return true;
-                }
-            }
-            else if constexpr (WhichDataType(FromDataType::type_id).isIPv6() && WhichDataType(ToDataType::type_id).isIPv4())
-            {
-                ret = [cast_ipv4_ipv6_default_on_conversion_error_value, requested_result_is_nullable](
-                                ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable * column_nullable, size_t)
-                        -> ColumnPtr
-                {
-                    if (!WhichDataType(result_type).isIPv4())
-                        throw Exception(
-                            ErrorCodes::TYPE_MISMATCH, "Wrong result type {}. Expected IPv4", result_type->getName());
-
-                    const auto * null_map = column_nullable ? &column_nullable->getNullMapData() : nullptr;
-                    if (requested_result_is_nullable)
-                        return convertIPv6ToIPv4<IPStringToNumExceptionMode::Null>(arguments[0].column, null_map);
-                    else if (cast_ipv4_ipv6_default_on_conversion_error_value)
-                        return convertIPv6ToIPv4<IPStringToNumExceptionMode::Default>(arguments[0].column, null_map);
-                    else
-                        return convertIPv6ToIPv4<IPStringToNumExceptionMode::Throw>(arguments[0].column, null_map);
-                };
-
-                return true;
-            }
-
-            if constexpr (WhichDataType(ToDataType::type_id).isStringOrFixedString())
-            {
-                if constexpr (WhichDataType(FromDataType::type_id).isEnum())
-                {
-                    ret = createEnumToStringWrapper<FromDataType>();
-                    return true;
-                }
-                else if (from_type->getCustomSerialization())
-                {
-                    ret = [](ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count) -> ColumnPtr
-                    {
-                        return ConvertImplGenericToString<typename ToDataType::ColumnType>::execute(arguments, result_type, input_rows_count);
-                    };
-                    return true;
-                }
-            }
-
-            return false;
-        };
-
-        if (callOnTwoTypeIndexes(from_type->getTypeId(), to_type->getTypeId(), make_custom_serialization_wrapper))
-            return ret;
-
-        if (callOnIndexAndDataType<void>(to_type->getTypeId(), make_default_wrapper))
-            return ret;
-
-        switch (to_type->getTypeId())
-        {
-            case TypeIndex::String:
-                return createStringWrapper(from_type);
-            case TypeIndex::FixedString:
-                return createFixedStringWrapper(from_type, checkAndGetDataType<DataTypeFixedString>(to_type.get())->getN());
-            case TypeIndex::Array:
-                return createArrayWrapper(from_type, static_cast<const DataTypeArray &>(*to_type));
-            case TypeIndex::Tuple:
-                return createTupleWrapper(from_type, checkAndGetDataType<DataTypeTuple>(to_type.get()));
-            case TypeIndex::Map:
-                return createMapWrapper(from_type, checkAndGetDataType<DataTypeMap>(to_type.get()));
-            case TypeIndex::Object:
-                return createObjectWrapper(from_type, checkAndGetDataType<DataTypeObject>(to_type.get()));
-            case TypeIndex::AggregateFunction:
-                return createAggregateFunctionWrapper(from_type, checkAndGetDataType<DataTypeAggregateFunction>(to_type.get()));
-            case TypeIndex::Interval:
-                return createIntervalWrapper(from_type, checkAndGetDataType<DataTypeInterval>(to_type.get())->getKind());
-            default:
-                break;
-        }
-
-        if (cast_type == CastType::accurateOrNull)
-            return createToNullableColumnWrapper();
-        else
-            throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Conversion from {} to {} is not supported",
-                from_type->getName(), to_type->getName());
-    }
-};
-
-class MonotonicityHelper
-{
-public:
-    using MonotonicityForRange = FunctionCastBase::MonotonicityForRange;
-
-    template <typename DataType>
-    static auto monotonicityForType(const DataType * const)
-    {
-        return FunctionTo<DataType>::Type::Monotonic::get;
-    }
-
-    static MonotonicityForRange getMonotonicityInformation(const DataTypePtr & from_type, const IDataType * to_type)
-    {
-        if (const auto * type = checkAndGetDataType<DataTypeUInt8>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeUInt16>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeUInt32>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeUInt64>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeUInt128>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeUInt256>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeInt8>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeInt16>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeInt32>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeInt64>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeInt128>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeInt256>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeFloat32>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeFloat64>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeDate>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeDate32>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeDateTime>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeString>(to_type))
-            return monotonicityForType(type);
-        if (isEnum(from_type))
-        {
-            if (const auto * type = checkAndGetDataType<DataTypeEnum8>(to_type))
-                return monotonicityForType(type);
-            if (const auto * type = checkAndGetDataType<DataTypeEnum16>(to_type))
-                return monotonicityForType(type);
-        }
-        /// other types like Null, FixedString, Array and Tuple have no monotonicity defined
-        return {};
-    }
-};
-
-}

From 277032e444652cfc18f62032a35af918bac39f4d Mon Sep 17 00:00:00 2001
From: "Mikhail f. Shiryaev" <felixoid@clickhouse.com>
Date: Thu, 14 Mar 2024 14:16:33 +0100
Subject: [PATCH 335/443] Move the least updating cctools into beginning,
 update it

---
 docker/packager/binary-builder/Dockerfile | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/docker/packager/binary-builder/Dockerfile b/docker/packager/binary-builder/Dockerfile
index 96c90403187..c9442accd7e 100644
--- a/docker/packager/binary-builder/Dockerfile
+++ b/docker/packager/binary-builder/Dockerfile
@@ -4,6 +4,9 @@ FROM clickhouse/fasttest:$FROM_TAG
 ENV CC=clang-${LLVM_VERSION}
 ENV CXX=clang++-${LLVM_VERSION}
 
+# If the cctools is updated, then first build it in the CI, then update here in a different commit
+COPY --from=clickhouse/cctools:d9e3596e706b /cctools /cctools
+
 # Rust toolchain and libraries
 ENV RUSTUP_HOME=/rust/rustup
 ENV CARGO_HOME=/rust/cargo
@@ -73,9 +76,6 @@ RUN curl -Lo /usr/bin/clang-tidy-cache \
         "https://raw.githubusercontent.com/matus-chochlik/ctcache/$CLANG_TIDY_SHA1/clang-tidy-cache" \
     && chmod +x /usr/bin/clang-tidy-cache
 
-# If the cctools is updated, then first build it in the CI, then update here in a different commit
-COPY --from=clickhouse/cctools:5a908f73878a /cctools /cctools
-
 RUN mkdir /workdir && chmod 777 /workdir
 WORKDIR /workdir
 

From d465835306c013b62e35c1330326505095b8b658 Mon Sep 17 00:00:00 2001
From: "Mikhail f. Shiryaev" <felixoid@clickhouse.com>
Date: Thu, 14 Mar 2024 14:08:43 +0100
Subject: [PATCH 336/443] Reorder hidden and shown checks in comment, change
 url of Mergeable check

---
 tests/ci/commit_status_helper.py | 20 +++++++++++---------
 1 file changed, 11 insertions(+), 9 deletions(-)

diff --git a/tests/ci/commit_status_helper.py b/tests/ci/commit_status_helper.py
index 1c2d8b2ade8..bda2db13991 100644
--- a/tests/ci/commit_status_helper.py
+++ b/tests/ci/commit_status_helper.py
@@ -18,8 +18,10 @@ from github.GithubObject import NotSet
 from github.IssueComment import IssueComment
 from github.Repository import Repository
 
-from ci_config import REQUIRED_CHECKS, CHECK_DESCRIPTIONS, CheckDescription
-from env_helper import GITHUB_JOB_URL, GITHUB_REPOSITORY, TEMP_PATH
+# isort: on
+
+from ci_config import CHECK_DESCRIPTIONS, REQUIRED_CHECKS, CheckDescription
+from env_helper import GITHUB_REPOSITORY, GITHUB_RUN_URL, TEMP_PATH
 from pr_info import SKIP_MERGEABLE_CHECK_LABEL, PRInfo
 from report import (
     ERROR,
@@ -259,6 +261,12 @@ def generate_status_comment(pr_info: PRInfo, statuses: CommitStatuses) -> str:
 
     result = [comment_body]
 
+    if visible_table_rows:
+        visible_table_rows.sort()
+        result.append(table_header)
+        result.extend(visible_table_rows)
+        result.append(table_footer)
+
     if hidden_table_rows:
         hidden_table_rows.sort()
         result.append(details_header)
@@ -267,12 +275,6 @@ def generate_status_comment(pr_info: PRInfo, statuses: CommitStatuses) -> str:
         result.append(table_footer)
         result.append(details_footer)
 
-    if visible_table_rows:
-        visible_table_rows.sort()
-        result.append(table_header)
-        result.extend(visible_table_rows)
-        result.append(table_footer)
-
     return "".join(result)
 
 
@@ -427,7 +429,7 @@ def set_mergeable_check(
         context=MERGEABLE_NAME,
         description=format_description(description),
         state=state,
-        target_url=GITHUB_JOB_URL(),
+        target_url=GITHUB_RUN_URL,
     )
 
 

From 03b9bca8a4dc7f27c20d8cca22cbbc234ed0468c Mon Sep 17 00:00:00 2001
From: "Mikhail f. Shiryaev" <felixoid@clickhouse.com>
Date: Thu, 14 Mar 2024 14:31:55 +0100
Subject: [PATCH 337/443] Terminate EC2 on spot event if runner isn't running

---
 tests/ci/worker/init_runner.sh | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/tests/ci/worker/init_runner.sh b/tests/ci/worker/init_runner.sh
index b211128cf10..de1d128dc87 100644
--- a/tests/ci/worker/init_runner.sh
+++ b/tests/ci/worker/init_runner.sh
@@ -138,13 +138,15 @@ check_spot_instance_is_old() {
 check_proceed_spot_termination() {
     # The function checks and proceeds spot instance termination if exists
     # The event for spot instance termination
+    local FORCE
+    FORCE=${1:-}
     if TERMINATION_DATA=$(curl -s --fail http://169.254.169.254/latest/meta-data/spot/instance-action); then
         # https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/spot-instance-termination-notices.html#instance-action-metadata
         _action=$(jq '.action' -r <<< "$TERMINATION_DATA")
         _time=$(jq '.time | fromdate' <<< "$TERMINATION_DATA")
         _until_action=$((_time - $(date +%s)))
         echo "Received the '$_action' event that will be effective in $_until_action seconds"
-        if (( _until_action <= 30 )); then
+        if (( _until_action <= 30 )) || [ "$FORCE" == "force" ]; then
             echo "The action $_action will be done in $_until_action, killing the runner and exit"
             local runner_pid
             runner_pid=$(pgrep Runner.Listener)
@@ -309,7 +311,7 @@ while true; do
         echo "Checking if the instance suppose to terminate"
         no_terminating_metadata || terminate_on_event
         check_spot_instance_is_old && terminate_and_exit
-        check_proceed_spot_termination
+        check_proceed_spot_termination force
 
         echo "Going to configure runner"
         sudo -u ubuntu ./config.sh --url $RUNNER_URL --token "$(get_runner_token)" \
@@ -319,7 +321,7 @@ while true; do
         echo "Another one check to avoid race between runner and infrastructure"
         no_terminating_metadata || terminate_on_event
         check_spot_instance_is_old && terminate_and_exit
-        check_proceed_spot_termination
+        check_proceed_spot_termination force
 
         echo "Run"
         sudo -u ubuntu \

From 82b089f4e95073e7048254bf5d4dc03a515f71a2 Mon Sep 17 00:00:00 2001
From: Anton Popov <anton@clickhouse.com>
Date: Thu, 14 Mar 2024 13:59:29 +0000
Subject: [PATCH 338/443] remove profile event

---
 src/Common/ProfileEvents.cpp    | 1 -
 src/Interpreters/Aggregator.cpp | 4 ----
 2 files changed, 5 deletions(-)

diff --git a/src/Common/ProfileEvents.cpp b/src/Common/ProfileEvents.cpp
index 8fd1e189977..e43e8910089 100644
--- a/src/Common/ProfileEvents.cpp
+++ b/src/Common/ProfileEvents.cpp
@@ -533,7 +533,6 @@ The server successfully detected this situation and will download merged part fr
     \
     M(AggregationPreallocatedElementsInHashTables, "How many elements were preallocated in hash tables for aggregation.") \
     M(AggregationHashTablesInitializedAsTwoLevel, "How many hash tables were inited as two-level for aggregation.") \
-    M(AggregationProcessedBlocks, "How many blocks were processed by Aggregator") \
     M(AggregationOptimizedEqualRangesOfKeys, "For how many blocks optimization of equal ranges of keys was applied") \
     \
     M(MetadataFromKeeperCacheHit, "Number of times an object storage metadata request was answered from cache without making request to Keeper") \
diff --git a/src/Interpreters/Aggregator.cpp b/src/Interpreters/Aggregator.cpp
index 7c9dac82eff..40b1c09a32e 100644
--- a/src/Interpreters/Aggregator.cpp
+++ b/src/Interpreters/Aggregator.cpp
@@ -53,7 +53,6 @@ namespace ProfileEvents
     extern const Event OverflowThrow;
     extern const Event OverflowBreak;
     extern const Event OverflowAny;
-    extern const Event AggregationProcessedBlocks;
     extern const Event AggregationOptimizedEqualRangesOfKeys;
 }
 
@@ -987,7 +986,6 @@ void Aggregator::executeOnBlockSmall(
 {
     /// `result` will destroy the states of aggregate functions in the destructor
     result.aggregator = this;
-    ProfileEvents::increment(ProfileEvents::AggregationProcessedBlocks);
 
     /// How to perform the aggregation?
     if (result.empty())
@@ -1521,7 +1519,6 @@ void NO_INLINE Aggregator::executeOnIntervalWithoutKey(
     /// `data_variants` will destroy the states of aggregate functions in the destructor
     data_variants.aggregator = this;
     data_variants.init(AggregatedDataVariants::Type::without_key);
-    ProfileEvents::increment(ProfileEvents::AggregationProcessedBlocks);
 
     AggregatedDataWithoutKey & res = data_variants.without_key;
 
@@ -1653,7 +1650,6 @@ bool Aggregator::executeOnBlock(Columns columns,
 {
     /// `result` will destroy the states of aggregate functions in the destructor
     result.aggregator = this;
-    ProfileEvents::increment(ProfileEvents::AggregationProcessedBlocks);
 
     /// How to perform the aggregation?
     if (result.empty())

From 200823c31108015b16ead859a346a59f83cc6e74 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Thu, 14 Mar 2024 15:04:19 +0100
Subject: [PATCH 339/443] Try fix docs check

---
 .../sql-reference/aggregate-functions/reference/varpop.md   | 4 ++--
 .../sql-reference/aggregate-functions/reference/varsamp.md  | 6 +++---
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/docs/en/sql-reference/aggregate-functions/reference/varpop.md b/docs/en/sql-reference/aggregate-functions/reference/varpop.md
index 76472f62789..2044b7e690b 100644
--- a/docs/en/sql-reference/aggregate-functions/reference/varpop.md
+++ b/docs/en/sql-reference/aggregate-functions/reference/varpop.md
@@ -63,8 +63,8 @@ covarPopStable(x, y)
 
 **Parameters**
 
-- `x`: The first data column. [String literal](../syntax#syntax-string-literal)
-- `y`: The second data column. [Expression](../syntax#syntax-expressions)
+- `x`: The first data column. [String literal](../../syntax#syntax-string-literal)
+- `y`: The second data column. [Expression](../../syntax#syntax-expressions)
 
 **Returned value**
 
diff --git a/docs/en/sql-reference/aggregate-functions/reference/varsamp.md b/docs/en/sql-reference/aggregate-functions/reference/varsamp.md
index e75cb075ff8..be669a16ae8 100644
--- a/docs/en/sql-reference/aggregate-functions/reference/varsamp.md
+++ b/docs/en/sql-reference/aggregate-functions/reference/varsamp.md
@@ -8,7 +8,7 @@ This page contains information on the `varSamp` and `varSampStable` ClickHouse f
 
 ## varSamp
 
-Calculate the sample variance of a data set. 
+Calculate the sample variance of a data set.
 
 **Syntax**
 
@@ -18,7 +18,7 @@ varSamp(expr)
 
 **Parameters**
 
-- `expr`: An expression representing the data set for which you want to calculate the sample variance. [Expression](../syntax#syntax-expressions)
+- `expr`: An expression representing the data set for which you want to calculate the sample variance. [Expression](../../syntax#syntax-expressions)
 
 **Returned value**
 
@@ -78,7 +78,7 @@ varSampStable(expr)
 
 **Parameters**
 
-- `expr`: An expression representing the data set for which you want to calculate the sample variance. [Expression](../syntax#syntax-expressions)
+- `expr`: An expression representing the data set for which you want to calculate the sample variance. [Expression](../../syntax#syntax-expressions)
 
 **Returned value**
 

From 02e81329794ed326853f1e039bca49e960ed469d Mon Sep 17 00:00:00 2001
From: Anton Popov <anton@clickhouse.com>
Date: Thu, 14 Mar 2024 14:05:19 +0000
Subject: [PATCH 340/443] fix test

---
 .../03008_optimize_equal_ranges.reference            | 12 ++++++++++--
 .../0_stateless/03008_optimize_equal_ranges.sql      |  9 ++++++---
 2 files changed, 16 insertions(+), 5 deletions(-)

diff --git a/tests/queries/0_stateless/03008_optimize_equal_ranges.reference b/tests/queries/0_stateless/03008_optimize_equal_ranges.reference
index 08f8008fca6..fc7a4f3c118 100644
--- a/tests/queries/0_stateless/03008_optimize_equal_ranges.reference
+++ b/tests/queries/0_stateless/03008_optimize_equal_ranges.reference
@@ -1,8 +1,16 @@
 0	30000
 1	30000
 2	30000
+0	30000
+1	30000
+2	30000
 0	449985000
 1	449985000
 2	449985000
-sum	1
-uniqExact	0
+0	449985000
+1	449985000
+2	449985000
+sum	1	1
+sum	16	1
+uniqExact	1	1
+uniqExact	16	0
diff --git a/tests/queries/0_stateless/03008_optimize_equal_ranges.sql b/tests/queries/0_stateless/03008_optimize_equal_ranges.sql
index c6143fb7f51..4d521420741 100644
--- a/tests/queries/0_stateless/03008_optimize_equal_ranges.sql
+++ b/tests/queries/0_stateless/03008_optimize_equal_ranges.sql
@@ -10,16 +10,19 @@ INSERT INTO t_optimize_equal_ranges SELECT 0, toString(number), number FROM numb
 INSERT INTO t_optimize_equal_ranges SELECT 1, toString(number), number FROM numbers(30000);
 INSERT INTO t_optimize_equal_ranges SELECT 2, toString(number), number FROM numbers(30000);
 
-SELECT a, uniqExact(b) FROM t_optimize_equal_ranges GROUP BY a ORDER BY a;
-SELECT a, sum(c) FROM t_optimize_equal_ranges GROUP BY a ORDER BY a;
+SELECT a, uniqExact(b) FROM t_optimize_equal_ranges GROUP BY a ORDER BY a SETTINGS max_threads = 16;
+SELECT a, uniqExact(b) FROM t_optimize_equal_ranges GROUP BY a ORDER BY a SETTINGS max_threads = 1;
+SELECT a, sum(c) FROM t_optimize_equal_ranges GROUP BY a ORDER BY a SETTINGS max_threads = 16;
+SELECT a, sum(c) FROM t_optimize_equal_ranges GROUP BY a ORDER BY a SETTINGS max_threads = 1;
 
 SYSTEM FLUSH LOGS;
 
 SELECT
     used_aggregate_functions[1] AS func,
+    Settings['max_threads'] AS threads,
     ProfileEvents['AggregationOptimizedEqualRangesOfKeys'] > 0
 FROM system.query_log
 WHERE type = 'QueryFinish' AND current_database = currentDatabase() AND query LIKE '%SELECT%FROM%t_optimize_equal_ranges%'
-ORDER BY func;
+ORDER BY func, threads;
 
 DROP TABLE t_optimize_equal_ranges;

From b21a5fec7b99bf223482491759f72357ec447480 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Thu, 14 Mar 2024 15:06:08 +0100
Subject: [PATCH 341/443] Fix broken list and thank clang tidy

---
 src/Client/QueryFuzzer.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Client/QueryFuzzer.cpp b/src/Client/QueryFuzzer.cpp
index 38e78157096..0a7cb1b36db 100644
--- a/src/Client/QueryFuzzer.cpp
+++ b/src/Client/QueryFuzzer.cpp
@@ -974,7 +974,7 @@ ASTPtr QueryFuzzer::reverseLiteralFuzzing(ASTPtr child)
             "toFixedString", /// Same as toDecimal
             "toInt128",
             "toInt256",
-            "toLowCardinality"
+            "toLowCardinality",
             "toNullable",
             "toUInt128",
             "toUInt256"};

From 026ac4deb14963077722ba527cabf76abd2ff0c0 Mon Sep 17 00:00:00 2001
From: Antonio Andelic <antonio@clickhouse.com>
Date: Thu, 14 Mar 2024 14:20:22 +0000
Subject: [PATCH 342/443] Fix heap-use-after-free for Merge table with alias

---
 src/Storages/StorageMerge.cpp                 |  3 +-
 ...t_storage_merge_aliases_analyzer.reference | 10 ++++
 ...25_test_storage_merge_aliases_analyzer.sql | 60 +++++++++++++++++++
 3 files changed, 72 insertions(+), 1 deletion(-)
 create mode 100644 tests/queries/0_stateless/01925_test_storage_merge_aliases_analyzer.reference
 create mode 100644 tests/queries/0_stateless/01925_test_storage_merge_aliases_analyzer.sql

diff --git a/src/Storages/StorageMerge.cpp b/src/Storages/StorageMerge.cpp
index 8410f0a8df8..e07bcf339c3 100644
--- a/src/Storages/StorageMerge.cpp
+++ b/src/Storages/StorageMerge.cpp
@@ -681,8 +681,9 @@ public:
         {
             if (column->hasExpression())
             {
+                auto column_name = column->getColumnName();
                 node = column->getExpressionOrThrow();
-                node->setAlias(column->getColumnName());
+                node->setAlias(column_name);
             }
             else
                 column->setColumnSource(replacement_table_expression);
diff --git a/tests/queries/0_stateless/01925_test_storage_merge_aliases_analyzer.reference b/tests/queries/0_stateless/01925_test_storage_merge_aliases_analyzer.reference
new file mode 100644
index 00000000000..b0fea25ed4b
--- /dev/null
+++ b/tests/queries/0_stateless/01925_test_storage_merge_aliases_analyzer.reference
@@ -0,0 +1,10 @@
+alias1
+1	4	16	23
+23	16	4	1
+2020-02-02	1	4	2	16	3	23
+alias2
+1	3	4	4
+4	4	3	1
+23	16	4	1
+2020-02-01	1	3	2	4	3	4
+2020-02-02	1	4	2	16	3	23
diff --git a/tests/queries/0_stateless/01925_test_storage_merge_aliases_analyzer.sql b/tests/queries/0_stateless/01925_test_storage_merge_aliases_analyzer.sql
new file mode 100644
index 00000000000..31035aa80cd
--- /dev/null
+++ b/tests/queries/0_stateless/01925_test_storage_merge_aliases_analyzer.sql
@@ -0,0 +1,60 @@
+-- Tags: no-parallel
+
+drop table if exists merge;
+set allow_experimental_analyzer = 1;
+create table merge
+(
+    dt Date,
+    colAlias0 Int32,
+    colAlias1 Int32,
+    col2 Int32,
+    colAlias2 UInt32,
+    col3 Int32,
+    colAlias3 UInt32
+)
+engine = Merge(currentDatabase(), '^alias_');
+
+drop table if exists alias_1;
+drop table if exists alias_2;
+
+create table alias_1
+(
+    dt Date,
+    col Int32,
+    colAlias0 UInt32 alias col,
+    colAlias1 UInt32 alias col3 + colAlias0,
+    col2 Int32,
+    colAlias2 Int32 alias colAlias1 + col2 + 10,
+    col3 Int32,
+    colAlias3 Int32 alias colAlias2 + colAlias1 + col3
+)
+engine = MergeTree()
+order by (dt);
+
+insert into alias_1 (dt, col, col2, col3) values ('2020-02-02', 1, 2, 3);
+
+select 'alias1';
+select colAlias0, colAlias1, colAlias2, colAlias3 from alias_1;
+select colAlias3, colAlias2, colAlias1, colAlias0 from merge;
+select * from merge;
+
+create table alias_2
+(
+    dt Date,
+    col Int32,
+    col2 Int32,
+    colAlias0 UInt32 alias col,
+    colAlias3 Int32 alias col3 + colAlias0,
+    colAlias1 UInt32 alias colAlias0 + col2,
+    colAlias2 Int32 alias colAlias0 + colAlias1,
+    col3 Int32
+)
+engine = MergeTree()
+order by (dt);
+
+insert into alias_2 (dt, col, col2, col3) values ('2020-02-01', 1, 2, 3);
+
+select 'alias2';
+select colAlias0, colAlias1, colAlias2, colAlias3 from alias_2;
+select colAlias3, colAlias2, colAlias1, colAlias0 from merge order by dt;
+select * from merge order by dt;

From e145115ef1a1cdb8f480cbbd51a7eab3f86a43d3 Mon Sep 17 00:00:00 2001
From: Dmitry Novik <n0vik@clickhouse.com>
Date: Thu, 14 Mar 2024 16:15:06 +0100
Subject: [PATCH 343/443] Fixup

---
 src/Storages/StorageMerge.cpp | 25 +++++--------------------
 1 file changed, 5 insertions(+), 20 deletions(-)

diff --git a/src/Storages/StorageMerge.cpp b/src/Storages/StorageMerge.cpp
index f0b9d58f3dd..7124cd7393e 100644
--- a/src/Storages/StorageMerge.cpp
+++ b/src/Storages/StorageMerge.cpp
@@ -56,17 +56,12 @@
 #include <Storages/checkAndGetLiteralArgument.h>
 #include <base/defines.h>
 #include <base/range.h>
-#include "Common/logger_useful.h"
 #include <Common/Exception.h>
 #include <Common/assert_cast.h>
 #include <Common/checkStackSize.h>
 #include <Common/typeid_cast.h>
-#include "Analyzer/QueryNode.h"
-#include "Core/QueryProcessingStage.h"
-#include "IO/WriteHelpers.h"
 #include <Core/NamesAndTypes.h>
 #include <Functions/FunctionFactory.h>
-#include <Poco/Logger.h>
 
 namespace DB
 {
@@ -803,13 +798,10 @@ QueryTreeNodePtr replaceTableExpressionAndRemoveJoin(
     const ContextPtr & context,
     const Names & required_column_names)
 {
-    LOG_DEBUG(&Poco::Logger::get("replaceTableExpressionAndRemoveJoin"), "BEFORE:\n{}", query->dumpTree());
     auto * query_node = query->as<QueryNode>();
     auto join_tree_type = query_node->getJoinTree()->getNodeType();
     auto modified_query = query_node->cloneAndReplace(original_table_expression, replacement_table_expression);
 
-    LOG_DEBUG(&Poco::Logger::get("replaceTableExpressionAndRemoveJoin"), "AFTER:\n{}", modified_query->dumpTree());
-
     // For the case when join tree is just a table or a table function we don't need to do anything more.
     if (join_tree_type == QueryTreeNodeType::TABLE || join_tree_type == QueryTreeNodeType::TABLE_FUNCTION)
         return modified_query;
@@ -969,8 +961,6 @@ SelectQueryInfo ReadFromMerge::getModifiedQueryInfo(const ContextPtr & modified_
                 column_name_to_node);
         }
 
-        LOG_DEBUG(&Poco::Logger::get("getModifiedQueryInfo"), "{}", modified_query_info.query_tree->dumpTree());
-
         modified_query_info.query = queryNodeToSelectQuery(modified_query_info.query_tree);
     }
     else
@@ -1058,8 +1048,6 @@ QueryPipelineBuilderPtr ReadFromMerge::createSources(
 
         Block pipe_header = builder->getHeader();
 
-        LOG_DEBUG(&Poco::Logger::get("createSources"), "Processed:{}\nStorage:{}", toString(processed_stage), toString(storage_stage));
-
         if (allow_experimental_analyzer)
         {
             String table_alias = modified_query_info.query_tree->as<QueryNode>()->getJoinTree()->as<TableNode>()->getAlias();
@@ -1067,7 +1055,8 @@ QueryPipelineBuilderPtr ReadFromMerge::createSources(
             String database_column = table_alias.empty() || processed_stage == QueryProcessingStage::FetchColumns ? "_database" : table_alias + "._database";
             String table_column = table_alias.empty() || processed_stage == QueryProcessingStage::FetchColumns ? "_table" : table_alias + "._table";
 
-            if (has_database_virtual_column && common_header.has(database_column) && (storage_stage == QueryProcessingStage::FetchColumns || dynamic_cast<const StorageDistributed *>(&storage_snapshot_->storage) != nullptr))
+            if (has_database_virtual_column && common_header.has(database_column)
+                && (storage_stage == QueryProcessingStage::FetchColumns || (dynamic_cast<const StorageDistributed *>(&storage_snapshot_->storage) != nullptr && !pipe_header.has("'" + database_name + "'_String"))))
             {
                 ColumnWithTypeAndName column;
                 column.name = database_column;
@@ -1082,7 +1071,8 @@ QueryPipelineBuilderPtr ReadFromMerge::createSources(
                                             { return std::make_shared<ExpressionTransform>(stream_header, adding_column_actions); });
             }
 
-            if (has_table_virtual_column && common_header.has(table_column) && (storage_stage == QueryProcessingStage::FetchColumns || dynamic_cast<const StorageDistributed *>(&storage_snapshot_->storage) != nullptr))
+            if (has_table_virtual_column && common_header.has(table_column)
+                && (storage_stage == QueryProcessingStage::FetchColumns || (dynamic_cast<const StorageDistributed *>(&storage_snapshot_->storage) != nullptr && !pipe_header.has("'" + table_name + "'_String"))))
             {
                 ColumnWithTypeAndName column;
                 column.name = table_column;
@@ -1166,8 +1156,6 @@ QueryPlan ReadFromMerge::createPlanForTable(
         storage_snapshot_,
         modified_query_info);
 
-    LOG_DEBUG(&Poco::Logger::get("createPlanForTable"), "Storage: {}", toString(storage_stage));
-
     QueryPlan plan;
 
     if (processed_stage <= storage_stage)
@@ -1545,8 +1533,7 @@ void ReadFromMerge::convertAndFilterSourceStream(
 
     ActionsDAG::MatchColumnsMode convert_actions_match_columns_mode = ActionsDAG::MatchColumnsMode::Name;
 
-    if (local_context->getSettingsRef().allow_experimental_analyzer
-        && (processed_stage == QueryProcessingStage::FetchColumns && dynamic_cast<const StorageDistributed *>(&snapshot->storage) != nullptr))
+    if (local_context->getSettingsRef().allow_experimental_analyzer && dynamic_cast<const StorageDistributed *>(&snapshot->storage) != nullptr)
         convert_actions_match_columns_mode = ActionsDAG::MatchColumnsMode::Position;
 
     if (row_policy_data_opt)
@@ -1554,8 +1541,6 @@ void ReadFromMerge::convertAndFilterSourceStream(
         row_policy_data_opt->addFilterTransform(builder);
     }
 
-    LOG_DEBUG(&Poco::Logger::get("convertAndFilterSourceStream"), "SOURCE:\n{}\nRESULT:\n{}", builder.getHeader().dumpStructure(), header.dumpStructure());
-
     auto convert_actions_dag = ActionsDAG::makeConvertingActions(builder.getHeader().getColumnsWithTypeAndName(),
                                                                 header.getColumnsWithTypeAndName(),
                                                                 convert_actions_match_columns_mode);

From d558c4dcb7ed2d80ca63d1bf2f9e09d580c843c0 Mon Sep 17 00:00:00 2001
From: Dmitry Novik <n0vik@clickhouse.com>
Date: Thu, 14 Mar 2024 16:17:58 +0100
Subject: [PATCH 344/443] Cleanup

---
 src/Storages/StorageMerge.cpp | 5 ++---
 src/Storages/StorageMerge.h   | 3 +--
 2 files changed, 3 insertions(+), 5 deletions(-)

diff --git a/src/Storages/StorageMerge.cpp b/src/Storages/StorageMerge.cpp
index 7124cd7393e..f95464f4bff 100644
--- a/src/Storages/StorageMerge.cpp
+++ b/src/Storages/StorageMerge.cpp
@@ -1121,7 +1121,7 @@ QueryPipelineBuilderPtr ReadFromMerge::createSources(
         /// Subordinary tables could have different but convertible types, like numeric types of different width.
         /// We must return streams with structure equals to structure of Merge table.
         convertAndFilterSourceStream(
-            header, modified_query_info, storage_snapshot_, aliases, row_policy_data_opt, context, *builder, storage_stage);
+            header, modified_query_info, storage_snapshot_, aliases, row_policy_data_opt, context, *builder);
     }
 
     return builder;
@@ -1473,8 +1473,7 @@ void ReadFromMerge::convertAndFilterSourceStream(
     const Aliases & aliases,
     const RowPolicyDataOpt & row_policy_data_opt,
     ContextPtr local_context,
-    QueryPipelineBuilder & builder,
-    QueryProcessingStage::Enum processed_stage [[maybe_unused]])
+    QueryPipelineBuilder & builder)
 {
     Block before_block_header = builder.getHeader();
 
diff --git a/src/Storages/StorageMerge.h b/src/Storages/StorageMerge.h
index 556649f622d..3aabd7e26e3 100644
--- a/src/Storages/StorageMerge.h
+++ b/src/Storages/StorageMerge.h
@@ -277,8 +277,7 @@ private:
         const Aliases & aliases,
         const RowPolicyDataOpt & row_policy_data_opt,
         ContextPtr context,
-        QueryPipelineBuilder & builder,
-        QueryProcessingStage::Enum processed_stage);
+        QueryPipelineBuilder & builder);
 
     StorageMerge::StorageListWithLocks getSelectedTables(
         ContextPtr query_context,

From c89010f80336384252e230158c85ae4f7ae1dae1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <benjamin.antal@clickhouse.com>
Date: Thu, 14 Mar 2024 15:22:01 +0000
Subject: [PATCH 345/443] Remove unused function template

---
 src/Dictionaries/RangeHashedDictionary_2.cpp | 114 -------------------
 1 file changed, 114 deletions(-)

diff --git a/src/Dictionaries/RangeHashedDictionary_2.cpp b/src/Dictionaries/RangeHashedDictionary_2.cpp
index 2329d621da4..d400fd1d830 100644
--- a/src/Dictionaries/RangeHashedDictionary_2.cpp
+++ b/src/Dictionaries/RangeHashedDictionary_2.cpp
@@ -118,120 +118,6 @@ size_t RangeHashedDictionary<dictionary_key_type>::getItemsShortCircuitImpl(
     return keys_found;
 }
 
-template <DictionaryKeyType dictionary_key_type>
-template <typename ValueType, bool is_nullable, typename DefaultValueExtractor>
-void RangeHashedDictionary<dictionary_key_type>::getItemsImpl(
-    const Attribute & attribute,
-    const Columns & key_columns,
-    typename RangeHashedDictionary<dictionary_key_type>::ValueSetterFunc<ValueType> && set_value,
-    DefaultValueExtractor & default_value_extractor) const
-{
-    const auto & attribute_container = std::get<AttributeContainerType<ValueType>>(attribute.container);
-
-    size_t keys_found = 0;
-
-    const ColumnPtr & range_column = key_columns.back();
-    auto key_columns_copy = key_columns;
-    key_columns_copy.pop_back();
-
-    DictionaryKeysArenaHolder<dictionary_key_type> arena_holder;
-    DictionaryKeysExtractor<dictionary_key_type> keys_extractor(key_columns_copy, arena_holder.getComplexKeyArena());
-    const size_t keys_size = keys_extractor.getKeysSize();
-
-    callOnRangeType(
-        dict_struct.range_min->type,
-        [&](const auto & types)
-        {
-            using Types = std::decay_t<decltype(types)>;
-            using RangeColumnType = typename Types::LeftType;
-            using RangeStorageType = typename RangeColumnType::ValueType;
-            using RangeInterval = Interval<RangeStorageType>;
-
-            const auto * range_column_typed = typeid_cast<const RangeColumnType *>(range_column.get());
-            if (!range_column_typed)
-                throw Exception(
-                    ErrorCodes::TYPE_MISMATCH,
-                    "Dictionary {} range column type should be equal to {}",
-                    getFullName(),
-                    dict_struct.range_min->type->getName());
-
-            const auto & range_column_data = range_column_typed->getData();
-
-            const auto & key_attribute_container = std::get<KeyAttributeContainerType<RangeStorageType>>(key_attribute.container);
-
-            for (size_t key_index = 0; key_index < keys_size; ++key_index)
-            {
-                auto key = keys_extractor.extractCurrentKey();
-                const auto it = key_attribute_container.find(key);
-
-                if (it)
-                {
-                    const auto date = range_column_data[key_index];
-                    const auto & interval_tree = it->getMapped();
-
-                    size_t value_index = 0;
-                    std::optional<RangeInterval> range;
-
-                    interval_tree.find(
-                        date,
-                        [&](auto & interval, auto & interval_value_index)
-                        {
-                            if (range)
-                            {
-                                if (likely(configuration.lookup_strategy == RangeHashedDictionaryLookupStrategy::min) && interval < *range)
-                                {
-                                    range = interval;
-                                    value_index = interval_value_index;
-                                }
-                                else if (configuration.lookup_strategy == RangeHashedDictionaryLookupStrategy::max && interval > *range)
-                                {
-                                    range = interval;
-                                    value_index = interval_value_index;
-                                }
-                            }
-                            else
-                            {
-                                range = interval;
-                                value_index = interval_value_index;
-                            }
-
-                            return true;
-                        });
-
-                    if (range.has_value())
-                    {
-                        ++keys_found;
-
-                        ValueType value = attribute_container[value_index];
-
-                        if constexpr (is_nullable)
-                        {
-                            bool is_null = (*attribute.is_value_nullable)[value_index];
-                            set_value(key_index, value, is_null);
-                        }
-                        else
-                        {
-                            set_value(key_index, value, false);
-                        }
-
-                        keys_extractor.rollbackCurrentKey();
-                        continue;
-                    }
-                }
-
-                if constexpr (is_nullable)
-                    set_value(key_index, default_value_extractor[key_index], default_value_extractor.isNullAt(key_index));
-                else
-                    set_value(key_index, default_value_extractor[key_index], false);
-
-                keys_extractor.rollbackCurrentKey();
-            }
-        });
-
-    query_count.fetch_add(keys_size, std::memory_order_relaxed);
-    found_count.fetch_add(keys_found, std::memory_order_relaxed);
-}
-
 #define INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL(DictionaryKeyType, IsNullable, ValueType) \
     template size_t RangeHashedDictionary<DictionaryKeyType>::getItemsShortCircuitImpl<ValueType, IsNullable>( \
         const Attribute & attribute, \

From b778b01408bf4057e030bd94e1d911c8f5efaca2 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Thu, 14 Mar 2024 16:39:32 +0100
Subject: [PATCH 346/443] Add settings

---
 src/Core/Settings.h                         | 3 +++
 src/Storages/StorageReplicatedMergeTree.cpp | 3 ++-
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/src/Core/Settings.h b/src/Core/Settings.h
index d70a6cf51c5..2755fa53866 100644
--- a/src/Core/Settings.h
+++ b/src/Core/Settings.h
@@ -838,6 +838,9 @@ class IColumn;
     M(Bool, compatibility_ignore_auto_increment_in_create_table, false, "Ignore AUTO_INCREMENT keyword in column declaration if true, otherwise return error. It simplifies migration from MySQL", 0) \
     M(Bool, multiple_joins_try_to_keep_original_names, false, "Do not add aliases to top level expression list on multiple joins rewrite", 0) \
     M(Bool, optimize_sorting_by_input_stream_properties, true, "Optimize sorting by sorting properties of input stream", 0) \
+    M(UInt64, keeper_max_retries, 10, "Max retries for general keeper operations", 0) \
+    M(UInt64, keeper_retry_initial_backoff_ms, 100, "Initial backoff timeout for general keeper operations", 0) \
+    M(UInt64, keeper_retry_max_backoff_ms, 5000, "Max backoff timeout for general keeper operations", 0) \
     M(UInt64, insert_keeper_max_retries, 20, "Max retries for keeper operations during insert", 0) \
     M(UInt64, insert_keeper_retry_initial_backoff_ms, 100, "Initial backoff timeout for keeper operations during insert", 0) \
     M(UInt64, insert_keeper_retry_max_backoff_ms, 10000, "Max backoff timeout for keeper operations during insert", 0) \
diff --git a/src/Storages/StorageReplicatedMergeTree.cpp b/src/Storages/StorageReplicatedMergeTree.cpp
index 71ac1fc4965..c41403e312b 100644
--- a/src/Storages/StorageReplicatedMergeTree.cpp
+++ b/src/Storages/StorageReplicatedMergeTree.cpp
@@ -7682,7 +7682,8 @@ void StorageReplicatedMergeTree::forcefullyRemoveBrokenOutdatedPartFromZooKeeper
 
     bool exists = false;
     String part_path = replica_path + "/parts/" + part_name;
-    ZooKeeperRetriesInfo retries_info{/* max_retries */5, /* initial_backoff_ms */100, /* max_backoff_ms */100};
+    const auto & settings = getContext()->getSettingsRef();
+    ZooKeeperRetriesInfo retries_info{settings.keeper_max_retries, settings.keeper_retry_initial_backoff_ms, settings.keeper_retry_max_backoff_ms};
     ZooKeeperRetriesControl retries_ctl("outdatedPartExists", log.load(), retries_info, nullptr);
 
     retries_ctl.retryLoop([&]() { exists = getZooKeeper()->exists(part_path); });

From 22ca96cf8d44aa907c0c3c463e4b0a5628312aa0 Mon Sep 17 00:00:00 2001
From: Antonio Andelic <antonio@clickhouse.com>
Date: Thu, 14 Mar 2024 16:05:01 +0000
Subject: [PATCH 347/443] Disable optimize_rewrite_sum_if_to_count_if if return
 is nullable

---
 src/Analyzer/Passes/SumIfToCountIfPass.cpp                     | 2 +-
 .../0_stateless/03010_sum_to_to_count_if_nullable.reference    | 0
 .../queries/0_stateless/03010_sum_to_to_count_if_nullable.sql  | 3 +++
 3 files changed, 4 insertions(+), 1 deletion(-)
 create mode 100644 tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.reference
 create mode 100644 tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.sql

diff --git a/src/Analyzer/Passes/SumIfToCountIfPass.cpp b/src/Analyzer/Passes/SumIfToCountIfPass.cpp
index 1a6ee9215a9..d374d92c1fb 100644
--- a/src/Analyzer/Passes/SumIfToCountIfPass.cpp
+++ b/src/Analyzer/Passes/SumIfToCountIfPass.cpp
@@ -32,7 +32,7 @@ public:
             return;
 
         auto * function_node = node->as<FunctionNode>();
-        if (!function_node || !function_node->isAggregateFunction())
+        if (!function_node || !function_node->isAggregateFunction() || function_node->getResultType()->isNullable())
             return;
 
         auto function_name = function_node->getFunctionName();
diff --git a/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.reference b/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.reference
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.sql b/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.sql
new file mode 100644
index 00000000000..394cd4f1ea5
--- /dev/null
+++ b/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.sql
@@ -0,0 +1,3 @@
+SET optimize_rewrite_sum_if_to_count_if = 1;
+SELECT (sumIf(toNullable(1), (number % 2) = 0), NULL) FROM numbers(10) SETTINGS allow_experimental_analyzer=0;
+SELECT (sumIf(toNullable(1), (number % 2) = 0), NULL) FROM numbers(10) SETTINGS allow_experimental_analyzer=1;
\ No newline at end of file

From 7c5ef07c7b9f20db16811eb60dbe27a36b821575 Mon Sep 17 00:00:00 2001
From: Peter <peter279k@gmail.com>
Date: Fri, 15 Mar 2024 00:18:37 +0800
Subject: [PATCH 348/443] Add checksum validating before extracting archive

---
 docs/en/getting-started/example-datasets/nyc-taxi.md | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/docs/en/getting-started/example-datasets/nyc-taxi.md b/docs/en/getting-started/example-datasets/nyc-taxi.md
index cac75fdc45a..516a6d54248 100644
--- a/docs/en/getting-started/example-datasets/nyc-taxi.md
+++ b/docs/en/getting-started/example-datasets/nyc-taxi.md
@@ -248,6 +248,9 @@ Some of the files might not download fully. Check the file sizes and re-download
 
 ``` bash
 $ curl -O https://datasets.clickhouse.com/trips_mergetree/partitions/trips_mergetree.tar
+# Validate the checksum
+$ md5sum trips_mergetree.tar
+# Checksum should be equal to: f3b8d469b41d9a82da064ded7245d12c
 $ tar xvf trips_mergetree.tar -C /var/lib/clickhouse # path to ClickHouse data directory
 $ # check permissions of unpacked data, fix if required
 $ sudo service clickhouse-server restart

From 8c22f8ec3c7b7b40d687e5eb137ba98bfe76c67b Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Thu, 14 Mar 2024 17:26:52 +0100
Subject: [PATCH 349/443] Update settings changes history

---
 src/Core/SettingsChangesHistory.h | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/src/Core/SettingsChangesHistory.h b/src/Core/SettingsChangesHistory.h
index e680c02671a..3b0af6ded27 100644
--- a/src/Core/SettingsChangesHistory.h
+++ b/src/Core/SettingsChangesHistory.h
@@ -93,6 +93,9 @@ static std::map<ClickHouseVersion, SettingsChangesHistory::SettingsChanges> sett
               {"input_format_json_use_string_type_for_ambiguous_paths_in_named_tuples_inference_from_objects", false, false, "Allow to use String type for ambiguous paths during named tuple inference from JSON objects"},
               {"throw_if_deduplication_in_dependent_materialized_views_enabled_with_async_insert", false, true, "Deduplication is dependent materialized view cannot work together with async inserts."},
               {"parallel_replicas_allow_in_with_subquery", false, true, "If true, subquery for IN will be executed on every follower replica"},
+              {"keeper_max_retries", 10, 10, "Max retries for general keeper operations"},
+              {"keeper_retry_initial_backoff_ms", 100, 100, "Initial backoff timeout for general keeper operations"},
+              {"keeper_retry_max_backoff_ms", 5000, 5000, "Max backoff timeout for general keeper operations"},
               }},
     {"24.2", {{"allow_suspicious_variant_types", true, false, "Don't allow creating Variant type with suspicious variants by default"},
               {"validate_experimental_and_suspicious_types_inside_nested_types", false, true, "Validate usage of experimental and suspicious types inside nested types"},

From 0b3b734c9b268e65c758e1f2092d176dfc1d9489 Mon Sep 17 00:00:00 2001
From: Dmitry Novik <n0vik@clickhouse.com>
Date: Thu, 14 Mar 2024 17:37:49 +0100
Subject: [PATCH 350/443] Analyzer: Fix planner context for subquery in
 StorageMerge

---
 src/Planner/PlannerContext.cpp              |  6 ++++++
 src/Planner/PlannerContext.h                |  8 ++++++--
 src/Storages/StorageMerge.cpp               | 22 ++++++++++++---------
 src/Storages/StorageMerge.h                 |  2 +-
 tests/analyzer_integration_broken_tests.txt |  1 -
 5 files changed, 26 insertions(+), 13 deletions(-)

diff --git a/src/Planner/PlannerContext.cpp b/src/Planner/PlannerContext.cpp
index f33255f0a44..f939b959ce7 100644
--- a/src/Planner/PlannerContext.cpp
+++ b/src/Planner/PlannerContext.cpp
@@ -48,6 +48,12 @@ PlannerContext::PlannerContext(ContextMutablePtr query_context_, GlobalPlannerCo
     , is_ast_level_optimization_allowed(!(query_context->getClientInfo().query_kind == ClientInfo::QueryKind::SECONDARY_QUERY || select_query_options_.ignore_ast_optimizations))
 {}
 
+PlannerContext::PlannerContext(ContextMutablePtr query_context_, PlannerContextPtr planner_context_)
+    : query_context(std::move(query_context_))
+    , global_planner_context(planner_context_->global_planner_context)
+    , is_ast_level_optimization_allowed(planner_context_->is_ast_level_optimization_allowed)
+{}
+
 TableExpressionData & PlannerContext::getOrCreateTableExpressionData(const QueryTreeNodePtr & table_expression_node)
 {
     auto [it, _] = table_expression_node_to_data.emplace(table_expression_node, TableExpressionData());
diff --git a/src/Planner/PlannerContext.h b/src/Planner/PlannerContext.h
index 4d9ba037cac..418240fa34e 100644
--- a/src/Planner/PlannerContext.h
+++ b/src/Planner/PlannerContext.h
@@ -75,12 +75,18 @@ private:
 
 using GlobalPlannerContextPtr = std::shared_ptr<GlobalPlannerContext>;
 
+class PlannerContext;
+using PlannerContextPtr = std::shared_ptr<PlannerContext>;
+
 class PlannerContext
 {
 public:
     /// Create planner context with query context and global planner context
     PlannerContext(ContextMutablePtr query_context_, GlobalPlannerContextPtr global_planner_context_, const SelectQueryOptions & select_query_options_);
 
+    /// Create planner with modified query_context
+    PlannerContext(ContextMutablePtr query_context_, PlannerContextPtr planner_context_);
+
     /// Get planner context query context
     ContextPtr getQueryContext() const
     {
@@ -191,6 +197,4 @@ private:
     PreparedSets prepared_sets;
 };
 
-using PlannerContextPtr = std::shared_ptr<PlannerContext>;
-
 }
diff --git a/src/Storages/StorageMerge.cpp b/src/Storages/StorageMerge.cpp
index 8410f0a8df8..4c53f67c76b 100644
--- a/src/Storages/StorageMerge.cpp
+++ b/src/Storages/StorageMerge.cpp
@@ -422,6 +422,7 @@ void ReadFromMerge::initializePipeline(QueryPipelineBuilder & pipeline, const Bu
     std::vector<std::unique_ptr<QueryPipelineBuilder>> pipelines;
 
     auto table_it = selected_tables.begin();
+    auto modified_context = Context::createCopy(context);
     for (size_t i = 0; i < selected_tables.size(); ++i, ++table_it)
     {
         auto & child_plan = child_plans->at(i);
@@ -438,7 +439,7 @@ void ReadFromMerge::initializePipeline(QueryPipelineBuilder & pipeline, const Bu
         if (child_plan.row_policy_data_opt)
             child_plan.row_policy_data_opt->extendNames(real_column_names);
 
-        auto modified_query_info = getModifiedQueryInfo(context, table, nested_storage_snaphsot, real_column_names, column_names_as_aliases, aliases);
+        auto modified_query_info = getModifiedQueryInfo(modified_context, table, nested_storage_snaphsot, real_column_names, column_names_as_aliases, aliases);
 
         auto source_pipeline = createSources(
             child_plan.plan,
@@ -547,9 +548,10 @@ std::vector<ReadFromMerge::ChildPlan> ReadFromMerge::createChildrenPlans(SelectQ
     }
 
     /// Settings will be modified when planning children tables.
-    auto modified_context = Context::createCopy(context);
     for (const auto & table : selected_tables)
     {
+        auto modified_context = Context::createCopy(context);
+
         size_t current_need_streams = tables_count >= num_streams ? 1 : (num_streams / tables_count);
         size_t current_streams = std::min(current_need_streams, remaining_streams);
         remaining_streams -= current_streams;
@@ -570,25 +572,25 @@ std::vector<ReadFromMerge::ChildPlan> ReadFromMerge::createChildrenPlans(SelectQ
         auto & aliases = res.back().table_aliases;
         auto & row_policy_data_opt = res.back().row_policy_data_opt;
         auto storage_metadata_snapshot = storage->getInMemoryMetadataPtr();
-        auto nested_storage_snaphsot = storage->getStorageSnapshot(storage_metadata_snapshot, context);
+        auto nested_storage_snaphsot = storage->getStorageSnapshot(storage_metadata_snapshot, modified_context);
 
         Names column_names_as_aliases;
         Names real_column_names = column_names;
 
         const auto & database_name = std::get<0>(table);
         const auto & table_name = std::get<3>(table);
-        auto row_policy_filter_ptr = context->getRowPolicyFilter(
+        auto row_policy_filter_ptr = modified_context->getRowPolicyFilter(
             database_name,
             table_name,
             RowPolicyFilterType::SELECT_FILTER);
         if (row_policy_filter_ptr)
         {
-            row_policy_data_opt = RowPolicyData(row_policy_filter_ptr, storage, context);
+            row_policy_data_opt = RowPolicyData(row_policy_filter_ptr, storage, modified_context);
             row_policy_data_opt->extendNames(real_column_names);
         }
 
         auto modified_query_info
-            = getModifiedQueryInfo(context, table, nested_storage_snaphsot, real_column_names, column_names_as_aliases, aliases);
+            = getModifiedQueryInfo(modified_context, table, nested_storage_snaphsot, real_column_names, column_names_as_aliases, aliases);
 
         if (!context->getSettingsRef().allow_experimental_analyzer)
         {
@@ -657,10 +659,9 @@ std::vector<ReadFromMerge::ChildPlan> ReadFromMerge::createChildrenPlans(SelectQ
             row_policy_data_opt,
             modified_context,
             current_streams);
+        res.back().plan.addInterpreterContext(modified_context);
     }
 
-    if (!res.empty())
-        res[0].plan.addInterpreterContext(modified_context);
 
     return res;
 }
@@ -863,7 +864,7 @@ QueryTreeNodePtr replaceTableExpressionAndRemoveJoin(
 
 }
 
-SelectQueryInfo ReadFromMerge::getModifiedQueryInfo(const ContextPtr & modified_context,
+SelectQueryInfo ReadFromMerge::getModifiedQueryInfo(const ContextMutablePtr & modified_context,
     const StorageWithLockAndName & storage_with_lock_and_name,
     const StorageSnapshotPtr & storage_snapshot_,
     Names required_column_names,
@@ -877,6 +878,9 @@ SelectQueryInfo ReadFromMerge::getModifiedQueryInfo(const ContextPtr & modified_
     if (modified_query_info.optimized_prewhere_info && !modified_query_info.prewhere_info)
         modified_query_info.prewhere_info = modified_query_info.optimized_prewhere_info;
 
+    if (modified_query_info.planner_context)
+        modified_query_info.planner_context = std::make_shared<PlannerContext>(modified_context, modified_query_info.planner_context);
+
     if (modified_query_info.table_expression)
     {
         auto replacement_table_expression = std::make_shared<TableNode>(storage, storage_lock, storage_snapshot_);
diff --git a/src/Storages/StorageMerge.h b/src/Storages/StorageMerge.h
index 556649f622d..c049d50f3b4 100644
--- a/src/Storages/StorageMerge.h
+++ b/src/Storages/StorageMerge.h
@@ -192,7 +192,7 @@ private:
 
     using Aliases = std::vector<AliasData>;
 
-    SelectQueryInfo getModifiedQueryInfo(const ContextPtr & modified_context,
+    SelectQueryInfo getModifiedQueryInfo(const ContextMutablePtr & modified_context,
         const StorageWithLockAndName & storage_with_lock_and_name,
         const StorageSnapshotPtr & storage_snapshot,
         Names required_column_names,
diff --git a/tests/analyzer_integration_broken_tests.txt b/tests/analyzer_integration_broken_tests.txt
index 31527dc3476..d2ef983f26d 100644
--- a/tests/analyzer_integration_broken_tests.txt
+++ b/tests/analyzer_integration_broken_tests.txt
@@ -1,4 +1,3 @@
 test_build_sets_from_multiple_threads/test.py::test_set
 test_concurrent_backups_s3/test.py::test_concurrent_backups
 test_distributed_type_object/test.py::test_distributed_type_object
-test_merge_table_over_distributed/test.py::test_global_in

From a6f1e09e69a583ab2f235f918e4bc0d92949d478 Mon Sep 17 00:00:00 2001
From: avogar <avogar@clickhouse.com>
Date: Thu, 14 Mar 2024 17:35:10 +0000
Subject: [PATCH 351/443] Test test test_system_clusters_actual_information
 flakiness

---
 .../test_system_clusters_actual_information/test.py | 13 -------------
 1 file changed, 13 deletions(-)

diff --git a/tests/integration/test_system_clusters_actual_information/test.py b/tests/integration/test_system_clusters_actual_information/test.py
index e90a6cdeb3f..c6e3262fd62 100644
--- a/tests/integration/test_system_clusters_actual_information/test.py
+++ b/tests/integration/test_system_clusters_actual_information/test.py
@@ -12,20 +12,11 @@ cluster = ClickHouseCluster(__file__)
 node = cluster.add_instance(
     "node", with_zookeeper=True, main_configs=["configs/remote_servers.xml"]
 )
-node_1 = cluster.add_instance("node_1", with_zookeeper=True)
-
 
 @pytest.fixture(scope="module")
 def started_cluster():
     try:
         cluster.start()
-        node_1.query_with_retry("DROP TABLE IF EXISTS replicated")
-
-        node_1.query_with_retry(
-            """CREATE TABLE replicated (id UInt32, date Date) ENGINE =
-            ReplicatedMergeTree('/clickhouse/tables/replicated', 'node_1')  ORDER BY id PARTITION BY toYYYYMM(date)"""
-        )
-
         node.query_with_retry(
             "CREATE TABLE distributed (id UInt32, date Date) ENGINE = Distributed('test_cluster', 'default', 'replicated')"
         )
@@ -37,8 +28,6 @@ def started_cluster():
 
 
 def test(started_cluster):
-    cluster.pause_container("node_1")
-
     node.query("SYSTEM RELOAD CONFIG")
     error = node.query_and_get_error(
         "SELECT count() FROM distributed SETTINGS receive_timeout=1, handshake_timeout_ms=1"
@@ -67,5 +56,3 @@ def test(started_cluster):
 
     assert recovery_time == 0
     assert errors_count == 0
-
-    cluster.unpause_container("node_1")

From 5c8f2bbda0124704fffc414e329b37d8245c42e4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <benjamin.antal@clickhouse.com>
Date: Thu, 14 Mar 2024 17:45:42 +0000
Subject: [PATCH 352/443] Split template instantiations into separate files

---
 src/Dictionaries/RangeHashedDictionary.h      |  1 +
 ... => RangeHashedDictionaryGetItemsImpl.txx} | 30 +++++++++----------
 ...ngeHashedDictionaryGetItemsImplDecimal.cpp | 10 +++++++
 ...RangeHashedDictionaryGetItemsImplFloat.cpp |  7 +++++
 .../RangeHashedDictionaryGetItemsImplInt.cpp  | 11 +++++++
 ...angeHashedDictionaryGetItemsImplOthers.cpp | 10 +++++++
 .../RangeHashedDictionaryGetItemsImplUInt.cpp | 11 +++++++
 ...hedDictionaryGetItemsShortCircuitImpl.txx} | 30 ++++++++-----------
 ...tionaryGetItemsShortCircuitImplDecimal.cpp | 10 +++++++
 ...ictionaryGetItemsShortCircuitImplFloat.cpp |  7 +++++
 ...dDictionaryGetItemsShortCircuitImplInt.cpp | 11 +++++++
 ...ctionaryGetItemsShortCircuitImplOthers.cpp | 10 +++++++
 ...DictionaryGetItemsShortCircuitImplUInt.cpp | 11 +++++++
 13 files changed, 126 insertions(+), 33 deletions(-)
 rename src/Dictionaries/{RangeHashedDictionary_3.cpp => RangeHashedDictionaryGetItemsImpl.txx} (98%)
 create mode 100644 src/Dictionaries/RangeHashedDictionaryGetItemsImplDecimal.cpp
 create mode 100644 src/Dictionaries/RangeHashedDictionaryGetItemsImplFloat.cpp
 create mode 100644 src/Dictionaries/RangeHashedDictionaryGetItemsImplInt.cpp
 create mode 100644 src/Dictionaries/RangeHashedDictionaryGetItemsImplOthers.cpp
 create mode 100644 src/Dictionaries/RangeHashedDictionaryGetItemsImplUInt.cpp
 rename src/Dictionaries/{RangeHashedDictionary_2.cpp => RangeHashedDictionaryGetItemsShortCircuitImpl.txx} (98%)
 create mode 100644 src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImplDecimal.cpp
 create mode 100644 src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImplFloat.cpp
 create mode 100644 src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImplInt.cpp
 create mode 100644 src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImplOthers.cpp
 create mode 100644 src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImplUInt.cpp

diff --git a/src/Dictionaries/RangeHashedDictionary.h b/src/Dictionaries/RangeHashedDictionary.h
index a5dedae97c4..23f7df8133e 100644
--- a/src/Dictionaries/RangeHashedDictionary.h
+++ b/src/Dictionaries/RangeHashedDictionary.h
@@ -292,6 +292,7 @@ private:
 extern template class RangeHashedDictionary<DictionaryKeyType::Simple>;
 extern template class RangeHashedDictionary<DictionaryKeyType::Complex>;
 
+
 namespace
 {
     template <typename F>
diff --git a/src/Dictionaries/RangeHashedDictionary_3.cpp b/src/Dictionaries/RangeHashedDictionaryGetItemsImpl.txx
similarity index 98%
rename from src/Dictionaries/RangeHashedDictionary_3.cpp
rename to src/Dictionaries/RangeHashedDictionaryGetItemsImpl.txx
index a3136d6f63d..9da2b0faf4a 100644
--- a/src/Dictionaries/RangeHashedDictionary_3.cpp
+++ b/src/Dictionaries/RangeHashedDictionaryGetItemsImpl.txx
@@ -1,5 +1,18 @@
 #include <Dictionaries/RangeHashedDictionary.h>
 
+#define INSTANTIATE_GET_ITEMS_IMPL(DictionaryKeyType, IsNullable, AttributeType, ValueType) \
+template void RangeHashedDictionary<DictionaryKeyType>::getItemsImpl<ValueType, IsNullable, DictionaryDefaultValueExtractor<AttributeType>>( \
+    const Attribute & attribute,\
+    const Columns & key_columns,\
+    typename RangeHashedDictionary<DictionaryKeyType>::ValueSetterFunc<ValueType> && set_value,\
+    DictionaryDefaultValueExtractor<AttributeType> & default_value_extractor) const;
+
+#define INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(AttributeType) \
+    INSTANTIATE_GET_ITEMS_IMPL(DictionaryKeyType::Simple, true, AttributeType, DictionaryValueType<AttributeType>) \
+    INSTANTIATE_GET_ITEMS_IMPL(DictionaryKeyType::Simple, false, AttributeType, DictionaryValueType<AttributeType>) \
+    INSTANTIATE_GET_ITEMS_IMPL(DictionaryKeyType::Complex, true, AttributeType, DictionaryValueType<AttributeType>) \
+    INSTANTIATE_GET_ITEMS_IMPL(DictionaryKeyType::Complex, false, AttributeType, DictionaryValueType<AttributeType>)
+
 namespace DB
 {
 
@@ -13,6 +26,7 @@ void RangeHashedDictionary<dictionary_key_type>::getItemsImpl(
 {
     const auto & attribute_container = std::get<AttributeContainerType<ValueType>>(attribute.container);
 
+
     size_t keys_found = 0;
 
     const ColumnPtr & range_column = key_columns.back();
@@ -116,20 +130,4 @@ void RangeHashedDictionary<dictionary_key_type>::getItemsImpl(
     query_count.fetch_add(keys_size, std::memory_order_relaxed);
     found_count.fetch_add(keys_found, std::memory_order_relaxed);
 }
-
-#define INSTANTIATE_GET_ITEMS_IMPL(DictionaryKeyType, IsNullable, AttributeType, ValueType) \
-template void RangeHashedDictionary<DictionaryKeyType>::getItemsImpl<ValueType, IsNullable, DictionaryDefaultValueExtractor<AttributeType>>( \
-    const Attribute & attribute,\
-    const Columns & key_columns,\
-    typename RangeHashedDictionary<DictionaryKeyType>::ValueSetterFunc<ValueType> && set_value,\
-    DictionaryDefaultValueExtractor<AttributeType> & default_value_extractor) const;
-
-#define INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(AttributeType) \
-    INSTANTIATE_GET_ITEMS_IMPL(DictionaryKeyType::Simple, true, AttributeType, DictionaryValueType<AttributeType>) \
-    INSTANTIATE_GET_ITEMS_IMPL(DictionaryKeyType::Simple, false, AttributeType, DictionaryValueType<AttributeType>) \
-    INSTANTIATE_GET_ITEMS_IMPL(DictionaryKeyType::Complex, true, AttributeType, DictionaryValueType<AttributeType>) \
-    INSTANTIATE_GET_ITEMS_IMPL(DictionaryKeyType::Complex, false, AttributeType, DictionaryValueType<AttributeType>)
-
-CALL_FOR_ALL_DICTIONARY_ATTRIBUTE_TYPES(INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE)
-
 }
diff --git a/src/Dictionaries/RangeHashedDictionaryGetItemsImplDecimal.cpp b/src/Dictionaries/RangeHashedDictionaryGetItemsImplDecimal.cpp
new file mode 100644
index 00000000000..f1ee4dd58e1
--- /dev/null
+++ b/src/Dictionaries/RangeHashedDictionaryGetItemsImplDecimal.cpp
@@ -0,0 +1,10 @@
+#include <Dictionaries/RangeHashedDictionaryGetItemsImpl.txx>
+
+namespace DB
+{
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(Decimal32);
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(Decimal64);
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(Decimal128);
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(Decimal256);
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(DateTime64);
+}
diff --git a/src/Dictionaries/RangeHashedDictionaryGetItemsImplFloat.cpp b/src/Dictionaries/RangeHashedDictionaryGetItemsImplFloat.cpp
new file mode 100644
index 00000000000..291a55a76db
--- /dev/null
+++ b/src/Dictionaries/RangeHashedDictionaryGetItemsImplFloat.cpp
@@ -0,0 +1,7 @@
+#include <Dictionaries/RangeHashedDictionaryGetItemsImpl.txx>
+
+namespace DB
+{
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(Float32);
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(Float64);
+}
diff --git a/src/Dictionaries/RangeHashedDictionaryGetItemsImplInt.cpp b/src/Dictionaries/RangeHashedDictionaryGetItemsImplInt.cpp
new file mode 100644
index 00000000000..a0748a9f486
--- /dev/null
+++ b/src/Dictionaries/RangeHashedDictionaryGetItemsImplInt.cpp
@@ -0,0 +1,11 @@
+#include <Dictionaries/RangeHashedDictionaryGetItemsImpl.txx>
+
+namespace DB
+{
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(Int8);
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(Int16);
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(Int32);
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(Int64);
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(Int128);
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(Int256);
+}
diff --git a/src/Dictionaries/RangeHashedDictionaryGetItemsImplOthers.cpp b/src/Dictionaries/RangeHashedDictionaryGetItemsImplOthers.cpp
new file mode 100644
index 00000000000..96e5bb54d0b
--- /dev/null
+++ b/src/Dictionaries/RangeHashedDictionaryGetItemsImplOthers.cpp
@@ -0,0 +1,10 @@
+#include <Dictionaries/RangeHashedDictionaryGetItemsImpl.txx>
+
+namespace DB
+{
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(UUID);
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(IPv4);
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(IPv6);
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(String);
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(Array);
+}
diff --git a/src/Dictionaries/RangeHashedDictionaryGetItemsImplUInt.cpp b/src/Dictionaries/RangeHashedDictionaryGetItemsImplUInt.cpp
new file mode 100644
index 00000000000..e60a7189a2d
--- /dev/null
+++ b/src/Dictionaries/RangeHashedDictionaryGetItemsImplUInt.cpp
@@ -0,0 +1,11 @@
+#include <Dictionaries/RangeHashedDictionaryGetItemsImpl.txx>
+
+namespace DB
+{
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(UInt8);
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(UInt16);
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(UInt32);
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(UInt64);
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(UInt128);
+INSTANTIATE_GET_ITEMS_IMPL_FOR_ATTRIBUTE_TYPE(UInt256);
+}
diff --git a/src/Dictionaries/RangeHashedDictionary_2.cpp b/src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImpl.txx
similarity index 98%
rename from src/Dictionaries/RangeHashedDictionary_2.cpp
rename to src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImpl.txx
index d400fd1d830..5807af519f9 100644
--- a/src/Dictionaries/RangeHashedDictionary_2.cpp
+++ b/src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImpl.txx
@@ -1,9 +1,21 @@
 #include <Dictionaries/RangeHashedDictionary.h>
 
+#define INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL(DictionaryKeyType, IsNullable, ValueType) \
+    template size_t RangeHashedDictionary<DictionaryKeyType>::getItemsShortCircuitImpl<ValueType, IsNullable>( \
+        const Attribute & attribute, \
+        const Columns & key_columns, \
+        typename RangeHashedDictionary<DictionaryKeyType>::ValueSetterFunc<ValueType> && set_value, \
+        IColumn::Filter & default_mask) const;
+
+#define INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(AttributeType) \
+    INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL(DictionaryKeyType::Simple, true, DictionaryValueType<AttributeType>) \
+    INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL(DictionaryKeyType::Simple, false, DictionaryValueType<AttributeType>) \
+    INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL(DictionaryKeyType::Complex, true, DictionaryValueType<AttributeType>) \
+    INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL(DictionaryKeyType::Complex, false, DictionaryValueType<AttributeType>)
+
 namespace DB
 {
 
-
 template <DictionaryKeyType dictionary_key_type>
 template <typename ValueType, bool is_nullable>
 size_t RangeHashedDictionary<dictionary_key_type>::getItemsShortCircuitImpl(
@@ -117,20 +129,4 @@ size_t RangeHashedDictionary<dictionary_key_type>::getItemsShortCircuitImpl(
     found_count.fetch_add(keys_found, std::memory_order_relaxed);
     return keys_found;
 }
-
-#define INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL(DictionaryKeyType, IsNullable, ValueType) \
-    template size_t RangeHashedDictionary<DictionaryKeyType>::getItemsShortCircuitImpl<ValueType, IsNullable>( \
-        const Attribute & attribute, \
-        const Columns & key_columns, \
-        typename RangeHashedDictionary<DictionaryKeyType>::ValueSetterFunc<ValueType> && set_value, \
-        IColumn::Filter & default_mask) const;
-
-#define INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(AttributeType) \
-    INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL(DictionaryKeyType::Simple, true, DictionaryValueType<AttributeType>) \
-    INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL(DictionaryKeyType::Simple, false, DictionaryValueType<AttributeType>) \
-    INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL(DictionaryKeyType::Complex, true, DictionaryValueType<AttributeType>) \
-    INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL(DictionaryKeyType::Complex, false, DictionaryValueType<AttributeType>)
-
-CALL_FOR_ALL_DICTIONARY_ATTRIBUTE_TYPES(INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE)
-
 }
diff --git a/src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImplDecimal.cpp b/src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImplDecimal.cpp
new file mode 100644
index 00000000000..298369e4735
--- /dev/null
+++ b/src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImplDecimal.cpp
@@ -0,0 +1,10 @@
+#include <Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImpl.txx>
+
+namespace DB
+{
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(Decimal32);
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(Decimal64);
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(Decimal128);
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(Decimal256);
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(DateTime64);
+}
diff --git a/src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImplFloat.cpp b/src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImplFloat.cpp
new file mode 100644
index 00000000000..e8e8da6c75e
--- /dev/null
+++ b/src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImplFloat.cpp
@@ -0,0 +1,7 @@
+#include <Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImpl.txx>
+
+namespace DB
+{
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(Float32);
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(Float64);
+}
diff --git a/src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImplInt.cpp b/src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImplInt.cpp
new file mode 100644
index 00000000000..c685b9b5331
--- /dev/null
+++ b/src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImplInt.cpp
@@ -0,0 +1,11 @@
+#include <Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImpl.txx>
+
+namespace DB
+{
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(Int8);
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(Int16);
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(Int32);
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(Int64);
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(Int128);
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(Int256);
+}
diff --git a/src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImplOthers.cpp b/src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImplOthers.cpp
new file mode 100644
index 00000000000..46ea141b59b
--- /dev/null
+++ b/src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImplOthers.cpp
@@ -0,0 +1,10 @@
+#include <Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImpl.txx>
+
+namespace DB
+{
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(UUID);
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(IPv4);
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(IPv6);
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(String);
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(Array);
+}
diff --git a/src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImplUInt.cpp b/src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImplUInt.cpp
new file mode 100644
index 00000000000..18421fd7e2d
--- /dev/null
+++ b/src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImplUInt.cpp
@@ -0,0 +1,11 @@
+#include <Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImpl.txx>
+
+namespace DB
+{
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(UInt8);
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(UInt16);
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(UInt32);
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(UInt64);
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(UInt128);
+INSTANTIATE_GET_ITEMS_SHORT_CIRCUIT_IMPL_FOR_ATTRIBUTE_TYPE(UInt256);
+}

From 7b79d92bbe4c4070135da9747be02edb8499c6a4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <benjamin.antal@clickhouse.com>
Date: Thu, 14 Mar 2024 17:49:23 +0000
Subject: [PATCH 353/443] Adjust large object check

---
 src/Dictionaries/RangeHashedDictionary.h | 1 -
 utils/check-style/check-large-objects.sh | 1 -
 2 files changed, 2 deletions(-)

diff --git a/src/Dictionaries/RangeHashedDictionary.h b/src/Dictionaries/RangeHashedDictionary.h
index 23f7df8133e..a5dedae97c4 100644
--- a/src/Dictionaries/RangeHashedDictionary.h
+++ b/src/Dictionaries/RangeHashedDictionary.h
@@ -292,7 +292,6 @@ private:
 extern template class RangeHashedDictionary<DictionaryKeyType::Simple>;
 extern template class RangeHashedDictionary<DictionaryKeyType::Complex>;
 
-
 namespace
 {
     template <typename F>
diff --git a/utils/check-style/check-large-objects.sh b/utils/check-style/check-large-objects.sh
index 5ef57ea4f6c..5b0e8e88df5 100755
--- a/utils/check-style/check-large-objects.sh
+++ b/utils/check-style/check-large-objects.sh
@@ -6,7 +6,6 @@ TU_EXCLUDES=(
     CastOverloadResolver
     AggregateFunctionUniq
     FunctionsConversion
-    RangeHashedDictionary_
 
     Aggregator
 )

From c6826145dbc6d21c90b3346a26c6c1c53323d138 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <benjamin.antal@clickhouse.com>
Date: Thu, 14 Mar 2024 17:59:58 +0000
Subject: [PATCH 354/443] Remove unnecessary empty lines

---
 src/Dictionaries/RangeHashedDictionary.h | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/src/Dictionaries/RangeHashedDictionary.h b/src/Dictionaries/RangeHashedDictionary.h
index a5dedae97c4..0469e82d7be 100644
--- a/src/Dictionaries/RangeHashedDictionary.h
+++ b/src/Dictionaries/RangeHashedDictionary.h
@@ -31,7 +31,6 @@
 #include <Functions/FunctionHelpers.h>
 #include <Interpreters/castColumn.h>
 
-
 namespace DB
 {
 
@@ -44,7 +43,6 @@ namespace ErrorCodes
     extern const int TYPE_MISMATCH;
 }
 
-
 enum class RangeHashedDictionaryLookupStrategy : uint8_t
 {
     min,
@@ -236,8 +234,6 @@ private:
 
     static Attribute createAttribute(const DictionaryAttribute & dictionary_attribute);
 
-
-
     template <typename ValueType>
     using ValueSetterFunc = std::function<void(size_t, const ValueType &, bool)>;
 

From e8da3bb2eb4ba0c54e948c1756c2d30e548d4432 Mon Sep 17 00:00:00 2001
From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com>
Date: Thu, 14 Mar 2024 19:06:02 +0100
Subject: [PATCH 355/443] Fix style

---
 .../integration/test_system_clusters_actual_information/test.py  | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tests/integration/test_system_clusters_actual_information/test.py b/tests/integration/test_system_clusters_actual_information/test.py
index c6e3262fd62..8b6436aeb5c 100644
--- a/tests/integration/test_system_clusters_actual_information/test.py
+++ b/tests/integration/test_system_clusters_actual_information/test.py
@@ -13,6 +13,7 @@ node = cluster.add_instance(
     "node", with_zookeeper=True, main_configs=["configs/remote_servers.xml"]
 )
 
+
 @pytest.fixture(scope="module")
 def started_cluster():
     try:

From 7e9d863c22e65ce34539670519de096a9f2261e6 Mon Sep 17 00:00:00 2001
From: Antonio Andelic <antonio@clickhouse.com>
Date: Thu, 14 Mar 2024 18:08:55 +0000
Subject: [PATCH 356/443] Update reference of the test

---
 .../0_stateless/03010_sum_to_to_count_if_nullable.reference     | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.reference b/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.reference
index e69de29bb2d..8627f639a03 100644
--- a/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.reference
+++ b/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.reference
@@ -0,0 +1,2 @@
+(5,NULL)
+(5,NULL)

From b65beba1fd439a1e98b028b85548873e8339335c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <benjamin.antal@clickhouse.com>
Date: Thu, 14 Mar 2024 18:13:06 +0000
Subject: [PATCH 357/443] Remove not used error code declaration

---
 src/Dictionaries/RangeHashedDictionary.cpp | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/src/Dictionaries/RangeHashedDictionary.cpp b/src/Dictionaries/RangeHashedDictionary.cpp
index 8299c3ad93a..30a0123ade6 100644
--- a/src/Dictionaries/RangeHashedDictionary.cpp
+++ b/src/Dictionaries/RangeHashedDictionary.cpp
@@ -3,11 +3,6 @@
 namespace DB
 {
 
-namespace ErrorCodes
-{
-    extern const int TYPE_MISMATCH;
-}
-
 template <DictionaryKeyType dictionary_key_type>
 ColumnPtr RangeHashedDictionary<dictionary_key_type>::getColumn(
     const std::string & attribute_name,

From b0008495294761e3394de0060dda61ee66808476 Mon Sep 17 00:00:00 2001
From: Antonio Andelic <antonio@clickhouse.com>
Date: Thu, 14 Mar 2024 18:15:12 +0000
Subject: [PATCH 358/443] Better

---
 src/Analyzer/Passes/SumIfToCountIfPass.cpp | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/src/Analyzer/Passes/SumIfToCountIfPass.cpp b/src/Analyzer/Passes/SumIfToCountIfPass.cpp
index d374d92c1fb..2c41b6dc467 100644
--- a/src/Analyzer/Passes/SumIfToCountIfPass.cpp
+++ b/src/Analyzer/Passes/SumIfToCountIfPass.cpp
@@ -32,7 +32,7 @@ public:
             return;
 
         auto * function_node = node->as<FunctionNode>();
-        if (!function_node || !function_node->isAggregateFunction() || function_node->getResultType()->isNullable())
+        if (!function_node || !function_node->isAggregateFunction())
             return;
 
         auto function_name = function_node->getFunctionName();
@@ -54,10 +54,10 @@ public:
             if (!constant_node)
                 return;
 
-            const auto & constant_value_literal = constant_node->getValue();
-            if (!isInt64OrUInt64FieldType(constant_value_literal.getType()))
+            if (auto constant_type = constant_node->getResultType(); !isUInt64(constant_type) && !isInt64(constant_type))
                 return;
 
+            const auto & constant_value_literal = constant_node->getValue();
             if (getSettings().aggregate_functions_null_for_empty)
                 return;
 

From e9f81170873bd06bf8c9875fa90654ec6fb0abc8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Thu, 14 Mar 2024 20:16:25 +0100
Subject: [PATCH 359/443] Increase memory limit for coverage builds

---
 CMakeLists.txt | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/CMakeLists.txt b/CMakeLists.txt
index 9ffb4789dc9..eff6dd3ff6a 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -56,13 +56,13 @@ option(ENABLE_CHECK_HEAVY_BUILDS "Don't allow C++ translation units to compile t
 if (ENABLE_CHECK_HEAVY_BUILDS)
     # set DATA (since RSS does not work since 2.6.x+) to 5G
     set (RLIMIT_DATA 5000000000)
-    # set VIRT (RLIMIT_AS) to 10G (DATA*10)
+    # set VIRT (RLIMIT_AS) to 10G (DATA*2)
     set (RLIMIT_AS 10000000000)
     # set CPU time limit to 1000 seconds
     set (RLIMIT_CPU 1000)
 
     # -fsanitize=memory and address are too heavy
-    if (SANITIZE)
+    if (SANITIZE OR SANITIZE_COVERAGE OR WITH_COVERAGE)
        set (RLIMIT_DATA 10000000000) # 10G
     endif()
 

From e8b3cc28518a409fbefe5206f676c2623d881484 Mon Sep 17 00:00:00 2001
From: Max Kainov <max.kainov@clickhouse.com>
Date: Thu, 14 Mar 2024 20:01:12 +0000
Subject: [PATCH 360/443] CI: skip hdfs tests for arm  #do_not_test  #batch_0 
 #no_merge_commit

 #ci_set_arm
---
 .../integration/test_allowed_url_from_config/test.py | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/tests/integration/test_allowed_url_from_config/test.py b/tests/integration/test_allowed_url_from_config/test.py
index 3106cf12702..fb7564ae9d3 100644
--- a/tests/integration/test_allowed_url_from_config/test.py
+++ b/tests/integration/test_allowed_url_from_config/test.py
@@ -1,3 +1,4 @@
+import platform
 import pytest
 from helpers.cluster import ClickHouseCluster
 
@@ -16,9 +17,11 @@ node5 = cluster.add_instance(
     "node5", main_configs=["configs/config_without_allowed_hosts.xml"]
 )
 node6 = cluster.add_instance("node6", main_configs=["configs/config_for_remote.xml"])
-node7 = cluster.add_instance(
-    "node7", main_configs=["configs/config_for_redirect.xml"], with_hdfs=True
-)
+
+if platform.processor() != "arm":
+    node7 = cluster.add_instance(
+        "node7", main_configs=["configs/config_for_redirect.xml"], with_hdfs=True
+    )
 
 
 @pytest.fixture(scope="module")
@@ -270,6 +273,7 @@ def test_table_function_remote(start_cluster):
     )
 
 
+@pytest.mark.skipif(platform.processor() == "arm", reason="skip for ARM")
 def test_redirect(start_cluster):
     hdfs_api = start_cluster.hdfs_api
 
@@ -284,6 +288,7 @@ def test_redirect(start_cluster):
     node7.query("DROP TABLE table_test_7_1")
 
 
+@pytest.mark.skipif(platform.processor() == "arm", reason="skip for ARM")
 def test_HDFS(start_cluster):
     assert "not allowed" in node7.query_and_get_error(
         "CREATE TABLE table_test_7_2 (word String) ENGINE=HDFS('http://hdfs1:50075/webhdfs/v1/simple_storage?op=OPEN&namenoderpcaddress=hdfs1:9000&offset=0', 'CSV')"
@@ -293,6 +298,7 @@ def test_HDFS(start_cluster):
     )
 
 
+@pytest.mark.skipif(platform.processor() == "arm", reason="skip for ARM")
 def test_schema_inference(start_cluster):
     error = node7.query_and_get_error("desc url('http://test.com`, 'TSVRaw'')")
     assert error.find("ReadWriteBufferFromHTTPBase") == -1

From e08eaebc9946d39d19b4e995cfdc962719c55c44 Mon Sep 17 00:00:00 2001
From: avogar <avogar@clickhouse.com>
Date: Thu, 14 Mar 2024 20:36:18 +0000
Subject: [PATCH 361/443] Add sanity check for poll_max_batch_size FileLog
 setting to avoid big untracked allocations

---
 src/Storages/FileLog/FileLogSettings.cpp                    | 6 ++++++
 .../03010_file_log_large_poll_batch_size.reference          | 0
 .../0_stateless/03010_file_log_large_poll_batch_size.sql    | 2 ++
 3 files changed, 8 insertions(+)
 create mode 100644 tests/queries/0_stateless/03010_file_log_large_poll_batch_size.reference
 create mode 100644 tests/queries/0_stateless/03010_file_log_large_poll_batch_size.sql

diff --git a/src/Storages/FileLog/FileLogSettings.cpp b/src/Storages/FileLog/FileLogSettings.cpp
index 2cd42c35870..8e245285b9a 100644
--- a/src/Storages/FileLog/FileLogSettings.cpp
+++ b/src/Storages/FileLog/FileLogSettings.cpp
@@ -11,6 +11,7 @@ namespace DB
 namespace ErrorCodes
 {
     extern const int UNKNOWN_SETTING;
+    extern const int INVALID_SETTING_VALUE;
 }
 
 IMPLEMENT_SETTINGS_TRAITS(FileLogSettingsTraits, LIST_OF_FILELOG_SETTINGS)
@@ -36,6 +37,11 @@ void FileLogSettings::loadFromQuery(ASTStorage & storage_def)
         settings_ast->is_standalone = false;
         storage_def.set(storage_def.settings, settings_ast);
     }
+
+    /// Check that batch size is not too high (the same as we check setting max_block_size).
+    constexpr UInt64 max_sane_block_rows_size = 4294967296; // 2^32
+    if (poll_max_batch_size > max_sane_block_rows_size)
+        throw Exception(ErrorCodes::INVALID_SETTING_VALUE, "Sanity check: 'poll_max_batch_size' value is too high ({})", poll_max_batch_size);
 }
 
 }
diff --git a/tests/queries/0_stateless/03010_file_log_large_poll_batch_size.reference b/tests/queries/0_stateless/03010_file_log_large_poll_batch_size.reference
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/queries/0_stateless/03010_file_log_large_poll_batch_size.sql b/tests/queries/0_stateless/03010_file_log_large_poll_batch_size.sql
new file mode 100644
index 00000000000..2663011f2ec
--- /dev/null
+++ b/tests/queries/0_stateless/03010_file_log_large_poll_batch_size.sql
@@ -0,0 +1,2 @@
+create table test (number UInt64) engine=FileLog('./user_files/data.jsonl', 'JSONEachRow') settings poll_max_batch_size=18446744073709; -- {serverError INVALID_SETTING_VALUE}
+

From 1cae77997a648414d2e16e806c322fae2f2da301 Mon Sep 17 00:00:00 2001
From: Nikolai Kochetov <nk@clickhouse.com>
Date: Thu, 14 Mar 2024 21:34:25 +0000
Subject: [PATCH 362/443] Try revert stage back

---
 src/Storages/StorageMerge.cpp | 8 +++++---
 src/Storages/StorageMerge.h   | 3 ++-
 2 files changed, 7 insertions(+), 4 deletions(-)

diff --git a/src/Storages/StorageMerge.cpp b/src/Storages/StorageMerge.cpp
index f95464f4bff..52362eb5cb8 100644
--- a/src/Storages/StorageMerge.cpp
+++ b/src/Storages/StorageMerge.cpp
@@ -1121,7 +1121,7 @@ QueryPipelineBuilderPtr ReadFromMerge::createSources(
         /// Subordinary tables could have different but convertible types, like numeric types of different width.
         /// We must return streams with structure equals to structure of Merge table.
         convertAndFilterSourceStream(
-            header, modified_query_info, storage_snapshot_, aliases, row_policy_data_opt, context, *builder);
+            header, modified_query_info, storage_snapshot_, aliases, row_policy_data_opt, context, *builder, processed_stage);
     }
 
     return builder;
@@ -1473,7 +1473,8 @@ void ReadFromMerge::convertAndFilterSourceStream(
     const Aliases & aliases,
     const RowPolicyDataOpt & row_policy_data_opt,
     ContextPtr local_context,
-    QueryPipelineBuilder & builder)
+    QueryPipelineBuilder & builder,
+    QueryProcessingStage::Enum processed_stage)
 {
     Block before_block_header = builder.getHeader();
 
@@ -1532,7 +1533,8 @@ void ReadFromMerge::convertAndFilterSourceStream(
 
     ActionsDAG::MatchColumnsMode convert_actions_match_columns_mode = ActionsDAG::MatchColumnsMode::Name;
 
-    if (local_context->getSettingsRef().allow_experimental_analyzer && dynamic_cast<const StorageDistributed *>(&snapshot->storage) != nullptr)
+    if (local_context->getSettingsRef().allow_experimental_analyzer
+        && (processed_stage != QueryProcessingStage::FetchColumns || dynamic_cast<const StorageDistributed *>(&snapshot->storage) != nullptr))
         convert_actions_match_columns_mode = ActionsDAG::MatchColumnsMode::Position;
 
     if (row_policy_data_opt)
diff --git a/src/Storages/StorageMerge.h b/src/Storages/StorageMerge.h
index 3aabd7e26e3..556649f622d 100644
--- a/src/Storages/StorageMerge.h
+++ b/src/Storages/StorageMerge.h
@@ -277,7 +277,8 @@ private:
         const Aliases & aliases,
         const RowPolicyDataOpt & row_policy_data_opt,
         ContextPtr context,
-        QueryPipelineBuilder & builder);
+        QueryPipelineBuilder & builder,
+        QueryProcessingStage::Enum processed_stage);
 
     StorageMerge::StorageListWithLocks getSelectedTables(
         ContextPtr query_context,

From e418f249b8a08484b10bbd5a60de5718686de2f3 Mon Sep 17 00:00:00 2001
From: Anton Popov <anton@clickhouse.com>
Date: Fri, 15 Mar 2024 00:40:28 +0000
Subject: [PATCH 363/443] fix test

---
 tests/queries/0_stateless/03008_optimize_equal_ranges.sql | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tests/queries/0_stateless/03008_optimize_equal_ranges.sql b/tests/queries/0_stateless/03008_optimize_equal_ranges.sql
index 4d521420741..6d769c7382a 100644
--- a/tests/queries/0_stateless/03008_optimize_equal_ranges.sql
+++ b/tests/queries/0_stateless/03008_optimize_equal_ranges.sql
@@ -5,6 +5,7 @@ CREATE TABLE t_optimize_equal_ranges (a UInt64, b String, c UInt64) ENGINE = Mer
 SET max_block_size = 1024;
 SET max_bytes_before_external_group_by = 0;
 SET optimize_aggregation_in_order = 0;
+SET optimize_use_projections = 0;
 
 INSERT INTO t_optimize_equal_ranges SELECT 0, toString(number), number FROM numbers(30000);
 INSERT INTO t_optimize_equal_ranges SELECT 1, toString(number), number FROM numbers(30000);

From 978fb783517639f495fbfbc67fe721cb8c8b2f5d Mon Sep 17 00:00:00 2001
From: peter279k <peter279k@gmail.com>
Date: Fri, 15 Mar 2024 11:18:21 +0800
Subject: [PATCH 364/443] Correct Criteo example dataset instruction section

---
 docs/en/getting-started/example-datasets/criteo.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/en/getting-started/example-datasets/criteo.md b/docs/en/getting-started/example-datasets/criteo.md
index a2e0fda0cb0..4becdb50731 100644
--- a/docs/en/getting-started/example-datasets/criteo.md
+++ b/docs/en/getting-started/example-datasets/criteo.md
@@ -55,7 +55,7 @@ CREATE TABLE criteo_log (
 ) ENGINE = Log;
 ```
 
-Download the data:
+Insert the data:
 
 ``` bash
 $ for i in {00..23}; do echo $i; zcat datasets/criteo/day_${i#0}.gz | sed -r 's/^/2000-01-'${i/00/24}'\t/' | clickhouse-client --host=example-perftest01j --query="INSERT INTO criteo_log FORMAT TabSeparated"; done

From a3da67ba9ae17440d4809e76ad19d2e308469e44 Mon Sep 17 00:00:00 2001
From: robot-clickhouse <robot-clickhouse@users.noreply.github.com>
Date: Fri, 15 Mar 2024 06:47:10 +0000
Subject: [PATCH 365/443] Update version_date.tsv and changelogs after
 v23.8.11.28-lts

---
 docs/changelogs/v23.8.11.28-lts.md   | 30 ++++++++++++++++++++++++++++
 utils/list-versions/version_date.tsv |  4 ++++
 2 files changed, 34 insertions(+)
 create mode 100644 docs/changelogs/v23.8.11.28-lts.md

diff --git a/docs/changelogs/v23.8.11.28-lts.md b/docs/changelogs/v23.8.11.28-lts.md
new file mode 100644
index 00000000000..acc284caa72
--- /dev/null
+++ b/docs/changelogs/v23.8.11.28-lts.md
@@ -0,0 +1,30 @@
+---
+sidebar_position: 1
+sidebar_label: 2024
+---
+
+# 2024 Changelog
+
+### ClickHouse release v23.8.11.28-lts (31879d2ab4c) FIXME as compared to v23.8.10.43-lts (a278225bba9)
+
+#### Improvement
+* Backported in [#60828](https://github.com/ClickHouse/ClickHouse/issues/60828): Update tzdata to 2024a. [#60768](https://github.com/ClickHouse/ClickHouse/pull/60768) ([Raúl Marín](https://github.com/Algunenano)).
+
+#### Bug Fix (user-visible misbehavior in an official stable release)
+
+* Fix buffer overflow in CompressionCodecMultiple [#60731](https://github.com/ClickHouse/ClickHouse/pull/60731) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
+* Remove nonsense from SQL/JSON [#60738](https://github.com/ClickHouse/ClickHouse/pull/60738) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
+* Fix crash in arrayEnumerateRanked [#60764](https://github.com/ClickHouse/ClickHouse/pull/60764) ([Raúl Marín](https://github.com/Algunenano)).
+* Fix crash when using input() in INSERT SELECT JOIN [#60765](https://github.com/ClickHouse/ClickHouse/pull/60765) ([Kruglov Pavel](https://github.com/Avogar)).
+* Remove recursion when reading from S3 [#60849](https://github.com/ClickHouse/ClickHouse/pull/60849) ([Antonio Andelic](https://github.com/antonio2368)).
+
+#### NO CL ENTRY
+
+* NO CL ENTRY:  'Use the current branch test-utils to build cctools'. [#61276](https://github.com/ClickHouse/ClickHouse/pull/61276) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
+
+#### NOT FOR CHANGELOG / INSIGNIFICANT
+
+* Cancel PipelineExecutor properly in case of exception in spawnThreads [#57104](https://github.com/ClickHouse/ClickHouse/pull/57104) ([Kruglov Pavel](https://github.com/Avogar)).
+* Detect io_uring in tests [#60373](https://github.com/ClickHouse/ClickHouse/pull/60373) ([Azat Khuzhin](https://github.com/azat)).
+* Cancel PipelineExecutor properly in case of exception in spawnThreads [#60499](https://github.com/ClickHouse/ClickHouse/pull/60499) ([Kruglov Pavel](https://github.com/Avogar)).
+
diff --git a/utils/list-versions/version_date.tsv b/utils/list-versions/version_date.tsv
index e372e407ce1..ad7c92d85d5 100644
--- a/utils/list-versions/version_date.tsv
+++ b/utils/list-versions/version_date.tsv
@@ -1,9 +1,11 @@
 v24.2.1.2248-stable	2024-02-29
+v24.1.6.52-stable	2024-03-07
 v24.1.5.6-stable	2024-02-14
 v24.1.4.20-stable	2024-02-14
 v24.1.3.31-stable	2024-02-09
 v24.1.2.5-stable	2024-02-02
 v24.1.1.2048-stable	2024-01-30
+v23.12.5.81-stable	2024-03-15
 v23.12.4.15-stable	2024-02-09
 v23.12.3.40-stable	2024-02-02
 v23.12.2.59-stable	2024-01-05
@@ -25,6 +27,7 @@ v23.9.4.11-stable	2023-11-08
 v23.9.3.12-stable	2023-10-31
 v23.9.2.56-stable	2023-10-19
 v23.9.1.1854-stable	2023-09-29
+v23.8.11.28-lts	2024-03-15
 v23.8.10.43-lts	2024-03-05
 v23.8.9.54-lts	2024-01-05
 v23.8.8.20-lts	2023-11-25
@@ -55,6 +58,7 @@ v23.4.4.16-stable	2023-06-17
 v23.4.3.48-stable	2023-06-12
 v23.4.2.11-stable	2023-05-02
 v23.4.1.1943-stable	2023-04-27
+v23.3.21.26-lts	2024-03-15
 v23.3.20.27-lts	2024-03-05
 v23.3.19.32-lts	2024-01-05
 v23.3.18.15-lts	2023-11-25

From e6f2cd080e2bc8b0fb47287c6749fe6bc45a14e7 Mon Sep 17 00:00:00 2001
From: robot-clickhouse <robot-clickhouse@users.noreply.github.com>
Date: Fri, 15 Mar 2024 06:47:48 +0000
Subject: [PATCH 366/443] Update version_date.tsv and changelogs after
 v23.3.21.26-lts

---
 docs/changelogs/v23.3.21.26-lts.md   | 24 ++++++++++++++++++++++++
 utils/list-versions/version_date.tsv |  2 ++
 2 files changed, 26 insertions(+)
 create mode 100644 docs/changelogs/v23.3.21.26-lts.md

diff --git a/docs/changelogs/v23.3.21.26-lts.md b/docs/changelogs/v23.3.21.26-lts.md
new file mode 100644
index 00000000000..b0f059c4907
--- /dev/null
+++ b/docs/changelogs/v23.3.21.26-lts.md
@@ -0,0 +1,24 @@
+---
+sidebar_position: 1
+sidebar_label: 2024
+---
+
+# 2024 Changelog
+
+### ClickHouse release v23.3.21.26-lts (d9672a3731f) FIXME as compared to v23.3.20.27-lts (cc974ba4f81)
+
+#### Bug Fix (user-visible misbehavior in an official stable release)
+
+* Fix reading from sparse columns after restart [#49660](https://github.com/ClickHouse/ClickHouse/pull/49660) ([Anton Popov](https://github.com/CurtizJ)).
+* Fix buffer overflow in CompressionCodecMultiple [#60731](https://github.com/ClickHouse/ClickHouse/pull/60731) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
+* Remove nonsense from SQL/JSON [#60738](https://github.com/ClickHouse/ClickHouse/pull/60738) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
+* Fix crash in arrayEnumerateRanked [#60764](https://github.com/ClickHouse/ClickHouse/pull/60764) ([Raúl Marín](https://github.com/Algunenano)).
+* Fix crash when using input() in INSERT SELECT JOIN [#60765](https://github.com/ClickHouse/ClickHouse/pull/60765) ([Kruglov Pavel](https://github.com/Avogar)).
+* Remove recursion when reading from S3 [#60849](https://github.com/ClickHouse/ClickHouse/pull/60849) ([Antonio Andelic](https://github.com/antonio2368)).
+
+#### NOT FOR CHANGELOG / INSIGNIFICANT
+
+* Cancel PipelineExecutor properly in case of exception in spawnThreads [#57104](https://github.com/ClickHouse/ClickHouse/pull/57104) ([Kruglov Pavel](https://github.com/Avogar)).
+* Detect io_uring in tests [#60373](https://github.com/ClickHouse/ClickHouse/pull/60373) ([Azat Khuzhin](https://github.com/azat)).
+* Cancel PipelineExecutor properly in case of exception in spawnThreads [#60499](https://github.com/ClickHouse/ClickHouse/pull/60499) ([Kruglov Pavel](https://github.com/Avogar)).
+
diff --git a/utils/list-versions/version_date.tsv b/utils/list-versions/version_date.tsv
index e372e407ce1..4ed577e0b61 100644
--- a/utils/list-versions/version_date.tsv
+++ b/utils/list-versions/version_date.tsv
@@ -1,4 +1,5 @@
 v24.2.1.2248-stable	2024-02-29
+v24.1.6.52-stable	2024-03-07
 v24.1.5.6-stable	2024-02-14
 v24.1.4.20-stable	2024-02-14
 v24.1.3.31-stable	2024-02-09
@@ -55,6 +56,7 @@ v23.4.4.16-stable	2023-06-17
 v23.4.3.48-stable	2023-06-12
 v23.4.2.11-stable	2023-05-02
 v23.4.1.1943-stable	2023-04-27
+v23.3.21.26-lts	2024-03-15
 v23.3.20.27-lts	2024-03-05
 v23.3.19.32-lts	2024-01-05
 v23.3.18.15-lts	2023-11-25

From 4e34887ab4d1bfe2355327da452d66d376dc6dae Mon Sep 17 00:00:00 2001
From: robot-clickhouse <robot-clickhouse@users.noreply.github.com>
Date: Fri, 15 Mar 2024 06:48:11 +0000
Subject: [PATCH 367/443] Update version_date.tsv and changelogs after
 v24.1.7.18-stable

---
 docker/keeper/Dockerfile             |  2 +-
 docker/server/Dockerfile.alpine      |  2 +-
 docker/server/Dockerfile.ubuntu      |  2 +-
 docs/changelogs/v24.1.7.18-stable.md | 26 ++++++++++++++++++++++++++
 utils/list-versions/version_date.tsv |  6 ++++++
 5 files changed, 35 insertions(+), 3 deletions(-)
 create mode 100644 docs/changelogs/v24.1.7.18-stable.md

diff --git a/docker/keeper/Dockerfile b/docker/keeper/Dockerfile
index 2f42854a972..17eee6d4287 100644
--- a/docker/keeper/Dockerfile
+++ b/docker/keeper/Dockerfile
@@ -34,7 +34,7 @@ RUN arch=${TARGETARCH:-amd64} \
 # lts / testing / prestable / etc
 ARG REPO_CHANNEL="stable"
 ARG REPOSITORY="https://packages.clickhouse.com/tgz/${REPO_CHANNEL}"
-ARG VERSION="24.2.1.2248"
+ARG VERSION="24.2.2.71"
 ARG PACKAGES="clickhouse-keeper"
 ARG DIRECT_DOWNLOAD_URLS=""
 
diff --git a/docker/server/Dockerfile.alpine b/docker/server/Dockerfile.alpine
index 7bd777de5b9..bd5fa313adc 100644
--- a/docker/server/Dockerfile.alpine
+++ b/docker/server/Dockerfile.alpine
@@ -32,7 +32,7 @@ RUN arch=${TARGETARCH:-amd64} \
 # lts / testing / prestable / etc
 ARG REPO_CHANNEL="stable"
 ARG REPOSITORY="https://packages.clickhouse.com/tgz/${REPO_CHANNEL}"
-ARG VERSION="24.2.1.2248"
+ARG VERSION="24.2.2.71"
 ARG PACKAGES="clickhouse-client clickhouse-server clickhouse-common-static"
 ARG DIRECT_DOWNLOAD_URLS=""
 
diff --git a/docker/server/Dockerfile.ubuntu b/docker/server/Dockerfile.ubuntu
index 03d01cfd5d7..256dcdc029f 100644
--- a/docker/server/Dockerfile.ubuntu
+++ b/docker/server/Dockerfile.ubuntu
@@ -27,7 +27,7 @@ RUN sed -i "s|http://archive.ubuntu.com|${apt_archive}|g" /etc/apt/sources.list
 
 ARG REPO_CHANNEL="stable"
 ARG REPOSITORY="deb [signed-by=/usr/share/keyrings/clickhouse-keyring.gpg] https://packages.clickhouse.com/deb ${REPO_CHANNEL} main"
-ARG VERSION="24.2.1.2248"
+ARG VERSION="24.2.2.71"
 ARG PACKAGES="clickhouse-client clickhouse-server clickhouse-common-static"
 
 # set non-empty deb_location_url url to create a docker image
diff --git a/docs/changelogs/v24.1.7.18-stable.md b/docs/changelogs/v24.1.7.18-stable.md
new file mode 100644
index 00000000000..603a83a67be
--- /dev/null
+++ b/docs/changelogs/v24.1.7.18-stable.md
@@ -0,0 +1,26 @@
+---
+sidebar_position: 1
+sidebar_label: 2024
+---
+
+# 2024 Changelog
+
+### ClickHouse release v24.1.7.18-stable (90925babd78) FIXME as compared to v24.1.6.52-stable (fa09f677bc9)
+
+#### Bug Fix (user-visible misbehavior in an official stable release)
+
+* Fix deadlock in parallel parsing when lots of rows are skipped due to errors [#60516](https://github.com/ClickHouse/ClickHouse/pull/60516) ([Kruglov Pavel](https://github.com/Avogar)).
+* Fix_max_query_size_for_kql_compound_operator: [#60534](https://github.com/ClickHouse/ClickHouse/pull/60534) ([Yong Wang](https://github.com/kashwy)).
+* Fix crash with different allow_experimental_analyzer value in subqueries [#60770](https://github.com/ClickHouse/ClickHouse/pull/60770) ([Dmitry Novik](https://github.com/novikd)).
+* Fix Keeper reconfig for standalone binary [#61233](https://github.com/ClickHouse/ClickHouse/pull/61233) ([Antonio Andelic](https://github.com/antonio2368)).
+
+#### CI Fix or Improvement (changelog entry is not required)
+
+* Backported in [#61043](https://github.com/ClickHouse/ClickHouse/issues/61043): Debug and fix markreleaseready. [#60611](https://github.com/ClickHouse/ClickHouse/pull/60611) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
+* Backported in [#61168](https://github.com/ClickHouse/ClickHouse/issues/61168): Just a preparation for the merge queue support. [#61099](https://github.com/ClickHouse/ClickHouse/pull/61099) ([Max K.](https://github.com/maxknv)).
+* Backported in [#61192](https://github.com/ClickHouse/ClickHouse/issues/61192): ... [#61185](https://github.com/ClickHouse/ClickHouse/pull/61185) ([Max K.](https://github.com/maxknv)).
+
+#### NOT FOR CHANGELOG / INSIGNIFICANT
+
+* Cancel PipelineExecutor properly in case of exception in spawnThreads [#60499](https://github.com/ClickHouse/ClickHouse/pull/60499) ([Kruglov Pavel](https://github.com/Avogar)).
+
diff --git a/utils/list-versions/version_date.tsv b/utils/list-versions/version_date.tsv
index e372e407ce1..32fbfee8274 100644
--- a/utils/list-versions/version_date.tsv
+++ b/utils/list-versions/version_date.tsv
@@ -1,9 +1,13 @@
+v24.2.2.71-stable	2024-03-15
 v24.2.1.2248-stable	2024-02-29
+v24.1.7.18-stable	2024-03-15
+v24.1.6.52-stable	2024-03-07
 v24.1.5.6-stable	2024-02-14
 v24.1.4.20-stable	2024-02-14
 v24.1.3.31-stable	2024-02-09
 v24.1.2.5-stable	2024-02-02
 v24.1.1.2048-stable	2024-01-30
+v23.12.5.81-stable	2024-03-15
 v23.12.4.15-stable	2024-02-09
 v23.12.3.40-stable	2024-02-02
 v23.12.2.59-stable	2024-01-05
@@ -25,6 +29,7 @@ v23.9.4.11-stable	2023-11-08
 v23.9.3.12-stable	2023-10-31
 v23.9.2.56-stable	2023-10-19
 v23.9.1.1854-stable	2023-09-29
+v23.8.11.28-lts	2024-03-15
 v23.8.10.43-lts	2024-03-05
 v23.8.9.54-lts	2024-01-05
 v23.8.8.20-lts	2023-11-25
@@ -55,6 +60,7 @@ v23.4.4.16-stable	2023-06-17
 v23.4.3.48-stable	2023-06-12
 v23.4.2.11-stable	2023-05-02
 v23.4.1.1943-stable	2023-04-27
+v23.3.21.26-lts	2024-03-15
 v23.3.20.27-lts	2024-03-05
 v23.3.19.32-lts	2024-01-05
 v23.3.18.15-lts	2023-11-25

From 5a56c439395c3864d0f95d7b3b8ef63ed069fab6 Mon Sep 17 00:00:00 2001
From: Pham Anh Tuan <tuanpa.vn.1991@gmail.com>
Date: Wed, 13 Mar 2024 15:06:11 +0000
Subject: [PATCH 368/443] add async metrics for virtual memory mappings

---
 src/Common/AsynchronousMetrics.cpp            | 53 +++++++++++++++++++
 src/Common/AsynchronousMetrics.h              |  3 ++
 ...ry_mappings_asynchronous_metrics.reference |  2 +
 ...l_memory_mappings_asynchronous_metrics.sql |  4 ++
 4 files changed, 62 insertions(+)
 create mode 100644 tests/queries/0_stateless/03010_virtual_memory_mappings_asynchronous_metrics.reference
 create mode 100644 tests/queries/0_stateless/03010_virtual_memory_mappings_asynchronous_metrics.sql

diff --git a/src/Common/AsynchronousMetrics.cpp b/src/Common/AsynchronousMetrics.cpp
index b24d9bcc301..0b9be18c84e 100644
--- a/src/Common/AsynchronousMetrics.cpp
+++ b/src/Common/AsynchronousMetrics.cpp
@@ -9,6 +9,7 @@
 #include <IO/MMappedFileCache.h>
 #include <IO/ReadHelpers.h>
 #include <base/errnoToString.h>
+#include <base/find_symbols.h>
 #include <base/getPageSize.h>
 #include <sys/resource.h>
 #include <chrono>
@@ -90,6 +91,9 @@ AsynchronousMetrics::AsynchronousMetrics(
         openFileIfExists("/sys/fs/cgroup/cpu/cpu.cfs_quota_us", cgroupcpu_cfs_quota);
     }
 
+    openFileIfExists("/proc/sys/vm/max_map_count", vm_max_map_count);
+    openFileIfExists("/proc/self/maps", vm_maps);
+
     openSensors();
     openBlockDevices();
     openEDAC();
@@ -1423,6 +1427,55 @@ void AsynchronousMetrics::update(TimePoint update_time, bool force_update)
         }
     }
 
+    if (vm_max_map_count)
+    {
+        try
+        {
+            vm_max_map_count->rewind();
+
+            uint64_t max_map_count = 0;
+            readText(max_map_count, *vm_max_map_count);
+            new_values["VMMaxMapCount"] = { max_map_count, "The maximum number of memory mappings a process may have (/proc/sys/vm/max_map_count)."};
+        }
+        catch (...)
+        {
+            tryLogCurrentException(__PRETTY_FUNCTION__);
+            openFileIfExists("/proc/sys/vm/max_map_count", vm_max_map_count);
+        }
+    }
+
+    if (vm_maps)
+    {
+        try
+        {
+            vm_maps->rewind();
+
+            uint64_t num_maps = 0;
+            while (!vm_maps->eof())
+            {
+                char * next_pos = find_first_symbols<'\n'>(vm_maps->position(), vm_maps->buffer().end());
+                vm_maps->position() = next_pos;
+
+                if (!vm_maps->hasPendingData())
+                    continue;
+
+                if (*vm_maps->position() == '\n')
+                {
+                    ++num_maps;
+                    ++vm_maps->position();
+                }
+            }
+            new_values["VMNumMaps"] = { num_maps,
+                "The current number of memory mappings of the process (/proc/self/maps)."
+                " If it is close to the maximum (VMMaxMapCount), you should increase the limit for vm.max_map_count in /etc/sysctl.conf"};
+        }
+        catch (...)
+        {
+            tryLogCurrentException(__PRETTY_FUNCTION__);
+            openFileIfExists("/proc/self/maps", vm_maps);
+        }
+    }
+
     try
     {
         for (size_t i = 0, size = thermal.size(); i < size; ++i)
diff --git a/src/Common/AsynchronousMetrics.h b/src/Common/AsynchronousMetrics.h
index 305e8136b8a..4b3d28e80c5 100644
--- a/src/Common/AsynchronousMetrics.h
+++ b/src/Common/AsynchronousMetrics.h
@@ -123,6 +123,9 @@ private:
     std::optional<ReadBufferFromFilePRead> cgroupcpu_cfs_quota TSA_GUARDED_BY(data_mutex);
     std::optional<ReadBufferFromFilePRead> cgroupcpu_max TSA_GUARDED_BY(data_mutex);
 
+    std::optional<ReadBufferFromFilePRead> vm_max_map_count TSA_GUARDED_BY(data_mutex);
+    std::optional<ReadBufferFromFilePRead> vm_maps TSA_GUARDED_BY(data_mutex);
+
     std::vector<std::unique_ptr<ReadBufferFromFilePRead>> thermal TSA_GUARDED_BY(data_mutex);
 
     std::unordered_map<String /* device name */,
diff --git a/tests/queries/0_stateless/03010_virtual_memory_mappings_asynchronous_metrics.reference b/tests/queries/0_stateless/03010_virtual_memory_mappings_asynchronous_metrics.reference
new file mode 100644
index 00000000000..aa47d0d46d4
--- /dev/null
+++ b/tests/queries/0_stateless/03010_virtual_memory_mappings_asynchronous_metrics.reference
@@ -0,0 +1,2 @@
+0
+0
diff --git a/tests/queries/0_stateless/03010_virtual_memory_mappings_asynchronous_metrics.sql b/tests/queries/0_stateless/03010_virtual_memory_mappings_asynchronous_metrics.sql
new file mode 100644
index 00000000000..e9aced83ded
--- /dev/null
+++ b/tests/queries/0_stateless/03010_virtual_memory_mappings_asynchronous_metrics.sql
@@ -0,0 +1,4 @@
+-- Tags: no-replicated-database
+
+SELECT least(value, 0) FROM system.asynchronous_metrics WHERE metric = 'VMMaxMapCount';
+SELECT least(value, 0) FROM system.asynchronous_metrics WHERE metric = 'VMNumMaps';

From e5a1c0e251e9da5b4e5f1d1b1646766a7c75ecdf Mon Sep 17 00:00:00 2001
From: robot-clickhouse <robot-clickhouse@users.noreply.github.com>
Date: Fri, 15 Mar 2024 06:50:44 +0000
Subject: [PATCH 369/443] Update version_date.tsv and changelogs after
 v24.2.2.71-stable

---
 docker/keeper/Dockerfile             |  2 +-
 docker/server/Dockerfile.alpine      |  2 +-
 docker/server/Dockerfile.ubuntu      |  2 +-
 docs/changelogs/v24.2.2.71-stable.md | 55 ++++++++++++++++++++++++++++
 utils/list-versions/version_date.tsv |  6 +++
 5 files changed, 64 insertions(+), 3 deletions(-)
 create mode 100644 docs/changelogs/v24.2.2.71-stable.md

diff --git a/docker/keeper/Dockerfile b/docker/keeper/Dockerfile
index 2f42854a972..17eee6d4287 100644
--- a/docker/keeper/Dockerfile
+++ b/docker/keeper/Dockerfile
@@ -34,7 +34,7 @@ RUN arch=${TARGETARCH:-amd64} \
 # lts / testing / prestable / etc
 ARG REPO_CHANNEL="stable"
 ARG REPOSITORY="https://packages.clickhouse.com/tgz/${REPO_CHANNEL}"
-ARG VERSION="24.2.1.2248"
+ARG VERSION="24.2.2.71"
 ARG PACKAGES="clickhouse-keeper"
 ARG DIRECT_DOWNLOAD_URLS=""
 
diff --git a/docker/server/Dockerfile.alpine b/docker/server/Dockerfile.alpine
index 7bd777de5b9..bd5fa313adc 100644
--- a/docker/server/Dockerfile.alpine
+++ b/docker/server/Dockerfile.alpine
@@ -32,7 +32,7 @@ RUN arch=${TARGETARCH:-amd64} \
 # lts / testing / prestable / etc
 ARG REPO_CHANNEL="stable"
 ARG REPOSITORY="https://packages.clickhouse.com/tgz/${REPO_CHANNEL}"
-ARG VERSION="24.2.1.2248"
+ARG VERSION="24.2.2.71"
 ARG PACKAGES="clickhouse-client clickhouse-server clickhouse-common-static"
 ARG DIRECT_DOWNLOAD_URLS=""
 
diff --git a/docker/server/Dockerfile.ubuntu b/docker/server/Dockerfile.ubuntu
index 03d01cfd5d7..256dcdc029f 100644
--- a/docker/server/Dockerfile.ubuntu
+++ b/docker/server/Dockerfile.ubuntu
@@ -27,7 +27,7 @@ RUN sed -i "s|http://archive.ubuntu.com|${apt_archive}|g" /etc/apt/sources.list
 
 ARG REPO_CHANNEL="stable"
 ARG REPOSITORY="deb [signed-by=/usr/share/keyrings/clickhouse-keyring.gpg] https://packages.clickhouse.com/deb ${REPO_CHANNEL} main"
-ARG VERSION="24.2.1.2248"
+ARG VERSION="24.2.2.71"
 ARG PACKAGES="clickhouse-client clickhouse-server clickhouse-common-static"
 
 # set non-empty deb_location_url url to create a docker image
diff --git a/docs/changelogs/v24.2.2.71-stable.md b/docs/changelogs/v24.2.2.71-stable.md
new file mode 100644
index 00000000000..b9aa5be626b
--- /dev/null
+++ b/docs/changelogs/v24.2.2.71-stable.md
@@ -0,0 +1,55 @@
+---
+sidebar_position: 1
+sidebar_label: 2024
+---
+
+# 2024 Changelog
+
+### ClickHouse release v24.2.2.71-stable (9293d361e72) FIXME as compared to v24.2.1.2248-stable (891689a4150)
+
+#### Improvement
+* Backported in [#60834](https://github.com/ClickHouse/ClickHouse/issues/60834): Update tzdata to 2024a. [#60768](https://github.com/ClickHouse/ClickHouse/pull/60768) ([Raúl Marín](https://github.com/Algunenano)).
+
+#### Bug Fix (user-visible misbehavior in an official stable release)
+
+* PartsSplitter invalid ranges for the same part [#60041](https://github.com/ClickHouse/ClickHouse/pull/60041) ([Maksim Kita](https://github.com/kitaisreal)).
+* Try to avoid calculation of scalar subqueries for CREATE TABLE. [#60464](https://github.com/ClickHouse/ClickHouse/pull/60464) ([Nikolai Kochetov](https://github.com/KochetovNicolai)).
+* Fix deadlock in parallel parsing when lots of rows are skipped due to errors [#60516](https://github.com/ClickHouse/ClickHouse/pull/60516) ([Kruglov Pavel](https://github.com/Avogar)).
+* Fix_max_query_size_for_kql_compound_operator: [#60534](https://github.com/ClickHouse/ClickHouse/pull/60534) ([Yong Wang](https://github.com/kashwy)).
+* Reduce the number of read rows from `system.numbers` [#60546](https://github.com/ClickHouse/ClickHouse/pull/60546) ([JackyWoo](https://github.com/JackyWoo)).
+* Don't output number tips for date types [#60577](https://github.com/ClickHouse/ClickHouse/pull/60577) ([Raúl Marín](https://github.com/Algunenano)).
+* Fix buffer overflow in CompressionCodecMultiple [#60731](https://github.com/ClickHouse/ClickHouse/pull/60731) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
+* Remove nonsense from SQL/JSON [#60738](https://github.com/ClickHouse/ClickHouse/pull/60738) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
+* Prevent setting custom metadata headers on unsupported multipart upload operations [#60748](https://github.com/ClickHouse/ClickHouse/pull/60748) ([Francisco J. Jurado Moreno](https://github.com/Beetelbrox)).
+* Fix crash in arrayEnumerateRanked [#60764](https://github.com/ClickHouse/ClickHouse/pull/60764) ([Raúl Marín](https://github.com/Algunenano)).
+* Fix crash when using input() in INSERT SELECT JOIN [#60765](https://github.com/ClickHouse/ClickHouse/pull/60765) ([Kruglov Pavel](https://github.com/Avogar)).
+* Fix crash with different allow_experimental_analyzer value in subqueries [#60770](https://github.com/ClickHouse/ClickHouse/pull/60770) ([Dmitry Novik](https://github.com/novikd)).
+* Remove recursion when reading from S3 [#60849](https://github.com/ClickHouse/ClickHouse/pull/60849) ([Antonio Andelic](https://github.com/antonio2368)).
+* Fix multiple bugs in groupArraySorted [#61203](https://github.com/ClickHouse/ClickHouse/pull/61203) ([Raúl Marín](https://github.com/Algunenano)).
+* Fix Keeper reconfig for standalone binary [#61233](https://github.com/ClickHouse/ClickHouse/pull/61233) ([Antonio Andelic](https://github.com/antonio2368)).
+
+#### CI Fix or Improvement (changelog entry is not required)
+
+* Backported in [#60758](https://github.com/ClickHouse/ClickHouse/issues/60758): Decoupled changes from [#60408](https://github.com/ClickHouse/ClickHouse/issues/60408). [#60553](https://github.com/ClickHouse/ClickHouse/pull/60553) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
+* Backported in [#60706](https://github.com/ClickHouse/ClickHouse/issues/60706): Eliminates the need to provide input args to docker server jobs to clean yml files. [#60602](https://github.com/ClickHouse/ClickHouse/pull/60602) ([Max K.](https://github.com/maxknv)).
+* Backported in [#61045](https://github.com/ClickHouse/ClickHouse/issues/61045): Debug and fix markreleaseready. [#60611](https://github.com/ClickHouse/ClickHouse/pull/60611) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
+* Backported in [#60721](https://github.com/ClickHouse/ClickHouse/issues/60721): Fix build_report job so that it's defined by ci_config only (not yml file). [#60613](https://github.com/ClickHouse/ClickHouse/pull/60613) ([Max K.](https://github.com/maxknv)).
+* Backported in [#60668](https://github.com/ClickHouse/ClickHouse/issues/60668): Do not await ci pending jobs on release branches decrease wait timeout to fit into gh job timeout. [#60652](https://github.com/ClickHouse/ClickHouse/pull/60652) ([Max K.](https://github.com/maxknv)).
+* Backported in [#60863](https://github.com/ClickHouse/ClickHouse/issues/60863): Set limited number of builds for "special build check" report in backports. [#60850](https://github.com/ClickHouse/ClickHouse/pull/60850) ([Max K.](https://github.com/maxknv)).
+* Backported in [#60946](https://github.com/ClickHouse/ClickHouse/issues/60946): ... [#60935](https://github.com/ClickHouse/ClickHouse/pull/60935) ([Max K.](https://github.com/maxknv)).
+* Backported in [#60972](https://github.com/ClickHouse/ClickHouse/issues/60972): ... [#60952](https://github.com/ClickHouse/ClickHouse/pull/60952) ([Max K.](https://github.com/maxknv)).
+* Backported in [#60980](https://github.com/ClickHouse/ClickHouse/issues/60980): ... [#60958](https://github.com/ClickHouse/ClickHouse/pull/60958) ([Max K.](https://github.com/maxknv)).
+* Backported in [#61170](https://github.com/ClickHouse/ClickHouse/issues/61170): Just a preparation for the merge queue support. [#61099](https://github.com/ClickHouse/ClickHouse/pull/61099) ([Max K.](https://github.com/maxknv)).
+* Backported in [#61181](https://github.com/ClickHouse/ClickHouse/issues/61181): ... [#61172](https://github.com/ClickHouse/ClickHouse/pull/61172) ([Max K.](https://github.com/maxknv)).
+* Backported in [#61228](https://github.com/ClickHouse/ClickHouse/issues/61228): ... [#61183](https://github.com/ClickHouse/ClickHouse/pull/61183) ([Han Fei](https://github.com/hanfei1991)).
+* Backported in [#61194](https://github.com/ClickHouse/ClickHouse/issues/61194): ... [#61185](https://github.com/ClickHouse/ClickHouse/pull/61185) ([Max K.](https://github.com/maxknv)).
+* Backported in [#61244](https://github.com/ClickHouse/ClickHouse/issues/61244): ... [#61214](https://github.com/ClickHouse/ClickHouse/pull/61214) ([Max K.](https://github.com/maxknv)).
+* Backported in [#61388](https://github.com/ClickHouse/ClickHouse/issues/61388):. [#61373](https://github.com/ClickHouse/ClickHouse/pull/61373) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
+
+#### NOT FOR CHANGELOG / INSIGNIFICANT
+
+* CI: make workflow yml abstract [#60421](https://github.com/ClickHouse/ClickHouse/pull/60421) ([Max K.](https://github.com/maxknv)).
+* Cancel PipelineExecutor properly in case of exception in spawnThreads [#60499](https://github.com/ClickHouse/ClickHouse/pull/60499) ([Kruglov Pavel](https://github.com/Avogar)).
+* General sanity in function `seriesOutliersDetectTukey` [#60535](https://github.com/ClickHouse/ClickHouse/pull/60535) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
+* Speed up cctools building [#61011](https://github.com/ClickHouse/ClickHouse/pull/61011) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
+
diff --git a/utils/list-versions/version_date.tsv b/utils/list-versions/version_date.tsv
index e372e407ce1..32fbfee8274 100644
--- a/utils/list-versions/version_date.tsv
+++ b/utils/list-versions/version_date.tsv
@@ -1,9 +1,13 @@
+v24.2.2.71-stable	2024-03-15
 v24.2.1.2248-stable	2024-02-29
+v24.1.7.18-stable	2024-03-15
+v24.1.6.52-stable	2024-03-07
 v24.1.5.6-stable	2024-02-14
 v24.1.4.20-stable	2024-02-14
 v24.1.3.31-stable	2024-02-09
 v24.1.2.5-stable	2024-02-02
 v24.1.1.2048-stable	2024-01-30
+v23.12.5.81-stable	2024-03-15
 v23.12.4.15-stable	2024-02-09
 v23.12.3.40-stable	2024-02-02
 v23.12.2.59-stable	2024-01-05
@@ -25,6 +29,7 @@ v23.9.4.11-stable	2023-11-08
 v23.9.3.12-stable	2023-10-31
 v23.9.2.56-stable	2023-10-19
 v23.9.1.1854-stable	2023-09-29
+v23.8.11.28-lts	2024-03-15
 v23.8.10.43-lts	2024-03-05
 v23.8.9.54-lts	2024-01-05
 v23.8.8.20-lts	2023-11-25
@@ -55,6 +60,7 @@ v23.4.4.16-stable	2023-06-17
 v23.4.3.48-stable	2023-06-12
 v23.4.2.11-stable	2023-05-02
 v23.4.1.1943-stable	2023-04-27
+v23.3.21.26-lts	2024-03-15
 v23.3.20.27-lts	2024-03-05
 v23.3.19.32-lts	2024-01-05
 v23.3.18.15-lts	2023-11-25

From 51f0930fe276da5d32a7a4df302f337ebfead00e Mon Sep 17 00:00:00 2001
From: robot-clickhouse <robot-clickhouse@users.noreply.github.com>
Date: Fri, 15 Mar 2024 06:53:49 +0000
Subject: [PATCH 370/443] Update version_date.tsv and changelogs after
 v23.12.5.81-stable

---
 docker/keeper/Dockerfile              |  2 +-
 docker/server/Dockerfile.alpine       |  2 +-
 docker/server/Dockerfile.ubuntu       |  2 +-
 docs/changelogs/v23.12.5.81-stable.md | 64 +++++++++++++++++++++++++++
 utils/list-versions/version_date.tsv  |  6 +++
 5 files changed, 73 insertions(+), 3 deletions(-)
 create mode 100644 docs/changelogs/v23.12.5.81-stable.md

diff --git a/docker/keeper/Dockerfile b/docker/keeper/Dockerfile
index 2f42854a972..17eee6d4287 100644
--- a/docker/keeper/Dockerfile
+++ b/docker/keeper/Dockerfile
@@ -34,7 +34,7 @@ RUN arch=${TARGETARCH:-amd64} \
 # lts / testing / prestable / etc
 ARG REPO_CHANNEL="stable"
 ARG REPOSITORY="https://packages.clickhouse.com/tgz/${REPO_CHANNEL}"
-ARG VERSION="24.2.1.2248"
+ARG VERSION="24.2.2.71"
 ARG PACKAGES="clickhouse-keeper"
 ARG DIRECT_DOWNLOAD_URLS=""
 
diff --git a/docker/server/Dockerfile.alpine b/docker/server/Dockerfile.alpine
index 7bd777de5b9..bd5fa313adc 100644
--- a/docker/server/Dockerfile.alpine
+++ b/docker/server/Dockerfile.alpine
@@ -32,7 +32,7 @@ RUN arch=${TARGETARCH:-amd64} \
 # lts / testing / prestable / etc
 ARG REPO_CHANNEL="stable"
 ARG REPOSITORY="https://packages.clickhouse.com/tgz/${REPO_CHANNEL}"
-ARG VERSION="24.2.1.2248"
+ARG VERSION="24.2.2.71"
 ARG PACKAGES="clickhouse-client clickhouse-server clickhouse-common-static"
 ARG DIRECT_DOWNLOAD_URLS=""
 
diff --git a/docker/server/Dockerfile.ubuntu b/docker/server/Dockerfile.ubuntu
index 03d01cfd5d7..256dcdc029f 100644
--- a/docker/server/Dockerfile.ubuntu
+++ b/docker/server/Dockerfile.ubuntu
@@ -27,7 +27,7 @@ RUN sed -i "s|http://archive.ubuntu.com|${apt_archive}|g" /etc/apt/sources.list
 
 ARG REPO_CHANNEL="stable"
 ARG REPOSITORY="deb [signed-by=/usr/share/keyrings/clickhouse-keyring.gpg] https://packages.clickhouse.com/deb ${REPO_CHANNEL} main"
-ARG VERSION="24.2.1.2248"
+ARG VERSION="24.2.2.71"
 ARG PACKAGES="clickhouse-client clickhouse-server clickhouse-common-static"
 
 # set non-empty deb_location_url url to create a docker image
diff --git a/docs/changelogs/v23.12.5.81-stable.md b/docs/changelogs/v23.12.5.81-stable.md
new file mode 100644
index 00000000000..0a0acd97d58
--- /dev/null
+++ b/docs/changelogs/v23.12.5.81-stable.md
@@ -0,0 +1,64 @@
+---
+sidebar_position: 1
+sidebar_label: 2024
+---
+
+# 2024 Changelog
+
+### ClickHouse release v23.12.5.81-stable (a0fbe3ae813) FIXME as compared to v23.12.4.15-stable (4233d111d20)
+
+#### Improvement
+* Backported in [#60290](https://github.com/ClickHouse/ClickHouse/issues/60290): Copy S3 file GCP fallback to buffer copy in case GCP returned `Internal Error` with `GATEWAY_TIMEOUT` HTTP error code. [#60164](https://github.com/ClickHouse/ClickHouse/pull/60164) ([Maksim Kita](https://github.com/kitaisreal)).
+* Backported in [#60830](https://github.com/ClickHouse/ClickHouse/issues/60830): Update tzdata to 2024a. [#60768](https://github.com/ClickHouse/ClickHouse/pull/60768) ([Raúl Marín](https://github.com/Algunenano)).
+
+#### Build/Testing/Packaging Improvement
+* Backported in [#59883](https://github.com/ClickHouse/ClickHouse/issues/59883): If you want to run initdb scripts every time when ClickHouse container is starting you shoud initialize environment varible CLICKHOUSE_ALWAYS_RUN_INITDB_SCRIPTS. [#59808](https://github.com/ClickHouse/ClickHouse/pull/59808) ([Alexander Nikolaev](https://github.com/AlexNik)).
+
+#### Bug Fix (user-visible misbehavior in an official stable release)
+
+* Fix_kql_issue_found_by_wingfuzz [#59626](https://github.com/ClickHouse/ClickHouse/pull/59626) ([Yong Wang](https://github.com/kashwy)).
+* Fix error "Read beyond last offset" for AsynchronousBoundedReadBuffer [#59630](https://github.com/ClickHouse/ClickHouse/pull/59630) ([Vitaly Baranov](https://github.com/vitlibar)).
+* Fix query start time on non initial queries [#59662](https://github.com/ClickHouse/ClickHouse/pull/59662) ([Raúl Marín](https://github.com/Algunenano)).
+* rabbitmq: fix having neither acked nor nacked messages [#59775](https://github.com/ClickHouse/ClickHouse/pull/59775) ([Kseniia Sumarokova](https://github.com/kssenii)).
+* Fix parsing of partition expressions surrounded by parens [#59901](https://github.com/ClickHouse/ClickHouse/pull/59901) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)).
+* Fix optimize_uniq_to_count removing the column alias [#60026](https://github.com/ClickHouse/ClickHouse/pull/60026) ([Raúl Marín](https://github.com/Algunenano)).
+* Fix cosineDistance crash with Nullable [#60150](https://github.com/ClickHouse/ClickHouse/pull/60150) ([Raúl Marín](https://github.com/Algunenano)).
+* Hide sensitive info for s3queue [#60233](https://github.com/ClickHouse/ClickHouse/pull/60233) ([Kseniia Sumarokova](https://github.com/kssenii)).
+* Fix deadlock in parallel parsing when lots of rows are skipped due to errors [#60516](https://github.com/ClickHouse/ClickHouse/pull/60516) ([Kruglov Pavel](https://github.com/Avogar)).
+* Fix_max_query_size_for_kql_compound_operator: [#60534](https://github.com/ClickHouse/ClickHouse/pull/60534) ([Yong Wang](https://github.com/kashwy)).
+* Reduce the number of read rows from `system.numbers` [#60546](https://github.com/ClickHouse/ClickHouse/pull/60546) ([JackyWoo](https://github.com/JackyWoo)).
+* Fix buffer overflow in CompressionCodecMultiple [#60731](https://github.com/ClickHouse/ClickHouse/pull/60731) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
+* Remove nonsense from SQL/JSON [#60738](https://github.com/ClickHouse/ClickHouse/pull/60738) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
+* Prevent setting custom metadata headers on unsupported multipart upload operations [#60748](https://github.com/ClickHouse/ClickHouse/pull/60748) ([Francisco J. Jurado Moreno](https://github.com/Beetelbrox)).
+* Fix crash in arrayEnumerateRanked [#60764](https://github.com/ClickHouse/ClickHouse/pull/60764) ([Raúl Marín](https://github.com/Algunenano)).
+* Fix crash when using input() in INSERT SELECT JOIN [#60765](https://github.com/ClickHouse/ClickHouse/pull/60765) ([Kruglov Pavel](https://github.com/Avogar)).
+* Fix crash with different allow_experimental_analyzer value in subqueries [#60770](https://github.com/ClickHouse/ClickHouse/pull/60770) ([Dmitry Novik](https://github.com/novikd)).
+* Remove recursion when reading from S3 [#60849](https://github.com/ClickHouse/ClickHouse/pull/60849) ([Antonio Andelic](https://github.com/antonio2368)).
+* Fix Keeper reconfig for standalone binary [#61233](https://github.com/ClickHouse/ClickHouse/pull/61233) ([Antonio Andelic](https://github.com/antonio2368)).
+
+#### CI Fix or Improvement (changelog entry is not required)
+
+* Backported in [#60767](https://github.com/ClickHouse/ClickHouse/issues/60767): Decoupled changes from [#60408](https://github.com/ClickHouse/ClickHouse/issues/60408). [#60553](https://github.com/ClickHouse/ClickHouse/pull/60553) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
+* Backported in [#60582](https://github.com/ClickHouse/ClickHouse/issues/60582): Arm and amd docker build jobs use similar job names and thus overwrite job reports - aarch64 and amd64 suffixes added to fix this. [#60554](https://github.com/ClickHouse/ClickHouse/pull/60554) ([Max K.](https://github.com/maxknv)).
+* Backported in [#61041](https://github.com/ClickHouse/ClickHouse/issues/61041): Debug and fix markreleaseready. [#60611](https://github.com/ClickHouse/ClickHouse/pull/60611) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
+* Backported in [#61030](https://github.com/ClickHouse/ClickHouse/issues/61030): ... [#61022](https://github.com/ClickHouse/ClickHouse/pull/61022) ([Max K.](https://github.com/maxknv)).
+* Backported in [#61224](https://github.com/ClickHouse/ClickHouse/issues/61224): ... [#61183](https://github.com/ClickHouse/ClickHouse/pull/61183) ([Han Fei](https://github.com/hanfei1991)).
+* Backported in [#61190](https://github.com/ClickHouse/ClickHouse/issues/61190): ... [#61185](https://github.com/ClickHouse/ClickHouse/pull/61185) ([Max K.](https://github.com/maxknv)).
+
+#### NO CL ENTRY
+
+* NO CL ENTRY:  'Revert "Backport [#59798](https://github.com/ClickHouse/ClickHouse/issues/59798) to 23.12: CI: do not reuse builds on release branches"'. [#59979](https://github.com/ClickHouse/ClickHouse/pull/59979) ([Max K.](https://github.com/maxknv)).
+
+#### NOT FOR CHANGELOG / INSIGNIFICANT
+
+* CI: move ci-specifics from job scripts to ci.py [#58516](https://github.com/ClickHouse/ClickHouse/pull/58516) ([Max K.](https://github.com/maxknv)).
+* Make ZooKeeper actually sequentialy consistent [#59735](https://github.com/ClickHouse/ClickHouse/pull/59735) ([Alexander Tokmakov](https://github.com/tavplubix)).
+* Fix special build reports in release branches [#59797](https://github.com/ClickHouse/ClickHouse/pull/59797) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
+* CI: do not reuse builds on release branches [#59798](https://github.com/ClickHouse/ClickHouse/pull/59798) ([Max K.](https://github.com/maxknv)).
+* Fix mark release ready [#59994](https://github.com/ClickHouse/ClickHouse/pull/59994) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
+* Ability to detect undead ZooKeeper sessions [#60044](https://github.com/ClickHouse/ClickHouse/pull/60044) ([Alexander Tokmakov](https://github.com/tavplubix)).
+* Detect io_uring in tests [#60373](https://github.com/ClickHouse/ClickHouse/pull/60373) ([Azat Khuzhin](https://github.com/azat)).
+* Cancel PipelineExecutor properly in case of exception in spawnThreads [#60499](https://github.com/ClickHouse/ClickHouse/pull/60499) ([Kruglov Pavel](https://github.com/Avogar)).
+* Remove broken test while we fix it [#60547](https://github.com/ClickHouse/ClickHouse/pull/60547) ([Raúl Marín](https://github.com/Algunenano)).
+* Speed up cctools building [#61011](https://github.com/ClickHouse/ClickHouse/pull/61011) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
+
diff --git a/utils/list-versions/version_date.tsv b/utils/list-versions/version_date.tsv
index e372e407ce1..32fbfee8274 100644
--- a/utils/list-versions/version_date.tsv
+++ b/utils/list-versions/version_date.tsv
@@ -1,9 +1,13 @@
+v24.2.2.71-stable	2024-03-15
 v24.2.1.2248-stable	2024-02-29
+v24.1.7.18-stable	2024-03-15
+v24.1.6.52-stable	2024-03-07
 v24.1.5.6-stable	2024-02-14
 v24.1.4.20-stable	2024-02-14
 v24.1.3.31-stable	2024-02-09
 v24.1.2.5-stable	2024-02-02
 v24.1.1.2048-stable	2024-01-30
+v23.12.5.81-stable	2024-03-15
 v23.12.4.15-stable	2024-02-09
 v23.12.3.40-stable	2024-02-02
 v23.12.2.59-stable	2024-01-05
@@ -25,6 +29,7 @@ v23.9.4.11-stable	2023-11-08
 v23.9.3.12-stable	2023-10-31
 v23.9.2.56-stable	2023-10-19
 v23.9.1.1854-stable	2023-09-29
+v23.8.11.28-lts	2024-03-15
 v23.8.10.43-lts	2024-03-05
 v23.8.9.54-lts	2024-01-05
 v23.8.8.20-lts	2023-11-25
@@ -55,6 +60,7 @@ v23.4.4.16-stable	2023-06-17
 v23.4.3.48-stable	2023-06-12
 v23.4.2.11-stable	2023-05-02
 v23.4.1.1943-stable	2023-04-27
+v23.3.21.26-lts	2024-03-15
 v23.3.20.27-lts	2024-03-05
 v23.3.19.32-lts	2024-01-05
 v23.3.18.15-lts	2023-11-25

From 7f97b11ce6095ae77d0baf1cb3f9517714c6bdc5 Mon Sep 17 00:00:00 2001
From: Zhuo Qiu <jewelz.q.915@gmail.com>
Date: Fri, 15 Mar 2024 15:46:09 +0800
Subject: [PATCH 371/443] only read _row_exists column when necessary

---
 src/Storages/MergeTree/MutateTask.cpp | 21 +++++++++++----------
 1 file changed, 11 insertions(+), 10 deletions(-)

diff --git a/src/Storages/MergeTree/MutateTask.cpp b/src/Storages/MergeTree/MutateTask.cpp
index 4d1e60f450e..bfdc109a89d 100644
--- a/src/Storages/MergeTree/MutateTask.cpp
+++ b/src/Storages/MergeTree/MutateTask.cpp
@@ -1018,8 +1018,8 @@ struct MutationContext
 
     scope_guard temporary_directory_lock;
 
-    /// Whether this mutation contains lightweight delete
-    bool has_lightweight_delete;
+    /// Whether we need to count lightweight delete rows in this mutation
+    bool count_lightweight_deleted_rows;
 };
 
 using MutationContextPtr = std::shared_ptr<MutationContext>;
@@ -1282,7 +1282,8 @@ bool PartMergerWriter::mutateOriginalPartAndPrepareProjections()
 
         ctx->out->write(cur_block);
 
-        if (ctx->has_lightweight_delete)
+        /// TODO: move this calculation to DELETE FROM mutation
+        if (ctx->count_lightweight_deleted_rows)
             existing_rows_count += MutationHelpers::getExistingRowsCount(cur_block);
 
         for (size_t i = 0, size = ctx->projections_to_build.size(); i < size; ++i)
@@ -1376,7 +1377,7 @@ bool PartMergerWriter::iterateThroughAllProjections()
 
 void PartMergerWriter::finalize()
 {
-    if (ctx->has_lightweight_delete)
+    if (ctx->count_lightweight_deleted_rows)
         ctx->new_data_part->existing_rows_count = existing_rows_count;
 }
 
@@ -2225,17 +2226,17 @@ bool MutateTask::prepare()
     if (ctx->mutating_pipeline_builder.initialized())
         ctx->execute_ttl_type = MutationHelpers::shouldExecuteTTL(ctx->metadata_snapshot, ctx->interpreter->getColumnDependencies());
 
-    if (ctx->updated_header.has(RowExistsColumn::name))
+    if (ctx->data->getSettings()->exclude_deleted_rows_for_part_size_in_merge && ctx->updated_header.has(RowExistsColumn::name))
     {
-        /// This mutation contains lightweight delete, reset existing_rows_count of new data part to 0
-        /// It will be updated while writing _row_exists column
-        ctx->has_lightweight_delete = true;
+        /// This mutation contains lightweight delete and we need to count the deleted rows,
+        /// Reset existing_rows_count of new data part to 0 and it will be updated while writing _row_exists column
+        ctx->count_lightweight_deleted_rows = true;
     }
     else
     {
-        ctx->has_lightweight_delete = false;
+        ctx->count_lightweight_deleted_rows = false;
 
-        /// This mutation does not contains lightweight delete, copy existing_rows_count from source part
+        /// No need to count deleted rows, copy existing_rows_count from source part
         ctx->new_data_part->existing_rows_count = ctx->source_part->existing_rows_count.value_or(ctx->source_part->rows_count);
     }
 

From be4554ba431c2c496c139e6b4869a68ca3ba58dc Mon Sep 17 00:00:00 2001
From: Zhuo Qiu <jewelz.q.915@gmail.com>
Date: Fri, 15 Mar 2024 16:31:50 +0800
Subject: [PATCH 372/443] fix test

---
 tests/queries/0_stateless/03001_consider_lwd_when_merge.sql | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/tests/queries/0_stateless/03001_consider_lwd_when_merge.sql b/tests/queries/0_stateless/03001_consider_lwd_when_merge.sql
index a65e8877020..988d7058f21 100644
--- a/tests/queries/0_stateless/03001_consider_lwd_when_merge.sql
+++ b/tests/queries/0_stateless/03001_consider_lwd_when_merge.sql
@@ -17,6 +17,9 @@ SELECT count() FROM system.parts WHERE database = currentDatabase() AND table =
 
 ALTER TABLE lwd_merge MODIFY SETTING exclude_deleted_rows_for_part_size_in_merge = 1;
 
+-- delete again because deleted rows will be counted in mutation
+DELETE FROM lwd_merge WHERE id % 100 == 0;
+
 OPTIMIZE TABLE lwd_merge;
 SELECT count() FROM system.parts WHERE database = currentDatabase() AND table = 'lwd_merge' AND active = 1;
 

From 4eda78440d696db68d654a5883cc20fb1b145365 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Fri, 15 Mar 2024 10:34:32 +0100
Subject: [PATCH 373/443] Remove C++ templates

---
 src/DataTypes/IDataType.h       |   2 +
 src/Functions/castOrDefault.cpp | 181 ++++++++++++++------------------
 2 files changed, 79 insertions(+), 104 deletions(-)

diff --git a/src/DataTypes/IDataType.h b/src/DataTypes/IDataType.h
index 55f584ef1e0..4403e3d9bd4 100644
--- a/src/DataTypes/IDataType.h
+++ b/src/DataTypes/IDataType.h
@@ -533,6 +533,8 @@ class DataTypeDateTime;
 class DataTypeDateTime64;
 
 template <is_decimal T> constexpr bool IsDataTypeDecimal<DataTypeDecimal<T>> = true;
+
+/// TODO: this is garbage, remove it.
 template <> inline constexpr bool IsDataTypeDecimal<DataTypeDateTime64> = true;
 
 template <typename T> constexpr bool IsDataTypeNumber<DataTypeNumber<T>> = true;
diff --git a/src/Functions/castOrDefault.cpp b/src/Functions/castOrDefault.cpp
index 970e6fd6f75..18bdea28029 100644
--- a/src/Functions/castOrDefault.cpp
+++ b/src/Functions/castOrDefault.cpp
@@ -173,25 +173,22 @@ private:
     bool keep_nullable;
 };
 
-template <typename Type, typename Name>
 class FunctionCastOrDefaultTyped final : public IFunction
 {
 public:
-    static constexpr auto name = Name::name;
-
-    static FunctionPtr create(ContextPtr context)
-    {
-        return std::make_shared<FunctionCastOrDefaultTyped>(context);
-    }
-
-    explicit FunctionCastOrDefaultTyped(ContextPtr context_)
-        : impl(context_)
+    explicit FunctionCastOrDefaultTyped(ContextPtr context_, String name_, DataTypePtr type_)
+        : impl(context_), name(std::move(name_)), type(std::move(type_)), which(type)
     {
     }
 
     String getName() const override { return name; }
 
 private:
+    FunctionCastOrDefault impl;
+    String name;
+    DataTypePtr type;
+    WhichDataType which;
+
     size_t getNumberOfArguments() const override { return 0; }
     bool isVariadic() const override { return true; }
 
@@ -209,10 +206,10 @@ private:
         FunctionArgumentDescriptors mandatory_args = {{"Value", nullptr, nullptr, nullptr}};
         FunctionArgumentDescriptors optional_args;
 
-        if constexpr (IsDataTypeDecimal<Type>)
+        if (isDecimal(type) || isDateTime64(type))
             mandatory_args.push_back({"scale", static_cast<FunctionArgumentDescriptor::TypeValidator>(&isNativeInteger), &isColumnConst, "const Integer"});
 
-        if (std::is_same_v<Type, DataTypeDateTime> || std::is_same_v<Type, DataTypeDateTime64>)
+        if (isDateTimeOrDateTime64(type))
             optional_args.push_back({"timezone", static_cast<FunctionArgumentDescriptor::TypeValidator>(&isString), isColumnConst, "const String"});
 
         optional_args.push_back({"default_value", nullptr, nullptr, nullptr});
@@ -224,7 +221,7 @@ private:
         size_t scale = 0;
         std::string time_zone;
 
-        if constexpr (IsDataTypeDecimal<Type>)
+        if (isDecimal(type))
         {
             const auto & scale_argument = arguments[additional_argument_index];
 
@@ -241,7 +238,7 @@ private:
             ++additional_argument_index;
         }
 
-        if constexpr (std::is_same_v<Type, DataTypeDateTime> || std::is_same_v<Type, DataTypeDateTime64>)
+        if (isDateTimeOrDateTime64(type))
         {
             if (additional_argument_index < arguments.size())
             {
@@ -251,16 +248,22 @@ private:
             }
         }
 
-        std::shared_ptr<Type> cast_type;
+        DataTypePtr cast_type;
 
-        if constexpr (std::is_same_v<Type, DataTypeDateTime64>)
-            cast_type = std::make_shared<Type>(scale, time_zone);
-        else if constexpr (IsDataTypeDecimal<Type>)
-            cast_type = std::make_shared<Type>(Type::maxPrecision(), scale);
-        else if constexpr (std::is_same_v<Type, DataTypeDateTime> || std::is_same_v<Type, DataTypeDateTime64>)
-            cast_type = std::make_shared<Type>(time_zone);
+        if (which.isDateTime64())
+            cast_type = std::make_shared<DataTypeDateTime64>(scale, time_zone);
+        else if (which.isDateTime())
+            cast_type = std::make_shared<DataTypeDateTime>(time_zone);
+        else if (which.isDecimal32())
+            cast_type = createDecimalMaxPrecision<Decimal32>(scale);
+        else if (which.isDecimal64())
+            cast_type = createDecimalMaxPrecision<Decimal64>(scale);
+        else if (which.isDecimal128())
+            cast_type = createDecimalMaxPrecision<Decimal128>(scale);
+        else if (which.isDecimal256())
+            cast_type = createDecimalMaxPrecision<Decimal256>(scale);
         else
-            cast_type = std::make_shared<Type>();
+            cast_type = type;
 
         ColumnWithTypeAndName type_argument =
         {
@@ -289,7 +292,8 @@ private:
 
     ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_size) const override
     {
-        size_t additional_arguments_size = IsDataTypeDecimal<Type> + (std::is_same_v<Type, DataTypeDateTime> || std::is_same_v<Type, DataTypeDateTime64>);
+        /// Scale and time zone
+        size_t additional_arguments_size = (which.isDecimal() || which.isDateTime64()) + which.isDateTimeOrDateTime64();
 
         ColumnWithTypeAndName second_argument =
         {
@@ -310,98 +314,67 @@ private:
 
         return impl.executeImpl(arguments_with_cast_type, result_type, input_rows_size);
     }
-
-    FunctionCastOrDefault impl;
 };
 
-struct NameToUInt8OrDefault { static constexpr auto name = "toUInt8OrDefault"; };
-struct NameToUInt16OrDefault { static constexpr auto name = "toUInt16OrDefault"; };
-struct NameToUInt32OrDefault { static constexpr auto name = "toUInt32OrDefault"; };
-struct NameToUInt64OrDefault { static constexpr auto name = "toUInt64OrDefault"; };
-struct NameToUInt256OrDefault { static constexpr auto name = "toUInt256OrDefault"; };
-struct NameToInt8OrDefault { static constexpr auto name = "toInt8OrDefault"; };
-struct NameToInt16OrDefault { static constexpr auto name = "toInt16OrDefault"; };
-struct NameToInt32OrDefault { static constexpr auto name = "toInt32OrDefault"; };
-struct NameToInt64OrDefault { static constexpr auto name = "toInt64OrDefault"; };
-struct NameToInt128OrDefault { static constexpr auto name = "toInt128OrDefault"; };
-struct NameToInt256OrDefault { static constexpr auto name = "toInt256OrDefault"; };
-struct NameToFloat32OrDefault { static constexpr auto name = "toFloat32OrDefault"; };
-struct NameToFloat64OrDefault { static constexpr auto name = "toFloat64OrDefault"; };
-struct NameToDateOrDefault { static constexpr auto name = "toDateOrDefault"; };
-struct NameToDate32OrDefault { static constexpr auto name = "toDate32OrDefault"; };
-struct NameToDateTimeOrDefault { static constexpr auto name = "toDateTimeOrDefault"; };
-struct NameToDateTime64OrDefault { static constexpr auto name = "toDateTime64OrDefault"; };
-struct NameToDecimal32OrDefault { static constexpr auto name = "toDecimal32OrDefault"; };
-struct NameToDecimal64OrDefault { static constexpr auto name = "toDecimal64OrDefault"; };
-struct NameToDecimal128OrDefault { static constexpr auto name = "toDecimal128OrDefault"; };
-struct NameToDecimal256OrDefault { static constexpr auto name = "toDecimal256OrDefault"; };
-struct NameToUUIDOrDefault { static constexpr auto name = "toUUIDOrDefault"; };
-struct NameToIPv4OrDefault { static constexpr auto name = "toIPv4OrDefault"; };
-struct NameToIPv6OrDefault { static constexpr auto name = "toIPv6OrDefault"; };
-
-using FunctionToUInt8OrDefault = FunctionCastOrDefaultTyped<DataTypeUInt8, NameToUInt8OrDefault>;
-using FunctionToUInt16OrDefault = FunctionCastOrDefaultTyped<DataTypeUInt16, NameToUInt16OrDefault>;
-using FunctionToUInt32OrDefault = FunctionCastOrDefaultTyped<DataTypeUInt32, NameToUInt32OrDefault>;
-using FunctionToUInt64OrDefault = FunctionCastOrDefaultTyped<DataTypeUInt64, NameToUInt64OrDefault>;
-using FunctionToUInt256OrDefault = FunctionCastOrDefaultTyped<DataTypeUInt256, NameToUInt256OrDefault>;
-
-using FunctionToInt8OrDefault = FunctionCastOrDefaultTyped<DataTypeInt8, NameToInt8OrDefault>;
-using FunctionToInt16OrDefault = FunctionCastOrDefaultTyped<DataTypeInt16, NameToInt16OrDefault>;
-using FunctionToInt32OrDefault = FunctionCastOrDefaultTyped<DataTypeInt32, NameToInt32OrDefault>;
-using FunctionToInt64OrDefault = FunctionCastOrDefaultTyped<DataTypeInt64, NameToInt64OrDefault>;
-using FunctionToInt128OrDefault = FunctionCastOrDefaultTyped<DataTypeInt128, NameToInt128OrDefault>;
-using FunctionToInt256OrDefault = FunctionCastOrDefaultTyped<DataTypeInt256, NameToInt256OrDefault>;
-
-using FunctionToFloat32OrDefault = FunctionCastOrDefaultTyped<DataTypeFloat32, NameToFloat32OrDefault>;
-using FunctionToFloat64OrDefault = FunctionCastOrDefaultTyped<DataTypeFloat64, NameToFloat64OrDefault>;
-
-using FunctionToDateOrDefault = FunctionCastOrDefaultTyped<DataTypeDate, NameToDateOrDefault>;
-using FunctionToDate32OrDefault = FunctionCastOrDefaultTyped<DataTypeDate32, NameToDate32OrDefault>;
-using FunctionToDateTimeOrDefault = FunctionCastOrDefaultTyped<DataTypeDateTime, NameToDateTimeOrDefault>;
-using FunctionToDateTime64OrDefault = FunctionCastOrDefaultTyped<DataTypeDateTime64, NameToDateTime64OrDefault>;
-
-using FunctionToDecimal32OrDefault = FunctionCastOrDefaultTyped<DataTypeDecimal<Decimal32>, NameToDecimal32OrDefault>;
-using FunctionToDecimal64OrDefault = FunctionCastOrDefaultTyped<DataTypeDecimal<Decimal64>, NameToDecimal64OrDefault>;
-using FunctionToDecimal128OrDefault = FunctionCastOrDefaultTyped<DataTypeDecimal<Decimal128>, NameToDecimal128OrDefault>;
-using FunctionToDecimal256OrDefault = FunctionCastOrDefaultTyped<DataTypeDecimal<Decimal256>, NameToDecimal256OrDefault>;
-
-using FunctionToUUIDOrDefault = FunctionCastOrDefaultTyped<DataTypeUUID, NameToUUIDOrDefault>;
-using FunctionToIPv4OrDefault = FunctionCastOrDefaultTyped<DataTypeIPv4, NameToIPv4OrDefault>;
-using FunctionToIPv6OrDefault = FunctionCastOrDefaultTyped<DataTypeIPv6, NameToIPv6OrDefault>;
-
 REGISTER_FUNCTION(CastOrDefault)
 {
     factory.registerFunction<FunctionCastOrDefault>();
 
-    factory.registerFunction<FunctionToUInt8OrDefault>();
-    factory.registerFunction<FunctionToUInt16OrDefault>();
-    factory.registerFunction<FunctionToUInt32OrDefault>();
-    factory.registerFunction<FunctionToUInt64OrDefault>();
-    factory.registerFunction<FunctionToUInt256OrDefault>();
+    factory.registerFunction("toUInt8OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt8OrDefault", std::make_shared<DataTypeUInt8>())); });
+    factory.registerFunction("toUInt16OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt16OrDefault", std::make_shared<DataTypeUInt16>())); });
+    factory.registerFunction("toUInt32OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt32OrDefault", std::make_shared<DataTypeUInt32>())); });
+    factory.registerFunction("toUInt64OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt64OrDefault", std::make_shared<DataTypeUInt64>())); });
+    factory.registerFunction("toUInt128OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt128OrDefault", std::make_shared<DataTypeUInt128>())); });
+    factory.registerFunction("toUInt256OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt256OrDefault", std::make_shared<DataTypeUInt256>())); });
 
-    factory.registerFunction<FunctionToInt8OrDefault>();
-    factory.registerFunction<FunctionToInt16OrDefault>();
-    factory.registerFunction<FunctionToInt32OrDefault>();
-    factory.registerFunction<FunctionToInt64OrDefault>();
-    factory.registerFunction<FunctionToInt128OrDefault>();
-    factory.registerFunction<FunctionToInt256OrDefault>();
+    factory.registerFunction("toInt8OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt8OrDefault", std::make_shared<DataTypeInt8>())); });
+    factory.registerFunction("toInt16OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt16OrDefault", std::make_shared<DataTypeInt16>())); });
+    factory.registerFunction("toInt32OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt32OrDefault", std::make_shared<DataTypeInt32>())); });
+    factory.registerFunction("toInt64OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt64OrDefault", std::make_shared<DataTypeInt64>())); });
+    factory.registerFunction("toInt128OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt128OrDefault", std::make_shared<DataTypeInt128>())); });
+    factory.registerFunction("toInt256OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt256OrDefault", std::make_shared<DataTypeInt256>())); });
 
-    factory.registerFunction<FunctionToFloat32OrDefault>();
-    factory.registerFunction<FunctionToFloat64OrDefault>();
+    factory.registerFunction("toFloat32OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toFloat32OrDefault", std::make_shared<DataTypeFloat32>())); });
+    factory.registerFunction("toFloat64OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toFloat64OrDefault", std::make_shared<DataTypeFloat64>())); });
 
-    factory.registerFunction<FunctionToDateOrDefault>();
-    factory.registerFunction<FunctionToDate32OrDefault>();
-    factory.registerFunction<FunctionToDateTimeOrDefault>();
-    factory.registerFunction<FunctionToDateTime64OrDefault>();
+    factory.registerFunction("toDateOrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toDateOrDefault", std::make_shared<DataTypeDate>())); });
+    factory.registerFunction("toDate32OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toDate32OrDefault", std::make_shared<DataTypeDate32>())); });
+    factory.registerFunction("toDateTimeOrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toDateTimeOrDefault", std::make_shared<DataTypeDateTime>())); });
+    factory.registerFunction("toDateTime64OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toDateTime64OrDefault", std::make_shared<DataTypeDateTime64>(3 /* default scale */))); });
 
-    factory.registerFunction<FunctionToDecimal32OrDefault>();
-    factory.registerFunction<FunctionToDecimal64OrDefault>();
-    factory.registerFunction<FunctionToDecimal128OrDefault>();
-    factory.registerFunction<FunctionToDecimal256OrDefault>();
+    factory.registerFunction("toDecimal32OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toDecimal32OrDefault", createDecimalMaxPrecision<Decimal32>(0))); });
+    factory.registerFunction("toDecimal64OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toDecimal64OrDefault", createDecimalMaxPrecision<Decimal64>(0))); });
+    factory.registerFunction("toDecimal128OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toDecimal128OrDefault", createDecimalMaxPrecision<Decimal128>(0))); });
+    factory.registerFunction("toDecimal256OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toDecimal256OrDefault", createDecimalMaxPrecision<Decimal256>(0))); });
 
-    factory.registerFunction<FunctionToUUIDOrDefault>();
-    factory.registerFunction<FunctionToIPv4OrDefault>();
-    factory.registerFunction<FunctionToIPv6OrDefault>();
+    factory.registerFunction("toUUIDOrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toUUIDOrDefault", std::make_shared<DataTypeUUID>())); });
+    factory.registerFunction("toIPv4OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toIPv4OrDefault", std::make_shared<DataTypeIPv4>())); });
+    factory.registerFunction("toIPv6OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toIPv6OrDefault", std::make_shared<DataTypeIPv6>())); });
 }
 
 }

From 4290b1fe2cfe6eace74c08601db884f0c1738bf8 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Fri, 15 Mar 2024 11:32:21 +0100
Subject: [PATCH 374/443] Garbage

---
 src/Functions/FunctionsConversion.cpp | 49 +++++++--------------------
 src/Functions/castOrDefault.cpp       |  2 +-
 2 files changed, 13 insertions(+), 38 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 42056067f00..f69d3e9146b 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -1980,13 +1980,6 @@ public:
 
     static constexpr bool to_datetime64 = std::is_same_v<ToDataType, DataTypeDateTime64>;
 
-    static constexpr bool to_string_or_fixed_string = std::is_same_v<ToDataType, DataTypeFixedString> ||
-                                                      std::is_same_v<ToDataType, DataTypeString>;
-
-    static constexpr bool to_date_or_datetime = std::is_same_v<ToDataType, DataTypeDate> ||
-                                                std::is_same_v<ToDataType, DataTypeDate32> ||
-                                                std::is_same_v<ToDataType, DataTypeDateTime>;
-
     static FunctionPtr create(ContextPtr context) { return std::make_shared<FunctionConvert>(context); }
     static FunctionPtr create() { return std::make_shared<FunctionConvert>(); }
 
@@ -2003,8 +1996,7 @@ public:
     bool isInjective(const ColumnsWithTypeAndName &) const override { return std::is_same_v<Name, NameToString>; }
     bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & arguments) const override
     {
-        /// TODO: We can make more optimizations here.
-        return !(to_date_or_datetime && isNumber(*arguments[0].type));
+        return !(IsDataTypeDateOrDateTime<ToDataType> && isNumber(*arguments[0].type));
     }
 
     using DefaultReturnTypeGetter = std::function<DataTypePtr(const ColumnsWithTypeAndName &)>;
@@ -2327,7 +2319,7 @@ private:
         }
 
         bool done = false;
-        if constexpr (to_string_or_fixed_string)
+        if constexpr (std::is_same_v<ToDataType, DataTypeFixedString> || std::is_same_v<ToDataType, DataTypeString>)
         {
             done = callOnIndexAndDataType<ToDataType>(from_type->getTypeId(), call, ConvertDefaultBehaviorTag{});
         }
@@ -3155,7 +3147,6 @@ public:
     }
 
 private:
-
     const char * cast_name;
     MonotonicityForRange monotonicity_for_range;
 
@@ -4623,26 +4614,12 @@ arguments, result_type, input_rows_count); \
             using Types = std::decay_t<decltype(types)>;
             using ToDataType = typename Types::LeftType;
 
-            if constexpr (
-                std::is_same_v<ToDataType, DataTypeUInt16> ||
-                std::is_same_v<ToDataType, DataTypeUInt32> ||
-                std::is_same_v<ToDataType, DataTypeUInt64> ||
-                std::is_same_v<ToDataType, DataTypeUInt128> ||
-                std::is_same_v<ToDataType, DataTypeUInt256> ||
-                std::is_same_v<ToDataType, DataTypeInt8> ||
-                std::is_same_v<ToDataType, DataTypeInt16> ||
-                std::is_same_v<ToDataType, DataTypeInt32> ||
-                std::is_same_v<ToDataType, DataTypeInt64> ||
-                std::is_same_v<ToDataType, DataTypeInt128> ||
-                std::is_same_v<ToDataType, DataTypeInt256> ||
-                std::is_same_v<ToDataType, DataTypeFloat32> ||
-                std::is_same_v<ToDataType, DataTypeFloat64> ||
-                std::is_same_v<ToDataType, DataTypeDate> ||
-                std::is_same_v<ToDataType, DataTypeDate32> ||
-                std::is_same_v<ToDataType, DataTypeDateTime> ||
-                std::is_same_v<ToDataType, DataTypeUUID> ||
-                std::is_same_v<ToDataType, DataTypeIPv4> ||
-                std::is_same_v<ToDataType, DataTypeIPv6>)
+            if constexpr (is_any_of<ToDataType,
+                DataTypeUInt16, DataTypeUInt32, DataTypeUInt64, DataTypeUInt128, DataTypeUInt256,
+                DataTypeInt8, DataTypeInt16, DataTypeInt32, DataTypeInt64, DataTypeInt128, DataTypeInt256,
+                DataTypeFloat32, DataTypeFloat64,
+                DataTypeDate, DataTypeDate32, DataTypeDateTime,
+                DataTypeUUID, DataTypeIPv4, DataTypeIPv6>)
             {
                 ret = createWrapper(from_type, checkAndGetDataType<ToDataType>(to_type.get()), requested_result_is_nullable);
                 return true;
@@ -4662,12 +4639,10 @@ arguments, result_type, input_rows_count); \
                 ret = createEnumWrapper(from_type, checkAndGetDataType<ToDataType>(to_type.get()));
                 return true;
             }
-            if constexpr (
-                std::is_same_v<ToDataType, DataTypeDecimal<Decimal32>> ||
-                std::is_same_v<ToDataType, DataTypeDecimal<Decimal64>> ||
-                std::is_same_v<ToDataType, DataTypeDecimal<Decimal128>> ||
-                std::is_same_v<ToDataType, DataTypeDecimal<Decimal256>> ||
-                std::is_same_v<ToDataType, DataTypeDateTime64>)
+            if constexpr (is_any_of<ToDataType,
+                DataTypeDecimal<Decimal32>, DataTypeDecimal<Decimal64>,
+                DataTypeDecimal<Decimal128>, DataTypeDecimal<Decimal256>,
+                DataTypeDateTime64>)
             {
                 ret = createDecimalWrapper(from_type, checkAndGetDataType<ToDataType>(to_type.get()), requested_result_is_nullable);
                 return true;
diff --git a/src/Functions/castOrDefault.cpp b/src/Functions/castOrDefault.cpp
index 18bdea28029..57cb03e0349 100644
--- a/src/Functions/castOrDefault.cpp
+++ b/src/Functions/castOrDefault.cpp
@@ -303,7 +303,7 @@ private:
         };
 
         ColumnsWithTypeAndName arguments_with_cast_type;
-        arguments_with_cast_type.reserve(arguments.size());
+        arguments_with_cast_type.reserve(arguments.size() + 1);
 
         arguments_with_cast_type.emplace_back(arguments[0]);
         arguments_with_cast_type.emplace_back(second_argument);

From 060f79862d1c13f9fb723b592e384db009a45823 Mon Sep 17 00:00:00 2001
From: Antonio Andelic <antonio@clickhouse.com>
Date: Fri, 15 Mar 2024 10:49:36 +0000
Subject: [PATCH 375/443] Fix

---
 src/Analyzer/Passes/SumIfToCountIfPass.cpp    |  2 +-
 ...3010_sum_to_to_count_if_nullable.reference | 66 +++++++++++++++++++
 .../03010_sum_to_to_count_if_nullable.sql     | 12 +++-
 3 files changed, 77 insertions(+), 3 deletions(-)

diff --git a/src/Analyzer/Passes/SumIfToCountIfPass.cpp b/src/Analyzer/Passes/SumIfToCountIfPass.cpp
index 2c41b6dc467..1c2097e7be9 100644
--- a/src/Analyzer/Passes/SumIfToCountIfPass.cpp
+++ b/src/Analyzer/Passes/SumIfToCountIfPass.cpp
@@ -54,7 +54,7 @@ public:
             if (!constant_node)
                 return;
 
-            if (auto constant_type = constant_node->getResultType(); !isUInt64(constant_type) && !isInt64(constant_type))
+            if (auto constant_type = constant_node->getResultType(); !isNativeInteger(constant_type))
                 return;
 
             const auto & constant_value_literal = constant_node->getValue();
diff --git a/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.reference b/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.reference
index 8627f639a03..89e5f639c66 100644
--- a/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.reference
+++ b/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.reference
@@ -1,2 +1,68 @@
 (5,NULL)
 (5,NULL)
+(5,NULL)
+QUERY id: 0
+  PROJECTION COLUMNS
+    (sumIf(toNullable(1), equals(modulo(number, 2), 0)), NULL) Tuple(Nullable(UInt64), Nullable(Nothing))
+  PROJECTION
+    LIST id: 1, nodes: 1
+      FUNCTION id: 2, function_name: tuple, function_type: ordinary, result_type: Tuple(Nullable(UInt64), Nullable(Nothing))
+        ARGUMENTS
+          LIST id: 3, nodes: 2
+            FUNCTION id: 4, function_name: sumIf, function_type: aggregate, result_type: Nullable(UInt64)
+              ARGUMENTS
+                LIST id: 5, nodes: 2
+                  CONSTANT id: 6, constant_value: UInt64_1, constant_value_type: Nullable(UInt8)
+                    EXPRESSION
+                      FUNCTION id: 7, function_name: toNullable, function_type: ordinary, result_type: Nullable(UInt8)
+                        ARGUMENTS
+                          LIST id: 8, nodes: 1
+                            CONSTANT id: 9, constant_value: UInt64_1, constant_value_type: UInt8
+                  FUNCTION id: 10, function_name: equals, function_type: ordinary, result_type: UInt8
+                    ARGUMENTS
+                      LIST id: 11, nodes: 2
+                        FUNCTION id: 12, function_name: modulo, function_type: ordinary, result_type: UInt8
+                          ARGUMENTS
+                            LIST id: 13, nodes: 2
+                              COLUMN id: 14, column_name: number, result_type: UInt64, source_id: 15
+                              CONSTANT id: 16, constant_value: UInt64_2, constant_value_type: UInt8
+                        CONSTANT id: 17, constant_value: UInt64_0, constant_value_type: UInt8
+            CONSTANT id: 18, constant_value: NULL, constant_value_type: Nullable(Nothing)
+  JOIN TREE
+    TABLE_FUNCTION id: 15, alias: __table1, table_function_name: numbers
+      ARGUMENTS
+        LIST id: 19, nodes: 1
+          CONSTANT id: 20, constant_value: UInt64_10, constant_value_type: UInt8
+(5,NULL)
+QUERY id: 0
+  PROJECTION COLUMNS
+    (sum(if(equals(modulo(number, 2), 0), toNullable(1), 0)), NULL) Tuple(Nullable(UInt64), Nullable(Nothing))
+  PROJECTION
+    LIST id: 1, nodes: 1
+      FUNCTION id: 2, function_name: tuple, function_type: ordinary, result_type: Tuple(Nullable(UInt64), Nullable(Nothing))
+        ARGUMENTS
+          LIST id: 3, nodes: 2
+            FUNCTION id: 4, function_name: sumOrNullIf, function_type: aggregate, result_type: Nullable(UInt64)
+              ARGUMENTS
+                LIST id: 5, nodes: 2
+                  CONSTANT id: 6, constant_value: UInt64_1, constant_value_type: Nullable(UInt8)
+                    EXPRESSION
+                      FUNCTION id: 7, function_name: toNullable, function_type: ordinary, result_type: Nullable(UInt8)
+                        ARGUMENTS
+                          LIST id: 8, nodes: 1
+                            CONSTANT id: 9, constant_value: UInt64_1, constant_value_type: UInt8
+                  FUNCTION id: 10, function_name: equals, function_type: ordinary, result_type: UInt8
+                    ARGUMENTS
+                      LIST id: 11, nodes: 2
+                        FUNCTION id: 12, function_name: modulo, function_type: ordinary, result_type: UInt8
+                          ARGUMENTS
+                            LIST id: 13, nodes: 2
+                              COLUMN id: 14, column_name: number, result_type: UInt64, source_id: 15
+                              CONSTANT id: 16, constant_value: UInt64_2, constant_value_type: UInt8
+                        CONSTANT id: 17, constant_value: UInt64_0, constant_value_type: UInt8
+            CONSTANT id: 18, constant_value: NULL, constant_value_type: Nullable(Nothing)
+  JOIN TREE
+    TABLE_FUNCTION id: 15, alias: __table1, table_function_name: numbers
+      ARGUMENTS
+        LIST id: 19, nodes: 1
+          CONSTANT id: 20, constant_value: UInt64_10, constant_value_type: UInt8
diff --git a/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.sql b/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.sql
index 394cd4f1ea5..b283a69a020 100644
--- a/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.sql
+++ b/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.sql
@@ -1,3 +1,11 @@
 SET optimize_rewrite_sum_if_to_count_if = 1;
-SELECT (sumIf(toNullable(1), (number % 2) = 0), NULL) FROM numbers(10) SETTINGS allow_experimental_analyzer=0;
-SELECT (sumIf(toNullable(1), (number % 2) = 0), NULL) FROM numbers(10) SETTINGS allow_experimental_analyzer=1;
\ No newline at end of file
+
+SET allow_experimental_analyzer = 0;
+SELECT (sumIf(toNullable(1), (number % 2) = 0), NULL) FROM numbers(10);
+SELECT (sum(if((number % 2) = 0, toNullable(1), 0)), NULL) FROM numbers(10);
+
+SET allow_experimental_analyzer = 1;
+SELECT (sumIf(toNullable(1), (number % 2) = 0), NULL) FROM numbers(10);
+EXPLAIN QUERY TREE SELECT (sumIf(toNullable(1), (number % 2) = 0), NULL) FROM numbers(10);
+SELECT (sum(if((number % 2) = 0, toNullable(1), 0)), NULL) FROM numbers(10);
+EXPLAIN QUERY TREE SELECT (sum(if((number % 2) = 0, toNullable(1), 0)), NULL) FROM numbers(10);
\ No newline at end of file

From ad4a25906842e89c526d2651954d1ac3e64fbfb7 Mon Sep 17 00:00:00 2001
From: "Mikhail f. Shiryaev" <felixoid@clickhouse.com>
Date: Fri, 15 Mar 2024 12:13:35 +0100
Subject: [PATCH 376/443] Restore automerge for approved PRs

---
 .github/workflows/pull_request.yml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/.github/workflows/pull_request.yml b/.github/workflows/pull_request.yml
index c065219f980..2dddde9aa14 100644
--- a/.github/workflows/pull_request.yml
+++ b/.github/workflows/pull_request.yml
@@ -172,6 +172,7 @@ jobs:
         run: |
           cd "$GITHUB_WORKSPACE/tests/ci"
           python3 finish_check.py
+          python3 merge_pr.py --check-approved
 
 
 #############################################################################################

From a5bd24205947aea5a93490bb698fa99856e00d89 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Fri, 15 Mar 2024 12:36:33 +0100
Subject: [PATCH 377/443] Even more garbage

---
 src/Functions/FunctionsConversion.cpp | 73 ++++++++++++---------------
 1 file changed, 32 insertions(+), 41 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index f69d3e9146b..7f130b0cc86 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -1183,26 +1183,26 @@ struct ConvertImpl
         {
             /// Conversion of DateTime to Date: throw off time component.
             /// Conversion of Date32 to Date.
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateImpl<date_time_overflow_behavior>, false>::execute(
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateImpl<date_time_overflow_behavior>, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count);
         }
         else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime> && std::is_same_v<ToDataType, DataTypeDate32>)
         {
             /// Conversion of DateTime to Date: throw off time component.
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Impl, false>::execute(
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Impl, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count);
         }
         else if constexpr ((std::is_same_v<FromDataType, DataTypeDate> || std::is_same_v<FromDataType, DataTypeDate32>)
             && std::is_same_v<ToDataType, DataTypeDateTime>)
         {
             /// Conversion from Date/Date32 to DateTime.
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeImpl<date_time_overflow_behavior>, false>::execute(
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeImpl<date_time_overflow_behavior>, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count);
         }
         else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime64> && std::is_same_v<ToDataType, DataTypeDate32>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<DataTypeDateTime64, DataTypeDate32, TransformDateTime64<ToDate32Impl>, false>::execute(
+            return DateTimeTransformImpl<DataTypeDateTime64, DataTypeDate32, TransformDateTime64<ToDate32Impl>, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count, additions);
         }
         /** Special case of converting Int8, Int16, (U)Int32 or (U)Int64 (and also, for convenience,
@@ -1220,7 +1220,7 @@ struct ConvertImpl
             && std::is_same_v<ToDataType, DataTypeDate>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform32Or64<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::execute(
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform32Or64<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count);
         }
         else if constexpr ((
@@ -1229,7 +1229,7 @@ struct ConvertImpl
             && std::is_same_v<ToDataType, DataTypeDate>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform8Or16Signed<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::execute(
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform8Or16Signed<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count);
         }
         else if constexpr ((
@@ -1240,7 +1240,7 @@ struct ConvertImpl
             && std::is_same_v<ToDataType, DataTypeDate>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform32Or64Signed<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::execute(
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform32Or64Signed<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count);
         }
         else if constexpr ((
@@ -1249,7 +1249,7 @@ struct ConvertImpl
             && std::is_same_v<ToDataType, DataTypeDate32>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform32Or64<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::execute(
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform32Or64<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count);
         }
         else if constexpr ((
@@ -1258,7 +1258,7 @@ struct ConvertImpl
             && std::is_same_v<ToDataType, DataTypeDate32>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform8Or16Signed<typename FromDataType::FieldType, UInt16>, false>::execute(
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform8Or16Signed<typename FromDataType::FieldType, UInt16>, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count);
         }
         else if constexpr ((
@@ -1269,7 +1269,7 @@ struct ConvertImpl
             && std::is_same_v<ToDataType, DataTypeDate32>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform32Or64Signed<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::execute(
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform32Or64Signed<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count);
         }
         /// Special case of converting Int8, Int16, Int32 or (U)Int64 (and also, for convenience, Float32, Float64) to DateTime.
@@ -1280,14 +1280,14 @@ struct ConvertImpl
             && std::is_same_v<ToDataType, DataTypeDateTime>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeTransformSigned<typename FromDataType::FieldType, UInt32, default_date_time_overflow_behavior>, false>::execute(
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeTransformSigned<typename FromDataType::FieldType, UInt32, default_date_time_overflow_behavior>, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count);
         }
         else if constexpr (std::is_same_v<FromDataType, DataTypeUInt64>
             && std::is_same_v<ToDataType, DataTypeDateTime>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeTransform64<typename FromDataType::FieldType, UInt32, default_date_time_overflow_behavior>, false>::execute(
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeTransform64<typename FromDataType::FieldType, UInt32, default_date_time_overflow_behavior>, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count);
         }
         else if constexpr ((
@@ -1297,7 +1297,7 @@ struct ConvertImpl
             && std::is_same_v<ToDataType, DataTypeDateTime>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeTransform64Signed<typename FromDataType::FieldType, UInt32, default_date_time_overflow_behavior>, false>::execute(
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTimeTransform64Signed<typename FromDataType::FieldType, UInt32, default_date_time_overflow_behavior>, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count);
         }
         else if constexpr ((
@@ -1308,15 +1308,15 @@ struct ConvertImpl
             && std::is_same_v<ToDataType, DataTypeDateTime64>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64TransformSigned<typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::execute(
-                arguments, result_type, input_rows_count, additions);
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64TransformSigned<typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::template execute<Additions>(
+                arguments, result_type, input_rows_count);
         }
         else if constexpr (std::is_same_v<FromDataType, DataTypeUInt64>
             && std::is_same_v<ToDataType, DataTypeDateTime64>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64TransformUnsigned<UInt64, default_date_time_overflow_behavior>, false>::execute(
-                arguments, result_type, input_rows_count, additions);
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64TransformUnsigned<UInt64, default_date_time_overflow_behavior>, false>::template execute<Additions>(
+                arguments, result_type, input_rows_count);
         }
         else if constexpr ((
                 std::is_same_v<FromDataType, DataTypeFloat32>
@@ -1324,22 +1324,22 @@ struct ConvertImpl
             && std::is_same_v<ToDataType, DataTypeDateTime64>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64TransformFloat<FromDataType, typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::execute(
-                arguments, result_type, input_rows_count, additions);
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64TransformFloat<FromDataType, typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::template execute<Additions>(
+                arguments, result_type, input_rows_count);
         }
         /// Conversion of DateTime64 to Date or DateTime: discards fractional part.
         else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime64>
             && std::is_same_v<ToDataType, DataTypeDate>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, TransformDateTime64<ToDateImpl<date_time_overflow_behavior>>, false>::execute(
+            return DateTimeTransformImpl<FromDataType, ToDataType, TransformDateTime64<ToDateImpl<date_time_overflow_behavior>>, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count, additions);
         }
         else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime64>
             && std::is_same_v<ToDataType, DataTypeDateTime>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, TransformDateTime64<ToDateTimeImpl<date_time_overflow_behavior>>, false>::execute(
+            return DateTimeTransformImpl<FromDataType, ToDataType, TransformDateTime64<ToDateTimeImpl<date_time_overflow_behavior>>, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count, additions);
         }
         /// Conversion of Date or DateTime to DateTime64: add zero sub-second part.
@@ -1350,8 +1350,8 @@ struct ConvertImpl
             && std::is_same_v<ToDataType, DataTypeDateTime64>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64Transform, false>::execute(
-                arguments, result_type, input_rows_count, additions);
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64Transform, false>::template execute<Additions>(
+                arguments, result_type, input_rows_count);
         }
         else if constexpr (IsDataTypeDateOrDateTime<FromDataType>
             && std::is_same_v<ToDataType, DataTypeString>)
@@ -1573,8 +1573,8 @@ struct ConvertImpl
             && std::is_same_v<FromDataType, DataTypeString>
             && std::is_same_v<ToDataType, DataTypeUInt32>)
         {
-            return ConvertImpl<FromDataType, DataTypeDateTime, Name, SpecialTag, date_time_overflow_behavior>::execute(
-                arguments, result_type, input_rows_count, additions);
+            return ConvertImpl<FromDataType, DataTypeDateTime, Name, SpecialTag, date_time_overflow_behavior>::template execute<Additions>(
+                arguments, result_type, input_rows_count);
         }
         else if constexpr ((std::is_same_v<FromDataType, DataTypeString> || std::is_same_v<FromDataType, DataTypeFixedString>)
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
@@ -1974,11 +1974,8 @@ public:
     using Monotonic = MonotonicityImpl;
 
     static constexpr auto name = Name::name;
-    static constexpr bool to_decimal =
-        std::is_same_v<Name, NameToDecimal32> || std::is_same_v<Name, NameToDecimal64>
-         || std::is_same_v<Name, NameToDecimal128> || std::is_same_v<Name, NameToDecimal256>;
-
     static constexpr bool to_datetime64 = std::is_same_v<ToDataType, DataTypeDateTime64>;
+    static constexpr bool to_decimal = IsDataTypeDecimal<ToDataType> && !to_datetime64;
 
     static FunctionPtr create(ContextPtr context) { return std::make_shared<FunctionConvert>(context); }
     static FunctionPtr create() { return std::make_shared<FunctionConvert>(); }
@@ -2319,7 +2316,7 @@ private:
         }
 
         bool done = false;
-        if constexpr (std::is_same_v<ToDataType, DataTypeFixedString> || std::is_same_v<ToDataType, DataTypeString>)
+        if constexpr (is_any_of<ToDataType, DataTypeString, DataTypeFixedString>)
         {
             done = callOnIndexAndDataType<ToDataType>(from_type->getTypeId(), call, ConvertDefaultBehaviorTag{});
         }
@@ -2371,13 +2368,8 @@ class FunctionConvertFromString : public IFunction
 {
 public:
     static constexpr auto name = Name::name;
-    static constexpr bool to_decimal =
-        std::is_same_v<ToDataType, DataTypeDecimal<Decimal32>> ||
-        std::is_same_v<ToDataType, DataTypeDecimal<Decimal64>> ||
-        std::is_same_v<ToDataType, DataTypeDecimal<Decimal128>> ||
-        std::is_same_v<ToDataType, DataTypeDecimal<Decimal256>>;
-
     static constexpr bool to_datetime64 = std::is_same_v<ToDataType, DataTypeDateTime64>;
+    static constexpr bool to_decimal = IsDataTypeDecimal<ToDataType> && !to_datetime64;
 
     static FunctionPtr create(ContextPtr) { return std::make_shared<FunctionConvertFromString>(); }
 
@@ -3203,13 +3195,12 @@ private:
             return createFunctionAdaptor(function, from_type);
         }
 
-        auto wrapper_cast_type = cast_type;
-
-        return [wrapper_cast_type, from_type_index, to_type, date_time_overflow_behavior]
-            (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *column_nullable, size_t input_rows_count)
+        return [wrapper_cast_type = cast_type, from_type_index, to_type, date_time_overflow_behavior]
+            (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable * column_nullable, size_t input_rows_count)
         {
             ColumnPtr result_column;
-            auto res = callOnIndexAndDataType<ToDataType>(from_type_index, [&](const auto & types) -> bool {
+            auto res = callOnIndexAndDataType<ToDataType>(from_type_index, [&](const auto & types) -> bool
+            {
                 using Types = std::decay_t<decltype(types)>;
                 using LeftDataType = typename Types::LeftType;
                 using RightDataType = typename Types::RightType;

From 61870fa4b5621d2c9d55a397ee463e285675eee2 Mon Sep 17 00:00:00 2001
From: "Mikhail f. Shiryaev" <felixoid@clickhouse.com>
Date: Fri, 15 Mar 2024 12:43:11 +0100
Subject: [PATCH 378/443] Disable broken SonarCloud

---
 .github/workflows/nightly.yml | 103 +++++++++++++++++-----------------
 1 file changed, 53 insertions(+), 50 deletions(-)

diff --git a/.github/workflows/nightly.yml b/.github/workflows/nightly.yml
index 93ac2be19b4..0ab02db8d7a 100644
--- a/.github/workflows/nightly.yml
+++ b/.github/workflows/nightly.yml
@@ -45,6 +45,7 @@ jobs:
     with:
       data: "${{ needs.RunConfig.outputs.data }}"
       set_latest: true
+
   SonarCloud:
     runs-on: [self-hosted, builder]
     env:
@@ -54,53 +55,55 @@ jobs:
       CC: clang-17
       CXX: clang++-17
     steps:
-      - name: Check out repository code
-        uses: ClickHouse/checkout@v1
-        with:
-          clear-repository: true
-          fetch-depth: 0  # Shallow clones should be disabled for a better relevancy of analysis
-          filter: tree:0
-          submodules: true
-      - name: Set up JDK 11
-        uses: actions/setup-java@v1
-        with:
-          java-version: 11
-      - name: Download and set up sonar-scanner
-        env:
-          SONAR_SCANNER_DOWNLOAD_URL: https://binaries.sonarsource.com/Distribution/sonar-scanner-cli/sonar-scanner-cli-${{ env.SONAR_SCANNER_VERSION }}-linux.zip
-        run: |
-          mkdir -p "$HOME/.sonar"
-          curl -sSLo "$HOME/.sonar/sonar-scanner.zip" "${{ env.SONAR_SCANNER_DOWNLOAD_URL }}"
-          unzip -o "$HOME/.sonar/sonar-scanner.zip" -d "$HOME/.sonar/"
-          echo "$HOME/.sonar/sonar-scanner-${{ env.SONAR_SCANNER_VERSION }}-linux/bin" >> "$GITHUB_PATH"
-      - name: Download and set up build-wrapper
-        env:
-          BUILD_WRAPPER_DOWNLOAD_URL: ${{ env.SONAR_SERVER_URL }}/static/cpp/build-wrapper-linux-x86.zip
-        run: |
-          curl -sSLo "$HOME/.sonar/build-wrapper-linux-x86.zip" "${{ env.BUILD_WRAPPER_DOWNLOAD_URL }}"
-          unzip -o "$HOME/.sonar/build-wrapper-linux-x86.zip" -d "$HOME/.sonar/"
-          echo "$HOME/.sonar/build-wrapper-linux-x86" >> "$GITHUB_PATH"
-      - name: Set Up Build Tools
-        run: |
-          sudo apt-get update
-          sudo apt-get install -yq git cmake ccache ninja-build python3 yasm nasm
-          sudo bash -c "$(wget -O - https://apt.llvm.org/llvm.sh)"
-      - name: Run build-wrapper
-        run: |
-          mkdir build
-          cd build
-          cmake ..
-          cd ..
-          build-wrapper-linux-x86-64 --out-dir ${{ env.BUILD_WRAPPER_OUT_DIR }} cmake --build build/
-      - name: Run sonar-scanner
-        env:
-          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-          SONAR_TOKEN: ${{ secrets.SONAR_TOKEN }}
-        run: |
-          sonar-scanner \
-            --define sonar.host.url="${{ env.SONAR_SERVER_URL }}" \
-            --define sonar.cfamily.build-wrapper-output="${{ env.BUILD_WRAPPER_OUT_DIR }}" \
-            --define sonar.projectKey="ClickHouse_ClickHouse" \
-            --define sonar.organization="clickhouse-java" \
-            --define sonar.cfamily.cpp23.enabled=true \
-            --define sonar.exclusions="**/*.java,**/*.ts,**/*.js,**/*.css,**/*.sql"
+      - name: Disabled
+        run: echo "The job is disabled since permanently broken"
+        # - name: Check out repository code
+        #   uses: ClickHouse/checkout@v1
+        #   with:
+        #     clear-repository: true
+        #     fetch-depth: 0  # Shallow clones should be disabled for a better relevancy of analysis
+        #     filter: tree:0
+        #     submodules: true
+        # - name: Set up JDK 11
+        #   uses: actions/setup-java@v1
+        #   with:
+        #     java-version: 11
+        # - name: Download and set up sonar-scanner
+        #   env:
+        #     SONAR_SCANNER_DOWNLOAD_URL: https://binaries.sonarsource.com/Distribution/sonar-scanner-cli/sonar-scanner-cli-${{ env.SONAR_SCANNER_VERSION }}-linux.zip
+        #   run: |
+        #     mkdir -p "$HOME/.sonar"
+        #     curl -sSLo "$HOME/.sonar/sonar-scanner.zip" "${{ env.SONAR_SCANNER_DOWNLOAD_URL }}"
+        #     unzip -o "$HOME/.sonar/sonar-scanner.zip" -d "$HOME/.sonar/"
+        #     echo "$HOME/.sonar/sonar-scanner-${{ env.SONAR_SCANNER_VERSION }}-linux/bin" >> "$GITHUB_PATH"
+        # - name: Download and set up build-wrapper
+        #   env:
+        #     BUILD_WRAPPER_DOWNLOAD_URL: ${{ env.SONAR_SERVER_URL }}/static/cpp/build-wrapper-linux-x86.zip
+        #   run: |
+        #     curl -sSLo "$HOME/.sonar/build-wrapper-linux-x86.zip" "${{ env.BUILD_WRAPPER_DOWNLOAD_URL }}"
+        #     unzip -o "$HOME/.sonar/build-wrapper-linux-x86.zip" -d "$HOME/.sonar/"
+        #     echo "$HOME/.sonar/build-wrapper-linux-x86" >> "$GITHUB_PATH"
+        # - name: Set Up Build Tools
+        #   run: |
+        #     sudo apt-get update
+        #     sudo apt-get install -yq git cmake ccache ninja-build python3 yasm nasm
+        #     sudo bash -c "$(wget -O - https://apt.llvm.org/llvm.sh)"
+        # - name: Run build-wrapper
+        #   run: |
+        #     mkdir build
+        #     cd build
+        #     cmake ..
+        #     cd ..
+        #     build-wrapper-linux-x86-64 --out-dir ${{ env.BUILD_WRAPPER_OUT_DIR }} cmake --build build/
+        # - name: Run sonar-scanner
+        #   env:
+        #     GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        #     SONAR_TOKEN: ${{ secrets.SONAR_TOKEN }}
+        #   run: |
+        #     sonar-scanner \
+        #       --define sonar.host.url="${{ env.SONAR_SERVER_URL }}" \
+        #       --define sonar.cfamily.build-wrapper-output="${{ env.BUILD_WRAPPER_OUT_DIR }}" \
+        #       --define sonar.projectKey="ClickHouse_ClickHouse" \
+        #       --define sonar.organization="clickhouse-java" \
+        #       --define sonar.cfamily.cpp23.enabled=true \
+        #       --define sonar.exclusions="**/*.java,**/*.ts,**/*.js,**/*.css,**/*.sql"

From 5787b7f7c8a38fd407d8c532b99f143685e55a78 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Fri, 15 Mar 2024 14:46:26 +0300
Subject: [PATCH 379/443] Update nightly.yml

---
 .github/workflows/nightly.yml | 62 -----------------------------------
 1 file changed, 62 deletions(-)

diff --git a/.github/workflows/nightly.yml b/.github/workflows/nightly.yml
index 0ab02db8d7a..515236bb826 100644
--- a/.github/workflows/nightly.yml
+++ b/.github/workflows/nightly.yml
@@ -45,65 +45,3 @@ jobs:
     with:
       data: "${{ needs.RunConfig.outputs.data }}"
       set_latest: true
-
-  SonarCloud:
-    runs-on: [self-hosted, builder]
-    env:
-      SONAR_SCANNER_VERSION: 4.8.0.2856
-      SONAR_SERVER_URL: "https://sonarcloud.io"
-      BUILD_WRAPPER_OUT_DIR: build_wrapper_output_directory # Directory where build-wrapper output will be placed
-      CC: clang-17
-      CXX: clang++-17
-    steps:
-      - name: Disabled
-        run: echo "The job is disabled since permanently broken"
-        # - name: Check out repository code
-        #   uses: ClickHouse/checkout@v1
-        #   with:
-        #     clear-repository: true
-        #     fetch-depth: 0  # Shallow clones should be disabled for a better relevancy of analysis
-        #     filter: tree:0
-        #     submodules: true
-        # - name: Set up JDK 11
-        #   uses: actions/setup-java@v1
-        #   with:
-        #     java-version: 11
-        # - name: Download and set up sonar-scanner
-        #   env:
-        #     SONAR_SCANNER_DOWNLOAD_URL: https://binaries.sonarsource.com/Distribution/sonar-scanner-cli/sonar-scanner-cli-${{ env.SONAR_SCANNER_VERSION }}-linux.zip
-        #   run: |
-        #     mkdir -p "$HOME/.sonar"
-        #     curl -sSLo "$HOME/.sonar/sonar-scanner.zip" "${{ env.SONAR_SCANNER_DOWNLOAD_URL }}"
-        #     unzip -o "$HOME/.sonar/sonar-scanner.zip" -d "$HOME/.sonar/"
-        #     echo "$HOME/.sonar/sonar-scanner-${{ env.SONAR_SCANNER_VERSION }}-linux/bin" >> "$GITHUB_PATH"
-        # - name: Download and set up build-wrapper
-        #   env:
-        #     BUILD_WRAPPER_DOWNLOAD_URL: ${{ env.SONAR_SERVER_URL }}/static/cpp/build-wrapper-linux-x86.zip
-        #   run: |
-        #     curl -sSLo "$HOME/.sonar/build-wrapper-linux-x86.zip" "${{ env.BUILD_WRAPPER_DOWNLOAD_URL }}"
-        #     unzip -o "$HOME/.sonar/build-wrapper-linux-x86.zip" -d "$HOME/.sonar/"
-        #     echo "$HOME/.sonar/build-wrapper-linux-x86" >> "$GITHUB_PATH"
-        # - name: Set Up Build Tools
-        #   run: |
-        #     sudo apt-get update
-        #     sudo apt-get install -yq git cmake ccache ninja-build python3 yasm nasm
-        #     sudo bash -c "$(wget -O - https://apt.llvm.org/llvm.sh)"
-        # - name: Run build-wrapper
-        #   run: |
-        #     mkdir build
-        #     cd build
-        #     cmake ..
-        #     cd ..
-        #     build-wrapper-linux-x86-64 --out-dir ${{ env.BUILD_WRAPPER_OUT_DIR }} cmake --build build/
-        # - name: Run sonar-scanner
-        #   env:
-        #     GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-        #     SONAR_TOKEN: ${{ secrets.SONAR_TOKEN }}
-        #   run: |
-        #     sonar-scanner \
-        #       --define sonar.host.url="${{ env.SONAR_SERVER_URL }}" \
-        #       --define sonar.cfamily.build-wrapper-output="${{ env.BUILD_WRAPPER_OUT_DIR }}" \
-        #       --define sonar.projectKey="ClickHouse_ClickHouse" \
-        #       --define sonar.organization="clickhouse-java" \
-        #       --define sonar.cfamily.cpp23.enabled=true \
-        #       --define sonar.exclusions="**/*.java,**/*.ts,**/*.js,**/*.css,**/*.sql"

From 16abbcd095c4034a49921288ff0d79835addbf16 Mon Sep 17 00:00:00 2001
From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com>
Date: Fri, 15 Mar 2024 12:54:13 +0100
Subject: [PATCH 380/443] Revert "Updated format settings references in the
 docs (datetime.md)"

---
 docs/en/sql-reference/data-types/datetime.md | 8 ++++----
 docs/ru/sql-reference/data-types/datetime.md | 8 ++++----
 2 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/docs/en/sql-reference/data-types/datetime.md b/docs/en/sql-reference/data-types/datetime.md
index a465106c2ff..1adff18f598 100644
--- a/docs/en/sql-reference/data-types/datetime.md
+++ b/docs/en/sql-reference/data-types/datetime.md
@@ -36,9 +36,9 @@ You can explicitly set a time zone for `DateTime`-type columns when creating a t
 
 The [clickhouse-client](../../interfaces/cli.md) applies the server time zone by default if a time zone isn’t explicitly set when initializing the data type. To use the client time zone, run `clickhouse-client` with the `--use_client_time_zone` parameter.
 
-ClickHouse outputs values depending on the value of the [date_time_output_format](../../operations/settings/settings-formats.md#date_time_output_format) setting. `YYYY-MM-DD hh:mm:ss` text format by default. Additionally, you can change the output with the [formatDateTime](../../sql-reference/functions/date-time-functions.md#formatdatetime) function.
+ClickHouse outputs values depending on the value of the [date_time_output_format](../../operations/settings/settings.md#settings-date_time_output_format) setting. `YYYY-MM-DD hh:mm:ss` text format by default. Additionally, you can change the output with the [formatDateTime](../../sql-reference/functions/date-time-functions.md#formatdatetime) function.
 
-When inserting data into ClickHouse, you can use different formats of date and time strings, depending on the value of the [date_time_input_format](../../operations/settings/settings-formats.md#date_time_input_format) setting.
+When inserting data into ClickHouse, you can use different formats of date and time strings, depending on the value of the [date_time_input_format](../../operations/settings/settings.md#settings-date_time_input_format) setting.
 
 ## Examples
 
@@ -147,8 +147,8 @@ Time shifts for multiple days. Some pacific islands changed their timezone offse
 - [Type conversion functions](../../sql-reference/functions/type-conversion-functions.md)
 - [Functions for working with dates and times](../../sql-reference/functions/date-time-functions.md)
 - [Functions for working with arrays](../../sql-reference/functions/array-functions.md)
-- [The `date_time_input_format` setting](../../operations/settings/settings-formats.md#date_time_input_format)
-- [The `date_time_output_format` setting](../../operations/settings/settings-formats.md#date_time_output_format)
+- [The `date_time_input_format` setting](../../operations/settings/settings-formats.md#settings-date_time_input_format)
+- [The `date_time_output_format` setting](../../operations/settings/settings-formats.md#settings-date_time_output_format)
 - [The `timezone` server configuration parameter](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-timezone)
 - [The `session_timezone` setting](../../operations/settings/settings.md#session_timezone)
 - [Operators for working with dates and times](../../sql-reference/operators/index.md#operators-datetime)
diff --git a/docs/ru/sql-reference/data-types/datetime.md b/docs/ru/sql-reference/data-types/datetime.md
index 25e87794147..57f24786bb7 100644
--- a/docs/ru/sql-reference/data-types/datetime.md
+++ b/docs/ru/sql-reference/data-types/datetime.md
@@ -27,9 +27,9 @@ DateTime([timezone])
 
 Консольный клиент ClickHouse по умолчанию использует часовой пояс сервера, если для значения `DateTime` часовой пояс не был задан в явном виде при инициализации типа данных. Чтобы использовать часовой пояс клиента, запустите [clickhouse-client](../../interfaces/cli.md) с параметром `--use_client_time_zone`.
 
-ClickHouse отображает значения в зависимости от значения параметра [date\_time\_output\_format](../../operations/settings/settings-formats.md#date_time_output_format). Текстовый формат по умолчанию `YYYY-MM-DD hh:mm:ss`. Кроме того, вы можете поменять отображение с помощью функции [formatDateTime](../../sql-reference/functions/date-time-functions.md#formatdatetime).
+ClickHouse отображает значения в зависимости от значения параметра [date\_time\_output\_format](../../operations/settings/index.md#settings-date_time_output_format). Текстовый формат по умолчанию `YYYY-MM-DD hh:mm:ss`. Кроме того, вы можете поменять отображение с помощью функции [formatDateTime](../../sql-reference/functions/date-time-functions.md#formatdatetime).
 
-При вставке данных в ClickHouse, можно использовать различные форматы даты и времени в зависимости от значения настройки [date_time_input_format](../../operations/settings/settings-formats.md#date_time_input_format).
+При вставке данных в ClickHouse, можно использовать различные форматы даты и времени в зависимости от значения настройки [date_time_input_format](../../operations/settings/index.md#settings-date_time_input_format).
 
 ## Примеры {#primery}
 
@@ -119,8 +119,8 @@ FROM dt
 -   [Функции преобразования типов](../../sql-reference/functions/type-conversion-functions.md)
 -   [Функции для работы с датой и временем](../../sql-reference/functions/date-time-functions.md)
 -   [Функции для работы с массивами](../../sql-reference/functions/array-functions.md)
--   [Настройка `date_time_input_format`](../../operations/settings/settings-formats.md#date_time_input_format)
--   [Настройка `date_time_output_format`](../../operations/settings/settings-formats.md#date_time_output_format)
+-   [Настройка `date_time_input_format`](../../operations/settings/index.md#settings-date_time_input_format)
+-   [Настройка `date_time_output_format`](../../operations/settings/index.md)
 -   [Конфигурационный параметр сервера `timezone`](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-timezone)
 -   [Параметр `session_timezone`](../../operations/settings/settings.md#session_timezone)
 -   [Операторы для работы с датой и временем](../../sql-reference/operators/index.md#operators-datetime)

From f91c45f562c0cdef32c9403e90eb1e52e818064c Mon Sep 17 00:00:00 2001
From: serxa <sergei@clickhouse.com>
Date: Fri, 15 Mar 2024 12:03:49 +0000
Subject: [PATCH 381/443] fix `01599_multiline_input_and_singleline_comments`
 properly

---
 ...multiline_input_and_singleline_comments.sh | 23 ++++++++++++-------
 1 file changed, 15 insertions(+), 8 deletions(-)

diff --git a/tests/queries/0_stateless/01599_multiline_input_and_singleline_comments.sh b/tests/queries/0_stateless/01599_multiline_input_and_singleline_comments.sh
index f1acd39136f..07c2e345009 100755
--- a/tests/queries/0_stateless/01599_multiline_input_and_singleline_comments.sh
+++ b/tests/queries/0_stateless/01599_multiline_input_and_singleline_comments.sh
@@ -2,7 +2,6 @@
 
 log_user 0
 
-# In some places `-timeout 1` is used to avoid expect to always wait for the whole timeout
 set timeout 60
 
 match_max 100000
@@ -14,15 +13,23 @@ expect ":) "
 
 # Make a query
 send -- "SELECT 1\r"
-expect -timeout 1 ":-] "
 send -- "-- xxx\r"
-expect -timeout 1 ":-] "
 send -- ", 2\r"
-expect -timeout 1 ":-] "
-send -- ";\r"
+send -- ";"
+
+# For some reason this sleep is required for this test to work properly
+sleep 1
+send -- "\r"
+
+expect {
+    "│ 1 │ 2 │" { }
+    timeout { exit 1 }
+}
 
-expect "│ 1 │ 2 │"
 expect ":) "
 
-send -- "\4"
-expect eof
+send -- ""
+expect {
+    eof { exit 0 }
+    timeout { exit 1 }
+}

From 5dad208b56d22a8f02fcfb72fcd822d87a8ca941 Mon Sep 17 00:00:00 2001
From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com>
Date: Fri, 15 Mar 2024 13:40:18 +0100
Subject: [PATCH 382/443] fix tests

---
 .../03009_storage_memory_circ_buffer_usage.reference   |  4 ++++
 .../03009_storage_memory_circ_buffer_usage.sql         | 10 ++++++----
 2 files changed, 10 insertions(+), 4 deletions(-)

diff --git a/tests/queries/0_stateless/03009_storage_memory_circ_buffer_usage.reference b/tests/queries/0_stateless/03009_storage_memory_circ_buffer_usage.reference
index 0bcc6fe4434..20dda4fa15a 100644
--- a/tests/queries/0_stateless/03009_storage_memory_circ_buffer_usage.reference
+++ b/tests/queries/0_stateless/03009_storage_memory_circ_buffer_usage.reference
@@ -1,12 +1,16 @@
+TESTING BYTES
 8192
 9216
 9216
 65536
+TESTING ROWS
 50
 1000
 1020
 1100
+TESTING NO CIRCULAR-BUFFER
 8192
 9216
 17408
 82944
+TESTING INVALID SETTINGS
diff --git a/tests/queries/0_stateless/03009_storage_memory_circ_buffer_usage.sql b/tests/queries/0_stateless/03009_storage_memory_circ_buffer_usage.sql
index 77aa1d0befc..fa4ba96277d 100644
--- a/tests/queries/0_stateless/03009_storage_memory_circ_buffer_usage.sql
+++ b/tests/queries/0_stateless/03009_storage_memory_circ_buffer_usage.sql
@@ -1,7 +1,9 @@
+SET max_block_size = 65409; -- Default value
+
 DROP TABLE IF EXISTS memory;
 CREATE TABLE memory (i UInt32) ENGINE = Memory SETTINGS min_bytes_to_keep = 4096, max_bytes_to_keep = 16384;
 
-/* TESTING BYTES */
+SELECT 'TESTING BYTES';
 /* 1. testing oldest block doesn't get deleted because of min-threshold */
 INSERT INTO memory SELECT * FROM numbers(0, 1600);
 SELECT total_bytes FROM system.tables WHERE name = 'memory' and database = currentDatabase();
@@ -21,7 +23,7 @@ SELECT total_bytes FROM system.tables WHERE name = 'memory' and database = curre
 DROP TABLE IF EXISTS memory;
 CREATE TABLE memory (i UInt32) ENGINE = Memory SETTINGS min_rows_to_keep = 100, max_rows_to_keep = 1000;
 
-/* TESTING ROWS */
+SELECT 'TESTING ROWS';
 /* 1. add normal number of rows */
 INSERT INTO memory SELECT * FROM numbers(0, 50);
 SELECT total_rows FROM system.tables WHERE name = 'memory' and database = currentDatabase();
@@ -38,7 +40,7 @@ SELECT total_rows FROM system.tables WHERE name = 'memory' and database = curren
 INSERT INTO memory SELECT * FROM numbers(3000, 1100);
 SELECT total_rows FROM system.tables WHERE name = 'memory' and database = currentDatabase();
 
-/* TESTING NO CIRCULAR-BUFFER */
+SELECT 'TESTING NO CIRCULAR-BUFFER';
 DROP TABLE IF EXISTS memory;
 CREATE TABLE memory (i UInt32) ENGINE = Memory;
 
@@ -54,7 +56,7 @@ SELECT total_bytes FROM system.tables WHERE name = 'memory' and database = curre
 INSERT INTO memory SELECT * FROM numbers(9000, 10000);
 SELECT total_bytes FROM system.tables WHERE name = 'memory' and database = currentDatabase();
 
-/* TESTING INVALID SETTINGS */
+SELECT 'TESTING INVALID SETTINGS';
 CREATE TABLE faulty_memory (i UInt32) ENGINE = Memory SETTINGS min_rows_to_keep = 100;  -- { serverError 452 }
 CREATE TABLE faulty_memory (i UInt32) ENGINE = Memory SETTINGS min_bytes_to_keep = 100; -- { serverError 452 }
 

From 20bad992a41a37b65a477dcd40bb1c70b5c4344c Mon Sep 17 00:00:00 2001
From: Max Kainov <max.kainov@clickhouse.com>
Date: Fri, 15 Mar 2024 12:46:15 +0000
Subject: [PATCH 383/443] CI: fix checkout action version

 #do_not_test
---
 .github/workflows/reusable_build.yml | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/.github/workflows/reusable_build.yml b/.github/workflows/reusable_build.yml
index d2fe6f5dbe7..80d78d93e1b 100644
--- a/.github/workflows/reusable_build.yml
+++ b/.github/workflows/reusable_build.yml
@@ -43,8 +43,7 @@ jobs:
     runs-on: [self-hosted, '${{inputs.runner_type}}']
     steps:
       - name: Check out repository code
-        # WIP: temporary try commit with limited perallelization of checkout
-        uses: ClickHouse/checkout@0be3f7b3098bae494d3ef5d29d2e0676fb606232
+        uses: ClickHouse/checkout@v1
         with:
           clear-repository: true
           ref: ${{ fromJson(inputs.data).git_ref }}

From bf8ae84cb6dba368966cb220b4ca5b3c6deef106 Mon Sep 17 00:00:00 2001
From: Max Kainov <max.kainov@clickhouse.com>
Date: Fri, 15 Mar 2024 13:43:03 +0000
Subject: [PATCH 384/443] fixup! CI: skip hdfs tests for arm  #do_not_test 
 #batch_0  #no_merge_commit

---
 tests/integration/helpers/cluster.py                  |  7 +++++++
 .../integration/test_allowed_url_from_config/test.py  | 11 +++++------
 2 files changed, 12 insertions(+), 6 deletions(-)

diff --git a/tests/integration/helpers/cluster.py b/tests/integration/helpers/cluster.py
index 52c0d8a8ee5..194c66fa5c3 100644
--- a/tests/integration/helpers/cluster.py
+++ b/tests/integration/helpers/cluster.py
@@ -1,8 +1,10 @@
 import base64
 import errno
+from functools import cached_property
 import http.client
 import logging
 import os
+import platform
 import stat
 import os.path as p
 import pprint
@@ -4746,3 +4748,8 @@ class ClickHouseKiller(object):
 
     def __exit__(self, exc_type, exc_val, exc_tb):
         self.clickhouse_node.start_clickhouse()
+
+
+@cached_property
+def is_arm():
+    return any(arch in platform.processor().lower() for arch in ("arm, aarch"))
diff --git a/tests/integration/test_allowed_url_from_config/test.py b/tests/integration/test_allowed_url_from_config/test.py
index fb7564ae9d3..df8934aa69b 100644
--- a/tests/integration/test_allowed_url_from_config/test.py
+++ b/tests/integration/test_allowed_url_from_config/test.py
@@ -1,6 +1,5 @@
-import platform
 import pytest
-from helpers.cluster import ClickHouseCluster
+from helpers.cluster import ClickHouseCluster, is_arm
 
 cluster = ClickHouseCluster(__file__)
 node1 = cluster.add_instance("node1", main_configs=["configs/config_with_hosts.xml"])
@@ -18,7 +17,7 @@ node5 = cluster.add_instance(
 )
 node6 = cluster.add_instance("node6", main_configs=["configs/config_for_remote.xml"])
 
-if platform.processor() != "arm":
+if not is_arm():
     node7 = cluster.add_instance(
         "node7", main_configs=["configs/config_for_redirect.xml"], with_hdfs=True
     )
@@ -273,7 +272,7 @@ def test_table_function_remote(start_cluster):
     )
 
 
-@pytest.mark.skipif(platform.processor() == "arm", reason="skip for ARM")
+@pytest.mark.skipif(is_arm(), reason="skip for ARM")
 def test_redirect(start_cluster):
     hdfs_api = start_cluster.hdfs_api
 
@@ -288,7 +287,7 @@ def test_redirect(start_cluster):
     node7.query("DROP TABLE table_test_7_1")
 
 
-@pytest.mark.skipif(platform.processor() == "arm", reason="skip for ARM")
+@pytest.mark.skipif(is_arm(), reason="skip for ARM")
 def test_HDFS(start_cluster):
     assert "not allowed" in node7.query_and_get_error(
         "CREATE TABLE table_test_7_2 (word String) ENGINE=HDFS('http://hdfs1:50075/webhdfs/v1/simple_storage?op=OPEN&namenoderpcaddress=hdfs1:9000&offset=0', 'CSV')"
@@ -298,7 +297,7 @@ def test_HDFS(start_cluster):
     )
 
 
-@pytest.mark.skipif(platform.processor() == "arm", reason="skip for ARM")
+@pytest.mark.skipif(is_arm(), reason="skip for ARM")
 def test_schema_inference(start_cluster):
     error = node7.query_and_get_error("desc url('http://test.com`, 'TSVRaw'')")
     assert error.find("ReadWriteBufferFromHTTPBase") == -1

From ff456ffb3399a024657930623b377f91def8ad61 Mon Sep 17 00:00:00 2001
From: Max K <max.kainov@clickhouse.com>
Date: Fri, 15 Mar 2024 15:08:03 +0100
Subject: [PATCH 385/443] Revert "CI: ARM integration tests: disable tests with
 HDFS "

---
 tests/integration/helpers/cluster.py                |  7 -------
 .../test_allowed_url_from_config/test.py            | 13 ++++---------
 2 files changed, 4 insertions(+), 16 deletions(-)

diff --git a/tests/integration/helpers/cluster.py b/tests/integration/helpers/cluster.py
index 194c66fa5c3..52c0d8a8ee5 100644
--- a/tests/integration/helpers/cluster.py
+++ b/tests/integration/helpers/cluster.py
@@ -1,10 +1,8 @@
 import base64
 import errno
-from functools import cached_property
 import http.client
 import logging
 import os
-import platform
 import stat
 import os.path as p
 import pprint
@@ -4748,8 +4746,3 @@ class ClickHouseKiller(object):
 
     def __exit__(self, exc_type, exc_val, exc_tb):
         self.clickhouse_node.start_clickhouse()
-
-
-@cached_property
-def is_arm():
-    return any(arch in platform.processor().lower() for arch in ("arm, aarch"))
diff --git a/tests/integration/test_allowed_url_from_config/test.py b/tests/integration/test_allowed_url_from_config/test.py
index df8934aa69b..3106cf12702 100644
--- a/tests/integration/test_allowed_url_from_config/test.py
+++ b/tests/integration/test_allowed_url_from_config/test.py
@@ -1,5 +1,5 @@
 import pytest
-from helpers.cluster import ClickHouseCluster, is_arm
+from helpers.cluster import ClickHouseCluster
 
 cluster = ClickHouseCluster(__file__)
 node1 = cluster.add_instance("node1", main_configs=["configs/config_with_hosts.xml"])
@@ -16,11 +16,9 @@ node5 = cluster.add_instance(
     "node5", main_configs=["configs/config_without_allowed_hosts.xml"]
 )
 node6 = cluster.add_instance("node6", main_configs=["configs/config_for_remote.xml"])
-
-if not is_arm():
-    node7 = cluster.add_instance(
-        "node7", main_configs=["configs/config_for_redirect.xml"], with_hdfs=True
-    )
+node7 = cluster.add_instance(
+    "node7", main_configs=["configs/config_for_redirect.xml"], with_hdfs=True
+)
 
 
 @pytest.fixture(scope="module")
@@ -272,7 +270,6 @@ def test_table_function_remote(start_cluster):
     )
 
 
-@pytest.mark.skipif(is_arm(), reason="skip for ARM")
 def test_redirect(start_cluster):
     hdfs_api = start_cluster.hdfs_api
 
@@ -287,7 +284,6 @@ def test_redirect(start_cluster):
     node7.query("DROP TABLE table_test_7_1")
 
 
-@pytest.mark.skipif(is_arm(), reason="skip for ARM")
 def test_HDFS(start_cluster):
     assert "not allowed" in node7.query_and_get_error(
         "CREATE TABLE table_test_7_2 (word String) ENGINE=HDFS('http://hdfs1:50075/webhdfs/v1/simple_storage?op=OPEN&namenoderpcaddress=hdfs1:9000&offset=0', 'CSV')"
@@ -297,7 +293,6 @@ def test_HDFS(start_cluster):
     )
 
 
-@pytest.mark.skipif(is_arm(), reason="skip for ARM")
 def test_schema_inference(start_cluster):
     error = node7.query_and_get_error("desc url('http://test.com`, 'TSVRaw'')")
     assert error.find("ReadWriteBufferFromHTTPBase") == -1

From 061cd5a53d141136adac7606a68042ce8355afe7 Mon Sep 17 00:00:00 2001
From: Dmitry Novik <n0vik@clickhouse.com>
Date: Fri, 15 Mar 2024 14:22:46 +0000
Subject: [PATCH 386/443] Fixup #ci_set_analyzer

---
 src/Storages/StorageMerge.cpp | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/src/Storages/StorageMerge.cpp b/src/Storages/StorageMerge.cpp
index 52362eb5cb8..4a3035dffce 100644
--- a/src/Storages/StorageMerge.cpp
+++ b/src/Storages/StorageMerge.cpp
@@ -1056,7 +1056,7 @@ QueryPipelineBuilderPtr ReadFromMerge::createSources(
             String table_column = table_alias.empty() || processed_stage == QueryProcessingStage::FetchColumns ? "_table" : table_alias + "._table";
 
             if (has_database_virtual_column && common_header.has(database_column)
-                && (storage_stage == QueryProcessingStage::FetchColumns || (dynamic_cast<const StorageDistributed *>(&storage_snapshot_->storage) != nullptr && !pipe_header.has("'" + database_name + "'_String"))))
+                && (storage_stage == QueryProcessingStage::FetchColumns || !pipe_header.has("'" + database_name + "'_String")))
             {
                 ColumnWithTypeAndName column;
                 column.name = database_column;
@@ -1072,7 +1072,7 @@ QueryPipelineBuilderPtr ReadFromMerge::createSources(
             }
 
             if (has_table_virtual_column && common_header.has(table_column)
-                && (storage_stage == QueryProcessingStage::FetchColumns || (dynamic_cast<const StorageDistributed *>(&storage_snapshot_->storage) != nullptr && !pipe_header.has("'" + table_name + "'_String"))))
+                && (storage_stage == QueryProcessingStage::FetchColumns || !pipe_header.has("'" + table_name + "'_String")))
             {
                 ColumnWithTypeAndName column;
                 column.name = table_column;
@@ -1121,7 +1121,7 @@ QueryPipelineBuilderPtr ReadFromMerge::createSources(
         /// Subordinary tables could have different but convertible types, like numeric types of different width.
         /// We must return streams with structure equals to structure of Merge table.
         convertAndFilterSourceStream(
-            header, modified_query_info, storage_snapshot_, aliases, row_policy_data_opt, context, *builder, processed_stage);
+            header, modified_query_info, storage_snapshot_, aliases, row_policy_data_opt, context, *builder, storage_stage);
     }
 
     return builder;

From 4d6aeaa151a7789fdcc8c1d250d1ed780fbd89c4 Mon Sep 17 00:00:00 2001
From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com>
Date: Fri, 15 Mar 2024 15:31:58 +0100
Subject: [PATCH 387/443] Reload CI


From f67eae6d7b8f3ecbe2becb7aa43a8d83431142f8 Mon Sep 17 00:00:00 2001
From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com>
Date: Fri, 15 Mar 2024 15:50:17 +0100
Subject: [PATCH 388/443] Update memory.md

---
 .../engines/table-engines/special/memory.md   | 40 ++++++++++++++-----
 1 file changed, 31 insertions(+), 9 deletions(-)

diff --git a/docs/en/engines/table-engines/special/memory.md b/docs/en/engines/table-engines/special/memory.md
index 8ebdd57ae67..19b5c798a76 100644
--- a/docs/en/engines/table-engines/special/memory.md
+++ b/docs/en/engines/table-engines/special/memory.md
@@ -22,8 +22,7 @@ Normally, using this table engine is not justified. However, it can be used for
 
 The Memory engine is used by the system for temporary tables with external query data (see the section “External data for processing a query”), and for implementing `GLOBAL IN` (see the section “IN operators”).
 
-Upper and lower bounds can be specified to limit Memory engine table size, effectively allowing it to act as a circular
-buffer (see [Engine Parameters](#engine-parameters)).
+Upper and lower bounds can be specified to limit Memory engine table size, effectively allowing it to act as a circular buffer (see [Engine Parameters](#engine-parameters)).
 
 ## Engine Parameters {#engine-parameters}
 
@@ -40,9 +39,6 @@ buffer (see [Engine Parameters](#engine-parameters)).
 
 ## Usage {#usage}
 
-``` sql
-CREATE TABLE memory (i UInt32) ENGINE = Memory;
-```
 
 **Initialize settings**
 ``` sql
@@ -56,16 +52,42 @@ CREATE TABLE memory (i UInt32) ENGINE = Memory SETTINGS min_rows_to_keep = 100,
 CREATE TABLE memory (i UInt32) ENGINE = Memory SETTINGS min_bytes_to_keep = 4096, max_bytes_to_keep = 16384;
 
 /* 1. testing oldest block doesn't get deleted due to min-threshold - 3000 rows */
-INSERT INTO memory SELECT * FROM numbers(0, 1600);
+INSERT INTO memory SELECT * FROM numbers(0, 1600); -- 8'192 bytes
 
 /* 2. adding block that doesn't get deleted */
-INSERT INTO memory SELECT * FROM numbers(1000, 100);
+INSERT INTO memory SELECT * FROM numbers(1000, 100); -- 1'024 bytes
 
 /* 3. testing oldest block gets deleted - 9216 bytes - 1100 */
-INSERT INTO memory SELECT * FROM numbers(9000, 1000);
+INSERT INTO memory SELECT * FROM numbers(9000, 1000); -- 8'192 bytes
 
 /* 4. checking a very large block overrides all */
-INSERT INTO memory SELECT * FROM numbers(9000, 10000);
+INSERT INTO memory SELECT * FROM numbers(9000, 10000); -- 65'536 bytes
+
+SELECT total_bytes, total_rows FROM system.tables WHERE name = 'memory' and database = currentDatabase();
+```
+
+``` text
+┌─total_bytes─┬─total_rows─┐
+│       65536 │      10000 │
+└─────────────┴────────────┘
+```
+
+also, for rows:
+
+``` sql
+CREATE TABLE memory (i UInt32) ENGINE = Memory SETTINGS min_rows_to_keep = 4000, max_rows_to_keep = 10000;
+
+/* 1. testing oldest block doesn't get deleted due to min-threshold - 3000 rows */
+INSERT INTO memory SELECT * FROM numbers(0, 1600); -- 1'600 rows
+
+/* 2. adding block that doesn't get deleted */
+INSERT INTO memory SELECT * FROM numbers(1000, 100); -- 100 rows
+
+/* 3. testing oldest block gets deleted - 9216 bytes - 1100 */
+INSERT INTO memory SELECT * FROM numbers(9000, 1000); -- 1'000 rows
+
+/* 4. checking a very large block overrides all */
+INSERT INTO memory SELECT * FROM numbers(9000, 10000); -- 10'000 rows
 
 SELECT total_bytes, total_rows FROM system.tables WHERE name = 'memory' and database = currentDatabase();
 ```

From 91d18649f4d98b2ebb1c4463f4511ecc30e56255 Mon Sep 17 00:00:00 2001
From: Han Fei <hanfei19910905@gmail.com>
Date: Fri, 15 Mar 2024 16:23:51 +0100
Subject: [PATCH 389/443] update comments

---
 programs/keeper/Keeper.cpp              | 3 +++
 src/Common/CgroupsMemoryUsageObserver.h | 3 ++-
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/programs/keeper/Keeper.cpp b/programs/keeper/Keeper.cpp
index 671dd50215b..9263719da2a 100644
--- a/programs/keeper/Keeper.cpp
+++ b/programs/keeper/Keeper.cpp
@@ -631,6 +631,9 @@ try
         if (wait_time != 0)
         {
             observer.emplace(std::chrono::seconds(wait_time));
+            /// Keeper use setting max_memory_usage_soft_limit to control memory limit. When memory usage
+            /// hits the limit, Keeper will refuse all the request util memory usage is lower.
+            /// We don't call setLimits() here because of this.
             observer->setOnMemoryAmountAvailableChanged([&]()
             {
                 main_config_reloader->reload();
diff --git a/src/Common/CgroupsMemoryUsageObserver.h b/src/Common/CgroupsMemoryUsageObserver.h
index 7ad0f882edd..1da49f8f777 100644
--- a/src/Common/CgroupsMemoryUsageObserver.h
+++ b/src/Common/CgroupsMemoryUsageObserver.h
@@ -17,7 +17,8 @@ namespace DB
 /// 2. Periodically reads the the maximum memory available to the process (which can change due to cgroups settings).
 ///    You can specify a callback to react on changes. The callback typically performs circular logic: It reloads the
 ///    configuration (e.g. server configuration file), which will check the memory amount again and re-calculate
-///    soft/hard limits (see 1.).
+///    soft/hard limits (see 1.) and update the value of setting `max_server_memory_usage` for clickhouse-server or
+///    `max_memory_usage_soft_limit` for clickhouse-keeper.
 #if defined(OS_LINUX)
 class CgroupsMemoryUsageObserver
 {

From 694f5416b9238a2d79b2aa7cc590ee743ca18e9d Mon Sep 17 00:00:00 2001
From: Han Fei <hanfei19910905@gmail.com>
Date: Fri, 15 Mar 2024 16:32:57 +0100
Subject: [PATCH 390/443] refine tests

---
 tests/integration/test_manipulate_statistic/test.py  | 2 +-
 tests/integration/test_memory_limit_observer/test.py | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/tests/integration/test_manipulate_statistic/test.py b/tests/integration/test_manipulate_statistic/test.py
index f1c00a61b07..7b96b392da8 100644
--- a/tests/integration/test_manipulate_statistic/test.py
+++ b/tests/integration/test_manipulate_statistic/test.py
@@ -6,7 +6,7 @@ from helpers.cluster import ClickHouseCluster
 cluster = ClickHouseCluster(__file__)
 
 node1 = cluster.add_instance(
-    "node1", user_configs=["config/config.xml"], with_zookeeper=True
+    "node1", user_configs=["config/config.xml"], with_zookeeper=False
 )
 
 
diff --git a/tests/integration/test_memory_limit_observer/test.py b/tests/integration/test_memory_limit_observer/test.py
index 408aa5e4018..6badb26c538 100644
--- a/tests/integration/test_memory_limit_observer/test.py
+++ b/tests/integration/test_memory_limit_observer/test.py
@@ -50,3 +50,4 @@ def test_observe_memory_limit(started_cluster):
         logging.debug(f"get new memory limit {new_max_mem}")
         if new_max_mem > original_max_mem:
             return
+    raise Exception('the memory limit does not increase as expected')

From 7a8399317398cc32faa42b24c0df9227eef530fd Mon Sep 17 00:00:00 2001
From: vdimir <vdimir@clickhouse.com>
Date: Fri, 15 Mar 2024 13:09:13 +0000
Subject: [PATCH 391/443] Convert test 02998_system_dns_cache_table to smoke

---
 .../02998_system_dns_cache_table.reference    |  3 ++-
 .../02998_system_dns_cache_table.sh           | 26 -------------------
 .../02998_system_dns_cache_table.sql          |  3 +++
 3 files changed, 5 insertions(+), 27 deletions(-)
 delete mode 100755 tests/queries/0_stateless/02998_system_dns_cache_table.sh
 create mode 100644 tests/queries/0_stateless/02998_system_dns_cache_table.sql

diff --git a/tests/queries/0_stateless/02998_system_dns_cache_table.reference b/tests/queries/0_stateless/02998_system_dns_cache_table.reference
index ed6cb000142..600d0bc0b39 100644
--- a/tests/queries/0_stateless/02998_system_dns_cache_table.reference
+++ b/tests/queries/0_stateless/02998_system_dns_cache_table.reference
@@ -1 +1,2 @@
-localhost	127.0.0.1	IPv4	1
+hostname	ip_address	ip_family	cached_at
+String	String	Enum8(\'IPv4\' = 0, \'IPv6\' = 1, \'UNIX_LOCAL\' = 2)	DateTime
diff --git a/tests/queries/0_stateless/02998_system_dns_cache_table.sh b/tests/queries/0_stateless/02998_system_dns_cache_table.sh
deleted file mode 100755
index b74fc00ab3b..00000000000
--- a/tests/queries/0_stateless/02998_system_dns_cache_table.sh
+++ /dev/null
@@ -1,26 +0,0 @@
-#!/usr/bin/env bash
-
-CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
-# shellcheck source=../shell_config.sh
-. "$CURDIR"/../shell_config.sh
-
-# Retries are necessary because the DNS cache may be flushed before second statement is executed
-i=0
-retries=5
-while [[ $i -lt $retries ]]; do
-    ${CLICKHOUSE_CURL} -sS --fail --data "SELECT * FROM url('http://localhost:8123/ping', CSV, 'auto', headers())" "${CLICKHOUSE_URL}" | grep -oP -q 'Ok.' || continue
-
-    RECORDS=$(${CLICKHOUSE_CURL} -sS --fail --data "SELECT hostname, ip_address, ip_family, (isNotNull(cached_at) AND cached_at > '1970-01-01 00:00:00') FROM system.dns_cache WHERE hostname = 'localhost' and ip_family = 'IPv4';" "${CLICKHOUSE_URL}")
-
-    if [[ -n "${RECORDS}" ]]; then
-       echo "${RECORDS}"
-       exit 0
-    fi
-
-    ((++i))
-    sleep 0.2
-done
-
-echo "All tries to fetch entries for localhost failed, no rows returned.
-Probably the DNS cache is disabled or the ClickHouse instance not responds to ping."
-exit 1
diff --git a/tests/queries/0_stateless/02998_system_dns_cache_table.sql b/tests/queries/0_stateless/02998_system_dns_cache_table.sql
new file mode 100644
index 00000000000..0ceb3d8a95a
--- /dev/null
+++ b/tests/queries/0_stateless/02998_system_dns_cache_table.sql
@@ -0,0 +1,3 @@
+SELECT hostname, ip_address, ip_family, cached_at FROM system.dns_cache
+LIMIT 0
+FORMAT TSVWithNamesAndTypes;

From 9bb697eb171f3e4d2baed381591fae5493beb80e Mon Sep 17 00:00:00 2001
From: Vitaly Baranov <vitlibar@clickhouse.com>
Date: Fri, 15 Mar 2024 17:43:48 +0100
Subject: [PATCH 392/443] Fix finishing a failed RESTORE.

---
 src/Backups/RestorerFromBackup.cpp | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/src/Backups/RestorerFromBackup.cpp b/src/Backups/RestorerFromBackup.cpp
index bae2f9aaa25..87c143f0fe2 100644
--- a/src/Backups/RestorerFromBackup.cpp
+++ b/src/Backups/RestorerFromBackup.cpp
@@ -101,10 +101,12 @@ RestorerFromBackup::RestorerFromBackup(
 
 RestorerFromBackup::~RestorerFromBackup()
 {
-    if (!futures.empty())
+    /// If an exception occurs we can come here to the destructor having some tasks still unfinished.
+    /// We have to wait until they finish.
+    if (getNumFutures() > 0)
     {
-        LOG_ERROR(log, "RestorerFromBackup must not be destroyed while {} tasks are still running", futures.size());
-        chassert(false && "RestorerFromBackup must not be destroyed while some tasks are still running");
+        LOG_INFO(log, "Waiting for {} tasks to finish", getNumFutures());
+        waitFutures();
     }
 }
 

From 4dbf3ee1bb8bf4339fcb0fa5da680f84a2cfed34 Mon Sep 17 00:00:00 2001
From: William Schoeffel <wschoeffel@securityscorecard.io>
Date: Fri, 15 Mar 2024 14:05:24 -0300
Subject: [PATCH 393/443] fix: remove duplicated check containing part

---
 src/Storages/MergeTree/MergeTreeData.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Storages/MergeTree/MergeTreeData.cpp b/src/Storages/MergeTree/MergeTreeData.cpp
index aacac5366a9..a22a1904345 100644
--- a/src/Storages/MergeTree/MergeTreeData.cpp
+++ b/src/Storages/MergeTree/MergeTreeData.cpp
@@ -6143,7 +6143,7 @@ MergeTreeData::MutableDataPartsVector MergeTreeData::tryLoadPartsToAttach(const
 
             LOG_DEBUG(log, "Found containing part {} for part {}", containing_part, part_info.dir_name);
 
-            if (!containing_part.empty() && containing_part != part_info.dir_name)
+            if (containing_part != part_info.dir_name)
                 part_info.disk->moveDirectory(fs::path(relative_data_path) / source_dir / part_info.dir_name,
                     fs::path(relative_data_path) / source_dir / ("inactive_" + part_info.dir_name));
             else

From c97731fb8c7ac9cab37445de41a4bf04838c4d77 Mon Sep 17 00:00:00 2001
From: Antonio Andelic <antonio@clickhouse.com>
Date: Fri, 15 Mar 2024 18:46:23 +0100
Subject: [PATCH 394/443] Correctly process last stacktrace

---
 .../postprocess-traces/postprocess-traces.pl  | 25 ++++++++++++-------
 1 file changed, 16 insertions(+), 9 deletions(-)

diff --git a/utils/postprocess-traces/postprocess-traces.pl b/utils/postprocess-traces/postprocess-traces.pl
index 476fb46418f..3e50f64d864 100755
--- a/utils/postprocess-traces/postprocess-traces.pl
+++ b/utils/postprocess-traces/postprocess-traces.pl
@@ -8,6 +8,19 @@ use Data::Dumper;
 my @current_stack = ();
 my $grouped_stacks = {};
 
+sub process_stacktrace
+{
+    my $group = \$grouped_stacks;
+    for my $frame (reverse @current_stack)
+    {
+        $$group->{count} ||= 0;
+        ++$$group->{count};
+        $group = \$$group->{children}{$frame};
+    }
+
+    @current_stack = ();
+}
+
 while (my $line = <>)
 {
     chomp $line;
@@ -21,18 +34,12 @@ while (my $line = <>)
 
     if ($line eq '')
     {
-        my $group = \$grouped_stacks;
-        for my $frame (reverse @current_stack)
-        {
-            $$group->{count} ||= 0;
-            ++$$group->{count};
-            $group = \$$group->{children}{$frame};
-        }
-
-        @current_stack = ();
+        process_stacktrace();
     }
 }
 
+process_stacktrace();
+
 sub print_group
 {
     my $group = shift;

From 8be4a3d7a569d6ea59ebfa4409c5ce13567b4a09 Mon Sep 17 00:00:00 2001
From: Nikita Taranov <nikita.taranov@clickhouse.com>
Date: Fri, 15 Mar 2024 20:28:46 +0100
Subject: [PATCH 395/443] Update BackupsWorker.cpp

---
 src/Backups/BackupsWorker.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Backups/BackupsWorker.cpp b/src/Backups/BackupsWorker.cpp
index 1b4279ca9a7..d0853300edb 100644
--- a/src/Backups/BackupsWorker.cpp
+++ b/src/Backups/BackupsWorker.cpp
@@ -327,7 +327,7 @@ public:
                 metric_active_threads = CurrentMetrics::RestoreThreadsActive;
                 metric_active_threads = CurrentMetrics::RestoreThreadsScheduled;
                 max_threads = num_restore_threads;
-                use_queue = (thread_pool_id != ThreadPoolId::RESTORE);
+                use_queue = true;
                 break;
             }
         }

From e1730c070ea815657463e8ba4ac34d41ef52d6a2 Mon Sep 17 00:00:00 2001
From: Han Fei <hanfei19910905@gmail.com>
Date: Fri, 15 Mar 2024 21:00:42 +0100
Subject: [PATCH 396/443] fix style

---
 tests/integration/test_memory_limit_observer/test.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/integration/test_memory_limit_observer/test.py b/tests/integration/test_memory_limit_observer/test.py
index 6badb26c538..fe3acd9a0cf 100644
--- a/tests/integration/test_memory_limit_observer/test.py
+++ b/tests/integration/test_memory_limit_observer/test.py
@@ -50,4 +50,4 @@ def test_observe_memory_limit(started_cluster):
         logging.debug(f"get new memory limit {new_max_mem}")
         if new_max_mem > original_max_mem:
             return
-    raise Exception('the memory limit does not increase as expected')
+    raise Exception("the memory limit does not increase as expected")

From bc6cd6e769c165a2aaffd06a2960bbce3cb616f0 Mon Sep 17 00:00:00 2001
From: Anton Popov <anton@clickhouse.com>
Date: Fri, 15 Mar 2024 22:30:49 +0000
Subject: [PATCH 397/443] fix test_polymorphic_parts

---
 tests/integration/test_polymorphic_parts/test.py | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/tests/integration/test_polymorphic_parts/test.py b/tests/integration/test_polymorphic_parts/test.py
index 01bc4804c9f..b91a72c5534 100644
--- a/tests/integration/test_polymorphic_parts/test.py
+++ b/tests/integration/test_polymorphic_parts/test.py
@@ -332,7 +332,13 @@ def test_different_part_types_on_replicas(start_cluster, table, part_type):
     for _ in range(3):
         insert_random_data(table, leader, 100)
 
-    leader.query("OPTIMIZE TABLE {} FINAL".format(table))
+    exec_query_with_retry(
+        leader,
+        "OPTIMIZE TABLE {} FINAL".format(table),
+        settings={"optimize_throw_if_noop": 1},
+        silent=True,
+    )
+
     follower.query("SYSTEM SYNC REPLICA {}".format(table), timeout=20)
 
     expected = "{}\t1\n".format(part_type)

From 3a8974e7e1bbad9d6ae49c3624609b69eedf4b1c Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 16 Mar 2024 16:23:12 +0100
Subject: [PATCH 398/443] A definitive guide to CAST

---
 src/Functions/formatDateTime.cpp              |   4 +-
 src/Functions/parseDateTime.cpp               |   2 +-
 .../03011_definitive_guide_to_cast.reference  |  56 ++++
 .../03011_definitive_guide_to_cast.sql        | 252 ++++++++++++++++++
 4 files changed, 311 insertions(+), 3 deletions(-)
 create mode 100644 tests/queries/0_stateless/03011_definitive_guide_to_cast.reference
 create mode 100644 tests/queries/0_stateless/03011_definitive_guide_to_cast.sql

diff --git a/src/Functions/formatDateTime.cpp b/src/Functions/formatDateTime.cpp
index 01ef2a733c8..87438365901 100644
--- a/src/Functions/formatDateTime.cpp
+++ b/src/Functions/formatDateTime.cpp
@@ -1832,10 +1832,10 @@ using FunctionFromUnixTimestampInJodaSyntax = FunctionFormatDateTimeImpl<NameFro
 REGISTER_FUNCTION(FormatDateTime)
 {
     factory.registerFunction<FunctionFormatDateTime>();
-    factory.registerAlias("DATE_FORMAT", FunctionFormatDateTime::name);
+    factory.registerAlias("DATE_FORMAT", FunctionFormatDateTime::name, FunctionFactory::CaseInsensitive);
 
     factory.registerFunction<FunctionFromUnixTimestamp>();
-    factory.registerAlias("FROM_UNIXTIME", FunctionFromUnixTimestamp::name);
+    factory.registerAlias("FROM_UNIXTIME", FunctionFromUnixTimestamp::name, FunctionFactory::CaseInsensitive);
 
     factory.registerFunction<FunctionFormatDateTimeInJodaSyntax>();
     factory.registerFunction<FunctionFromUnixTimestampInJodaSyntax>();
diff --git a/src/Functions/parseDateTime.cpp b/src/Functions/parseDateTime.cpp
index 18882177c90..7a0d7c75774 100644
--- a/src/Functions/parseDateTime.cpp
+++ b/src/Functions/parseDateTime.cpp
@@ -1942,7 +1942,7 @@ namespace
 REGISTER_FUNCTION(ParseDateTime)
 {
     factory.registerFunction<FunctionParseDateTime>();
-    factory.registerAlias("TO_UNIXTIME", FunctionParseDateTime::name);
+    factory.registerAlias("TO_UNIXTIME", FunctionParseDateTime::name, FunctionFactory::CaseInsensitive);
     factory.registerFunction<FunctionParseDateTimeOrZero>();
     factory.registerFunction<FunctionParseDateTimeOrNull>();
     factory.registerAlias("str_to_date", FunctionParseDateTimeOrNull::name, FunctionFactory::CaseInsensitive);
diff --git a/tests/queries/0_stateless/03011_definitive_guide_to_cast.reference b/tests/queries/0_stateless/03011_definitive_guide_to_cast.reference
new file mode 100644
index 00000000000..f8f37fa7807
--- /dev/null
+++ b/tests/queries/0_stateless/03011_definitive_guide_to_cast.reference
@@ -0,0 +1,56 @@
+123
+2009-02-14 00:31:30
+[1,2,3]
+123	Nullable(UInt8)
+\N	Nullable(UInt8)
+1
+255
+['Hello','wo\'rld\\']
+Hello	wo\\\'rld\\\\
+wo\'rld\\	wo\\\'rld\\\\
+123
+123
+123
+1	-1
+[]	[]	Array(Nothing)	Array(Array(Array(Tuple(UInt64, String))))
+123
+123
+123
+123
+123
+123
+123
+String	123
+123	UInt8
+200	UInt8
+123
+123
+1.1
+1.10000000000000016387
+18446744073709551615
+[1.1,2.3]
+[1.10000000000000016387,2.29999999999999967236]
+Row 1:
+──────
+CAST('1.1', 'Decimal(30, 20)'): 1.1
+CAST('1.1', 'Decimal(30, 20)'): 1.1
+CAST(plus(1, 1), 'UInt8'):      2
+-1
+\N
+0
+255
+123
+Hello\0\0\0\0\0
+Hello\0\0\0\0\0
+123.45
+2024-04-25 01:02:03
+2024-04-25 01:02:03.000000
+2024-04-25 01:02:03
+2024-04-25 01:02:03.000000
+2024-03-16 16:22:33
+2024-03-16 16:22:33
+2024-04-25	2024-01-01 02:03:04	1	12
+2024-04-25	2024-01-01 02:03:04.000000	2009-02-14 00:31:30
+2024-04-25	2024-01-01 02:03:04.000000	2009-02-14 00:31:30
+1986-04-25 13:00:00
+14
diff --git a/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql b/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql
new file mode 100644
index 00000000000..771123b153a
--- /dev/null
+++ b/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql
@@ -0,0 +1,252 @@
+SET session_timezone = 'Europe/Amsterdam';
+
+-- Type conversion functions and operators.
+
+
+-- 1. SQL standard CAST operator: `CAST(value AS Type)`.
+
+SELECT CAST(123 AS String);
+
+-- It convert between various data types, including parameterized data types
+
+SELECT CAST(1234567890 AS DateTime('Europe/Amsterdam'));
+
+-- and composite data types:
+
+SELECT CAST('[1, 2, 3]' AS Array(UInt8));
+
+-- It's return type depends on the setting `cast_keep_nullable`. If it is enabled, if the source argument type is Nullable, the resulting data type will be also Nullable, even if it is not written explicitly:
+
+SET cast_keep_nullable = 1;
+SELECT CAST(x AS UInt8) AS y, toTypeName(y) FROM VALUES('x Nullable(String)', ('123'), ('NULL'));
+
+SET cast_keep_nullable = 0;
+SELECT CAST(x AS UInt8) AS y, toTypeName(y) FROM VALUES('x Nullable(String)', ('123'), ('NULL')); -- { serverError CANNOT_PARSE_TEXT }
+
+-- There are various type conversion rules, some worth noting.
+
+-- Conversion between numeric types can involve implementation defined overflow:
+
+SELECT CAST(257 AS UInt8);
+SELECT CAST(-1 AS UInt8);
+
+-- Conversion from string acts like parsing, and for composite data types like Array, Tuple, it works in the same way as from the `Values` data format:
+
+SELECT CAST($$['Hello', 'wo\'rld\\']$$ AS Array(String));
+
+-- '
+-- While for simple data types it does not interpret escape sequences:
+
+SELECT arrayJoin(CAST($$['Hello', 'wo\'rld\\']$$ AS Array(String))) AS x, CAST($$wo\'rld\\$$ AS FixedString(9)) AS y;
+
+-- The operator is case-insensitive:
+
+SELECT CAST(123 AS String);
+SELECT cast(123 AS String);
+SELECT Cast(123 AS String);
+
+-- Conversion from a floating point value to an integer will involve truncation towards zero:
+
+SELECT CAST(1.9, 'Int64'), CAST(-1.9, 'Int64');
+
+-- Conversion from NULL into a non-Nullable type will throw an exception, as well as conversions from denormal floating point numbers (NaN, inf, -inf) to an integer, or conversion between arrays of different dimensions.
+
+-- However, you might find it amusing that an empty array of Nothing data type can be converted to arrays of any dimensions:
+
+SELECT [] AS x, CAST(x AS Array(Array(Array(Tuple(UInt64, String))))) AS y, toTypeName(x), toTypeName(y);
+
+
+-- 2. The functional form of this operator: `CAST(value, 'Type')`:
+
+SELECT CAST(123, 'String');
+
+-- This form is equivalent. Keep in mind that the type has to be a constant expression:
+
+SELECT CAST(123, 'Str'||'ing'); -- this works.
+
+-- This does not work: SELECT materialize('String') AS type, CAST(123, type);
+
+-- It is also case-insensitive:
+
+SELECT CasT(123, 'String');
+
+-- The functional form exists for the consistency of implementation (as every operator also exists in the functional form and the functional form is represented in the query's Abstract Syntax Tree). Anyway, the functional form also makes sense for users, when they need to construct a data type name from a constant expression, or when they want to generate a query programmatically.
+
+-- It's worth noting that the operator form does not allow to specify the type name as a string literal:
+
+-- This does not work: SELECT CAST(123 AS 'String');
+
+-- By only allowing it as an identifier, either bare word:
+
+SELECT CAST(123 AS String);
+
+-- Or as a MySQL or PostgreSQL quoted identifiers:
+
+SELECT CAST(123 AS `String`);
+SELECT CAST(123 AS "String");
+
+-- While the functional form only allows the type name as a string literal:
+
+SELECT CAST(123, 'String'); -- works
+SELECT CAST(123, String); -- { serverError UNKNOWN_IDENTIFIER }
+
+-- However, you can cheat:
+
+SELECT 'String' AS String, CAST(123, String);
+
+
+-- 3. The internal function `_CAST` which is different from `CAST` only by being not dependent on the value of `cast_keep_nullable` setting and other settings.
+
+-- This is needed when ClickHouse has to persist an expression for future use, like in table definitions, including primary and partition key and other indices.
+
+-- The function is not intended for being used directly. When a user uses a regular `CAST` operator or function in a table definition, it is transparently converted to `_CAST` to persist its behavior. However, the user can still use the internal version directly:
+
+SELECT _CAST(x, 'UInt8') AS y, toTypeName(y) FROM VALUES('x Nullable(String)', ('123'), ('456'));
+
+-- There is no operator form of this function:
+
+--  does not work, here UInt8 is interpreted as an alias for the value:
+SELECT _CAST(123 AS UInt8); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH }
+
+SELECT CAST(123 AS UInt8); -- works
+
+
+-- 4. PostgreSQL-style cast syntax `::`
+
+SELECT 123::String;
+
+-- It has a difference from the `CAST` operator: if it is applied to a simple literal value, instead of performing a type conversion, it invokes the SQL parser directly on the corresponding text fragment of the query. The most important case will be the floating-point and decimal types.
+
+-- In this example, we parse `1.1` as Decimal and not involving any type conversion:
+
+SELECT 1.1::Decimal(30, 20);
+
+-- In this example, `1.1` is first parsed as usual, yielding a Float64 value, and then converted to Decimal, producing a wrong result:
+
+SELECT CAST(1.1 AS Decimal(30, 20));
+
+-- We can change this behavior in the future.
+
+-- Another example:
+
+SELECT -1::UInt64; -- { serverError CANNOT_PARSE_NUMBER }
+
+SELECT CAST(-1 AS UInt64); -- conversion with overflow
+
+-- For composite data types, if a value is a literal, it is parsed directly:
+
+SELECT [1.1, 2.3]::Array(Decimal(30, 20));
+
+-- But if the value contains expressions, the usage of the `::` operator will be equivalent to invoking the CAST operator on the expression:
+
+SELECT [1.1, 2.3 + 0]::Array(Decimal(30, 20));
+
+-- The automatic column name for the result of an application of the `::` operator may be the same as for the result of an application of the CAST operator to a string containing the corresponding fragment of the query or to a corresponding expression:
+
+SELECT 1.1::Decimal(30, 20), CAST('1.1' AS Decimal(30, 20)), (1+1)::UInt8 FORMAT Vertical;
+
+-- The operator has the highest priority among others:
+
+SELECT 1-1::String; -- { serverError ILLEGAL_TYPE_OF_ARGUMENT }
+
+-- But one interesting example is the unary minus. Here the minus is not an operator, but part of the numeric literal:
+
+SELECT -1::String;
+
+-- Here it is an operator:
+
+SELECT 1 AS x, -x::String; -- { serverError ILLEGAL_TYPE_OF_ARGUMENT }
+
+
+-- 5. Accurate casting functions: `accurateCast`, `accurateCastOrNull`, `accurateCastOrDefault`.
+
+-- These functions check if the value is exactly representable in the target data type.
+
+-- The function `accurateCast` performs the conversion or throws an exception if the value is not exactly representable:
+
+SELECT accurateCast(1.123456789, 'Float32'); -- { serverError CANNOT_CONVERT_TYPE }
+
+-- The function `accurateCastOrNull` always wraps the target type into Nullable, and returns NULL if the value is not exactly representable:
+
+SELECT accurateCastOrNull(1.123456789, 'Float32');
+
+-- The function `accurateCastOrDefault` takes an additional parameter, which must be of the target type, and returns it if the value is not exactly representable:
+
+SELECT accurateCastOrDefault(-1, 'UInt64', 0::UInt64);
+
+-- These functions are case-sensitive and there are no corresponding operators:
+
+SELECT ACCURATECAST(1, 'String'); -- { serverError UNKNOWN_FUNCTION }.
+
+
+-- 6. Explicit conversion functions:
+
+-- `toString`, `toFixedString`,
+-- `toUInt8`, `toUInt16`, `toUInt32`, `toUInt64`, `toUInt128`, `toUInt256`,
+-- `toInt8`, `toInt16`, `toInt32`, `toInt64`, `toInt128`, `toInt256`,
+-- `toFloat32`, `toFloat64`,
+-- `toDecimal32`, `toDecimal64`, `toDecimal128`, `toDecimal256`,
+-- `toDate`, `toDate32`, `toDateTime`, `toDateTime64`,
+-- `toUUID`, `toIPv4`, `toIPv6`,
+-- `toIntervalNanosecond`, `toIntervalMicrosecond`, `toIntervalMillisecond`,
+-- `toIntervalSecond`, `toIntervalMinute`, `toIntervalHour`,
+-- `toIntervalDay`, `toIntervalWeek`, `toIntervalMonth`, `toIntervalQuarter`, `toIntervalYear`
+
+-- These functions work under the same rules as the CAST operator, and can be thought as a elementary implementation parts of that operator. They allow implementation defined overflow while converting between numeric types.
+
+SELECT toUInt8(-1);
+
+-- These are ClickHouse-native conversion functions. They take an argument with the input value, and for some of data types (`FixedString`, `DateTime`, `DateTime64`, `Decimal`s) the subsequent arguments are constant expressions, defining the parameters of these data types, or the rules to interpret the source value.
+
+SELECT toFloat64(123); -- no arguments
+SELECT toFixedString('Hello', 10) FORMAT TSV; -- the parameter of the FixedString data type, the function returns FixedString(10)
+SELECT toFixedString('Hello', 5 + 5) FORMAT TSV; -- it can be a constant expression
+
+SELECT toDecimal32('123.456', 2); -- the scale of the Decimal data type
+
+SELECT toDateTime('2024-04-25 01:02:03', 'Europe/Amsterdam'); -- the time zone of DateTime
+SELECT toDateTime64('2024-04-25 01:02:03', 6, 'Europe/Amsterdam'); -- the scale of DateTime64 and its time zone
+
+-- The length of FixedString and the scale of Decimal and DateTime64 types are mandatory arguments, while the time zone of the DateTime data type is optional.
+
+-- If the time zone is not specified, the time zone of the argument's data type is used, and if the argument is not a date time, the session time zone is used.
+
+SELECT toDateTime('2024-04-25 01:02:03');
+SELECT toDateTime64('2024-04-25 01:02:03', 6);
+
+-- Here the time zone can be specified as the rule of interpreration of the value during conversion:
+
+SELECT toString(now(), 'Europe/Amsterdam');
+SELECT toString(now());
+
+
+-- 7. SQL-compatibility type-defining operators:
+
+SELECT DATE '2024-04-25', TIMESTAMP '2024-01-01 02:03:04', INTERVAL 1 MINUTE, INTERVAL '12 hour';
+
+-- These operators are interpreted as the corresponding explicit conversion functions.
+
+
+-- 8. SQL-compatibility aliases for explicit conversion functions:
+
+SELECT DATE('2024-04-25'), TIMESTAMP('2024-01-01 02:03:04'), FROM_UNIXTIME(1234567890);
+
+-- These functions exist for compatibility with MySQL. They are case-insensive.
+
+SELECT date '2024-04-25', timeSTAMP('2024-01-01 02:03:04'), From_Unixtime(1234567890);
+
+
+-- 9. Specialized conversion functions:
+
+-- `parseDateTimeBestEffort`, `parseDateTimeBestEffortUS`, `parseDateTime64BestEffort`, `parseDateTime64BestEffortUS`, `toUnixTimestamp`
+
+-- These functions are similar to explicit conversion functions, but provide special rules on how the conversion is performed.
+
+SELECT parseDateTimeBestEffort('25 Apr 1986 1pm');
+
+
+-- 10. Functions for converting between different components or rounding of date and time data types.
+
+SELECT toDayOfMonth(toDateTime(1234567890));
+
+-- These functions are coverted in a separate topic.

From f5a13a023bc1d19b688ec98902411252796f96d2 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Sat, 16 Mar 2024 18:58:16 +0100
Subject: [PATCH 399/443] Rename CacheGuard -> CachePriorityGuard (because it
 is more correct)

---
 src/Interpreters/Cache/EvictionCandidates.cpp |  2 +-
 src/Interpreters/Cache/EvictionCandidates.h   |  2 +-
 src/Interpreters/Cache/FileCache.cpp          |  6 ++--
 src/Interpreters/Cache/FileCache.h            |  8 ++---
 src/Interpreters/Cache/Guards.h               | 22 ++++++-------
 src/Interpreters/Cache/IFileCachePriority.h   | 26 +++++++--------
 .../Cache/LRUFileCachePriority.cpp            | 26 +++++++--------
 src/Interpreters/Cache/LRUFileCachePriority.h | 32 +++++++++----------
 src/Interpreters/Cache/QueryLimit.cpp         | 12 +++----
 src/Interpreters/Cache/QueryLimit.h           | 12 +++----
 .../Cache/SLRUFileCachePriority.cpp           | 24 +++++++-------
 .../Cache/SLRUFileCachePriority.h             | 22 ++++++-------
 12 files changed, 97 insertions(+), 97 deletions(-)

diff --git a/src/Interpreters/Cache/EvictionCandidates.cpp b/src/Interpreters/Cache/EvictionCandidates.cpp
index 7dceab4f95f..f1ae2baa347 100644
--- a/src/Interpreters/Cache/EvictionCandidates.cpp
+++ b/src/Interpreters/Cache/EvictionCandidates.cpp
@@ -32,7 +32,7 @@ void EvictionCandidates::add(LockedKey & locked_key, const FileSegmentMetadataPt
     ++candidates_size;
 }
 
-void EvictionCandidates::evict(FileCacheQueryLimit::QueryContext * query_context, const CacheGuard::Lock & lock)
+void EvictionCandidates::evict(FileCacheQueryLimit::QueryContext * query_context, const CachePriorityGuard::Lock & lock)
 {
     if (candidates.empty())
         return;
diff --git a/src/Interpreters/Cache/EvictionCandidates.h b/src/Interpreters/Cache/EvictionCandidates.h
index 0557962d97f..e817d33d5fe 100644
--- a/src/Interpreters/Cache/EvictionCandidates.h
+++ b/src/Interpreters/Cache/EvictionCandidates.h
@@ -11,7 +11,7 @@ public:
 
     void add(LockedKey & locked_key, const FileSegmentMetadataPtr & candidate);
 
-    void evict(FileCacheQueryLimit::QueryContext * query_context, const CacheGuard::Lock &);
+    void evict(FileCacheQueryLimit::QueryContext * query_context, const CachePriorityGuard::Lock &);
 
     size_t size() const { return candidates_size; }
 
diff --git a/src/Interpreters/Cache/FileCache.cpp b/src/Interpreters/Cache/FileCache.cpp
index ea40ffcfa3c..65b6a3a172d 100644
--- a/src/Interpreters/Cache/FileCache.cpp
+++ b/src/Interpreters/Cache/FileCache.cpp
@@ -183,13 +183,13 @@ void FileCache::initialize()
     is_initialized = true;
 }
 
-CacheGuard::Lock FileCache::lockCache() const
+CachePriorityGuard::Lock FileCache::lockCache() const
 {
     ProfileEventTimeIncrement<Microseconds> watch(ProfileEvents::FilesystemCacheLockCacheMicroseconds);
     return cache_guard.lock();
 }
 
-CacheGuard::Lock FileCache::tryLockCache(std::optional<std::chrono::milliseconds> acquire_timeout) const
+CachePriorityGuard::Lock FileCache::tryLockCache(std::optional<std::chrono::milliseconds> acquire_timeout) const
 {
     return acquire_timeout.has_value() ? cache_guard.tryLockFor(acquire_timeout.value()) : cache_guard.tryLock();
 }
@@ -706,7 +706,7 @@ KeyMetadata::iterator FileCache::addFileSegment(
     size_t size,
     FileSegment::State state,
     const CreateFileSegmentSettings & create_settings,
-    const CacheGuard::Lock * lock)
+    const CachePriorityGuard::Lock * lock)
 {
     /// Create a file_segment_metadata and put it in `files` map by [key][offset].
 
diff --git a/src/Interpreters/Cache/FileCache.h b/src/Interpreters/Cache/FileCache.h
index 007c4fd9483..8ea5f4dab40 100644
--- a/src/Interpreters/Cache/FileCache.h
+++ b/src/Interpreters/Cache/FileCache.h
@@ -173,8 +173,8 @@ public:
 
     void deactivateBackgroundOperations();
 
-    CacheGuard::Lock lockCache() const;
-    CacheGuard::Lock tryLockCache(std::optional<std::chrono::milliseconds> acquire_timeout = std::nullopt) const;
+    CachePriorityGuard::Lock lockCache() const;
+    CachePriorityGuard::Lock tryLockCache(std::optional<std::chrono::milliseconds> acquire_timeout = std::nullopt) const;
 
     std::vector<FileSegment::Info> sync();
 
@@ -208,7 +208,7 @@ private:
     CacheMetadata metadata;
 
     FileCachePriorityPtr main_priority;
-    mutable CacheGuard cache_guard;
+    mutable CachePriorityGuard cache_guard;
 
     struct HitsCountStash
     {
@@ -280,7 +280,7 @@ private:
         size_t size,
         FileSegment::State state,
         const CreateFileSegmentSettings & create_settings,
-        const CacheGuard::Lock *);
+        const CachePriorityGuard::Lock *);
 };
 
 }
diff --git a/src/Interpreters/Cache/Guards.h b/src/Interpreters/Cache/Guards.h
index 0ac7cb80483..6193ee38755 100644
--- a/src/Interpreters/Cache/Guards.h
+++ b/src/Interpreters/Cache/Guards.h
@@ -10,17 +10,17 @@ namespace DB
  * 2. KeyGuard::Lock (hold till the end of the method)
  *
  * FileCache::tryReserve
- * 1. CacheGuard::Lock
+ * 1. CachePriorityGuard::Lock
  * 2. KeyGuard::Lock (taken without metadata lock)
  * 3. any number of KeyGuard::Lock's for files which are going to be evicted (taken via metadata lock)
  *
  * FileCache::removeIfExists
- * 1. CacheGuard::Lock
+ * 1. CachePriorityGuard::Lock
  * 2. KeyGuard::Lock (taken via metadata lock)
  * 3. FileSegmentGuard::Lock
  *
  * FileCache::removeAllReleasable
- * 1. CacheGuard::Lock
+ * 1. CachePriorityGuard::Lock
  * 2. any number of KeyGuard::Lock's locks (takken via metadata lock), but at a moment of time only one key lock can be hold
  * 3. FileSegmentGuard::Lock
  *
@@ -34,23 +34,23 @@ namespace DB
  * 2. FileSegmentGuard::Lock
  *
  * FileSegment::complete
- * 1. CacheGuard::Lock
+ * 1. CachePriorityGuard::Lock
  * 2. KeyGuard::Lock (taken without metadata lock)
  * 3. FileSegmentGuard::Lock
  *
  * Rules:
- * 1. Priority of locking: CacheGuard::Lock > CacheMetadataGuard::Lock > KeyGuard::Lock > FileSegmentGuard::Lock
- * 2. If we take more than one key lock at a moment of time, we need to take CacheGuard::Lock (example: tryReserve())
+ * 1. Priority of locking: CachePriorityGuard::Lock > CacheMetadataGuard::Lock > KeyGuard::Lock > FileSegmentGuard::Lock
+ * 2. If we take more than one key lock at a moment of time, we need to take CachePriorityGuard::Lock (example: tryReserve())
  *
  *
- *                                 _CacheGuard_
+ *                                 _CachePriorityGuard_
  *                                 1. FileCache::tryReserve
  *                                 2. FileCache::removeIfExists(key)
  *                                 3. FileCache::removeAllReleasable
  *                                 4. FileSegment::complete
  *
  *             _KeyGuard_                                      _CacheMetadataGuard_
- *             1. all from CacheGuard                          1. getOrSet/get/set
+ *             1. all from CachePriorityGuard                          1. getOrSet/get/set
  *             2. getOrSet/get/Set
  *
  * *This table does not include locks taken for introspection and system tables.
@@ -59,11 +59,11 @@ namespace DB
 /**
  * Cache priority queue guard.
  */
-struct CacheGuard : private boost::noncopyable
+struct CachePriorityGuard : private boost::noncopyable
 {
     using Mutex = std::timed_mutex;
-    /// struct is used (not keyword `using`) to make CacheGuard::Lock non-interchangable with other guards locks
-    /// so, we wouldn't be able to pass CacheGuard::Lock to a function which accepts KeyGuard::Lock, for example
+    /// struct is used (not keyword `using`) to make CachePriorityGuard::Lock non-interchangable with other guards locks
+    /// so, we wouldn't be able to pass CachePriorityGuard::Lock to a function which accepts KeyGuard::Lock, for example
     struct Lock : public std::unique_lock<Mutex>
     {
         using Base = std::unique_lock<Mutex>;
diff --git a/src/Interpreters/Cache/IFileCachePriority.h b/src/Interpreters/Cache/IFileCachePriority.h
index bc036166940..58011780323 100644
--- a/src/Interpreters/Cache/IFileCachePriority.h
+++ b/src/Interpreters/Cache/IFileCachePriority.h
@@ -43,11 +43,11 @@ public:
 
         virtual EntryPtr getEntry() const = 0;
 
-        virtual size_t increasePriority(const CacheGuard::Lock &) = 0;
+        virtual size_t increasePriority(const CachePriorityGuard::Lock &) = 0;
 
         virtual void updateSize(int64_t size) = 0;
 
-        virtual void remove(const CacheGuard::Lock &) = 0;
+        virtual void remove(const CachePriorityGuard::Lock &) = 0;
 
         virtual void invalidate() = 0;
 
@@ -57,13 +57,13 @@ public:
 
     virtual ~IFileCachePriority() = default;
 
-    size_t getElementsLimit(const CacheGuard::Lock &) const { return max_elements; }
+    size_t getElementsLimit(const CachePriorityGuard::Lock &) const { return max_elements; }
 
-    size_t getSizeLimit(const CacheGuard::Lock &) const { return max_size; }
+    size_t getSizeLimit(const CachePriorityGuard::Lock &) const { return max_size; }
 
-    virtual size_t getSize(const CacheGuard::Lock &) const = 0;
+    virtual size_t getSize(const CachePriorityGuard::Lock &) const = 0;
 
-    virtual size_t getElementsCount(const CacheGuard::Lock &) const = 0;
+    virtual size_t getElementsCount(const CachePriorityGuard::Lock &) const = 0;
 
     /// Throws exception if there is not enough size to fit it.
     virtual IteratorPtr add( /// NOLINT
@@ -71,7 +71,7 @@ public:
         size_t offset,
         size_t size,
         const UserInfo & user,
-        const CacheGuard::Lock &,
+        const CachePriorityGuard::Lock &,
         bool best_effort = false) = 0;
 
     /// `reservee` is the entry for which are reserving now.
@@ -79,11 +79,11 @@ public:
     /// for the corresponding file segment.
     virtual bool canFit( /// NOLINT
         size_t size,
-        const CacheGuard::Lock &,
+        const CachePriorityGuard::Lock &,
         IteratorPtr reservee = nullptr,
         bool best_effort = false) const = 0;
 
-    virtual void shuffle(const CacheGuard::Lock &) = 0;
+    virtual void shuffle(const CachePriorityGuard::Lock &) = 0;
 
     struct IPriorityDump
     {
@@ -91,9 +91,9 @@ public:
     };
     using PriorityDumpPtr = std::shared_ptr<IPriorityDump>;
 
-    virtual PriorityDumpPtr dump(const CacheGuard::Lock &) = 0;
+    virtual PriorityDumpPtr dump(const CachePriorityGuard::Lock &) = 0;
 
-    using FinalizeEvictionFunc = std::function<void(const CacheGuard::Lock & lk)>;
+    using FinalizeEvictionFunc = std::function<void(const CachePriorityGuard::Lock & lk)>;
     virtual bool collectCandidatesForEviction(
         size_t size,
         FileCacheReserveStat & stat,
@@ -101,9 +101,9 @@ public:
         IFileCachePriority::IteratorPtr reservee,
         FinalizeEvictionFunc & finalize_eviction_func,
         const UserID & user_id,
-        const CacheGuard::Lock &) = 0;
+        const CachePriorityGuard::Lock &) = 0;
 
-    virtual bool modifySizeLimits(size_t max_size_, size_t max_elements_, double size_ratio_, const CacheGuard::Lock &) = 0;
+    virtual bool modifySizeLimits(size_t max_size_, size_t max_elements_, double size_ratio_, const CachePriorityGuard::Lock &) = 0;
 
 protected:
     IFileCachePriority(size_t max_size_, size_t max_elements_);
diff --git a/src/Interpreters/Cache/LRUFileCachePriority.cpp b/src/Interpreters/Cache/LRUFileCachePriority.cpp
index bce03b60024..08e65b577ca 100644
--- a/src/Interpreters/Cache/LRUFileCachePriority.cpp
+++ b/src/Interpreters/Cache/LRUFileCachePriority.cpp
@@ -43,13 +43,13 @@ IFileCachePriority::IteratorPtr LRUFileCachePriority::add( /// NOLINT
     size_t offset,
     size_t size,
     const UserInfo &,
-    const CacheGuard::Lock & lock,
+    const CachePriorityGuard::Lock & lock,
     bool)
 {
     return std::make_shared<LRUIterator>(add(std::make_shared<Entry>(key_metadata->key, offset, size, key_metadata), lock));
 }
 
-LRUFileCachePriority::LRUIterator LRUFileCachePriority::add(EntryPtr entry, const CacheGuard::Lock & lock)
+LRUFileCachePriority::LRUIterator LRUFileCachePriority::add(EntryPtr entry, const CachePriorityGuard::Lock & lock)
 {
     if (entry->size == 0)
     {
@@ -93,7 +93,7 @@ LRUFileCachePriority::LRUIterator LRUFileCachePriority::add(EntryPtr entry, cons
     return LRUIterator(this, iterator);
 }
 
-LRUFileCachePriority::LRUQueue::iterator LRUFileCachePriority::remove(LRUQueue::iterator it, const CacheGuard::Lock &)
+LRUFileCachePriority::LRUQueue::iterator LRUFileCachePriority::remove(LRUQueue::iterator it, const CachePriorityGuard::Lock &)
 {
     /// If size is 0, entry is invalidated, current_elements_num was already updated.
     const auto & entry = **it;
@@ -150,7 +150,7 @@ bool LRUFileCachePriority::LRUIterator::operator ==(const LRUIterator & other) c
     return cache_priority == other.cache_priority && iterator == other.iterator;
 }
 
-void LRUFileCachePriority::iterate(IterateFunc && func, const CacheGuard::Lock & lock)
+void LRUFileCachePriority::iterate(IterateFunc && func, const CachePriorityGuard::Lock & lock)
 {
     for (auto it = queue.begin(); it != queue.end();)
     {
@@ -201,7 +201,7 @@ void LRUFileCachePriority::iterate(IterateFunc && func, const CacheGuard::Lock &
 
 bool LRUFileCachePriority::canFit( /// NOLINT
     size_t size,
-    const CacheGuard::Lock & lock,
+    const CachePriorityGuard::Lock & lock,
     IteratorPtr,
     bool) const
 {
@@ -212,7 +212,7 @@ bool LRUFileCachePriority::canFit(
     size_t size,
     size_t released_size_assumption,
     size_t released_elements_assumption,
-    const CacheGuard::Lock &) const
+    const CachePriorityGuard::Lock &) const
 {
     return (max_size == 0 || (state->current_size + size - released_size_assumption <= max_size))
         && (max_elements == 0 || state->current_elements_num + 1 - released_elements_assumption <= max_elements);
@@ -225,7 +225,7 @@ bool LRUFileCachePriority::collectCandidatesForEviction(
     IFileCachePriority::IteratorPtr,
     FinalizeEvictionFunc &,
     const UserID &,
-    const CacheGuard::Lock & lock)
+    const CachePriorityGuard::Lock & lock)
 {
     if (canFit(size, lock))
         return true;
@@ -264,7 +264,7 @@ bool LRUFileCachePriority::collectCandidatesForEviction(
     return can_fit();
 }
 
-LRUFileCachePriority::LRUIterator LRUFileCachePriority::move(LRUIterator & it, LRUFileCachePriority & other, const CacheGuard::Lock &)
+LRUFileCachePriority::LRUIterator LRUFileCachePriority::move(LRUIterator & it, LRUFileCachePriority & other, const CachePriorityGuard::Lock &)
 {
     const auto & entry = *it.getEntry();
     if (entry.size == 0)
@@ -297,7 +297,7 @@ LRUFileCachePriority::LRUIterator LRUFileCachePriority::move(LRUIterator & it, L
     return LRUIterator(this, it.iterator);
 }
 
-IFileCachePriority::PriorityDumpPtr LRUFileCachePriority::dump(const CacheGuard::Lock & lock)
+IFileCachePriority::PriorityDumpPtr LRUFileCachePriority::dump(const CachePriorityGuard::Lock & lock)
 {
     std::vector<FileSegmentInfo> res;
     iterate([&](LockedKey &, const FileSegmentMetadataPtr & segment_metadata)
@@ -309,7 +309,7 @@ IFileCachePriority::PriorityDumpPtr LRUFileCachePriority::dump(const CacheGuard:
 }
 
 bool LRUFileCachePriority::modifySizeLimits(
-    size_t max_size_, size_t max_elements_, double /* size_ratio_ */, const CacheGuard::Lock & lock)
+    size_t max_size_, size_t max_elements_, double /* size_ratio_ */, const CachePriorityGuard::Lock & lock)
 {
     if (max_size == max_size_ && max_elements == max_elements_)
         return false; /// Nothing to change.
@@ -353,7 +353,7 @@ bool LRUFileCachePriority::modifySizeLimits(
     return true;
 }
 
-void LRUFileCachePriority::LRUIterator::remove(const CacheGuard::Lock & lock)
+void LRUFileCachePriority::LRUIterator::remove(const CachePriorityGuard::Lock & lock)
 {
     assertValid();
     cache_priority->remove(iterator, lock);
@@ -389,7 +389,7 @@ void LRUFileCachePriority::LRUIterator::updateSize(int64_t size)
     entry->size += size;
 }
 
-size_t LRUFileCachePriority::LRUIterator::increasePriority(const CacheGuard::Lock &)
+size_t LRUFileCachePriority::LRUIterator::increasePriority(const CachePriorityGuard::Lock &)
 {
     assertValid();
     cache_priority->queue.splice(cache_priority->queue.end(), cache_priority->queue, iterator);
@@ -402,7 +402,7 @@ void LRUFileCachePriority::LRUIterator::assertValid() const
         throw Exception(ErrorCodes::LOGICAL_ERROR, "Attempt to use invalid iterator");
 }
 
-void LRUFileCachePriority::shuffle(const CacheGuard::Lock &)
+void LRUFileCachePriority::shuffle(const CachePriorityGuard::Lock &)
 {
     std::vector<LRUQueue::iterator> its;
     its.reserve(queue.size());
diff --git a/src/Interpreters/Cache/LRUFileCachePriority.h b/src/Interpreters/Cache/LRUFileCachePriority.h
index a74a4b8b621..dcd4ee0a24c 100644
--- a/src/Interpreters/Cache/LRUFileCachePriority.h
+++ b/src/Interpreters/Cache/LRUFileCachePriority.h
@@ -24,13 +24,13 @@ protected:
 public:
     LRUFileCachePriority(size_t max_size_, size_t max_elements_, StatePtr state_ = nullptr);
 
-    size_t getSize(const CacheGuard::Lock &) const override { return state->current_size; }
+    size_t getSize(const CachePriorityGuard::Lock &) const override { return state->current_size; }
 
-    size_t getElementsCount(const CacheGuard::Lock &) const override { return state->current_elements_num; }
+    size_t getElementsCount(const CachePriorityGuard::Lock &) const override { return state->current_elements_num; }
 
     bool canFit( /// NOLINT
         size_t size,
-        const CacheGuard::Lock &,
+        const CachePriorityGuard::Lock &,
         IteratorPtr reservee = nullptr,
         bool best_effort = false) const override;
 
@@ -39,7 +39,7 @@ public:
         size_t offset,
         size_t size,
         const UserInfo & user,
-        const CacheGuard::Lock &,
+        const CachePriorityGuard::Lock &,
         bool best_effort = false) override;
 
     bool collectCandidatesForEviction(
@@ -49,9 +49,9 @@ public:
         IFileCachePriority::IteratorPtr reservee,
         FinalizeEvictionFunc & finalize_eviction_func,
         const UserID & user_id,
-        const CacheGuard::Lock &) override;
+        const CachePriorityGuard::Lock &) override;
 
-    void shuffle(const CacheGuard::Lock &) override;
+    void shuffle(const CachePriorityGuard::Lock &) override;
 
     struct LRUPriorityDump : public IPriorityDump
     {
@@ -59,11 +59,11 @@ public:
         explicit LRUPriorityDump(const std::vector<FileSegmentInfo> & infos_) : infos(infos_) {}
         void merge(const LRUPriorityDump & other) { infos.insert(infos.end(), other.infos.begin(), other.infos.end()); }
     };
-    PriorityDumpPtr dump(const CacheGuard::Lock &) override;
+    PriorityDumpPtr dump(const CachePriorityGuard::Lock &) override;
 
-    void pop(const CacheGuard::Lock & lock) { remove(queue.begin(), lock); }
+    void pop(const CachePriorityGuard::Lock & lock) { remove(queue.begin(), lock); }
 
-    bool modifySizeLimits(size_t max_size_, size_t max_elements_, double size_ratio_, const CacheGuard::Lock &) override;
+    bool modifySizeLimits(size_t max_size_, size_t max_elements_, double size_ratio_, const CachePriorityGuard::Lock &) override;
 
 private:
     class LRUIterator;
@@ -77,9 +77,9 @@ private:
     void updateElementsCount(int64_t num);
     void updateSize(int64_t size);
 
-    bool canFit(size_t size, size_t released_size_assumption, size_t released_elements_assumption, const CacheGuard::Lock &) const;
+    bool canFit(size_t size, size_t released_size_assumption, size_t released_elements_assumption, const CachePriorityGuard::Lock &) const;
 
-    LRUQueue::iterator remove(LRUQueue::iterator it, const CacheGuard::Lock &);
+    LRUQueue::iterator remove(LRUQueue::iterator it, const CachePriorityGuard::Lock &);
 
     enum class IterationResult
     {
@@ -88,10 +88,10 @@ private:
         REMOVE_AND_CONTINUE,
     };
     using IterateFunc = std::function<IterationResult(LockedKey &, const FileSegmentMetadataPtr &)>;
-    void iterate(IterateFunc && func, const CacheGuard::Lock &);
+    void iterate(IterateFunc && func, const CachePriorityGuard::Lock &);
 
-    LRUIterator move(LRUIterator & it, LRUFileCachePriority & other, const CacheGuard::Lock &);
-    LRUIterator add(EntryPtr entry, const CacheGuard::Lock &);
+    LRUIterator move(LRUIterator & it, LRUFileCachePriority & other, const CachePriorityGuard::Lock &);
+    LRUIterator add(EntryPtr entry, const CachePriorityGuard::Lock &);
 };
 
 class LRUFileCachePriority::LRUIterator : public IFileCachePriority::Iterator
@@ -108,9 +108,9 @@ public:
 
     EntryPtr getEntry() const override { return *iterator; }
 
-    size_t increasePriority(const CacheGuard::Lock &) override;
+    size_t increasePriority(const CachePriorityGuard::Lock &) override;
 
-    void remove(const CacheGuard::Lock &) override;
+    void remove(const CachePriorityGuard::Lock &) override;
 
     void invalidate() override;
 
diff --git a/src/Interpreters/Cache/QueryLimit.cpp b/src/Interpreters/Cache/QueryLimit.cpp
index ba4f5017772..9421005dc92 100644
--- a/src/Interpreters/Cache/QueryLimit.cpp
+++ b/src/Interpreters/Cache/QueryLimit.cpp
@@ -16,7 +16,7 @@ static bool isQueryInitialized()
         && !CurrentThread::getQueryId().empty();
 }
 
-FileCacheQueryLimit::QueryContextPtr FileCacheQueryLimit::tryGetQueryContext(const CacheGuard::Lock &)
+FileCacheQueryLimit::QueryContextPtr FileCacheQueryLimit::tryGetQueryContext(const CachePriorityGuard::Lock &)
 {
     if (!isQueryInitialized())
         return nullptr;
@@ -25,7 +25,7 @@ FileCacheQueryLimit::QueryContextPtr FileCacheQueryLimit::tryGetQueryContext(con
     return (query_iter == query_map.end()) ? nullptr : query_iter->second;
 }
 
-void FileCacheQueryLimit::removeQueryContext(const std::string & query_id, const CacheGuard::Lock &)
+void FileCacheQueryLimit::removeQueryContext(const std::string & query_id, const CachePriorityGuard::Lock &)
 {
     auto query_iter = query_map.find(query_id);
     if (query_iter == query_map.end())
@@ -41,7 +41,7 @@ void FileCacheQueryLimit::removeQueryContext(const std::string & query_id, const
 FileCacheQueryLimit::QueryContextPtr FileCacheQueryLimit::getOrSetQueryContext(
     const std::string & query_id,
     const ReadSettings & settings,
-    const CacheGuard::Lock &)
+    const CachePriorityGuard::Lock &)
 {
     if (query_id.empty())
         return nullptr;
@@ -70,7 +70,7 @@ void FileCacheQueryLimit::QueryContext::add(
     size_t offset,
     size_t size,
     const FileCache::UserInfo & user,
-    const CacheGuard::Lock & lock)
+    const CachePriorityGuard::Lock & lock)
 {
     auto it = getPriority().add(key_metadata, offset, size, user, lock);
     auto [_, inserted] = records.emplace(FileCacheKeyAndOffset{key_metadata->key, offset}, it);
@@ -87,7 +87,7 @@ void FileCacheQueryLimit::QueryContext::add(
 void FileCacheQueryLimit::QueryContext::remove(
     const Key & key,
     size_t offset,
-    const CacheGuard::Lock & lock)
+    const CachePriorityGuard::Lock & lock)
 {
     auto record = records.find({key, offset});
     if (record == records.end())
@@ -100,7 +100,7 @@ void FileCacheQueryLimit::QueryContext::remove(
 IFileCachePriority::IteratorPtr FileCacheQueryLimit::QueryContext::tryGet(
     const Key & key,
     size_t offset,
-    const CacheGuard::Lock &)
+    const CachePriorityGuard::Lock &)
 {
     auto it = records.find({key, offset});
     if (it == records.end())
diff --git a/src/Interpreters/Cache/QueryLimit.h b/src/Interpreters/Cache/QueryLimit.h
index 419126601f0..7553eff82ba 100644
--- a/src/Interpreters/Cache/QueryLimit.h
+++ b/src/Interpreters/Cache/QueryLimit.h
@@ -13,14 +13,14 @@ public:
     class QueryContext;
     using QueryContextPtr = std::shared_ptr<QueryContext>;
 
-    QueryContextPtr tryGetQueryContext(const CacheGuard::Lock & lock);
+    QueryContextPtr tryGetQueryContext(const CachePriorityGuard::Lock & lock);
 
     QueryContextPtr getOrSetQueryContext(
         const std::string & query_id,
         const ReadSettings & settings,
-        const CacheGuard::Lock &);
+        const CachePriorityGuard::Lock &);
 
-    void removeQueryContext(const std::string & query_id, const CacheGuard::Lock &);
+    void removeQueryContext(const std::string & query_id, const CachePriorityGuard::Lock &);
 
     class QueryContext
     {
@@ -38,19 +38,19 @@ public:
         Priority::IteratorPtr tryGet(
             const Key & key,
             size_t offset,
-            const CacheGuard::Lock &);
+            const CachePriorityGuard::Lock &);
 
         void add(
             KeyMetadataPtr key_metadata,
             size_t offset,
             size_t size,
             const FileCacheUserInfo & user,
-            const CacheGuard::Lock &);
+            const CachePriorityGuard::Lock &);
 
         void remove(
             const Key & key,
             size_t offset,
-            const CacheGuard::Lock &);
+            const CachePriorityGuard::Lock &);
 
     private:
         using Records = std::unordered_map<FileCacheKeyAndOffset, Priority::IteratorPtr, FileCacheKeyAndOffsetHash>;
diff --git a/src/Interpreters/Cache/SLRUFileCachePriority.cpp b/src/Interpreters/Cache/SLRUFileCachePriority.cpp
index 43f1c1012ba..1767cb94be7 100644
--- a/src/Interpreters/Cache/SLRUFileCachePriority.cpp
+++ b/src/Interpreters/Cache/SLRUFileCachePriority.cpp
@@ -34,19 +34,19 @@ SLRUFileCachePriority::SLRUFileCachePriority(
         probationary_queue.max_size, protected_queue.max_elements);
 }
 
-size_t SLRUFileCachePriority::getSize(const CacheGuard::Lock & lock) const
+size_t SLRUFileCachePriority::getSize(const CachePriorityGuard::Lock & lock) const
 {
     return protected_queue.getSize(lock) + probationary_queue.getSize(lock);
 }
 
-size_t SLRUFileCachePriority::getElementsCount(const CacheGuard::Lock & lock) const
+size_t SLRUFileCachePriority::getElementsCount(const CachePriorityGuard::Lock & lock) const
 {
     return protected_queue.getElementsCount(lock) + probationary_queue.getElementsCount(lock);
 }
 
 bool SLRUFileCachePriority::canFit( /// NOLINT
     size_t size,
-    const CacheGuard::Lock & lock,
+    const CachePriorityGuard::Lock & lock,
     IteratorPtr reservee,
     bool best_effort) const
 {
@@ -70,7 +70,7 @@ IFileCachePriority::IteratorPtr SLRUFileCachePriority::add( /// NOLINT
     size_t offset,
     size_t size,
     const UserInfo &,
-    const CacheGuard::Lock & lock,
+    const CachePriorityGuard::Lock & lock,
     bool is_startup)
 {
     if (is_startup)
@@ -103,7 +103,7 @@ bool SLRUFileCachePriority::collectCandidatesForEviction(
     IFileCachePriority::IteratorPtr reservee,
     FinalizeEvictionFunc & finalize_eviction_func,
     const UserID & user_id,
-    const CacheGuard::Lock & lock)
+    const CachePriorityGuard::Lock & lock)
 {
     /// If `it` is nullptr, then it is the first space reservation attempt
     /// for a corresponding file segment, so it will be directly put into probationary queue.
@@ -143,7 +143,7 @@ bool SLRUFileCachePriority::collectCandidatesForEviction(
         && !probationary_queue.collectCandidatesForEviction(size_to_downgrade, stat, res, reservee, noop, user_id, lock))
         return false;
 
-    finalize_eviction_func = [=, this](const CacheGuard::Lock & lk) mutable
+    finalize_eviction_func = [=, this](const CachePriorityGuard::Lock & lk) mutable
     {
         for (const auto & [key, key_candidates] : *downgrade_candidates)
         {
@@ -159,7 +159,7 @@ bool SLRUFileCachePriority::collectCandidatesForEviction(
     return true;
 }
 
-void SLRUFileCachePriority::increasePriority(SLRUIterator & iterator, const CacheGuard::Lock & lock)
+void SLRUFileCachePriority::increasePriority(SLRUIterator & iterator, const CachePriorityGuard::Lock & lock)
 {
     /// If entry is already in protected queue,
     /// we only need to increase its priority within the protected queue.
@@ -242,7 +242,7 @@ void SLRUFileCachePriority::increasePriority(SLRUIterator & iterator, const Cach
     iterator.is_protected = true;
 }
 
-IFileCachePriority::PriorityDumpPtr SLRUFileCachePriority::dump(const CacheGuard::Lock & lock)
+IFileCachePriority::PriorityDumpPtr SLRUFileCachePriority::dump(const CachePriorityGuard::Lock & lock)
 {
     auto res = dynamic_pointer_cast<LRUFileCachePriority::LRUPriorityDump>(probationary_queue.dump(lock));
     auto part_res = dynamic_pointer_cast<LRUFileCachePriority::LRUPriorityDump>(protected_queue.dump(lock));
@@ -250,14 +250,14 @@ IFileCachePriority::PriorityDumpPtr SLRUFileCachePriority::dump(const CacheGuard
     return res;
 }
 
-void SLRUFileCachePriority::shuffle(const CacheGuard::Lock & lock)
+void SLRUFileCachePriority::shuffle(const CachePriorityGuard::Lock & lock)
 {
     protected_queue.shuffle(lock);
     probationary_queue.shuffle(lock);
 }
 
 bool SLRUFileCachePriority::modifySizeLimits(
-    size_t max_size_, size_t max_elements_, double size_ratio_, const CacheGuard::Lock & lock)
+    size_t max_size_, size_t max_elements_, double size_ratio_, const CachePriorityGuard::Lock & lock)
 {
     if (max_size == max_size_ && max_elements == max_elements_ && size_ratio == size_ratio_)
         return false; /// Nothing to change.
@@ -287,7 +287,7 @@ SLRUFileCachePriority::EntryPtr SLRUFileCachePriority::SLRUIterator::getEntry()
     return entry;
 }
 
-size_t SLRUFileCachePriority::SLRUIterator::increasePriority(const CacheGuard::Lock & lock)
+size_t SLRUFileCachePriority::SLRUIterator::increasePriority(const CachePriorityGuard::Lock & lock)
 {
     assertValid();
     cache_priority->increasePriority(*this, lock);
@@ -306,7 +306,7 @@ void SLRUFileCachePriority::SLRUIterator::invalidate()
     lru_iterator.invalidate();
 }
 
-void SLRUFileCachePriority::SLRUIterator::remove(const CacheGuard::Lock & lock)
+void SLRUFileCachePriority::SLRUIterator::remove(const CachePriorityGuard::Lock & lock)
 {
     assertValid();
     lru_iterator.remove(lock);
diff --git a/src/Interpreters/Cache/SLRUFileCachePriority.h b/src/Interpreters/Cache/SLRUFileCachePriority.h
index d97fa80a6c7..d81ce1bc480 100644
--- a/src/Interpreters/Cache/SLRUFileCachePriority.h
+++ b/src/Interpreters/Cache/SLRUFileCachePriority.h
@@ -21,13 +21,13 @@ public:
         LRUFileCachePriority::StatePtr probationary_state_ = nullptr,
         LRUFileCachePriority::StatePtr protected_state_ = nullptr);
 
-    size_t getSize(const CacheGuard::Lock & lock) const override;
+    size_t getSize(const CachePriorityGuard::Lock & lock) const override;
 
-    size_t getElementsCount(const CacheGuard::Lock &) const override;
+    size_t getElementsCount(const CachePriorityGuard::Lock &) const override;
 
     bool canFit( /// NOLINT
         size_t size,
-        const CacheGuard::Lock &,
+        const CachePriorityGuard::Lock &,
         IteratorPtr reservee = nullptr,
         bool best_effort = false) const override;
 
@@ -36,7 +36,7 @@ public:
         size_t offset,
         size_t size,
         const UserInfo & user,
-        const CacheGuard::Lock &,
+        const CachePriorityGuard::Lock &,
         bool is_startup = false) override;
 
     bool collectCandidatesForEviction(
@@ -46,13 +46,13 @@ public:
         IFileCachePriority::IteratorPtr reservee,
         FinalizeEvictionFunc & finalize_eviction_func,
         const UserID & user_id,
-        const CacheGuard::Lock &) override;
+        const CachePriorityGuard::Lock &) override;
 
-    void shuffle(const CacheGuard::Lock &) override;
+    void shuffle(const CachePriorityGuard::Lock &) override;
 
-    PriorityDumpPtr dump(const CacheGuard::Lock &) override;
+    PriorityDumpPtr dump(const CachePriorityGuard::Lock &) override;
 
-    bool modifySizeLimits(size_t max_size_, size_t max_elements_, double size_ratio_, const CacheGuard::Lock &) override;
+    bool modifySizeLimits(size_t max_size_, size_t max_elements_, double size_ratio_, const CachePriorityGuard::Lock &) override;
 
 private:
     double size_ratio;
@@ -60,7 +60,7 @@ private:
     LRUFileCachePriority probationary_queue;
     LoggerPtr log = getLogger("SLRUFileCachePriority");
 
-    void increasePriority(SLRUIterator & iterator, const CacheGuard::Lock & lock);
+    void increasePriority(SLRUIterator & iterator, const CachePriorityGuard::Lock & lock);
 };
 
 class SLRUFileCachePriority::SLRUIterator : public IFileCachePriority::Iterator
@@ -74,9 +74,9 @@ public:
 
     EntryPtr getEntry() const override;
 
-    size_t increasePriority(const CacheGuard::Lock &) override;
+    size_t increasePriority(const CachePriorityGuard::Lock &) override;
 
-    void remove(const CacheGuard::Lock &) override;
+    void remove(const CachePriorityGuard::Lock &) override;
 
     void invalidate() override;
 

From 3fcede709e39db1525c8ed58bc6deb39814edd83 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 16 Mar 2024 19:03:42 +0100
Subject: [PATCH 400/443] Fix error

---
 .../0_stateless/03011_definitive_guide_to_cast.sql        | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql b/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql
index 771123b153a..7cc662eee29 100644
--- a/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql
+++ b/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql
@@ -212,12 +212,12 @@ SELECT toDateTime64('2024-04-25 01:02:03', 6, 'Europe/Amsterdam'); -- the scale
 -- If the time zone is not specified, the time zone of the argument's data type is used, and if the argument is not a date time, the session time zone is used.
 
 SELECT toDateTime('2024-04-25 01:02:03');
-SELECT toDateTime64('2024-04-25 01:02:03', 6);
+SELECT toDateTime64('2024-04-25 01:02:03.456789', 6);
 
--- Here the time zone can be specified as the rule of interpreration of the value during conversion:
+-- Here the time zone can be specified as the rule of interpretation of the value during conversion:
 
-SELECT toString(now(), 'Europe/Amsterdam');
-SELECT toString(now());
+SELECT toString(1710612085::DateTime, 'America/Los_Angeles');
+SELECT toString(1710612085::DateTime);
 
 
 -- 7. SQL-compatibility type-defining operators:

From acd4e49dc2cbc61d03cc1177af40c5c0ce11243d Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 16 Mar 2024 19:35:56 +0100
Subject: [PATCH 401/443] More guides

---
 .../03011_definitive_guide_to_cast.reference  | 31 ++++--
 .../03011_definitive_guide_to_cast.sql        | 97 ++++++++++++++++++-
 2 files changed, 118 insertions(+), 10 deletions(-)

diff --git a/tests/queries/0_stateless/03011_definitive_guide_to_cast.reference b/tests/queries/0_stateless/03011_definitive_guide_to_cast.reference
index f8f37fa7807..f20fe0b7425 100644
--- a/tests/queries/0_stateless/03011_definitive_guide_to_cast.reference
+++ b/tests/queries/0_stateless/03011_definitive_guide_to_cast.reference
@@ -8,11 +8,23 @@
 ['Hello','wo\'rld\\']
 Hello	wo\\\'rld\\\\
 wo\'rld\\	wo\\\'rld\\\\
-123
-123
-123
+133	210
+210
+[123,456]
 1	-1
 []	[]	Array(Nothing)	Array(Array(Array(Tuple(UInt64, String))))
+1970-01-01 01:00:00
+2009-02-14 00:31:30.123456
+1970-01-01 00:59:59.888889
+2009-02-14 00:31:30
+1970-01-01 01:00:00
+2299-12-31 23:59:59.000000
+2009-02-14
+2009-02-14
+123\0\0
+123
+123
+123
 123
 123
 123
@@ -46,9 +58,16 @@ Hello\0\0\0\0\0
 2024-04-25 01:02:03
 2024-04-25 01:02:03.000000
 2024-04-25 01:02:03
-2024-04-25 01:02:03.000000
-2024-03-16 16:22:33
-2024-03-16 16:22:33
+2024-04-25 01:02:03.456789
+2024-03-16 11:01:25
+2024-03-16 19:01:25
+123	\N	\N	\N
+123	0	0	0
+Nullable(UInt8)	UInt8
+123
+123
+123
+\N
 2024-04-25	2024-01-01 02:03:04	1	12
 2024-04-25	2024-01-01 02:03:04.000000	2009-02-14 00:31:30
 2024-04-25	2024-01-01 02:03:04.000000	2009-02-14 00:31:30
diff --git a/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql b/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql
index 7cc662eee29..91487666a54 100644
--- a/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql
+++ b/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql
@@ -39,11 +39,25 @@ SELECT CAST($$['Hello', 'wo\'rld\\']$$ AS Array(String));
 
 SELECT arrayJoin(CAST($$['Hello', 'wo\'rld\\']$$ AS Array(String))) AS x, CAST($$wo\'rld\\$$ AS FixedString(9)) AS y;
 
--- The operator is case-insensitive:
+-- As conversion from String is similar to direct parsing rather than conversion from other types,
+-- it can be stricter for numbers by not tolerating overflows in some cases:
 
-SELECT CAST(123 AS String);
-SELECT cast(123 AS String);
-SELECT Cast(123 AS String);
+SELECT CAST(-123 AS UInt8), CAST(1234 AS UInt8);
+
+SELECT CAST('-123' AS UInt8); -- { serverError CANNOT_PARSE_NUMBER }
+
+-- In some cases it still allows overflows, but it is implementation defined:
+
+SELECT CAST('1234' AS UInt8);
+
+-- Parsing from a string does not tolerate extra whitespace characters:
+
+SELECT CAST(' 123' AS UInt8); -- { serverError CANNOT_PARSE_TEXT }
+SELECT CAST('123 ' AS UInt8); -- { serverError CANNOT_PARSE_TEXT }
+
+-- But for composite data types, it involves a more featured parser, that take care of whitespace inside the data structures:
+
+SELECT CAST('[ 123 ,456, ]' AS Array(UInt16));
 
 -- Conversion from a floating point value to an integer will involve truncation towards zero:
 
@@ -55,6 +69,54 @@ SELECT CAST(1.9, 'Int64'), CAST(-1.9, 'Int64');
 
 SELECT [] AS x, CAST(x AS Array(Array(Array(Tuple(UInt64, String))))) AS y, toTypeName(x), toTypeName(y);
 
+-- Conversion between numbers and DateTime/Date data types interprets the number as the number of seconds/days from the Unix epoch,
+-- where Unix epoch starts from 1970-01-01T00:00:00Z (the midnight of Gregorian year 1970 in UTC),
+-- and the number of seconds don't count the coordination seconds, as in Unix.
+
+-- For example, it is 1 AM in Amsterdam:
+
+SELECT CAST(0 AS DateTime('Europe/Amsterdam'));
+
+-- The numbers can be fractional and negative (for DateTime64):
+
+SELECT CAST(1234567890.123456 AS DateTime64(6, 'Europe/Amsterdam'));
+SELECT CAST(-0.111111 AS DateTime64(6, 'Europe/Amsterdam'));
+
+-- If the result does not fit in the range of the corresponding time data types, it is truncated and saturated to the boundaries:
+
+SELECT CAST(1234567890.123456 AS DateTime('Europe/Amsterdam'));
+SELECT CAST(-1 AS DateTime('Europe/Amsterdam'));
+
+SELECT CAST(1e20 AS DateTime64(6, 'Europe/Amsterdam'));
+
+-- A special case is DateTime64(9) - the maximum resolution, where is does not cover the usual range,
+-- and in this case, it throws an exception on overflow (I don't mind if we change this behavior in the future):
+
+ SELECT CAST(1e20 AS DateTime64(9, 'Europe/Amsterdam')); -- { serverError DECIMAL_OVERFLOW }
+
+-- If a number is converted to a Date data type, the value is interpreted as the number of days since the Unix epoch,
+-- but if the number is larger than the range of the data type, it is interpreted as a unix timestamp
+-- (the number of seconds since the Unix epoch), similarly how it is done for the DateTime data type,
+-- for convenience (while the internal representation of Date is the number of days,
+-- often people want the unix timestamp to be also parsed into the Date data type):
+
+SELECT CAST(14289 AS Date);
+SELECT CAST(1234567890 AS Date);
+
+-- When converting to a FixedString, if the length of the result data type is larger than the value, the result is padded with zero bytes:
+
+SELECT CAST('123' AS FixedString(5)) FORMAT TSV;
+
+-- But if it does not fit, an exception is thrown:
+
+SELECT CAST('12345' AS FixedString(3)) FORMAT TSV; -- { serverError TOO_LARGE_STRING_SIZE }
+
+-- The operator is case-insensitive:
+
+SELECT CAST(123 AS String);
+SELECT cast(123 AS String);
+SELECT Cast(123 AS String);
+
 
 -- 2. The functional form of this operator: `CAST(value, 'Type')`:
 
@@ -219,6 +281,33 @@ SELECT toDateTime64('2024-04-25 01:02:03.456789', 6);
 SELECT toString(1710612085::DateTime, 'America/Los_Angeles');
 SELECT toString(1710612085::DateTime);
 
+-- Functions converting to numeric types, date and datetime, IP and UUID, also have versions with -OrNull and -OrZero fallbacks,
+-- that don't throw exceptions on parsing errors.
+-- They use the same rules to the accurateCast operator:
+
+SELECT toUInt8OrNull('123'), toUInt8OrNull('-123'), toUInt8OrNull('1234'), toUInt8OrNull(' 123');
+SELECT toUInt8OrZero('123'), toUInt8OrZero('-123'), toUInt8OrZero('1234'), toUInt8OrZero(' 123');
+
+SELECT toTypeName(toUInt8OrNull('123')), toTypeName(toUInt8OrZero('123'));
+
+-- These functions are only applicable to string data types.
+-- Although it is a room for extension:
+
+SELECT toUInt8OrNull(123); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT }
+
+-- String and FixedString work:
+
+SELECT toUInt8OrNull(123::FixedString(3));
+
+-- For the FixedString data type trailing zero bytes are allowed, because they are the padding for FixedString:
+
+SELECT toUInt8OrNull('123'::FixedString(4));
+SELECT toUInt8OrNull('123\0'::FixedString(4));
+
+-- While for String, they don't:
+
+SELECT toUInt8OrNull('123\0');
+
 
 -- 7. SQL-compatibility type-defining operators:
 

From 2d8676d61a58925ace94e4e71893c93fee9551c9 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 16 Mar 2024 19:38:32 +0100
Subject: [PATCH 402/443] Update 03011_definitive_guide_to_cast.sql

---
 .../03011_definitive_guide_to_cast.sql        | 30 +++++++++----------
 1 file changed, 15 insertions(+), 15 deletions(-)

diff --git a/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql b/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql
index 91487666a54..335b97c9db6 100644
--- a/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql
+++ b/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql
@@ -7,7 +7,7 @@ SET session_timezone = 'Europe/Amsterdam';
 
 SELECT CAST(123 AS String);
 
--- It convert between various data types, including parameterized data types
+-- It converts between various data types, including parameterized data types
 
 SELECT CAST(1234567890 AS DateTime('Europe/Amsterdam'));
 
@@ -15,7 +15,7 @@ SELECT CAST(1234567890 AS DateTime('Europe/Amsterdam'));
 
 SELECT CAST('[1, 2, 3]' AS Array(UInt8));
 
--- It's return type depends on the setting `cast_keep_nullable`. If it is enabled, if the source argument type is Nullable, the resulting data type will be also Nullable, even if it is not written explicitly:
+-- Its return type depends on the setting `cast_keep_nullable`. If it is enabled, if the source argument type is Nullable, the resulting data type will be also Nullable, even if it is not written explicitly:
 
 SET cast_keep_nullable = 1;
 SELECT CAST(x AS UInt8) AS y, toTypeName(y) FROM VALUES('x Nullable(String)', ('123'), ('NULL'));
@@ -25,7 +25,7 @@ SELECT CAST(x AS UInt8) AS y, toTypeName(y) FROM VALUES('x Nullable(String)', ('
 
 -- There are various type conversion rules, some worth noting.
 
--- Conversion between numeric types can involve implementation defined overflow:
+-- Conversion between numeric types can involve implementation-defined overflow:
 
 SELECT CAST(257 AS UInt8);
 SELECT CAST(-1 AS UInt8);
@@ -35,7 +35,7 @@ SELECT CAST(-1 AS UInt8);
 SELECT CAST($$['Hello', 'wo\'rld\\']$$ AS Array(String));
 
 -- '
--- While for simple data types it does not interpret escape sequences:
+-- While for simple data types, it does not interpret escape sequences:
 
 SELECT arrayJoin(CAST($$['Hello', 'wo\'rld\\']$$ AS Array(String))) AS x, CAST($$wo\'rld\\$$ AS FixedString(9)) AS y;
 
@@ -159,9 +159,9 @@ SELECT 'String' AS String, CAST(123, String);
 
 -- 3. The internal function `_CAST` which is different from `CAST` only by being not dependent on the value of `cast_keep_nullable` setting and other settings.
 
--- This is needed when ClickHouse has to persist an expression for future use, like in table definitions, including primary and partition key and other indices.
+-- This is needed when ClickHouse has to persist an expression for future use, like in table definitions, including primary and partition keys and other indices.
 
--- The function is not intended for being used directly. When a user uses a regular `CAST` operator or function in a table definition, it is transparently converted to `_CAST` to persist its behavior. However, the user can still use the internal version directly:
+-- The function is not intended to be used directly. When a user uses a regular `CAST` operator or function in a table definition, it is transparently converted to `_CAST` to persist its behavior. However, the user can still use the internal version directly:
 
 SELECT _CAST(x, 'UInt8') AS y, toTypeName(y) FROM VALUES('x Nullable(String)', ('123'), ('456'));
 
@@ -179,7 +179,7 @@ SELECT 123::String;
 
 -- It has a difference from the `CAST` operator: if it is applied to a simple literal value, instead of performing a type conversion, it invokes the SQL parser directly on the corresponding text fragment of the query. The most important case will be the floating-point and decimal types.
 
--- In this example, we parse `1.1` as Decimal and not involving any type conversion:
+-- In this example, we parse `1.1` as Decimal and do not involve any type conversion:
 
 SELECT 1.1::Decimal(30, 20);
 
@@ -211,7 +211,7 @@ SELECT 1.1::Decimal(30, 20), CAST('1.1' AS Decimal(30, 20)), (1+1)::UInt8 FORMAT
 
 SELECT 1-1::String; -- { serverError ILLEGAL_TYPE_OF_ARGUMENT }
 
--- But one interesting example is the unary minus. Here the minus is not an operator, but part of the numeric literal:
+-- But one interesting example is the unary minus. Here the minus is not an operator but part of the numeric literal:
 
 SELECT -1::String;
 
@@ -236,7 +236,7 @@ SELECT accurateCastOrNull(1.123456789, 'Float32');
 
 SELECT accurateCastOrDefault(-1, 'UInt64', 0::UInt64);
 
--- These functions are case-sensitive and there are no corresponding operators:
+-- These functions are case-sensitive, and there are no corresponding operators:
 
 SELECT ACCURATECAST(1, 'String'); -- { serverError UNKNOWN_FUNCTION }.
 
@@ -254,11 +254,11 @@ SELECT ACCURATECAST(1, 'String'); -- { serverError UNKNOWN_FUNCTION }.
 -- `toIntervalSecond`, `toIntervalMinute`, `toIntervalHour`,
 -- `toIntervalDay`, `toIntervalWeek`, `toIntervalMonth`, `toIntervalQuarter`, `toIntervalYear`
 
--- These functions work under the same rules as the CAST operator, and can be thought as a elementary implementation parts of that operator. They allow implementation defined overflow while converting between numeric types.
+-- These functions work under the same rules as the CAST operator and can be thought as elementary implementation parts of that operator. They allow implementation-defined overflow while converting between numeric types.
 
 SELECT toUInt8(-1);
 
--- These are ClickHouse-native conversion functions. They take an argument with the input value, and for some of data types (`FixedString`, `DateTime`, `DateTime64`, `Decimal`s) the subsequent arguments are constant expressions, defining the parameters of these data types, or the rules to interpret the source value.
+-- These are ClickHouse-native conversion functions. They take an argument with the input value, and for some of the data types (`FixedString`, `DateTime`, `DateTime64`, `Decimal`s), the subsequent arguments are constant expressions, defining the parameters of these data types, or the rules to interpret the source value.
 
 SELECT toFloat64(123); -- no arguments
 SELECT toFixedString('Hello', 10) FORMAT TSV; -- the parameter of the FixedString data type, the function returns FixedString(10)
@@ -276,7 +276,7 @@ SELECT toDateTime64('2024-04-25 01:02:03', 6, 'Europe/Amsterdam'); -- the scale
 SELECT toDateTime('2024-04-25 01:02:03');
 SELECT toDateTime64('2024-04-25 01:02:03.456789', 6);
 
--- Here the time zone can be specified as the rule of interpretation of the value during conversion:
+-- Here, the time zone can be specified as the rule of interpretation of the value during conversion:
 
 SELECT toString(1710612085::DateTime, 'America/Los_Angeles');
 SELECT toString(1710612085::DateTime);
@@ -320,7 +320,7 @@ SELECT DATE '2024-04-25', TIMESTAMP '2024-01-01 02:03:04', INTERVAL 1 MINUTE, IN
 
 SELECT DATE('2024-04-25'), TIMESTAMP('2024-01-01 02:03:04'), FROM_UNIXTIME(1234567890);
 
--- These functions exist for compatibility with MySQL. They are case-insensive.
+-- These functions exist for compatibility with MySQL. They are case-insensitive.
 
 SELECT date '2024-04-25', timeSTAMP('2024-01-01 02:03:04'), From_Unixtime(1234567890);
 
@@ -329,7 +329,7 @@ SELECT date '2024-04-25', timeSTAMP('2024-01-01 02:03:04'), From_Unixtime(123456
 
 -- `parseDateTimeBestEffort`, `parseDateTimeBestEffortUS`, `parseDateTime64BestEffort`, `parseDateTime64BestEffortUS`, `toUnixTimestamp`
 
--- These functions are similar to explicit conversion functions, but provide special rules on how the conversion is performed.
+-- These functions are similar to explicit conversion functions but provide special rules on how the conversion is performed.
 
 SELECT parseDateTimeBestEffort('25 Apr 1986 1pm');
 
@@ -338,4 +338,4 @@ SELECT parseDateTimeBestEffort('25 Apr 1986 1pm');
 
 SELECT toDayOfMonth(toDateTime(1234567890));
 
--- These functions are coverted in a separate topic.
+-- These functions are covered in a separate topic.

From 175efee0891223c6242f34e6bf08301a625ef407 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 16 Mar 2024 19:50:11 +0100
Subject: [PATCH 403/443] More guides

---
 .../0_stateless/03011_definitive_guide_to_cast.reference     | 2 ++
 tests/queries/0_stateless/03011_definitive_guide_to_cast.sql | 5 +++++
 2 files changed, 7 insertions(+)

diff --git a/tests/queries/0_stateless/03011_definitive_guide_to_cast.reference b/tests/queries/0_stateless/03011_definitive_guide_to_cast.reference
index f20fe0b7425..b71011d75ff 100644
--- a/tests/queries/0_stateless/03011_definitive_guide_to_cast.reference
+++ b/tests/queries/0_stateless/03011_definitive_guide_to_cast.reference
@@ -61,6 +61,8 @@ Hello\0\0\0\0\0
 2024-04-25 01:02:03.456789
 2024-03-16 11:01:25
 2024-03-16 19:01:25
+2024-03-16 19:01:25
+2024-03-16 11:01:25
 123	\N	\N	\N
 123	0	0	0
 Nullable(UInt8)	UInt8
diff --git a/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql b/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql
index 335b97c9db6..e79572e1ddb 100644
--- a/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql
+++ b/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql
@@ -281,6 +281,11 @@ SELECT toDateTime64('2024-04-25 01:02:03.456789', 6);
 SELECT toString(1710612085::DateTime, 'America/Los_Angeles');
 SELECT toString(1710612085::DateTime);
 
+-- In the case when the time zone is not the part of the resulting data type, but a rule of interpretation of the source value,
+-- it can be non-constant. Let's clarify: in this example, the resulting data type is a String; it does not have a time zone parameter:
+
+SELECT toString(1710612085::DateTime, tz) FROM Values('tz String', 'Europe/Amsterdam', 'America/Los_Angeles');
+
 -- Functions converting to numeric types, date and datetime, IP and UUID, also have versions with -OrNull and -OrZero fallbacks,
 -- that don't throw exceptions on parsing errors.
 -- They use the same rules to the accurateCast operator:

From 65eb6f135c19459ace827035dd518b371c64ee49 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 16 Mar 2024 19:52:10 +0100
Subject: [PATCH 404/443] More guides

---
 tests/queries/0_stateless/03011_definitive_guide_to_cast.sql | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql b/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql
index e79572e1ddb..76819ad8313 100644
--- a/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql
+++ b/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql
@@ -55,7 +55,7 @@ SELECT CAST('1234' AS UInt8);
 SELECT CAST(' 123' AS UInt8); -- { serverError CANNOT_PARSE_TEXT }
 SELECT CAST('123 ' AS UInt8); -- { serverError CANNOT_PARSE_TEXT }
 
--- But for composite data types, it involves a more featured parser, that take care of whitespace inside the data structures:
+-- But for composite data types, it involves a more featured parser, that takes care of whitespace inside the data structures:
 
 SELECT CAST('[ 123 ,456, ]' AS Array(UInt16));
 

From e5f15b6ac4894f32c7dd82091a3ee8260ef05f86 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sat, 16 Mar 2024 23:00:57 +0100
Subject: [PATCH 405/443] Fix errors

---
 src/Functions/FunctionsConversion.cpp | 26 +++++++++++++-------------
 1 file changed, 13 insertions(+), 13 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 7f130b0cc86..0deb2960d34 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -364,9 +364,9 @@ struct ToDateTime64TransformUnsigned
 {
     static constexpr auto name = "toDateTime64";
 
-    const DateTime64::NativeType scale_multiplier = 1;
+    const DateTime64::NativeType scale_multiplier;
 
-    ToDateTime64TransformUnsigned(UInt32 scale = 0) /// NOLINT
+    ToDateTime64TransformUnsigned(UInt32 scale) /// NOLINT
         : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
     {}
 
@@ -388,9 +388,9 @@ struct ToDateTime64TransformSigned
 {
     static constexpr auto name = "toDateTime64";
 
-    const DateTime64::NativeType scale_multiplier = 1;
+    const DateTime64::NativeType scale_multiplier;
 
-    ToDateTime64TransformSigned(UInt32 scale = 0) /// NOLINT
+    ToDateTime64TransformSigned(UInt32 scale) /// NOLINT
         : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
     {}
 
@@ -412,9 +412,9 @@ struct ToDateTime64TransformFloat
 {
     static constexpr auto name = "toDateTime64";
 
-    const UInt32 scale = 1;
+    const UInt32 scale;
 
-    ToDateTime64TransformFloat(UInt32 scale_ = 0) /// NOLINT
+    ToDateTime64TransformFloat(UInt32 scale_) /// NOLINT
         : scale(scale_)
     {}
 
@@ -439,7 +439,7 @@ struct FromDateTime64Transform
 {
     static constexpr auto name = Transform::name;
 
-    const DateTime64::NativeType scale_multiplier = 1;
+    const DateTime64::NativeType scale_multiplier;
 
     FromDateTime64Transform(UInt32 scale) /// NOLINT
         : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
@@ -456,9 +456,9 @@ struct ToDateTime64Transform
 {
     static constexpr auto name = "toDateTime64";
 
-    const DateTime64::NativeType scale_multiplier = 1;
+    const DateTime64::NativeType scale_multiplier;
 
-    ToDateTime64Transform(UInt32 scale = 0) /// NOLINT
+    ToDateTime64Transform(UInt32 scale) /// NOLINT
         : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
     {}
 
@@ -1309,14 +1309,14 @@ struct ConvertImpl
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
             return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64TransformSigned<typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::template execute<Additions>(
-                arguments, result_type, input_rows_count);
+                arguments, result_type, input_rows_count, additions);
         }
         else if constexpr (std::is_same_v<FromDataType, DataTypeUInt64>
             && std::is_same_v<ToDataType, DataTypeDateTime64>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
             return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64TransformUnsigned<UInt64, default_date_time_overflow_behavior>, false>::template execute<Additions>(
-                arguments, result_type, input_rows_count);
+                arguments, result_type, input_rows_count, additions);
         }
         else if constexpr ((
                 std::is_same_v<FromDataType, DataTypeFloat32>
@@ -1325,7 +1325,7 @@ struct ConvertImpl
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
             return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64TransformFloat<FromDataType, typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::template execute<Additions>(
-                arguments, result_type, input_rows_count);
+                arguments, result_type, input_rows_count, additions);
         }
         /// Conversion of DateTime64 to Date or DateTime: discards fractional part.
         else if constexpr (std::is_same_v<FromDataType, DataTypeDateTime64>
@@ -1351,7 +1351,7 @@ struct ConvertImpl
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
             return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTime64Transform, false>::template execute<Additions>(
-                arguments, result_type, input_rows_count);
+                arguments, result_type, input_rows_count, additions);
         }
         else if constexpr (IsDataTypeDateOrDateTime<FromDataType>
             && std::is_same_v<ToDataType, DataTypeString>)

From f1be9e67070a25362e4e67ce78540ce95cc4e952 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 17 Mar 2024 00:17:59 +0100
Subject: [PATCH 406/443] Trash

---
 src/Functions/FunctionsConversion.cpp | 94 ++++++++++-----------------
 1 file changed, 34 insertions(+), 60 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 0deb2960d34..088076eeeb4 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -181,12 +181,12 @@ struct ToDateTimeImpl
 
 /// Implementation of toDate function.
 
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+template <typename FromType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
 struct ToDateTransform32Or64
 {
     static constexpr auto name = "toDate";
 
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
+    static NO_SANITIZE_UNDEFINED UInt16 execute(const FromType & from, const DateLUTImpl & time_zone)
     {
         if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
         {
@@ -203,12 +203,12 @@ struct ToDateTransform32Or64
 };
 
 
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+template <typename FromType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
 struct ToDateTransform32Or64Signed
 {
     static constexpr auto name = "toDate";
 
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
+    static NO_SANITIZE_UNDEFINED Int32 execute(const FromType & from, const DateLUTImpl & time_zone)
     {
         // TODO: decide narrow or extended range based on FromType
         if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
@@ -222,17 +222,17 @@ struct ToDateTransform32Or64Signed
                 return 0;
         }
         return (from <= DATE_LUT_MAX_DAY_NUM)
-            ? static_cast<ToType>(from)
+            ? static_cast<Int32>(from)
             : time_zone.toDayNum(std::min(time_t(from), time_t(MAX_DATE_TIMESTAMP)));
     }
 };
 
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+template <typename FromType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
 struct ToDateTransform8Or16Signed
 {
     static constexpr auto name = "toDate";
 
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
+    static NO_SANITIZE_UNDEFINED UInt16 execute(const FromType & from, const DateLUTImpl &)
     {
         if (from < 0)
         {
@@ -247,15 +247,15 @@ struct ToDateTransform8Or16Signed
 
 /// Implementation of toDate32 function.
 
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+template <typename FromType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
 struct ToDate32Transform32Or64
 {
     static constexpr auto name = "toDate32";
 
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
+    static NO_SANITIZE_UNDEFINED Int32 execute(const FromType & from, const DateLUTImpl & time_zone)
     {
         if (from < DATE_LUT_MAX_EXTEND_DAY_NUM)
-            return static_cast<ToType>(from);
+            return static_cast<Int32>(from);
         else
         {
             if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
@@ -268,12 +268,12 @@ struct ToDate32Transform32Or64
     }
 };
 
-template <typename FromType, typename ToType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
+template <typename FromType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
 struct ToDate32Transform32Or64Signed
 {
     static constexpr auto name = "toDate32";
 
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl & time_zone)
+    static NO_SANITIZE_UNDEFINED Int32 execute(const FromType & from, const DateLUTImpl & time_zone)
     {
         static const Int32 daynum_min_offset = -static_cast<Int32>(time_zone.getDayNumOffsetEpoch());
 
@@ -287,17 +287,17 @@ struct ToDate32Transform32Or64Signed
             return daynum_min_offset;
 
         return (from < DATE_LUT_MAX_EXTEND_DAY_NUM)
-            ? static_cast<ToType>(from)
+            ? static_cast<Int32>(from)
             : time_zone.toDayNum(std::min(time_t(Int64(from)), time_t(MAX_DATETIME64_TIMESTAMP)));
     }
 };
 
-template <typename FromType, typename ToType>
+template <typename FromType>
 struct ToDate32Transform8Or16Signed
 {
     static constexpr auto name = "toDate32";
 
-    static NO_SANITIZE_UNDEFINED ToType execute(const FromType & from, const DateLUTImpl &)
+    static NO_SANITIZE_UNDEFINED Int32 execute(const FromType & from, const DateLUTImpl &)
     {
         return from;
     }
@@ -383,6 +383,7 @@ struct ToDateTime64TransformUnsigned
             return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(std::min<time_t>(from, MAX_DATETIME64_TIMESTAMP), 0, scale_multiplier);
     }
 };
+
 template <typename FromType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
 struct ToDateTime64TransformSigned
 {
@@ -407,6 +408,7 @@ struct ToDateTime64TransformSigned
         return DecimalUtils::decimalFromComponentsWithMultiplier<DateTime64>(from, 0, scale_multiplier);
     }
 };
+
 template <typename FromDataType, typename FromType, FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior>
 struct ToDateTime64TransformFloat
 {
@@ -432,26 +434,6 @@ struct ToDateTime64TransformFloat
     }
 };
 
-/** Conversion of DateTime64 to Date or DateTime: discards fractional part.
-  */
-template <typename Transform>
-struct FromDateTime64Transform
-{
-    static constexpr auto name = Transform::name;
-
-    const DateTime64::NativeType scale_multiplier;
-
-    FromDateTime64Transform(UInt32 scale) /// NOLINT
-        : scale_multiplier(DecimalUtils::scaleMultiplier<DateTime64::NativeType>(scale))
-    {}
-
-    auto execute(DateTime64::NativeType dt, const DateLUTImpl & time_zone) const
-    {
-        const auto c = DecimalUtils::splitWithScaleMultiplier(DateTime64(dt), scale_multiplier);
-        return Transform::execute(static_cast<UInt32>(c.whole), time_zone);
-    }
-};
-
 struct ToDateTime64Transform
 {
     static constexpr auto name = "toDateTime64";
@@ -972,7 +954,7 @@ struct ConvertThroughParsing
             size_t next_offset = std::is_same_v<FromDataType, DataTypeString> ? (*offsets)[i] : (current_offset + fixed_string_size);
             size_t string_size = std::is_same_v<FromDataType, DataTypeString> ? next_offset - current_offset - 1 : fixed_string_size;
 
-            ReadBufferFromMemory read_buffer(&(*chars)[current_offset], string_size);
+            ReadBufferFromMemory read_buffer(chars->data() + current_offset, string_size);
 
             if constexpr (exception_mode == ConvertFromStringExceptionMode::Throw)
             {
@@ -1087,23 +1069,15 @@ struct ConvertThroughParsing
                         parsed = SerializationDecimal<typename ToDataType::FieldType>::tryReadText(
                             vec_to[i], read_buffer, ToDataType::maxPrecision(), col_to->getScale());
                     }
+                    else if (std::is_same_v<FromDataType, DataTypeFixedString> && std::is_same_v<ToDataType, DataTypeIPv6>
+                            && fixed_string_size == IPV6_BINARY_LENGTH)
+                    {
+                        readBinary(vec_to[i], read_buffer);
+                        parsed = true;
+                    }
                     else
                     {
-                        /// we want to utilize constexpr condition here, which is not mixable with value comparison
-                        do
-                        {
-                            if constexpr (std::is_same_v<FromDataType, DataTypeFixedString> && std::is_same_v<ToDataType, DataTypeIPv6>)
-                            {
-                                if (fixed_string_size == IPV6_BINARY_LENGTH)
-                                {
-                                    readBinary(vec_to[i], read_buffer);
-                                    parsed = true;
-                                    break;
-                                }
-                            }
-
-                            parsed = tryParseImpl<ToDataType>(vec_to[i], read_buffer, local_time_zone, precise_float_parsing);
-                        } while (false);
+                        parsed = tryParseImpl<ToDataType>(vec_to[i], read_buffer, local_time_zone, precise_float_parsing);
                     }
                 }
 
@@ -1220,7 +1194,7 @@ struct ConvertImpl
             && std::is_same_v<ToDataType, DataTypeDate>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform32Or64<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::template execute<Additions>(
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform32Or64<typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count);
         }
         else if constexpr ((
@@ -1229,7 +1203,7 @@ struct ConvertImpl
             && std::is_same_v<ToDataType, DataTypeDate>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform8Or16Signed<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::template execute<Additions>(
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform8Or16Signed<typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count);
         }
         else if constexpr ((
@@ -1240,7 +1214,7 @@ struct ConvertImpl
             && std::is_same_v<ToDataType, DataTypeDate>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform32Or64Signed<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::template execute<Additions>(
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform32Or64Signed<typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count);
         }
         else if constexpr ((
@@ -1249,7 +1223,7 @@ struct ConvertImpl
             && std::is_same_v<ToDataType, DataTypeDate32>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform32Or64<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::template execute<Additions>(
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform32Or64<typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count);
         }
         else if constexpr ((
@@ -1258,7 +1232,7 @@ struct ConvertImpl
             && std::is_same_v<ToDataType, DataTypeDate32>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform8Or16Signed<typename FromDataType::FieldType, UInt16>, false>::template execute<Additions>(
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform8Or16Signed<typename FromDataType::FieldType>, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count);
         }
         else if constexpr ((
@@ -1269,7 +1243,7 @@ struct ConvertImpl
             && std::is_same_v<ToDataType, DataTypeDate32>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform32Or64Signed<typename FromDataType::FieldType, UInt16, default_date_time_overflow_behavior>, false>::template execute<Additions>(
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform32Or64Signed<typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count);
         }
         /// Special case of converting Int8, Int16, Int32 or (U)Int64 (and also, for convenience, Float32, Float64) to DateTime.
@@ -2475,7 +2449,7 @@ public:
     }
 
     template <typename ConvertToDataType>
-    ColumnPtr executeInternal(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count, UInt32 scale = 0) const
+    ColumnPtr executeInternal(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count, UInt32 scale) const
     {
         const IDataType * from_type = arguments[0].type.get();
 
@@ -2510,7 +2484,7 @@ public:
 
             if (scale == 0)
             {
-                result_column = executeInternal<DataTypeDateTime>(arguments, result_type, input_rows_count);
+                result_column = executeInternal<DataTypeDateTime>(arguments, result_type, input_rows_count, 0);
             }
             else
             {
@@ -2519,7 +2493,7 @@ public:
         }
         else
         {
-            result_column = executeInternal<ToDataType>(arguments, result_type, input_rows_count);
+            result_column = executeInternal<ToDataType>(arguments, result_type, input_rows_count, 0);
         }
 
         if (!result_column)

From ced06cb163bdf9280cde5ee49f9ab3eb506228e3 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 17 Mar 2024 00:18:38 +0100
Subject: [PATCH 407/443] More guides

---
 .../03011_definitive_guide_to_cast.reference         |  4 ++++
 .../0_stateless/03011_definitive_guide_to_cast.sql   | 12 +++++++++++-
 2 files changed, 15 insertions(+), 1 deletion(-)

diff --git a/tests/queries/0_stateless/03011_definitive_guide_to_cast.reference b/tests/queries/0_stateless/03011_definitive_guide_to_cast.reference
index b71011d75ff..7c875a24b6d 100644
--- a/tests/queries/0_stateless/03011_definitive_guide_to_cast.reference
+++ b/tests/queries/0_stateless/03011_definitive_guide_to_cast.reference
@@ -50,6 +50,8 @@ CAST(plus(1, 1), 'UInt8'):      2
 -1
 \N
 0
+0
+1970-01-01 01:00:00
 255
 123
 Hello\0\0\0\0\0
@@ -65,6 +67,8 @@ Hello\0\0\0\0\0
 2024-03-16 11:01:25
 123	\N	\N	\N
 123	0	0	0
+123	10	10	10
+123	0	0	0
 Nullable(UInt8)	UInt8
 123
 123
diff --git a/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql b/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql
index 76819ad8313..708db0adce0 100644
--- a/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql
+++ b/tests/queries/0_stateless/03011_definitive_guide_to_cast.sql
@@ -236,6 +236,14 @@ SELECT accurateCastOrNull(1.123456789, 'Float32');
 
 SELECT accurateCastOrDefault(-1, 'UInt64', 0::UInt64);
 
+-- If this parameter is omitted, it is assumed to be the default value of the corresponding data type:
+
+SELECT accurateCastOrDefault(-1, 'UInt64');
+SELECT accurateCastOrDefault(-1, 'DateTime');
+
+-- Unfortunately, this does not work as expected: SELECT accurateCastOrDefault(-1, $$Enum8('None' = 1, 'Hello' = 2, 'World' = 3)$$);
+-- https://github.com/ClickHouse/ClickHouse/issues/61495
+
 -- These functions are case-sensitive, and there are no corresponding operators:
 
 SELECT ACCURATECAST(1, 'String'); -- { serverError UNKNOWN_FUNCTION }.
@@ -286,12 +294,14 @@ SELECT toString(1710612085::DateTime);
 
 SELECT toString(1710612085::DateTime, tz) FROM Values('tz String', 'Europe/Amsterdam', 'America/Los_Angeles');
 
--- Functions converting to numeric types, date and datetime, IP and UUID, also have versions with -OrNull and -OrZero fallbacks,
+-- Functions converting to numeric types, date and datetime, IP and UUID, also have versions with -OrNull, -OrZero, and -OrDefault fallbacks,
 -- that don't throw exceptions on parsing errors.
 -- They use the same rules to the accurateCast operator:
 
 SELECT toUInt8OrNull('123'), toUInt8OrNull('-123'), toUInt8OrNull('1234'), toUInt8OrNull(' 123');
 SELECT toUInt8OrZero('123'), toUInt8OrZero('-123'), toUInt8OrZero('1234'), toUInt8OrZero(' 123');
+SELECT toUInt8OrDefault('123', 10), toUInt8OrDefault('-123', 10), toUInt8OrDefault('1234', 10), toUInt8OrDefault(' 123', 10);
+SELECT toUInt8OrDefault('123'), toUInt8OrDefault('-123'), toUInt8OrDefault('1234'), toUInt8OrDefault(' 123');
 
 SELECT toTypeName(toUInt8OrNull('123')), toTypeName(toUInt8OrZero('123'));
 

From d627fbef557422cff10dda9f890d6b805fda5a19 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 17 Mar 2024 00:26:07 +0100
Subject: [PATCH 408/443] Fix errors

---
 src/Functions/FunctionsConversion.cpp | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 088076eeeb4..ada7e4ac58d 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -208,7 +208,7 @@ struct ToDateTransform32Or64Signed
 {
     static constexpr auto name = "toDate";
 
-    static NO_SANITIZE_UNDEFINED Int32 execute(const FromType & from, const DateLUTImpl & time_zone)
+    static NO_SANITIZE_UNDEFINED UInt16 execute(const FromType & from, const DateLUTImpl & time_zone)
     {
         // TODO: decide narrow or extended range based on FromType
         if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
@@ -222,7 +222,7 @@ struct ToDateTransform32Or64Signed
                 return 0;
         }
         return (from <= DATE_LUT_MAX_DAY_NUM)
-            ? static_cast<Int32>(from)
+            ? static_cast<UInt16>(from)
             : time_zone.toDayNum(std::min(time_t(from), time_t(MAX_DATE_TIMESTAMP)));
     }
 };
@@ -255,7 +255,9 @@ struct ToDate32Transform32Or64
     static NO_SANITIZE_UNDEFINED Int32 execute(const FromType & from, const DateLUTImpl & time_zone)
     {
         if (from < DATE_LUT_MAX_EXTEND_DAY_NUM)
+        {
             return static_cast<Int32>(from);
+        }
         else
         {
             if constexpr (date_time_overflow_behavior == FormatSettings::DateTimeOverflowBehavior::Throw)
@@ -1214,7 +1216,7 @@ struct ConvertImpl
             && std::is_same_v<ToDataType, DataTypeDate>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform32Or64Signed<typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::template execute<Additions>(
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform32Or64Signed<typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count);
         }
         else if constexpr ((
@@ -1243,7 +1245,7 @@ struct ConvertImpl
             && std::is_same_v<ToDataType, DataTypeDate32>
             && std::is_same_v<SpecialTag, ConvertDefaultBehaviorTag>)
         {
-            return DateTimeTransformImpl<FromDataType, ToDataType, ToDateTransform32Or64Signed<typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::template execute<Additions>(
+            return DateTimeTransformImpl<FromDataType, ToDataType, ToDate32Transform32Or64Signed<typename FromDataType::FieldType, default_date_time_overflow_behavior>, false>::template execute<Additions>(
                 arguments, result_type, input_rows_count);
         }
         /// Special case of converting Int8, Int16, Int32 or (U)Int64 (and also, for convenience, Float32, Float64) to DateTime.

From fac040e08eda9ad958785965f45ebc038155b2de Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 17 Mar 2024 00:34:32 +0100
Subject: [PATCH 409/443] Add documentation

---
 src/Functions/castOrDefault.cpp | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/src/Functions/castOrDefault.cpp b/src/Functions/castOrDefault.cpp
index 57cb03e0349..ac04a883c11 100644
--- a/src/Functions/castOrDefault.cpp
+++ b/src/Functions/castOrDefault.cpp
@@ -329,7 +329,18 @@ REGISTER_FUNCTION(CastOrDefault)
     factory.registerFunction("toUInt64OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
         std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt64OrDefault", std::make_shared<DataTypeUInt64>())); });
     factory.registerFunction("toUInt128OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt128OrDefault", std::make_shared<DataTypeUInt128>())); });
+        std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt128OrDefault", std::make_shared<DataTypeUInt128>())); },
+        FunctionDocumentation{
+            .description=R"(
+Converts a string in the first argument of the function to UInt128 by parsing it.
+If it cannot parse the value, returns the default value, which can be provided as the second function argument, and if provided, must be of UInt128 type.
+If the default value is not provided in the second argument, it is assumed to be zero.
+)",
+            .examples{
+                {"Successful conversion", "SELECT toUInt128OrDefault('1', 2::UInt128)", "1"},
+                {"Default value", "SELECT toUInt128OrDefault('upyachka', 123456789012345678901234567890::UInt128)", "123456789012345678901234567890"}},
+            .categories{"ConversionFunctions"}
+        });
     factory.registerFunction("toUInt256OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
         std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt256OrDefault", std::make_shared<DataTypeUInt256>())); });
 

From 96f53df7ada476384b0665f52a7c22cd33ce2415 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 17 Mar 2024 12:02:02 +0100
Subject: [PATCH 410/443] Fix error

---
 src/Functions/castOrDefault.cpp | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/src/Functions/castOrDefault.cpp b/src/Functions/castOrDefault.cpp
index ac04a883c11..c858dfdd589 100644
--- a/src/Functions/castOrDefault.cpp
+++ b/src/Functions/castOrDefault.cpp
@@ -221,7 +221,7 @@ private:
         size_t scale = 0;
         std::string time_zone;
 
-        if (isDecimal(type))
+        if (isDecimal(type) || isDateTime64(type))
         {
             const auto & scale_argument = arguments[additional_argument_index];
 
@@ -338,7 +338,8 @@ If the default value is not provided in the second argument, it is assumed to be
 )",
             .examples{
                 {"Successful conversion", "SELECT toUInt128OrDefault('1', 2::UInt128)", "1"},
-                {"Default value", "SELECT toUInt128OrDefault('upyachka', 123456789012345678901234567890::UInt128)", "123456789012345678901234567890"}},
+                {"Default value", "SELECT toUInt128OrDefault('upyachka', 123456789012345678901234567890::UInt128)", "123456789012345678901234567890"},
+                {"Implicit default value", "SELECT toUInt128OrDefault('upyachka')", "0"}},
             .categories{"ConversionFunctions"}
         });
     factory.registerFunction("toUInt256OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(

From 921ec12192fc2461bc828653bd9c3900daee74c8 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 17 Mar 2024 12:29:45 +0100
Subject: [PATCH 411/443] Maybe not worse

---
 src/Functions/FunctionsConversion.cpp | 86 ++++++++-------------------
 1 file changed, 24 insertions(+), 62 deletions(-)

diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index ada7e4ac58d..0d245f809bb 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -53,6 +53,7 @@
 #include <Functions/TransformDateTime64.h>
 #include <Functions/FunctionsCodingIP.h>
 #include <Functions/CastOverloadResolver.h>
+#include <Functions/castTypeToEither.h>
 #include <DataTypes/DataTypeLowCardinality.h>
 #include <Columns/ColumnLowCardinality.h>
 #include <Interpreters/Context.h>
@@ -4776,67 +4777,6 @@ arguments, result_type, input_rows_count); \
     }
 };
 
-class MonotonicityHelper
-{
-public:
-    using MonotonicityForRange = FunctionCast::MonotonicityForRange;
-
-    template <typename DataType>
-    static auto monotonicityForType(const DataType * const)
-    {
-        return FunctionTo<DataType>::Type::Monotonic::get;
-    }
-
-    static MonotonicityForRange getMonotonicityInformation(const DataTypePtr & from_type, const IDataType * to_type)
-    {
-        if (const auto * type = checkAndGetDataType<DataTypeUInt8>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeUInt16>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeUInt32>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeUInt64>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeUInt128>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeUInt256>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeInt8>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeInt16>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeInt32>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeInt64>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeInt128>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeInt256>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeFloat32>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeFloat64>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeDate>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeDate32>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeDateTime>(to_type))
-            return monotonicityForType(type);
-        if (const auto * type = checkAndGetDataType<DataTypeString>(to_type))
-            return monotonicityForType(type);
-        if (isEnum(from_type))
-        {
-            if (const auto * type = checkAndGetDataType<DataTypeEnum8>(to_type))
-                return monotonicityForType(type);
-            if (const auto * type = checkAndGetDataType<DataTypeEnum16>(to_type))
-                return monotonicityForType(type);
-        }
-        /// other types like Null, FixedString, Array and Tuple have no monotonicity defined
-        return {};
-    }
-};
-
 }
 
 
@@ -4853,7 +4793,29 @@ FunctionBasePtr createFunctionBaseCast(
     for (size_t i = 0; i < arguments.size(); ++i)
         data_types[i] = arguments[i].type;
 
-    auto monotonicity = MonotonicityHelper::getMonotonicityInformation(arguments.front().type, return_type.get());
+    FunctionCast::MonotonicityForRange monotonicity;
+
+    if (isEnum(arguments.front().type)
+        && castTypeToEither<DataTypeEnum8, DataTypeEnum16>(return_type.get(), [&](auto & type)
+        {
+            monotonicity = FunctionTo<std::decay_t<decltype(type)>>::Type::Monotonic::get;
+            return true;
+        }))
+    {
+    }
+    else if (castTypeToEither<
+        DataTypeUInt8, DataTypeUInt16, DataTypeUInt32, DataTypeUInt64, DataTypeUInt128, DataTypeUInt256,
+        DataTypeInt8, DataTypeInt16, DataTypeInt32, DataTypeInt64, DataTypeInt128, DataTypeInt256,
+        DataTypeFloat32, DataTypeFloat64,
+        DataTypeDate, DataTypeDate32, DataTypeDateTime,
+        DataTypeString>(return_type.get(), [&](auto & type)
+        {
+            monotonicity = FunctionTo<std::decay_t<decltype(type)>>::Type::Monotonic::get;
+            return true;
+        }))
+    {
+    }
+
     return std::make_unique<FunctionCast>(context, name, std::move(monotonicity), data_types, return_type, diagnostic, cast_type);
 }
 

From 2af6d35752e064bfd7859b23a52948c222f7c716 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 17 Mar 2024 12:53:59 +0100
Subject: [PATCH 412/443] Less crap

---
 src/Functions/FunctionFactory.cpp          |  12 +++
 src/Functions/FunctionFactory.h            |   7 ++
 src/Functions/array/emptyArray.cpp         |   3 +-
 src/Functions/castOrDefault.cpp            | 100 ++++++++++-----------
 src/Functions/coverage.cpp                 |   6 +-
 src/Functions/currentProfiles.cpp          |   6 +-
 src/Functions/fromUnixTimestamp64Micro.cpp |   3 +-
 src/Functions/fromUnixTimestamp64Milli.cpp |   3 +-
 src/Functions/fromUnixTimestamp64Nano.cpp  |   3 +-
 src/Functions/snowflake.cpp                |  12 +--
 src/Functions/toUnixTimestamp64Micro.cpp   |   3 +-
 src/Functions/toUnixTimestamp64Milli.cpp   |   3 +-
 src/Functions/toUnixTimestamp64Nano.cpp    |   3 +-
 13 files changed, 86 insertions(+), 78 deletions(-)

diff --git a/src/Functions/FunctionFactory.cpp b/src/Functions/FunctionFactory.cpp
index 6a7274376b9..004ef745a93 100644
--- a/src/Functions/FunctionFactory.cpp
+++ b/src/Functions/FunctionFactory.cpp
@@ -49,6 +49,18 @@ void FunctionFactory::registerFunction(
     }
 }
 
+void FunctionFactory::registerFunction(
+    const std::string & name,
+    FunctionSimpleCreator creator,
+    FunctionDocumentation doc,
+    CaseSensitiveness case_sensitiveness)
+{
+    registerFunction(name, [my_creator = std::move(creator)](ContextPtr context)
+    {
+        return std::make_unique<FunctionToOverloadResolverAdaptor>(my_creator(context));
+    }, std::move(doc), std::move(case_sensitiveness));
+}
+
 
 FunctionOverloadResolverPtr FunctionFactory::getImpl(
     const std::string & name,
diff --git a/src/Functions/FunctionFactory.h b/src/Functions/FunctionFactory.h
index 588cae64e16..c2fd34f0488 100644
--- a/src/Functions/FunctionFactory.h
+++ b/src/Functions/FunctionFactory.h
@@ -17,6 +17,7 @@ namespace DB
 {
 
 using FunctionCreator = std::function<FunctionOverloadResolverPtr(ContextPtr)>;
+using FunctionSimpleCreator = std::function<FunctionPtr(ContextPtr)>;
 using FunctionFactoryData = std::pair<FunctionCreator, FunctionDocumentation>;
 
 /** Creates function by name.
@@ -66,6 +67,12 @@ public:
         FunctionDocumentation doc = {},
         CaseSensitiveness case_sensitiveness = CaseSensitive);
 
+    void registerFunction(
+        const std::string & name,
+        FunctionSimpleCreator creator,
+        FunctionDocumentation doc = {},
+        CaseSensitiveness case_sensitiveness = CaseSensitive);
+
     FunctionDocumentation getDocumentation(const std::string & name) const;
 
 private:
diff --git a/src/Functions/array/emptyArray.cpp b/src/Functions/array/emptyArray.cpp
index 684f8af162a..77c191b3adc 100644
--- a/src/Functions/array/emptyArray.cpp
+++ b/src/Functions/array/emptyArray.cpp
@@ -49,8 +49,7 @@ private:
 void registerFunction(FunctionFactory & factory, const String & element_type)
 {
     factory.registerFunction(FunctionEmptyArray::getNameImpl(element_type),
-        [element_type](ContextPtr){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-            std::make_shared<FunctionEmptyArray>(element_type)); });
+        [element_type](ContextPtr){ return std::make_shared<FunctionEmptyArray>(element_type); });
 }
 
 }
diff --git a/src/Functions/castOrDefault.cpp b/src/Functions/castOrDefault.cpp
index c858dfdd589..b5653fca1e9 100644
--- a/src/Functions/castOrDefault.cpp
+++ b/src/Functions/castOrDefault.cpp
@@ -320,16 +320,16 @@ REGISTER_FUNCTION(CastOrDefault)
 {
     factory.registerFunction<FunctionCastOrDefault>();
 
-    factory.registerFunction("toUInt8OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt8OrDefault", std::make_shared<DataTypeUInt8>())); });
-    factory.registerFunction("toUInt16OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt16OrDefault", std::make_shared<DataTypeUInt16>())); });
-    factory.registerFunction("toUInt32OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt32OrDefault", std::make_shared<DataTypeUInt32>())); });
-    factory.registerFunction("toUInt64OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt64OrDefault", std::make_shared<DataTypeUInt64>())); });
-    factory.registerFunction("toUInt128OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt128OrDefault", std::make_shared<DataTypeUInt128>())); },
+    factory.registerFunction("toUInt8OrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt8OrDefault", std::make_shared<DataTypeUInt8>()); });
+    factory.registerFunction("toUInt16OrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt16OrDefault", std::make_shared<DataTypeUInt16>()); });
+    factory.registerFunction("toUInt32OrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt32OrDefault", std::make_shared<DataTypeUInt32>()); });
+    factory.registerFunction("toUInt64OrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt64OrDefault", std::make_shared<DataTypeUInt64>()); });
+    factory.registerFunction("toUInt128OrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt128OrDefault", std::make_shared<DataTypeUInt128>()); },
         FunctionDocumentation{
             .description=R"(
 Converts a string in the first argument of the function to UInt128 by parsing it.
@@ -342,51 +342,51 @@ If the default value is not provided in the second argument, it is assumed to be
                 {"Implicit default value", "SELECT toUInt128OrDefault('upyachka')", "0"}},
             .categories{"ConversionFunctions"}
         });
-    factory.registerFunction("toUInt256OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt256OrDefault", std::make_shared<DataTypeUInt256>())); });
+    factory.registerFunction("toUInt256OrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt256OrDefault", std::make_shared<DataTypeUInt256>()); });
 
-    factory.registerFunction("toInt8OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt8OrDefault", std::make_shared<DataTypeInt8>())); });
-    factory.registerFunction("toInt16OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt16OrDefault", std::make_shared<DataTypeInt16>())); });
-    factory.registerFunction("toInt32OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt32OrDefault", std::make_shared<DataTypeInt32>())); });
-    factory.registerFunction("toInt64OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt64OrDefault", std::make_shared<DataTypeInt64>())); });
-    factory.registerFunction("toInt128OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt128OrDefault", std::make_shared<DataTypeInt128>())); });
-    factory.registerFunction("toInt256OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt256OrDefault", std::make_shared<DataTypeInt256>())); });
+    factory.registerFunction("toInt8OrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt8OrDefault", std::make_shared<DataTypeInt8>()); });
+    factory.registerFunction("toInt16OrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt16OrDefault", std::make_shared<DataTypeInt16>()); });
+    factory.registerFunction("toInt32OrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt32OrDefault", std::make_shared<DataTypeInt32>()); });
+    factory.registerFunction("toInt64OrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt64OrDefault", std::make_shared<DataTypeInt64>()); });
+    factory.registerFunction("toInt128OrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt128OrDefault", std::make_shared<DataTypeInt128>()); });
+    factory.registerFunction("toInt256OrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt256OrDefault", std::make_shared<DataTypeInt256>()); });
 
-    factory.registerFunction("toFloat32OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toFloat32OrDefault", std::make_shared<DataTypeFloat32>())); });
-    factory.registerFunction("toFloat64OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toFloat64OrDefault", std::make_shared<DataTypeFloat64>())); });
+    factory.registerFunction("toFloat32OrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toFloat32OrDefault", std::make_shared<DataTypeFloat32>()); });
+    factory.registerFunction("toFloat64OrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toFloat64OrDefault", std::make_shared<DataTypeFloat64>()); });
 
-    factory.registerFunction("toDateOrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toDateOrDefault", std::make_shared<DataTypeDate>())); });
-    factory.registerFunction("toDate32OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toDate32OrDefault", std::make_shared<DataTypeDate32>())); });
-    factory.registerFunction("toDateTimeOrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toDateTimeOrDefault", std::make_shared<DataTypeDateTime>())); });
-    factory.registerFunction("toDateTime64OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toDateTime64OrDefault", std::make_shared<DataTypeDateTime64>(3 /* default scale */))); });
+    factory.registerFunction("toDateOrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDateOrDefault", std::make_shared<DataTypeDate>()); });
+    factory.registerFunction("toDate32OrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDate32OrDefault", std::make_shared<DataTypeDate32>()); });
+    factory.registerFunction("toDateTimeOrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDateTimeOrDefault", std::make_shared<DataTypeDateTime>()); });
+    factory.registerFunction("toDateTime64OrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDateTime64OrDefault", std::make_shared<DataTypeDateTime64>(3 /* default scale */)); });
 
-    factory.registerFunction("toDecimal32OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toDecimal32OrDefault", createDecimalMaxPrecision<Decimal32>(0))); });
-    factory.registerFunction("toDecimal64OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toDecimal64OrDefault", createDecimalMaxPrecision<Decimal64>(0))); });
-    factory.registerFunction("toDecimal128OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toDecimal128OrDefault", createDecimalMaxPrecision<Decimal128>(0))); });
-    factory.registerFunction("toDecimal256OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toDecimal256OrDefault", createDecimalMaxPrecision<Decimal256>(0))); });
+    factory.registerFunction("toDecimal32OrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDecimal32OrDefault", createDecimalMaxPrecision<Decimal32>(0)); });
+    factory.registerFunction("toDecimal64OrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDecimal64OrDefault", createDecimalMaxPrecision<Decimal64>(0)); });
+    factory.registerFunction("toDecimal128OrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDecimal128OrDefault", createDecimalMaxPrecision<Decimal128>(0)); });
+    factory.registerFunction("toDecimal256OrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDecimal256OrDefault", createDecimalMaxPrecision<Decimal256>(0)); });
 
-    factory.registerFunction("toUUIDOrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toUUIDOrDefault", std::make_shared<DataTypeUUID>())); });
-    factory.registerFunction("toIPv4OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toIPv4OrDefault", std::make_shared<DataTypeIPv4>())); });
-    factory.registerFunction("toIPv6OrDefault", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-        std::make_shared<FunctionCastOrDefaultTyped>(context, "toIPv6OrDefault", std::make_shared<DataTypeIPv6>())); });
+    factory.registerFunction("toUUIDOrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toUUIDOrDefault", std::make_shared<DataTypeUUID>()); });
+    factory.registerFunction("toIPv4OrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toIPv4OrDefault", std::make_shared<DataTypeIPv4>()); });
+    factory.registerFunction("toIPv6OrDefault", [](ContextPtr context){
+        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toIPv6OrDefault", std::make_shared<DataTypeIPv6>()); });
 }
 
 }
diff --git a/src/Functions/coverage.cpp b/src/Functions/coverage.cpp
index a1a43d0cf58..97f807e22b7 100644
--- a/src/Functions/coverage.cpp
+++ b/src/Functions/coverage.cpp
@@ -93,7 +93,7 @@ public:
 
 REGISTER_FUNCTION(Coverage)
 {
-    factory.registerFunction("coverageCurrent", [](ContextPtr){ return std::make_unique<FunctionToOverloadResolverAdaptor>(std::make_shared<FunctionCoverage>(Kind::Current)); },
+    factory.registerFunction("coverageCurrent", [](ContextPtr){ return std::make_shared<FunctionCoverage>(Kind::Current); },
         FunctionDocumentation
         {
             .description=R"(
@@ -124,7 +124,7 @@ See https://clang.llvm.org/docs/SanitizerCoverage.html for more information.
             .categories{"Introspection"}
         });
 
-    factory.registerFunction("coverageCumulative", [](ContextPtr){ return std::make_unique<FunctionToOverloadResolverAdaptor>(std::make_shared<FunctionCoverage>(Kind::Cumulative)); },
+    factory.registerFunction("coverageCumulative", [](ContextPtr){ return std::make_shared<FunctionCoverage>(Kind::Cumulative); },
         FunctionDocumentation
         {
             .description=R"(
@@ -140,7 +140,7 @@ See the `coverageCurrent` function for the details.
             .categories{"Introspection"}
         });
 
-    factory.registerFunction("coverageAll", [](ContextPtr){ return std::make_unique<FunctionToOverloadResolverAdaptor>(std::make_shared<FunctionCoverage>(Kind::All)); },
+    factory.registerFunction("coverageAll", [](ContextPtr){ return std::make_shared<FunctionCoverage>(Kind::All); },
         FunctionDocumentation
         {
             .description=R"(
diff --git a/src/Functions/currentProfiles.cpp b/src/Functions/currentProfiles.cpp
index 77c8a20ccee..8f14943e011 100644
--- a/src/Functions/currentProfiles.cpp
+++ b/src/Functions/currentProfiles.cpp
@@ -98,9 +98,9 @@ namespace
 
 REGISTER_FUNCTION(Profiles)
 {
-    factory.registerFunction("currentProfiles", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(std::make_shared<FunctionProfiles>(context, Kind::currentProfiles)); });
-    factory.registerFunction("enabledProfiles", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(std::make_shared<FunctionProfiles>(context, Kind::enabledProfiles)); });
-    factory.registerFunction("defaultProfiles", [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(std::make_shared<FunctionProfiles>(context, Kind::defaultProfiles)); });
+    factory.registerFunction("currentProfiles", [](ContextPtr context){ return std::make_shared<FunctionProfiles>(context, Kind::currentProfiles); });
+    factory.registerFunction("enabledProfiles", [](ContextPtr context){ return std::make_shared<FunctionProfiles>(context, Kind::enabledProfiles); });
+    factory.registerFunction("defaultProfiles", [](ContextPtr context){ return std::make_shared<FunctionProfiles>(context, Kind::defaultProfiles); });
 }
 
 }
diff --git a/src/Functions/fromUnixTimestamp64Micro.cpp b/src/Functions/fromUnixTimestamp64Micro.cpp
index 191e2137a0d..d96e0232335 100644
--- a/src/Functions/fromUnixTimestamp64Micro.cpp
+++ b/src/Functions/fromUnixTimestamp64Micro.cpp
@@ -7,8 +7,7 @@ namespace DB
 REGISTER_FUNCTION(FromUnixTimestamp64Micro)
 {
     factory.registerFunction("fromUnixTimestamp64Micro",
-        [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-            std::make_shared<FunctionFromUnixTimestamp64>(6, "fromUnixTimestamp64Micro", context)); });
+        [](ContextPtr context){ return std::make_shared<FunctionFromUnixTimestamp64>(6, "fromUnixTimestamp64Micro", context); });
 }
 
 }
diff --git a/src/Functions/fromUnixTimestamp64Milli.cpp b/src/Functions/fromUnixTimestamp64Milli.cpp
index c6d4fcd30a2..aa77e8043c1 100644
--- a/src/Functions/fromUnixTimestamp64Milli.cpp
+++ b/src/Functions/fromUnixTimestamp64Milli.cpp
@@ -7,8 +7,7 @@ namespace DB
 REGISTER_FUNCTION(FromUnixTimestamp64Milli)
 {
     factory.registerFunction("fromUnixTimestamp64Milli",
-        [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-            std::make_shared<FunctionFromUnixTimestamp64>(3, "fromUnixTimestamp64Milli", context)); });
+        [](ContextPtr context){ return std::make_shared<FunctionFromUnixTimestamp64>(3, "fromUnixTimestamp64Milli", context); });
 }
 
 }
diff --git a/src/Functions/fromUnixTimestamp64Nano.cpp b/src/Functions/fromUnixTimestamp64Nano.cpp
index 2b5a7addbfc..f9d69219933 100644
--- a/src/Functions/fromUnixTimestamp64Nano.cpp
+++ b/src/Functions/fromUnixTimestamp64Nano.cpp
@@ -7,8 +7,7 @@ namespace DB
 REGISTER_FUNCTION(FromUnixTimestamp64Nano)
 {
     factory.registerFunction("fromUnixTimestamp64Nano",
-        [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-            std::make_shared<FunctionFromUnixTimestamp64>(9, "fromUnixTimestamp64Nano", context)); });
+        [](ContextPtr context){ return std::make_shared<FunctionFromUnixTimestamp64>(9, "fromUnixTimestamp64Nano", context); });
 }
 
 }
diff --git a/src/Functions/snowflake.cpp b/src/Functions/snowflake.cpp
index f2dd1f1c51d..4a2d502a31a 100644
--- a/src/Functions/snowflake.cpp
+++ b/src/Functions/snowflake.cpp
@@ -249,28 +249,24 @@ public:
 REGISTER_FUNCTION(DateTimeToSnowflake)
 {
     factory.registerFunction("dateTimeToSnowflake",
-        [](ContextPtr){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-            std::make_shared<FunctionDateTimeToSnowflake>("dateTimeToSnowflake")); });
+        [](ContextPtr){ return std::make_shared<FunctionDateTimeToSnowflake>("dateTimeToSnowflake"); });
 }
 
 REGISTER_FUNCTION(DateTime64ToSnowflake)
 {
     factory.registerFunction("dateTime64ToSnowflake",
-        [](ContextPtr){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-            std::make_shared<FunctionDateTime64ToSnowflake>("dateTime64ToSnowflake")); });
+        [](ContextPtr){ return std::make_shared<FunctionDateTime64ToSnowflake>("dateTime64ToSnowflake"); });
 }
 
 REGISTER_FUNCTION(SnowflakeToDateTime)
 {
     factory.registerFunction("snowflakeToDateTime",
-        [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-            std::make_shared<FunctionSnowflakeToDateTime>("snowflakeToDateTime", context)); });
+        [](ContextPtr context){ return std::make_shared<FunctionSnowflakeToDateTime>("snowflakeToDateTime", context); });
 }
 REGISTER_FUNCTION(SnowflakeToDateTime64)
 {
     factory.registerFunction("snowflakeToDateTime64",
-        [](ContextPtr context){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-            std::make_shared<FunctionSnowflakeToDateTime64>("snowflakeToDateTime64", context)); });
+        [](ContextPtr context){ return std::make_shared<FunctionSnowflakeToDateTime64>("snowflakeToDateTime64", context); });
 }
 
 }
diff --git a/src/Functions/toUnixTimestamp64Micro.cpp b/src/Functions/toUnixTimestamp64Micro.cpp
index fd35e2a7a73..964ad5a2c18 100644
--- a/src/Functions/toUnixTimestamp64Micro.cpp
+++ b/src/Functions/toUnixTimestamp64Micro.cpp
@@ -7,8 +7,7 @@ namespace DB
 REGISTER_FUNCTION(ToUnixTimestamp64Micro)
 {
     factory.registerFunction("toUnixTimestamp64Micro",
-        [](ContextPtr){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-            std::make_shared<FunctionToUnixTimestamp64>(6, "toUnixTimestamp64Micro")); });
+        [](ContextPtr){ return std::make_shared<FunctionToUnixTimestamp64>(6, "toUnixTimestamp64Micro"); });
 }
 
 }
diff --git a/src/Functions/toUnixTimestamp64Milli.cpp b/src/Functions/toUnixTimestamp64Milli.cpp
index e6a680f941a..bc92a6d1fe3 100644
--- a/src/Functions/toUnixTimestamp64Milli.cpp
+++ b/src/Functions/toUnixTimestamp64Milli.cpp
@@ -7,8 +7,7 @@ namespace DB
 REGISTER_FUNCTION(ToUnixTimestamp64Milli)
 {
     factory.registerFunction("toUnixTimestamp64Milli",
-        [](ContextPtr){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-            std::make_shared<FunctionToUnixTimestamp64>(3, "toUnixTimestamp64Milli")); });
+        [](ContextPtr){ return std::make_shared<FunctionToUnixTimestamp64>(3, "toUnixTimestamp64Milli"); });
 }
 
 }
diff --git a/src/Functions/toUnixTimestamp64Nano.cpp b/src/Functions/toUnixTimestamp64Nano.cpp
index 257f011603c..8829b00bf56 100644
--- a/src/Functions/toUnixTimestamp64Nano.cpp
+++ b/src/Functions/toUnixTimestamp64Nano.cpp
@@ -7,8 +7,7 @@ namespace DB
 REGISTER_FUNCTION(ToUnixTimestamp64Nano)
 {
     factory.registerFunction("toUnixTimestamp64Nano",
-        [](ContextPtr){ return std::make_unique<FunctionToOverloadResolverAdaptor>(
-            std::make_shared<FunctionToUnixTimestamp64>(9, "toUnixTimestamp64Nano")); });
+        [](ContextPtr){ return std::make_shared<FunctionToUnixTimestamp64>(9, "toUnixTimestamp64Nano"); });
 }
 
 }

From 427a8b3264cd9000b6f7ab13f13ceac0878d000a Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 17 Mar 2024 13:20:49 +0100
Subject: [PATCH 413/443] Less crap

---
 programs/local/LocalServer.cpp        |  4 ++++
 src/Functions/CRC.cpp                 | 21 ++++++++-------------
 src/Functions/FunctionFQDN.cpp        |  2 +-
 src/Functions/FunctionFactory.h       | 21 ++++++---------------
 src/Functions/FunctionsConversion.cpp | 15 ++++++---------
 src/Functions/FunctionsRound.cpp      | 10 +++++-----
 src/Functions/caseWithExpression.cpp  |  4 +---
 src/Functions/getFuzzerData.h         |  8 ++------
 src/Functions/multiIf.cpp             |  6 ++----
 src/Functions/toFixedString.h         |  2 --
 10 files changed, 35 insertions(+), 58 deletions(-)

diff --git a/programs/local/LocalServer.cpp b/programs/local/LocalServer.cpp
index 50f9c242712..0dd3823cc6d 100644
--- a/programs/local/LocalServer.cpp
+++ b/programs/local/LocalServer.cpp
@@ -1013,13 +1013,16 @@ extern "C" int LLVMFuzzerInitialize(int * pargc, char *** pargv)
 
     auto it = argv.begin() + 1;
     for (; *it; ++it)
+    {
         if (strcmp(*it, "--") == 0)
         {
             ++it;
             break;
         }
+    }
 
     while (*it)
+    {
         if (strncmp(*it, "--", 2) != 0)
         {
             *(p++) = *it;
@@ -1027,6 +1030,7 @@ extern "C" int LLVMFuzzerInitialize(int * pargc, char *** pargv)
         }
         else
             ++it;
+    }
 
     *pargc = static_cast<int>(p - &(*pargv)[0]);
     *p = nullptr;
diff --git a/src/Functions/CRC.cpp b/src/Functions/CRC.cpp
index ba13fcf78f1..49d6dd6fa52 100644
--- a/src/Functions/CRC.cpp
+++ b/src/Functions/CRC.cpp
@@ -51,7 +51,7 @@ struct CRC32IEEEImpl : public CRCImpl<UInt32, CRC32_IEEE>
     static constexpr auto name = "CRC32IEEE";
 };
 
-struct CRC32ZLIBImpl
+struct CRC32ZLibImpl
 {
     using ReturnType = UInt32;
     static constexpr auto name = "CRC32";
@@ -133,13 +133,14 @@ private:
     }
 };
 
-template <class T>
+template <typename T>
 using FunctionCRC = FunctionStringOrArrayToT<CRCFunctionWrapper<T>, T, typename T::ReturnType>;
+
 // The same as IEEE variant, but uses 0xffffffff as initial value
 // This is the default
 //
-// (And zlib is used here, since it has optimized version)
-using FunctionCRC32ZLIB = FunctionCRC<CRC32ZLIBImpl>;
+// (And ZLib is used here, since it has optimized version)
+using FunctionCRC32ZLib = FunctionCRC<CRC32ZLibImpl>;
 // Uses CRC-32-IEEE 802.3 polynomial
 using FunctionCRC32IEEE = FunctionCRC<CRC32IEEEImpl>;
 // Uses CRC-64-ECMA polynomial
@@ -147,17 +148,11 @@ using FunctionCRC64ECMA = FunctionCRC<CRC64ECMAImpl>;
 
 }
 
-template <class T>
-void registerFunctionCRCImpl(FunctionFactory & factory)
-{
-    factory.registerFunction<T>(T::name, {}, FunctionFactory::CaseInsensitive);
-}
-
 REGISTER_FUNCTION(CRC)
 {
-    registerFunctionCRCImpl<FunctionCRC32ZLIB>(factory);
-    registerFunctionCRCImpl<FunctionCRC32IEEE>(factory);
-    registerFunctionCRCImpl<FunctionCRC64ECMA>(factory);
+    factory.registerFunction<FunctionCRC32ZLib>({}, FunctionFactory::CaseInsensitive);
+    factory.registerFunction<FunctionCRC32IEEE>({}, FunctionFactory::CaseInsensitive);
+    factory.registerFunction<FunctionCRC64ECMA>({}, FunctionFactory::CaseInsensitive);
 }
 
 }
diff --git a/src/Functions/FunctionFQDN.cpp b/src/Functions/FunctionFQDN.cpp
index b054ff8e1d7..108a96216fd 100644
--- a/src/Functions/FunctionFQDN.cpp
+++ b/src/Functions/FunctionFQDN.cpp
@@ -47,7 +47,7 @@ public:
 REGISTER_FUNCTION(FQDN)
 {
     factory.registerFunction<FunctionFQDN>({}, FunctionFactory::CaseInsensitive);
-    factory.registerFunction<FunctionFQDN>("fullHostName");
+    factory.registerAlias("fullHostName", "FQDN");
 }
 
 }
diff --git a/src/Functions/FunctionFactory.h b/src/Functions/FunctionFactory.h
index c2fd34f0488..bb43d4719b8 100644
--- a/src/Functions/FunctionFactory.h
+++ b/src/Functions/FunctionFactory.h
@@ -35,15 +35,6 @@ public:
         registerFunction<Function>(Function::name, std::move(doc), case_sensitiveness);
     }
 
-    template <typename Function>
-    void registerFunction(const std::string & name, FunctionDocumentation doc = {}, CaseSensitiveness case_sensitiveness = CaseSensitive)
-    {
-        if constexpr (std::is_base_of_v<IFunction, Function>)
-            registerFunction(name, &adaptFunctionToOverloadResolver<Function>, std::move(doc), case_sensitiveness);
-        else
-            registerFunction(name, &Function::create, std::move(doc), case_sensitiveness);
-    }
-
     /// This function is used by YQL - innovative transactional DBMS that depends on ClickHouse by source code.
     std::vector<std::string> getAllNames() const;
 
@@ -81,17 +72,17 @@ private:
     Functions functions;
     Functions case_insensitive_functions;
 
-    template <typename Function>
-    static FunctionOverloadResolverPtr adaptFunctionToOverloadResolver(ContextPtr context)
-    {
-        return std::make_unique<FunctionToOverloadResolverAdaptor>(Function::create(context));
-    }
-
     const Functions & getMap() const override { return functions; }
 
     const Functions & getCaseInsensitiveMap() const override { return case_insensitive_functions; }
 
     String getFactoryName() const override { return "FunctionFactory"; }
+
+    template <typename Function>
+    void registerFunction(const std::string & name, FunctionDocumentation doc = {}, CaseSensitiveness case_sensitiveness = CaseSensitive)
+    {
+        registerFunction(name, &Function::create, std::move(doc), case_sensitiveness);
+    }
 };
 
 const String & getFunctionCanonicalNameIfAny(const String & name);
diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp
index 0d245f809bb..ceff4f3fd7e 100644
--- a/src/Functions/FunctionsConversion.cpp
+++ b/src/Functions/FunctionsConversion.cpp
@@ -1955,9 +1955,6 @@ public:
     static constexpr bool to_decimal = IsDataTypeDecimal<ToDataType> && !to_datetime64;
 
     static FunctionPtr create(ContextPtr context) { return std::make_shared<FunctionConvert>(context); }
-    static FunctionPtr create() { return std::make_shared<FunctionConvert>(); }
-
-    FunctionConvert() = default;
     explicit FunctionConvert(ContextPtr context_) : context(context_) {}
 
     String getName() const override
@@ -3295,9 +3292,9 @@ arguments, result_type, input_rows_count); \
         };
     }
 
-    static WrapperType createStringWrapper(const DataTypePtr & from_type)
+    WrapperType createStringWrapper(const DataTypePtr & from_type) const
     {
-        FunctionPtr function = FunctionToString::create();
+        FunctionPtr function = FunctionToString::create(context);
         return createFunctionAdaptor(function, from_type);
     }
 
@@ -3318,9 +3315,9 @@ arguments, result_type, input_rows_count); \
 
 #define GENERATE_INTERVAL_CASE(INTERVAL_KIND) \
             case IntervalKind::Kind::INTERVAL_KIND: \
-                return createFunctionAdaptor(FunctionConvert<DataTypeInterval, NameToInterval##INTERVAL_KIND, PositiveMonotonicity>::create(), from_type);
+                return createFunctionAdaptor(FunctionConvert<DataTypeInterval, NameToInterval##INTERVAL_KIND, PositiveMonotonicity>::create(context), from_type);
 
-    static WrapperType createIntervalWrapper(const DataTypePtr & from_type, IntervalKind kind)
+    WrapperType createIntervalWrapper(const DataTypePtr & from_type, IntervalKind kind) const
     {
         switch (kind.kind)
         {
@@ -4207,7 +4204,7 @@ arguments, result_type, input_rows_count); \
             return createStringToEnumWrapper<ColumnFixedString, EnumType>();
         else if (isNativeNumber(from_type) || isEnum(from_type))
         {
-            auto function = Function::create();
+            auto function = Function::create(context);
             return createFunctionAdaptor(function, from_type);
         }
         else
@@ -4846,7 +4843,7 @@ REGISTER_FUNCTION(Conversion)
     /// MySQL compatibility alias. Cannot be registered as alias,
     /// because we don't want it to be normalized to toDate in queries,
     /// otherwise CREATE DICTIONARY query breaks.
-    factory.registerFunction<FunctionToDate>("DATE", {}, FunctionFactory::CaseInsensitive);
+    factory.registerFunction("DATE", &FunctionToDate::create, {}, FunctionFactory::CaseInsensitive);
 
     factory.registerFunction<FunctionToDate32>();
     factory.registerFunction<FunctionToDateTime>();
diff --git a/src/Functions/FunctionsRound.cpp b/src/Functions/FunctionsRound.cpp
index 02fe1d659de..059476acb40 100644
--- a/src/Functions/FunctionsRound.cpp
+++ b/src/Functions/FunctionsRound.cpp
@@ -7,11 +7,11 @@ namespace DB
 
 REGISTER_FUNCTION(Round)
 {
-    factory.registerFunction<FunctionRound>("round", {}, FunctionFactory::CaseInsensitive);
-    factory.registerFunction<FunctionRoundBankers>("roundBankers", {}, FunctionFactory::CaseSensitive);
-    factory.registerFunction<FunctionFloor>("floor", {}, FunctionFactory::CaseInsensitive);
-    factory.registerFunction<FunctionCeil>("ceil", {}, FunctionFactory::CaseInsensitive);
-    factory.registerFunction<FunctionTrunc>("trunc", {}, FunctionFactory::CaseInsensitive);
+    factory.registerFunction<FunctionRound>({}, FunctionFactory::CaseInsensitive);
+    factory.registerFunction<FunctionRoundBankers>({}, FunctionFactory::CaseSensitive);
+    factory.registerFunction<FunctionFloor>({}, FunctionFactory::CaseInsensitive);
+    factory.registerFunction<FunctionCeil>({}, FunctionFactory::CaseInsensitive);
+    factory.registerFunction<FunctionTrunc>({}, FunctionFactory::CaseInsensitive);
     factory.registerFunction<FunctionRoundDown>();
 
     /// Compatibility aliases.
diff --git a/src/Functions/caseWithExpression.cpp b/src/Functions/caseWithExpression.cpp
index 9547cd200b2..71fccc8436e 100644
--- a/src/Functions/caseWithExpression.cpp
+++ b/src/Functions/caseWithExpression.cpp
@@ -113,9 +113,7 @@ REGISTER_FUNCTION(CaseWithExpression)
     factory.registerFunction<FunctionCaseWithExpression>();
 
     /// These are obsolete function names.
-    factory.registerFunction<FunctionCaseWithExpression>("caseWithExpr");
+    factory.registerAlias("caseWithExpr", "caseWithExpression");
 }
 
 }
-
-
diff --git a/src/Functions/getFuzzerData.h b/src/Functions/getFuzzerData.h
index 635ca2bdce9..8d7b3c090c4 100644
--- a/src/Functions/getFuzzerData.h
+++ b/src/Functions/getFuzzerData.h
@@ -7,6 +7,7 @@
 
 namespace DB
 {
+
 class FunctionGetFuzzerData : public IFunction
 {
     inline static String fuzz_data;
@@ -14,12 +15,7 @@ class FunctionGetFuzzerData : public IFunction
 public:
     static constexpr auto name = "getFuzzerData";
 
-    inline static FunctionPtr create(ContextPtr) { return create(); }
-
-    static FunctionPtr create()
-    {
-        return std::make_shared<FunctionGetFuzzerData>();
-    }
+    inline static FunctionPtr create(ContextPtr) { return std::make_shared<FunctionGetFuzzerData>(); }
 
     inline String getName() const override { return name; }
 
diff --git a/src/Functions/multiIf.cpp b/src/Functions/multiIf.cpp
index 81304f3afbd..49c45d0c0be 100644
--- a/src/Functions/multiIf.cpp
+++ b/src/Functions/multiIf.cpp
@@ -549,8 +549,8 @@ REGISTER_FUNCTION(MultiIf)
     factory.registerFunction<FunctionMultiIf>();
 
     /// These are obsolete function names.
-    factory.registerFunction<FunctionMultiIf>("caseWithoutExpr");
-    factory.registerFunction<FunctionMultiIf>("caseWithoutExpression");
+    factory.registerAlias("caseWithoutExpr", "multiIf");
+    factory.registerAlias("caseWithoutExpression", "multiIf");
 }
 
 FunctionOverloadResolverPtr createInternalMultiIfOverloadResolver(bool allow_execute_multiif_columnar, bool allow_experimental_variant_type, bool use_variant_as_common_type)
@@ -559,5 +559,3 @@ FunctionOverloadResolverPtr createInternalMultiIfOverloadResolver(bool allow_exe
 }
 
 }
-
-
diff --git a/src/Functions/toFixedString.h b/src/Functions/toFixedString.h
index 7bee666c5dd..9c7ffc48004 100644
--- a/src/Functions/toFixedString.h
+++ b/src/Functions/toFixedString.h
@@ -34,7 +34,6 @@ class FunctionToFixedString : public IFunction
 public:
     static constexpr auto name = "toFixedString";
     static FunctionPtr create(ContextPtr) { return std::make_shared<FunctionToFixedString>(); }
-    static FunctionPtr create() { return std::make_shared<FunctionToFixedString>(); }
 
     String getName() const override
     {
@@ -158,4 +157,3 @@ public:
 };
 
 }
-

From 7c35f1d07e5d9c344bdb3a14e6f03efd76212e4e Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 17 Mar 2024 13:23:09 +0100
Subject: [PATCH 414/443] Less crap

---
 src/Functions/getFuzzerData.cpp | 48 ++++++++++++++++++++++++++++++++-
 src/Functions/getFuzzerData.h   | 46 -------------------------------
 2 files changed, 47 insertions(+), 47 deletions(-)
 delete mode 100644 src/Functions/getFuzzerData.h

diff --git a/src/Functions/getFuzzerData.cpp b/src/Functions/getFuzzerData.cpp
index 6d748619926..5c536477401 100644
--- a/src/Functions/getFuzzerData.cpp
+++ b/src/Functions/getFuzzerData.cpp
@@ -1,13 +1,59 @@
 #ifdef FUZZING_MODE
-#include <Functions/getFuzzerData.h>
+
+#include <Functions/IFunction.h>
+#include <Functions/FunctionFactory.h>
+#include <DataTypes/DataTypeString.h>
+#include <Core/Field.h>
+
 
 namespace DB
 {
 
+namespace
+{
+
+class FunctionGetFuzzerData : public IFunction
+{
+    inline static String fuzz_data;
+
+public:
+    static constexpr auto name = "getFuzzerData";
+
+    inline static FunctionPtr create(ContextPtr) { return std::make_shared<FunctionGetFuzzerData>(); }
+
+    inline String getName() const override { return name; }
+
+    inline size_t getNumberOfArguments() const override { return 0; }
+
+    DataTypePtr getReturnTypeImpl(const DataTypes & /*arguments*/) const override
+    {
+        return std::make_shared<DataTypeString>();
+    }
+
+    inline bool isDeterministic() const override { return false; }
+
+    bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return false; }
+
+    ColumnPtr executeImpl(const ColumnsWithTypeAndName &,
+                          const DataTypePtr &,
+                          size_t input_rows_count) const override
+    {
+        return DataTypeString().createColumnConst(input_rows_count, fuzz_data);
+    }
+
+    static void update(const String & fuzz_data_)
+    {
+        fuzz_data = fuzz_data_;
+    }
+};
+
+}
+
 REGISTER_FUNCTION(GetFuzzerData)
 {
     factory.registerFunction<FunctionGetFuzzerData>();
 }
 
 }
+
 #endif
diff --git a/src/Functions/getFuzzerData.h b/src/Functions/getFuzzerData.h
deleted file mode 100644
index 8d7b3c090c4..00000000000
--- a/src/Functions/getFuzzerData.h
+++ /dev/null
@@ -1,46 +0,0 @@
-#pragma once
-
-#include <Functions/IFunction.h>
-#include <Functions/FunctionFactory.h>
-#include <DataTypes/DataTypeString.h>
-#include <Core/Field.h>
-
-namespace DB
-{
-
-class FunctionGetFuzzerData : public IFunction
-{
-    inline static String fuzz_data;
-
-public:
-    static constexpr auto name = "getFuzzerData";
-
-    inline static FunctionPtr create(ContextPtr) { return std::make_shared<FunctionGetFuzzerData>(); }
-
-    inline String getName() const override { return name; }
-
-    inline size_t getNumberOfArguments() const override { return 0; }
-
-    DataTypePtr getReturnTypeImpl(const DataTypes & /*arguments*/) const override
-    {
-        return std::make_shared<DataTypeString>();
-    }
-
-    inline bool isDeterministic() const override { return false; }
-
-    bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return false; }
-
-    ColumnPtr executeImpl(const ColumnsWithTypeAndName &,
-                          const DataTypePtr &,
-                          size_t input_rows_count) const override
-    {
-        return DataTypeString().createColumnConst(input_rows_count, fuzz_data);
-    }
-
-    static void update(const String & fuzz_data_)
-    {
-        fuzz_data = fuzz_data_;
-    }
-};
-
-}

From eec0bf2f52e0f7465de2e4b78af5886d13d14893 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 17 Mar 2024 13:34:34 +0100
Subject: [PATCH 415/443] Fix style

---
 src/Functions/castOrDefault.cpp | 100 ++++++++++++++++----------------
 1 file changed, 50 insertions(+), 50 deletions(-)

diff --git a/src/Functions/castOrDefault.cpp b/src/Functions/castOrDefault.cpp
index b5653fca1e9..44b39811882 100644
--- a/src/Functions/castOrDefault.cpp
+++ b/src/Functions/castOrDefault.cpp
@@ -320,16 +320,16 @@ REGISTER_FUNCTION(CastOrDefault)
 {
     factory.registerFunction<FunctionCastOrDefault>();
 
-    factory.registerFunction("toUInt8OrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt8OrDefault", std::make_shared<DataTypeUInt8>()); });
-    factory.registerFunction("toUInt16OrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt16OrDefault", std::make_shared<DataTypeUInt16>()); });
-    factory.registerFunction("toUInt32OrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt32OrDefault", std::make_shared<DataTypeUInt32>()); });
-    factory.registerFunction("toUInt64OrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt64OrDefault", std::make_shared<DataTypeUInt64>()); });
-    factory.registerFunction("toUInt128OrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt128OrDefault", std::make_shared<DataTypeUInt128>()); },
+    factory.registerFunction("toUInt8OrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt8OrDefault", std::make_shared<DataTypeUInt8>()); });
+    factory.registerFunction("toUInt16OrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt16OrDefault", std::make_shared<DataTypeUInt16>()); });
+    factory.registerFunction("toUInt32OrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt32OrDefault", std::make_shared<DataTypeUInt32>()); });
+    factory.registerFunction("toUInt64OrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt64OrDefault", std::make_shared<DataTypeUInt64>()); });
+    factory.registerFunction("toUInt128OrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt128OrDefault", std::make_shared<DataTypeUInt128>()); },
         FunctionDocumentation{
             .description=R"(
 Converts a string in the first argument of the function to UInt128 by parsing it.
@@ -342,51 +342,51 @@ If the default value is not provided in the second argument, it is assumed to be
                 {"Implicit default value", "SELECT toUInt128OrDefault('upyachka')", "0"}},
             .categories{"ConversionFunctions"}
         });
-    factory.registerFunction("toUInt256OrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt256OrDefault", std::make_shared<DataTypeUInt256>()); });
+    factory.registerFunction("toUInt256OrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toUInt256OrDefault", std::make_shared<DataTypeUInt256>()); });
 
-    factory.registerFunction("toInt8OrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt8OrDefault", std::make_shared<DataTypeInt8>()); });
-    factory.registerFunction("toInt16OrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt16OrDefault", std::make_shared<DataTypeInt16>()); });
-    factory.registerFunction("toInt32OrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt32OrDefault", std::make_shared<DataTypeInt32>()); });
-    factory.registerFunction("toInt64OrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt64OrDefault", std::make_shared<DataTypeInt64>()); });
-    factory.registerFunction("toInt128OrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt128OrDefault", std::make_shared<DataTypeInt128>()); });
-    factory.registerFunction("toInt256OrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt256OrDefault", std::make_shared<DataTypeInt256>()); });
+    factory.registerFunction("toInt8OrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt8OrDefault", std::make_shared<DataTypeInt8>()); });
+    factory.registerFunction("toInt16OrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt16OrDefault", std::make_shared<DataTypeInt16>()); });
+    factory.registerFunction("toInt32OrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt32OrDefault", std::make_shared<DataTypeInt32>()); });
+    factory.registerFunction("toInt64OrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt64OrDefault", std::make_shared<DataTypeInt64>()); });
+    factory.registerFunction("toInt128OrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt128OrDefault", std::make_shared<DataTypeInt128>()); });
+    factory.registerFunction("toInt256OrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toInt256OrDefault", std::make_shared<DataTypeInt256>()); });
 
-    factory.registerFunction("toFloat32OrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toFloat32OrDefault", std::make_shared<DataTypeFloat32>()); });
-    factory.registerFunction("toFloat64OrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toFloat64OrDefault", std::make_shared<DataTypeFloat64>()); });
+    factory.registerFunction("toFloat32OrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toFloat32OrDefault", std::make_shared<DataTypeFloat32>()); });
+    factory.registerFunction("toFloat64OrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toFloat64OrDefault", std::make_shared<DataTypeFloat64>()); });
 
-    factory.registerFunction("toDateOrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDateOrDefault", std::make_shared<DataTypeDate>()); });
-    factory.registerFunction("toDate32OrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDate32OrDefault", std::make_shared<DataTypeDate32>()); });
-    factory.registerFunction("toDateTimeOrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDateTimeOrDefault", std::make_shared<DataTypeDateTime>()); });
-    factory.registerFunction("toDateTime64OrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDateTime64OrDefault", std::make_shared<DataTypeDateTime64>(3 /* default scale */)); });
+    factory.registerFunction("toDateOrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDateOrDefault", std::make_shared<DataTypeDate>()); });
+    factory.registerFunction("toDate32OrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDate32OrDefault", std::make_shared<DataTypeDate32>()); });
+    factory.registerFunction("toDateTimeOrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDateTimeOrDefault", std::make_shared<DataTypeDateTime>()); });
+    factory.registerFunction("toDateTime64OrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDateTime64OrDefault", std::make_shared<DataTypeDateTime64>(3 /* default scale */)); });
 
-    factory.registerFunction("toDecimal32OrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDecimal32OrDefault", createDecimalMaxPrecision<Decimal32>(0)); });
-    factory.registerFunction("toDecimal64OrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDecimal64OrDefault", createDecimalMaxPrecision<Decimal64>(0)); });
-    factory.registerFunction("toDecimal128OrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDecimal128OrDefault", createDecimalMaxPrecision<Decimal128>(0)); });
-    factory.registerFunction("toDecimal256OrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDecimal256OrDefault", createDecimalMaxPrecision<Decimal256>(0)); });
+    factory.registerFunction("toDecimal32OrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDecimal32OrDefault", createDecimalMaxPrecision<Decimal32>(0)); });
+    factory.registerFunction("toDecimal64OrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDecimal64OrDefault", createDecimalMaxPrecision<Decimal64>(0)); });
+    factory.registerFunction("toDecimal128OrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDecimal128OrDefault", createDecimalMaxPrecision<Decimal128>(0)); });
+    factory.registerFunction("toDecimal256OrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toDecimal256OrDefault", createDecimalMaxPrecision<Decimal256>(0)); });
 
-    factory.registerFunction("toUUIDOrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toUUIDOrDefault", std::make_shared<DataTypeUUID>()); });
-    factory.registerFunction("toIPv4OrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toIPv4OrDefault", std::make_shared<DataTypeIPv4>()); });
-    factory.registerFunction("toIPv6OrDefault", [](ContextPtr context){
-        return std::make_shared<FunctionCastOrDefaultTyped>(context, "toIPv6OrDefault", std::make_shared<DataTypeIPv6>()); });
+    factory.registerFunction("toUUIDOrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toUUIDOrDefault", std::make_shared<DataTypeUUID>()); });
+    factory.registerFunction("toIPv4OrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toIPv4OrDefault", std::make_shared<DataTypeIPv4>()); });
+    factory.registerFunction("toIPv6OrDefault", [](ContextPtr context)
+        { return std::make_shared<FunctionCastOrDefaultTyped>(context, "toIPv6OrDefault", std::make_shared<DataTypeIPv6>()); });
 }
 
 }

From 938a54c85c9b82bd51607080239574b0bb83d311 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 17 Mar 2024 14:12:48 +0100
Subject: [PATCH 416/443] Update reference

---
 .../02415_all_new_functions_must_be_documented.reference      | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference b/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference
index e15002da69c..8b85ac48c16 100644
--- a/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference
+++ b/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference
@@ -194,10 +194,7 @@ blockSerializedSize
 blockSize
 buildId
 byteSize
-caseWithExpr
 caseWithExpression
-caseWithoutExpr
-caseWithoutExpression
 catboostEvaluate
 cbrt
 ceil
@@ -312,7 +309,6 @@ fromUnixTimestamp64Micro
 fromUnixTimestamp64Milli
 fromUnixTimestamp64Nano
 fromUnixTimestampInJodaSyntax
-fullHostName
 fuzzBits
 gccMurmurHash
 gcd

From 4527b3e0d4b1e02285fcfc3579c8c48c5b7657b2 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 17 Mar 2024 15:17:58 +0100
Subject: [PATCH 417/443] Remove useless code

---
 CMakeLists.txt     |  5 -----
 cmake/fuzzer.cmake | 17 -----------------
 2 files changed, 22 deletions(-)
 delete mode 100644 cmake/fuzzer.cmake

diff --git a/CMakeLists.txt b/CMakeLists.txt
index eff6dd3ff6a..957bb3f71de 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -110,11 +110,6 @@ endif()
 # - sanitize.cmake
 add_library(global-libs INTERFACE)
 
-# We don't want to instrument everything with fuzzer, but only specific targets (see below),
-# also, since we build our own llvm, we specifically don't want to instrument
-# libFuzzer library itself - it would result in infinite recursion
-#include (cmake/fuzzer.cmake)
-
 include (cmake/sanitize.cmake)
 
 option(ENABLE_COLORED_BUILD "Enable colors in compiler output" ON)
diff --git a/cmake/fuzzer.cmake b/cmake/fuzzer.cmake
deleted file mode 100644
index dd0c4b080fe..00000000000
--- a/cmake/fuzzer.cmake
+++ /dev/null
@@ -1,17 +0,0 @@
-# see ./CMakeLists.txt for variable declaration
-if (FUZZER)
-    if (FUZZER STREQUAL "libfuzzer")
-        # NOTE: Eldar Zaitov decided to name it "libfuzzer" instead of "fuzzer" to keep in mind another possible fuzzer backends.
-        # NOTE: no-link means that all the targets are built with instrumentation for fuzzer, but only some of them
-        # (tests) have entry point for fuzzer and it's not checked.
-        set (CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} ${SAN_FLAGS} -fsanitize=fuzzer-no-link -DFUZZER=1")
-        set (CMAKE_C_FLAGS "${CMAKE_C_FLAGS} ${SAN_FLAGS} -fsanitize=fuzzer-no-link -DFUZZER=1")
-
-        # NOTE: oss-fuzz can change LIB_FUZZING_ENGINE variable
-        if (NOT LIB_FUZZING_ENGINE)
-            set (LIB_FUZZING_ENGINE "-fsanitize=fuzzer")
-        endif ()
-    else ()
-        message (FATAL_ERROR "Unknown fuzzer type: ${FUZZER}")
-    endif ()
-endif()

From 0a3e42401ce11036a64a2d7a7c30f425d8b700f2 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 17 Mar 2024 15:44:36 +0100
Subject: [PATCH 418/443] Fix fuzzers

---
 .../aggregate_function_state_deserialization_fuzzer.cpp     | 3 +++
 src/Common/tests/gtest_global_context.cpp                   | 6 ------
 src/Common/tests/gtest_global_context.h                     | 2 --
 src/DataTypes/fuzzers/data_type_deserialization_fuzzer.cpp  | 3 +++
 src/Formats/fuzzers/format_fuzzer.cpp                       | 3 +++
 src/Interpreters/fuzzers/execute_query_fuzzer.cpp           | 3 +++
 6 files changed, 12 insertions(+), 8 deletions(-)

diff --git a/src/AggregateFunctions/fuzzers/aggregate_function_state_deserialization_fuzzer.cpp b/src/AggregateFunctions/fuzzers/aggregate_function_state_deserialization_fuzzer.cpp
index 290da81944d..425364efb9c 100644
--- a/src/AggregateFunctions/fuzzers/aggregate_function_state_deserialization_fuzzer.cpp
+++ b/src/AggregateFunctions/fuzzers/aggregate_function_state_deserialization_fuzzer.cpp
@@ -27,6 +27,9 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t * data, size_t size)
 
         auto initialize = [&]() mutable
         {
+            if (context)
+                return true;
+
             shared_context = Context::createShared();
             context = Context::createGlobal(shared_context.get());
             context->makeGlobalContext();
diff --git a/src/Common/tests/gtest_global_context.cpp b/src/Common/tests/gtest_global_context.cpp
index ec86c953c5b..0c1556766a9 100644
--- a/src/Common/tests/gtest_global_context.cpp
+++ b/src/Common/tests/gtest_global_context.cpp
@@ -10,9 +10,3 @@ ContextHolder & getMutableContext()
     static ContextHolder holder;
     return holder;
 }
-
-void destroyContext()
-{
-    auto & holder = getMutableContext();
-    return holder.destroy();
-}
diff --git a/src/Common/tests/gtest_global_context.h b/src/Common/tests/gtest_global_context.h
index f846a0dbe4f..7ae8bb32f70 100644
--- a/src/Common/tests/gtest_global_context.h
+++ b/src/Common/tests/gtest_global_context.h
@@ -28,5 +28,3 @@ struct ContextHolder
 const ContextHolder & getContext();
 
 ContextHolder & getMutableContext();
-
-void destroyContext();
diff --git a/src/DataTypes/fuzzers/data_type_deserialization_fuzzer.cpp b/src/DataTypes/fuzzers/data_type_deserialization_fuzzer.cpp
index e40734e0a57..0ae325871fb 100644
--- a/src/DataTypes/fuzzers/data_type_deserialization_fuzzer.cpp
+++ b/src/DataTypes/fuzzers/data_type_deserialization_fuzzer.cpp
@@ -24,6 +24,9 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t * data, size_t size)
 
         auto initialize = [&]() mutable
         {
+            if (context)
+                return true;
+
             shared_context = Context::createShared();
             context = Context::createGlobal(shared_context.get());
             context->makeGlobalContext();
diff --git a/src/Formats/fuzzers/format_fuzzer.cpp b/src/Formats/fuzzers/format_fuzzer.cpp
index 583d1173a01..46661e4828c 100644
--- a/src/Formats/fuzzers/format_fuzzer.cpp
+++ b/src/Formats/fuzzers/format_fuzzer.cpp
@@ -32,6 +32,9 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t * data, size_t size)
 
         auto initialize = [&]() mutable
         {
+            if (context)
+                return true;
+
             shared_context = Context::createShared();
             context = Context::createGlobal(shared_context.get());
             context->makeGlobalContext();
diff --git a/src/Interpreters/fuzzers/execute_query_fuzzer.cpp b/src/Interpreters/fuzzers/execute_query_fuzzer.cpp
index edff202d547..a02ce66e6b5 100644
--- a/src/Interpreters/fuzzers/execute_query_fuzzer.cpp
+++ b/src/Interpreters/fuzzers/execute_query_fuzzer.cpp
@@ -25,6 +25,9 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t * data, size_t size)
 
         auto initialize = [&]() mutable
         {
+            if (context)
+                return true;
+
             shared_context = Context::createShared();
             context = Context::createGlobal(shared_context.get());
             context->makeGlobalContext();

From 44e918bc67efe0b1c36059693ee2eb709d31c59c Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 17 Mar 2024 16:22:33 +0100
Subject: [PATCH 419/443] Revive `getFuzzerData`

---
 CMakeLists.txt                             |  9 ++++-----
 programs/local/LocalServer.cpp             |  6 ------
 programs/main.cpp                          | 15 +++++++--------
 src/Interpreters/AsynchronousMetricLog.cpp |  1 -
 4 files changed, 11 insertions(+), 20 deletions(-)

diff --git a/CMakeLists.txt b/CMakeLists.txt
index 957bb3f71de..8c4e16eace2 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -549,7 +549,9 @@ if (ENABLE_RUST)
     endif()
 endif()
 
-if (CMAKE_BUILD_TYPE_UC STREQUAL "RELWITHDEBINFO" AND NOT SANITIZE AND NOT SANITIZE_COVERAGE AND OS_LINUX AND (ARCH_AMD64 OR ARCH_AARCH64))
+if (CMAKE_BUILD_TYPE_UC STREQUAL "RELWITHDEBINFO"
+    AND NOT SANITIZE AND NOT SANITIZE_COVERAGE AND NOT ENABLE_FUZZING
+    AND OS_LINUX AND (ARCH_AMD64 OR ARCH_AARCH64))
     set(CHECK_LARGE_OBJECT_SIZES_DEFAULT ON)
 else ()
     set(CHECK_LARGE_OBJECT_SIZES_DEFAULT OFF)
@@ -572,10 +574,7 @@ if (FUZZER)
             if (NOT(target_type STREQUAL "INTERFACE_LIBRARY" OR target_type STREQUAL "UTILITY"))
                 target_compile_options(${target} PRIVATE "-fsanitize=fuzzer-no-link")
             endif()
-            # clickhouse fuzzer isn't working correctly
-            # initial PR https://github.com/ClickHouse/ClickHouse/pull/27526
-            #if (target MATCHES ".+_fuzzer" OR target STREQUAL "clickhouse")
-            if (target_type STREQUAL "EXECUTABLE" AND target MATCHES ".+_fuzzer")
+            if (target_type STREQUAL "EXECUTABLE" AND (target MATCHES ".+_fuzzer" OR target STREQUAL "clickhouse"))
                 message(STATUS "${target} instrumented with fuzzer")
                 target_link_libraries(${target} PUBLIC ch_contrib::fuzzer)
                 # Add to fuzzers bundle
diff --git a/programs/local/LocalServer.cpp b/programs/local/LocalServer.cpp
index 50f9c242712..10fbda0fe46 100644
--- a/programs/local/LocalServer.cpp
+++ b/programs/local/LocalServer.cpp
@@ -1000,12 +1000,6 @@ extern "C" int LLVMFuzzerInitialize(int * pargc, char *** pargv)
 {
     std::vector<char *> argv(*pargv, *pargv + (*pargc + 1));
 
-    if (!isClickhouseApp("local", argv))
-    {
-        std::cerr << "\033[31m" << "ClickHouse compiled in fuzzing mode, only clickhouse local is available." << "\033[0m" << std::endl;
-        exit(1);
-    }
-
     /// As a user you can add flags to clickhouse binary in fuzzing mode as follows
     /// clickhouse local <set of clickhouse-local specific flag> -- <set of libfuzzer flags>
 
diff --git a/programs/main.cpp b/programs/main.cpp
index c5f1b30f60e..0a35594bd30 100644
--- a/programs/main.cpp
+++ b/programs/main.cpp
@@ -68,7 +68,6 @@ namespace
 using MainFunc = int (*)(int, char**);
 
 #if !defined(FUZZING_MODE)
-
 /// Add an item here to register new application
 std::pair<std::string_view, MainFunc> clickhouse_applications[] =
 {
@@ -105,13 +104,6 @@ std::pair<std::string_view, MainFunc> clickhouse_applications[] =
     {"restart", mainEntryClickHouseRestart},
 };
 
-/// Add an item here to register a new short name
-std::pair<std::string_view, std::string_view> clickhouse_short_names[] =
-{
-    {"chl", "local"},
-    {"chc", "client"},
-};
-
 int printHelp(int, char **)
 {
     std::cerr << "Use one of the following commands:" << std::endl;
@@ -121,6 +113,13 @@ int printHelp(int, char **)
 }
 #endif
 
+/// Add an item here to register a new short name
+std::pair<std::string_view, std::string_view> clickhouse_short_names[] =
+{
+    {"chl", "local"},
+    {"chc", "client"},
+};
+
 
 enum class InstructionFail
 {
diff --git a/src/Interpreters/AsynchronousMetricLog.cpp b/src/Interpreters/AsynchronousMetricLog.cpp
index f905f72e7a7..5cf7f951eec 100644
--- a/src/Interpreters/AsynchronousMetricLog.cpp
+++ b/src/Interpreters/AsynchronousMetricLog.cpp
@@ -8,7 +8,6 @@
 #include <Interpreters/AsynchronousMetricLog.h>
 #include <Parsers/parseQuery.h>
 #include <Parsers/ExpressionElementParsers.h>
-#include <Common/AsynchronousMetrics.h>
 
 
 namespace DB

From 01136bbc3beb01eb1f150747412db2030115507a Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 17 Mar 2024 19:53:58 +0100
Subject: [PATCH 420/443] Limit backtracking in parser

---
 programs/compressor/Compressor.cpp            |   2 +-
 programs/format/Format.cpp                    |   2 +-
 programs/keeper-client/KeeperClient.cpp       |   3 +-
 src/Access/AccessEntityIO.cpp                 |   2 +-
 src/Access/RowPolicyCache.cpp                 |   2 +-
 src/Access/UsersConfigAccessStorage.cpp       |   2 +-
 .../parseAggregateFunctionParameters.cpp      |   3 +-
 src/Backups/BackupInfo.cpp                    |   2 +-
 src/Backups/RestorerFromBackup.cpp            |   4 +-
 src/Client/ClientBase.cpp                     |  24 +--
 src/Client/ClientBase.h                       |   2 +-
 src/Client/QueryFuzzer.cpp                    |   6 +-
 .../NamedCollections/NamedCollectionUtils.cpp |   2 +-
 .../tests/gtest_compressionCodec.cpp          |   2 +-
 src/Core/Defines.h                            |   2 +
 src/Core/Settings.h                           |   1 +
 src/Core/SettingsChangesHistory.h             |   1 +
 src/DataTypes/DataTypeFactory.cpp             |   5 +-
 src/Databases/DDLDependencyVisitor.cpp        |   3 +-
 src/Databases/DatabaseDictionary.cpp          |   4 +-
 src/Databases/DatabaseFilesystem.cpp          |   2 +-
 src/Databases/DatabaseHDFS.cpp                |   2 +-
 src/Databases/DatabaseOnDisk.cpp              |  18 +-
 src/Databases/DatabaseOrdinary.cpp            |   2 +-
 src/Databases/DatabaseReplicated.cpp          |   7 +-
 src/Databases/DatabaseS3.cpp                  |   2 +-
 src/Databases/DatabasesCommon.cpp             |   5 +-
 src/Databases/DatabasesCommon.h               |   3 +-
 src/Databases/MySQL/DatabaseMySQL.cpp         |  14 +-
 .../MySQL/tryConvertStringLiterals.cpp        |   2 +-
 .../MySQL/tryParseTableIDFromDDL.cpp          |   2 +-
 .../MySQL/tryQuoteUnrecognizedTokens.cpp      |   2 +-
 src/Databases/SQLite/DatabaseSQLite.cpp       |   6 +-
 .../tests/gtest_dictionary_configuration.cpp  |   8 +-
 src/Formats/SchemaInferenceUtils.cpp          |   2 +-
 src/Functions/FunctionSQLJSON.h               |   9 +-
 .../UserDefinedSQLObjectsBackup.cpp           |   2 +-
 .../UserDefinedSQLObjectsDiskStorage.cpp      |   3 +-
 .../UserDefinedSQLObjectsZooKeeperStorage.cpp |   3 +-
 src/Functions/formatQuery.cpp                 | 168 +++++++++---------
 ...InterpreterShowCreateAccessEntityQuery.cpp |   2 +-
 src/Interpreters/AsynchronousMetricLog.cpp    |  10 +-
 src/Interpreters/DDLTask.cpp                  |   5 +-
 .../IInterpreterUnionOrSelectQuery.cpp        |   2 +-
 src/Interpreters/InterpreterCreateQuery.cpp   |  10 +-
 src/Interpreters/InterpreterDeleteQuery.cpp   |   3 +-
 .../InterpreterKillQueryQuery.cpp             |   2 +-
 src/Interpreters/InterpreterSelectQuery.cpp   |   4 +-
 .../JoinToSubqueryTransformVisitor.cpp        |   2 +-
 .../MySQL/tests/gtest_create_rewritten.cpp    |   2 +-
 src/Interpreters/SystemLog.cpp                |   4 +-
 src/Interpreters/executeQuery.cpp             |  10 +-
 .../getCustomKeyFilterForParallelReplicas.cpp |   2 +-
 src/Interpreters/loadMetadata.cpp             |   4 +-
 .../parseColumnsListForTableFunction.cpp      |   4 +-
 .../tests/gtest_comparison_graph.cpp          |   2 +-
 .../tests/gtest_cycle_aliases.cpp             |  10 +-
 .../tests/gtest_table_overrides.cpp           |   4 +-
 src/Parsers/ExpressionListParsers.cpp         |   2 +-
 src/Parsers/IParser.cpp                       |  33 ++++
 src/Parsers/IParser.h                         |  15 +-
 .../KustoFunctions/IParserKQLFunction.cpp     |   8 +-
 .../Kusto/KustoFunctions/IParserKQLFunction.h |   4 +-
 .../KustoFunctions/KQLCastingFunctions.cpp    |   2 +-
 .../KustoFunctions/KQLDynamicFunctions.cpp    |  10 +-
 .../Kusto/KustoFunctions/KQLIPFunctions.cpp   |  32 ++--
 .../KustoFunctions/KQLStringFunctions.cpp     |   4 +-
 src/Parsers/Kusto/ParserKQLDistinct.cpp       |   2 +-
 src/Parsers/Kusto/ParserKQLExtend.cpp         |   4 +-
 src/Parsers/Kusto/ParserKQLFilter.cpp         |   2 +-
 src/Parsers/Kusto/ParserKQLLimit.cpp          |   2 +-
 src/Parsers/Kusto/ParserKQLMVExpand.cpp       |  16 +-
 src/Parsers/Kusto/ParserKQLMVExpand.h         |   2 +-
 src/Parsers/Kusto/ParserKQLMakeSeries.cpp     |  20 +--
 src/Parsers/Kusto/ParserKQLMakeSeries.h       |   2 +-
 src/Parsers/Kusto/ParserKQLPrint.cpp          |   2 +-
 src/Parsers/Kusto/ParserKQLProject.cpp        |   2 +-
 src/Parsers/Kusto/ParserKQLQuery.cpp          |  18 +-
 src/Parsers/Kusto/ParserKQLQuery.h            |   6 +-
 src/Parsers/Kusto/ParserKQLSort.cpp           |   2 +-
 src/Parsers/Kusto/ParserKQLStatement.cpp      |   2 +-
 src/Parsers/Kusto/ParserKQLSummarize.cpp      |   8 +-
 src/Parsers/Kusto/parseKQLQuery.cpp           |  23 ++-
 src/Parsers/Kusto/parseKQLQuery.h             |  26 +--
 .../tests/gtest_alter_command_parser.cpp      |   2 +-
 .../MySQL/tests/gtest_alter_parser.cpp        |   2 +-
 .../MySQL/tests/gtest_column_parser.cpp       |   4 +-
 .../MySQL/tests/gtest_constraint_parser.cpp   |  10 +-
 .../MySQL/tests/gtest_create_parser.cpp       |   8 +-
 .../MySQL/tests/gtest_index_parser.cpp        |  50 +++---
 .../tests/gtest_partition_options_parser.cpp  |  26 +--
 .../MySQL/tests/gtest_partition_parser.cpp    |  21 ++-
 .../MySQL/tests/gtest_reference_parser.cpp    |  21 ++-
 .../MySQL/tests/gtest_subpartition_parser.cpp |   5 +-
 .../tests/gtest_table_options_parser.cpp      |   4 +-
 src/Parsers/PRQL/ParserPRQLQuery.cpp          |   4 +-
 src/Parsers/PRQL/ParserPRQLQuery.h            |   3 +-
 src/Parsers/ParserAlterQuery.cpp              |   3 +-
 src/Parsers/ParserCreateQuery.h               |   2 +-
 src/Parsers/QueryParameterVisitor.cpp         |   2 +-
 src/Parsers/TokenIterator.h                   |  12 --
 src/Parsers/examples/create_parser.cpp        |   2 +-
 src/Parsers/examples/select_parser.cpp        |   2 +-
 src/Parsers/fuzzers/select_parser_fuzzer.cpp  |   3 +-
 src/Parsers/parseQuery.cpp                    |  28 +--
 src/Parsers/parseQuery.h                      |  16 +-
 src/Parsers/tests/gtest_Parser.cpp            |  13 +-
 src/Parsers/tests/gtest_common.cpp            |   8 +-
 src/Parsers/tests/gtest_dictionary_parser.cpp |  18 +-
 src/Parsers/tests/gtest_format_hiliting.cpp   |   2 +-
 src/Planner/PlannerJoinTree.cpp               |   2 +-
 src/Planner/Utils.cpp                         |   2 +-
 .../Impl/ConstantExpressionTemplate.cpp       |   2 +-
 .../Formats/Impl/MySQLDumpRowInputFormat.cpp  |   3 +-
 .../Formats/Impl/ValuesBlockInputFormat.cpp   |   4 +-
 .../QueryPlan/ReadFromMergeTree.cpp           |   2 +-
 src/Server/GRPCServer.cpp                     |   2 +-
 src/Server/PostgreSQLHandler.cpp              |   1 +
 src/Storages/ColumnsDescription.cpp           |   4 +-
 src/Storages/ConstraintsDescription.cpp       |   2 +-
 src/Storages/IndicesDescription.cpp           |   2 +-
 src/Storages/KeyDescription.cpp               |   2 +-
 src/Storages/MergeTree/IMergeTreeDataPart.cpp |   2 +-
 .../MergeTreeDataPartWriterCompact.cpp        |   2 +-
 .../MergeTreeDataPartWriterOnDisk.cpp         |   4 +-
 .../MergeTree/MergeTreeDataPartWriterWide.cpp |   2 +-
 .../MergeTree/MergeTreeDataSelectExecutor.cpp |   2 +-
 .../ReplicatedMergeTreeTableMetadata.cpp      |   4 +-
 src/Storages/MutationCommands.cpp             |   2 +-
 .../StorageMaterializedPostgreSQL.cpp         |   2 +-
 src/Storages/ProjectionsDescription.cpp       |   2 +-
 .../System/StorageSystemDDLWorkerQueue.cpp    |   3 +-
 .../System/attachInformationSchemaTables.cpp  |   2 +-
 src/Storages/TTLDescription.cpp               |   2 +-
 src/Storages/getStructureOfRemoteTable.cpp    |   4 +-
 ..._transform_query_for_external_database.cpp |   4 +-
 src/TableFunctions/Hive/TableFunctionHive.cpp |   4 +-
 src/TableFunctions/TableFunctionExplain.cpp   |   8 +-
 138 files changed, 549 insertions(+), 466 deletions(-)
 create mode 100644 src/Parsers/IParser.cpp

diff --git a/programs/compressor/Compressor.cpp b/programs/compressor/Compressor.cpp
index 7125fdc744f..050bb495024 100644
--- a/programs/compressor/Compressor.cpp
+++ b/programs/compressor/Compressor.cpp
@@ -143,7 +143,7 @@ int mainEntryClickHouseCompressor(int argc, char ** argv)
             ParserCodec codec_parser;
 
             std::string codecs_line = boost::algorithm::join(codecs, ",");
-            auto ast = parseQuery(codec_parser, "(" + codecs_line + ")", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+            auto ast = parseQuery(codec_parser, "(" + codecs_line + ")", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
             codec = CompressionCodecFactory::instance().get(ast, nullptr);
         }
         else
diff --git a/programs/format/Format.cpp b/programs/format/Format.cpp
index c92106e2f90..50f801f2560 100644
--- a/programs/format/Format.cpp
+++ b/programs/format/Format.cpp
@@ -234,7 +234,7 @@ int mainEntryClickHouseFormat(int argc, char ** argv)
                 size_t approx_query_length = multiple ? find_first_symbols<';'>(pos, end) - pos : end - pos;
 
                 ASTPtr res = parseQueryAndMovePosition(
-                    parser, pos, end, "query", multiple, cmd_settings.max_query_size, cmd_settings.max_parser_depth);
+                    parser, pos, end, "query", multiple, cmd_settings.max_query_size, cmd_settings.max_parser_depth, cmd_settings.max_parser_backtracks);
 
                 std::unique_ptr<ReadBuffer> insert_query_payload = nullptr;
                 /// If the query is INSERT ... VALUES, then we will try to parse the data.
diff --git a/programs/keeper-client/KeeperClient.cpp b/programs/keeper-client/KeeperClient.cpp
index 595fc65e50e..8297fab5ed9 100644
--- a/programs/keeper-client/KeeperClient.cpp
+++ b/programs/keeper-client/KeeperClient.cpp
@@ -44,7 +44,7 @@ String KeeperClient::executeFourLetterCommand(const String & command)
 std::vector<String> KeeperClient::getCompletions(const String & prefix) const
 {
     Tokens tokens(prefix.data(), prefix.data() + prefix.size(), 0, false);
-    IParser::Pos pos(tokens, 0);
+    IParser::Pos pos(tokens, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
 
     if (pos->type != TokenType::BareWord)
         return registered_commands_and_four_letter_words;
@@ -278,6 +278,7 @@ bool KeeperClient::processQueryText(const String & text)
                 /* allow_multi_statements = */ true,
                 /* max_query_size = */ 0,
                 /* max_parser_depth = */ 0,
+                /* max_parser_backtracks = */ 0,
                 /* skip_insignificant = */ false);
 
             if (!res)
diff --git a/src/Access/AccessEntityIO.cpp b/src/Access/AccessEntityIO.cpp
index 80bb63b04bf..b0dfd74c53b 100644
--- a/src/Access/AccessEntityIO.cpp
+++ b/src/Access/AccessEntityIO.cpp
@@ -62,7 +62,7 @@ AccessEntityPtr deserializeAccessEntityImpl(const String & definition)
     const char * end = begin + definition.size();
     while (pos < end)
     {
-        queries.emplace_back(parseQueryAndMovePosition(parser, pos, end, "", true, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH));
+        queries.emplace_back(parseQueryAndMovePosition(parser, pos, end, "", true, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS));
         while (isWhitespaceASCII(*pos) || *pos == ';')
             ++pos;
     }
diff --git a/src/Access/RowPolicyCache.cpp b/src/Access/RowPolicyCache.cpp
index 13140099a63..c1c4928d0da 100644
--- a/src/Access/RowPolicyCache.cpp
+++ b/src/Access/RowPolicyCache.cpp
@@ -86,7 +86,7 @@ void RowPolicyCache::PolicyInfo::setPolicy(const RowPolicyPtr & policy_)
         try
         {
             ParserExpression parser;
-            parsed_filters[filter_type_i] = parseQuery(parser, filter, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+            parsed_filters[filter_type_i] = parseQuery(parser, filter, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
         }
         catch (...)
         {
diff --git a/src/Access/UsersConfigAccessStorage.cpp b/src/Access/UsersConfigAccessStorage.cpp
index e9b2e1397ab..b4b843fc77e 100644
--- a/src/Access/UsersConfigAccessStorage.cpp
+++ b/src/Access/UsersConfigAccessStorage.cpp
@@ -66,7 +66,7 @@ namespace
 
         String error_message;
         const char * pos = string_query.data();
-        auto ast = tryParseQuery(parser, pos, pos + string_query.size(), error_message, false, "", false, 0, 0);
+        auto ast = tryParseQuery(parser, pos, pos + string_query.size(), error_message, false, "", false, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS, true);
 
         if (!ast)
             throw Exception(ErrorCodes::BAD_ARGUMENTS, "Failed to parse grant query. Error: {}", error_message);
diff --git a/src/AggregateFunctions/parseAggregateFunctionParameters.cpp b/src/AggregateFunctions/parseAggregateFunctionParameters.cpp
index db1efe224d1..593be1e0a79 100644
--- a/src/AggregateFunctions/parseAggregateFunctionParameters.cpp
+++ b/src/AggregateFunctions/parseAggregateFunctionParameters.cpp
@@ -81,7 +81,8 @@ void getAggregateFunctionNameAndParametersArray(
     ParserExpressionList params_parser(false);
     ASTPtr args_ast = parseQuery(params_parser,
         parameters_str.data(), parameters_str.data() + parameters_str.size(),
-        "parameters of aggregate function in " + error_context, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+        "parameters of aggregate function in " + error_context,
+        0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
 
     if (args_ast->children.empty())
         throw Exception(ErrorCodes::BAD_ARGUMENTS, "Incorrect list of parameters to aggregate function {}",
diff --git a/src/Backups/BackupInfo.cpp b/src/Backups/BackupInfo.cpp
index 2bff400d4fe..461f613ecd2 100644
--- a/src/Backups/BackupInfo.cpp
+++ b/src/Backups/BackupInfo.cpp
@@ -25,7 +25,7 @@ String BackupInfo::toString() const
 BackupInfo BackupInfo::fromString(const String & str)
 {
     ParserIdentifierWithOptionalParameters parser;
-    ASTPtr ast = parseQuery(parser, str, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+    ASTPtr ast = parseQuery(parser, str, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
     return fromAST(*ast);
 }
 
diff --git a/src/Backups/RestorerFromBackup.cpp b/src/Backups/RestorerFromBackup.cpp
index 87c143f0fe2..e20e8eb66c6 100644
--- a/src/Backups/RestorerFromBackup.cpp
+++ b/src/Backups/RestorerFromBackup.cpp
@@ -424,7 +424,7 @@ void RestorerFromBackup::findTableInBackupImpl(const QualifiedTableName & table_
     readStringUntilEOF(create_query_str, *read_buffer);
     read_buffer.reset();
     ParserCreateQuery create_parser;
-    ASTPtr create_table_query = parseQuery(create_parser, create_query_str, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+    ASTPtr create_table_query = parseQuery(create_parser, create_query_str, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
     applyCustomStoragePolicy(create_table_query);
     renameDatabaseAndTableNameInCreateQuery(create_table_query, renaming_map, context->getGlobalContext());
     String create_table_query_str = serializeAST(*create_table_query);
@@ -534,7 +534,7 @@ void RestorerFromBackup::findDatabaseInBackupImpl(const String & database_name_i
         readStringUntilEOF(create_query_str, *read_buffer);
         read_buffer.reset();
         ParserCreateQuery create_parser;
-        ASTPtr create_database_query = parseQuery(create_parser, create_query_str, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+        ASTPtr create_database_query = parseQuery(create_parser, create_query_str, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
         renameDatabaseAndTableNameInCreateQuery(create_database_query, renaming_map, context->getGlobalContext());
         String create_database_query_str = serializeAST(*create_database_query);
 
diff --git a/src/Client/ClientBase.cpp b/src/Client/ClientBase.cpp
index 48962880b8f..d561a64895b 100644
--- a/src/Client/ClientBase.cpp
+++ b/src/Client/ClientBase.cpp
@@ -345,7 +345,7 @@ ASTPtr ClientBase::parseQuery(const char *& pos, const char * end, bool allow_mu
     if (dialect == Dialect::kusto)
         parser = std::make_unique<ParserKQLStatement>(end, global_context->getSettings().allow_settings_after_format_in_insert);
     else if (dialect == Dialect::prql)
-        parser = std::make_unique<ParserPRQLQuery>(max_length, settings.max_parser_depth);
+        parser = std::make_unique<ParserPRQLQuery>(max_length, settings.max_parser_depth, settings.max_parser_backtracks);
     else
         parser = std::make_unique<ParserQuery>(end, global_context->getSettings().allow_settings_after_format_in_insert);
 
@@ -353,9 +353,9 @@ ASTPtr ClientBase::parseQuery(const char *& pos, const char * end, bool allow_mu
     {
         String message;
         if (dialect == Dialect::kusto)
-            res = tryParseKQLQuery(*parser, pos, end, message, true, "", allow_multi_statements, max_length, settings.max_parser_depth);
+            res = tryParseKQLQuery(*parser, pos, end, message, true, "", allow_multi_statements, max_length, settings.max_parser_depth, settings.max_parser_backtracks, true);
         else
-            res = tryParseQuery(*parser, pos, end, message, true, "", allow_multi_statements, max_length, settings.max_parser_depth);
+            res = tryParseQuery(*parser, pos, end, message, true, "", allow_multi_statements, max_length, settings.max_parser_depth, settings.max_parser_backtracks, true);
 
         if (!res)
         {
@@ -366,9 +366,9 @@ ASTPtr ClientBase::parseQuery(const char *& pos, const char * end, bool allow_mu
     else
     {
         if (dialect == Dialect::kusto)
-            res = parseKQLQueryAndMovePosition(*parser, pos, end, "", allow_multi_statements, max_length, settings.max_parser_depth);
+            res = parseKQLQueryAndMovePosition(*parser, pos, end, "", allow_multi_statements, max_length, settings.max_parser_depth, settings.max_parser_backtracks);
         else
-            res = parseQueryAndMovePosition(*parser, pos, end, "", allow_multi_statements, max_length, settings.max_parser_depth);
+            res = parseQueryAndMovePosition(*parser, pos, end, "", allow_multi_statements, max_length, settings.max_parser_depth, settings.max_parser_backtracks);
     }
 
     if (is_interactive)
@@ -385,12 +385,12 @@ ASTPtr ClientBase::parseQuery(const char *& pos, const char * end, bool allow_mu
 
 
 /// Consumes trailing semicolons and tries to consume the same-line trailing comment.
-void ClientBase::adjustQueryEnd(const char *& this_query_end, const char * all_queries_end, uint32_t max_parser_depth)
+void ClientBase::adjustQueryEnd(const char *& this_query_end, const char * all_queries_end, uint32_t max_parser_depth, uint32_t max_parser_backtracks)
 {
     // We have to skip the trailing semicolon that might be left
     // after VALUES parsing or just after a normal semicolon-terminated query.
     Tokens after_query_tokens(this_query_end, all_queries_end);
-    IParser::Pos after_query_iterator(after_query_tokens, max_parser_depth);
+    IParser::Pos after_query_iterator(after_query_tokens, max_parser_depth, max_parser_backtracks);
     while (after_query_iterator.isValid() && after_query_iterator->type == TokenType::Semicolon)
     {
         this_query_end = after_query_iterator->end;
@@ -1984,6 +1984,7 @@ MultiQueryProcessingStage ClientBase::analyzeMultiQueryText(
         return MultiQueryProcessingStage::QUERIES_END;
 
     unsigned max_parser_depth = static_cast<unsigned>(global_context->getSettingsRef().max_parser_depth);
+    unsigned max_parser_backtracks = static_cast<unsigned>(global_context->getSettingsRef().max_parser_backtracks);
 
     // If there are only comments left until the end of file, we just
     // stop. The parser can't handle this situation because it always
@@ -1994,7 +1995,7 @@ MultiQueryProcessingStage ClientBase::analyzeMultiQueryText(
     // and it makes more sense to treat them as such.
     {
         Tokens tokens(this_query_begin, all_queries_end);
-        IParser::Pos token_iterator(tokens, max_parser_depth);
+        IParser::Pos token_iterator(tokens, max_parser_depth, max_parser_backtracks);
         if (!token_iterator.isValid())
             return MultiQueryProcessingStage::QUERIES_END;
     }
@@ -2015,7 +2016,7 @@ MultiQueryProcessingStage ClientBase::analyzeMultiQueryText(
         if (ignore_error)
         {
             Tokens tokens(this_query_begin, all_queries_end);
-            IParser::Pos token_iterator(tokens, max_parser_depth);
+            IParser::Pos token_iterator(tokens, max_parser_depth, max_parser_backtracks);
             while (token_iterator->type != TokenType::Semicolon && token_iterator.isValid())
                 ++token_iterator;
             this_query_begin = token_iterator->end;
@@ -2055,7 +2056,7 @@ MultiQueryProcessingStage ClientBase::analyzeMultiQueryText(
     // after we have processed the query. But even this guess is
     // beneficial so that we see proper trailing comments in "echo" and
     // server log.
-    adjustQueryEnd(this_query_end, all_queries_end, max_parser_depth);
+    adjustQueryEnd(this_query_end, all_queries_end, max_parser_depth, max_parser_backtracks);
     return MultiQueryProcessingStage::EXECUTE_QUERY;
 }
 
@@ -2251,7 +2252,8 @@ bool ClientBase::executeMultiQuery(const String & all_queries_text)
                     this_query_end = insert_ast->end;
                     adjustQueryEnd(
                         this_query_end, all_queries_end,
-                        static_cast<unsigned>(global_context->getSettingsRef().max_parser_depth));
+                        static_cast<unsigned>(global_context->getSettingsRef().max_parser_depth),
+                        static_cast<unsigned>(global_context->getSettingsRef().max_parser_backtracks));
                 }
 
                 // Report error.
diff --git a/src/Client/ClientBase.h b/src/Client/ClientBase.h
index dd08e7c059b..7a9e9666e67 100644
--- a/src/Client/ClientBase.h
+++ b/src/Client/ClientBase.h
@@ -94,7 +94,7 @@ protected:
     void processParsedSingleQuery(const String & full_query, const String & query_to_execute,
         ASTPtr parsed_query, std::optional<bool> echo_query_ = {}, bool report_error = false);
 
-    static void adjustQueryEnd(const char *& this_query_end, const char * all_queries_end, uint32_t max_parser_depth);
+    static void adjustQueryEnd(const char *& this_query_end, const char * all_queries_end, uint32_t max_parser_depth, uint32_t max_parser_backtracks);
     ASTPtr parseQuery(const char *& pos, const char * end, bool allow_multi_statements) const;
     static void setupSignalHandler();
 
diff --git a/src/Client/QueryFuzzer.cpp b/src/Client/QueryFuzzer.cpp
index 0a7cb1b36db..7be01686258 100644
--- a/src/Client/QueryFuzzer.cpp
+++ b/src/Client/QueryFuzzer.cpp
@@ -569,7 +569,8 @@ void QueryFuzzer::fuzzColumnDeclaration(ASTColumnDeclaration & column)
         auto data_type = fuzzDataType(DataTypeFactory::instance().get(column.type));
 
         ParserDataType parser;
-        column.type = parseQuery(parser, data_type->getName(), DBMS_DEFAULT_MAX_QUERY_SIZE, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+        column.type = parseQuery(parser, data_type->getName(),
+            DBMS_DEFAULT_MAX_QUERY_SIZE, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
     }
 }
 
@@ -821,7 +822,8 @@ static ASTPtr tryParseInsertQuery(const String & full_query)
     ParserInsertQuery parser(end, false);
     String message;
 
-    return tryParseQuery(parser, pos, end, message, false, "", false, DBMS_DEFAULT_MAX_QUERY_SIZE, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+    return tryParseQuery(parser, pos, end, message, false, "", false,
+        DBMS_DEFAULT_MAX_QUERY_SIZE, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS, true);
 }
 
 ASTs QueryFuzzer::getInsertQueriesForFuzzedTables(const String & full_query)
diff --git a/src/Common/NamedCollections/NamedCollectionUtils.cpp b/src/Common/NamedCollections/NamedCollectionUtils.cpp
index fe0f42467c7..9b569390b3c 100644
--- a/src/Common/NamedCollections/NamedCollectionUtils.cpp
+++ b/src/Common/NamedCollections/NamedCollectionUtils.cpp
@@ -302,7 +302,7 @@ private:
         readStringUntilEOF(query, in);
 
         ParserCreateNamedCollectionQuery parser;
-        auto ast = parseQuery(parser, query, "in file " + path, 0, settings.max_parser_depth);
+        auto ast = parseQuery(parser, query, "in file " + path, 0, settings.max_parser_depth, settings.max_parser_backtracks);
         const auto & create_query = ast->as<const ASTCreateNamedCollectionQuery &>();
         return create_query;
     }
diff --git a/src/Compression/tests/gtest_compressionCodec.cpp b/src/Compression/tests/gtest_compressionCodec.cpp
index 24f16a55c25..16573e035e0 100644
--- a/src/Compression/tests/gtest_compressionCodec.cpp
+++ b/src/Compression/tests/gtest_compressionCodec.cpp
@@ -442,7 +442,7 @@ CompressionCodecPtr makeCodec(const std::string & codec_string, const DataTypePt
 {
     const std::string codec_statement = "(" + codec_string + ")";
     Tokens tokens(codec_statement.begin().base(), codec_statement.end().base());
-    IParser::Pos token_iterator(tokens, 0);
+    IParser::Pos token_iterator(tokens, 0, 0);
 
     Expected expected;
     ASTPtr codec_ast;
diff --git a/src/Core/Defines.h b/src/Core/Defines.h
index cc6f49aa361..a8dd26519c2 100644
--- a/src/Core/Defines.h
+++ b/src/Core/Defines.h
@@ -63,6 +63,8 @@ static constexpr auto DBMS_DEFAULT_LOCK_ACQUIRE_TIMEOUT_SEC = 120;
 
 /// Default limit on recursion depth of recursive descend parser.
 static constexpr auto DBMS_DEFAULT_MAX_PARSER_DEPTH = 1000;
+/// Default limit on the amount of backtracking of recursive descend parser.
+static constexpr auto DBMS_DEFAULT_MAX_PARSER_BACKTRACKS = 1000000;
 
 /// Default limit on query size.
 static constexpr auto DBMS_DEFAULT_MAX_QUERY_SIZE = 262144;
diff --git a/src/Core/Settings.h b/src/Core/Settings.h
index 48f6b4d621c..e6adb00137f 100644
--- a/src/Core/Settings.h
+++ b/src/Core/Settings.h
@@ -607,6 +607,7 @@ class IColumn;
     M(Bool, use_compact_format_in_distributed_parts_names, true, "Changes format of directories names for distributed table insert parts.", 0) \
     M(Bool, validate_polygons, true, "Throw exception if polygon is invalid in function pointInPolygon (e.g. self-tangent, self-intersecting). If the setting is false, the function will accept invalid polygons but may silently return wrong result.", 0) \
     M(UInt64, max_parser_depth, DBMS_DEFAULT_MAX_PARSER_DEPTH, "Maximum parser depth (recursion depth of recursive descend parser).", 0) \
+    M(UInt64, max_parser_backtracks, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS, "Maximum parser backtracking (how many times it tries different alternatives in the recursive descend parsing process).", 0) \
     M(Bool, allow_settings_after_format_in_insert, false, "Allow SETTINGS after FORMAT, but note, that this is not always safe (note: this is a compatibility setting).", 0) \
     M(Seconds, periodic_live_view_refresh, 60, "Interval after which periodically refreshed live view is forced to refresh.", 0) \
     M(Bool, transform_null_in, false, "If enabled, NULL values will be matched with 'IN' operator as if they are considered equal.", 0) \
diff --git a/src/Core/SettingsChangesHistory.h b/src/Core/SettingsChangesHistory.h
index 4914f97a6fb..8e2b2915c2a 100644
--- a/src/Core/SettingsChangesHistory.h
+++ b/src/Core/SettingsChangesHistory.h
@@ -95,6 +95,7 @@ static std::map<ClickHouseVersion, SettingsChangesHistory::SettingsChanges> sett
               {"throw_if_deduplication_in_dependent_materialized_views_enabled_with_async_insert", false, true, "Deduplication is dependent materialized view cannot work together with async inserts."},
               {"parallel_replicas_allow_in_with_subquery", false, true, "If true, subquery for IN will be executed on every follower replica"},
               {"filesystem_cache_reserve_space_wait_lock_timeout_milliseconds", 1000, 1000, "Wait time to lock cache for sapce reservation in filesystem cache"},
+              {"max_parser_backtracks", 0, 1000000, "Limiting the complexity of parsing"},
               }},
     {"24.2", {{"allow_suspicious_variant_types", true, false, "Don't allow creating Variant type with suspicious variants by default"},
               {"validate_experimental_and_suspicious_types_inside_nested_types", false, true, "Validate usage of experimental and suspicious types inside nested types"},
diff --git a/src/DataTypes/DataTypeFactory.cpp b/src/DataTypes/DataTypeFactory.cpp
index d154b386ace..844384f3c95 100644
--- a/src/DataTypes/DataTypeFactory.cpp
+++ b/src/DataTypes/DataTypeFactory.cpp
@@ -56,13 +56,14 @@ DataTypePtr DataTypeFactory::getImpl(const String & full_name) const
     {
         String out_err;
         const char * start = full_name.data();
-        ast = tryParseQuery(parser, start, start + full_name.size(), out_err, false, "data type", false, DBMS_DEFAULT_MAX_QUERY_SIZE, data_type_max_parse_depth);
+        ast = tryParseQuery(parser, start, start + full_name.size(), out_err, false, "data type", false,
+            DBMS_DEFAULT_MAX_QUERY_SIZE, data_type_max_parse_depth, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS, true);
         if (!ast)
             return nullptr;
     }
     else
     {
-        ast = parseQuery(parser, full_name.data(), full_name.data() + full_name.size(), "data type", false, data_type_max_parse_depth);
+        ast = parseQuery(parser, full_name.data(), full_name.data() + full_name.size(), "data type", false, data_type_max_parse_depth, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
     }
 
     return getImpl<nullptr_on_error>(ast);
diff --git a/src/Databases/DDLDependencyVisitor.cpp b/src/Databases/DDLDependencyVisitor.cpp
index cb85119e3b0..75a01a6190f 100644
--- a/src/Databases/DDLDependencyVisitor.cpp
+++ b/src/Databases/DDLDependencyVisitor.cpp
@@ -444,8 +444,9 @@ namespace
             ParserSelectWithUnionQuery parser;
             String description = fmt::format("Query for ClickHouse dictionary {}", data.table_name);
             String fixed_query = removeWhereConditionPlaceholder(query);
+            const Settings & settings = data.context->getSettingsRef();
             ASTPtr select = parseQuery(parser, fixed_query, description,
-                                       data.context->getSettingsRef().max_query_size, data.context->getSettingsRef().max_parser_depth);
+                settings.max_query_size, settings.max_parser_depth, settings.max_parser_backtracks);
 
             DDLDependencyVisitor::Visitor visitor{data};
             visitor.visit(select);
diff --git a/src/Databases/DatabaseDictionary.cpp b/src/Databases/DatabaseDictionary.cpp
index 9a65c7a46ef..76fdb4fa961 100644
--- a/src/Databases/DatabaseDictionary.cpp
+++ b/src/Databases/DatabaseDictionary.cpp
@@ -115,7 +115,7 @@ ASTPtr DatabaseDictionary::getCreateTableQueryImpl(const String & table_name, Co
     const char * pos = query.data();
     std::string error_message;
     auto ast = tryParseQuery(parser, pos, pos + query.size(), error_message,
-            /* hilite = */ false, "", /* allow_multi_statements = */ false, 0, settings.max_parser_depth);
+        /* hilite = */ false, "", /* allow_multi_statements = */ false, 0, settings.max_parser_depth, settings.max_parser_backtracks, true);
 
     if (!ast && throw_on_error)
         throw Exception::createDeprecated(error_message, ErrorCodes::SYNTAX_ERROR);
@@ -134,7 +134,7 @@ ASTPtr DatabaseDictionary::getCreateDatabaseQuery() const
     }
     auto settings = getContext()->getSettingsRef();
     ParserCreateQuery parser;
-    return parseQuery(parser, query.data(), query.data() + query.size(), "", 0, settings.max_parser_depth);
+    return parseQuery(parser, query.data(), query.data() + query.size(), "", 0, settings.max_parser_depth, settings.max_parser_backtracks);
 }
 
 void DatabaseDictionary::shutdown()
diff --git a/src/Databases/DatabaseFilesystem.cpp b/src/Databases/DatabaseFilesystem.cpp
index 5af1e1ae0d2..05af0acf978 100644
--- a/src/Databases/DatabaseFilesystem.cpp
+++ b/src/Databases/DatabaseFilesystem.cpp
@@ -187,7 +187,7 @@ ASTPtr DatabaseFilesystem::getCreateDatabaseQuery() const
     const String query = fmt::format("CREATE DATABASE {} ENGINE = Filesystem('{}')", backQuoteIfNeed(getDatabaseName()), path);
 
     ParserCreateQuery parser;
-    ASTPtr ast = parseQuery(parser, query.data(), query.data() + query.size(), "", 0, settings.max_parser_depth);
+    ASTPtr ast = parseQuery(parser, query.data(), query.data() + query.size(), "", 0, settings.max_parser_depth, settings.max_parser_backtracks);
 
     if (const auto database_comment = getDatabaseComment(); !database_comment.empty())
     {
diff --git a/src/Databases/DatabaseHDFS.cpp b/src/Databases/DatabaseHDFS.cpp
index 3a1e6b16ccf..2688ff2443c 100644
--- a/src/Databases/DatabaseHDFS.cpp
+++ b/src/Databases/DatabaseHDFS.cpp
@@ -183,7 +183,7 @@ ASTPtr DatabaseHDFS::getCreateDatabaseQuery() const
     ParserCreateQuery parser;
 
     const String query = fmt::format("CREATE DATABASE {} ENGINE = HDFS('{}')", backQuoteIfNeed(getDatabaseName()), source);
-    ASTPtr ast = parseQuery(parser, query.data(), query.data() + query.size(), "", 0, settings.max_parser_depth);
+    ASTPtr ast = parseQuery(parser, query.data(), query.data() + query.size(), "", 0, settings.max_parser_depth, settings.max_parser_backtracks);
 
     if (const auto database_comment = getDatabaseComment(); !database_comment.empty())
     {
diff --git a/src/Databases/DatabaseOnDisk.cpp b/src/Databases/DatabaseOnDisk.cpp
index fcb073644c5..dcfc1916450 100644
--- a/src/Databases/DatabaseOnDisk.cpp
+++ b/src/Databases/DatabaseOnDisk.cpp
@@ -526,7 +526,7 @@ ASTPtr DatabaseOnDisk::getCreateDatabaseQuery() const
         /// If database.sql doesn't exist, then engine is Ordinary
         String query = "CREATE DATABASE " + backQuoteIfNeed(getDatabaseName()) + " ENGINE = Ordinary";
         ParserCreateQuery parser;
-        ast = parseQuery(parser, query.data(), query.data() + query.size(), "", 0, settings.max_parser_depth);
+        ast = parseQuery(parser, query.data(), query.data() + query.size(), "", 0, settings.max_parser_depth, settings.max_parser_backtracks);
     }
 
     if (const auto database_comment = getDatabaseComment(); !database_comment.empty())
@@ -707,7 +707,7 @@ ASTPtr DatabaseOnDisk::parseQueryFromMetadata(
     const char * pos = query.data();
     std::string error_message;
     auto ast = tryParseQuery(parser, pos, pos + query.size(), error_message, /* hilite = */ false,
-                             "in file " + metadata_file_path, /* allow_multi_statements = */ false, 0, settings.max_parser_depth);
+        "in file " + metadata_file_path, /* allow_multi_statements = */ false, 0, settings.max_parser_depth, settings.max_parser_backtracks, true);
 
     if (!ast && throw_on_error)
         throw Exception::createDeprecated(error_message, ErrorCodes::SYNTAX_ERROR);
@@ -765,12 +765,14 @@ ASTPtr DatabaseOnDisk::getCreateQueryFromStorage(const String & table_name, cons
     auto ast_storage = std::make_shared<ASTStorage>();
     ast_storage->set(ast_storage->engine, ast_engine);
 
-    unsigned max_parser_depth = static_cast<unsigned>(getContext()->getSettingsRef().max_parser_depth);
-    auto create_table_query = DB::getCreateQueryFromStorage(storage,
-                                                            ast_storage,
-                                                            false,
-                                                            max_parser_depth,
-                                                            throw_on_error);
+    const Settings & settings = getContext()->getSettingsRef();
+    auto create_table_query = DB::getCreateQueryFromStorage(
+        storage,
+        ast_storage,
+        false,
+        static_cast<unsigned>(settings.max_parser_depth),
+        static_cast<unsigned>(settings.max_parser_backtracks),
+        throw_on_error);
 
     create_table_query->set(create_table_query->as<ASTCreateQuery>()->comment,
                             std::make_shared<ASTLiteral>("SYSTEM TABLE is built on the fly."));
diff --git a/src/Databases/DatabaseOrdinary.cpp b/src/Databases/DatabaseOrdinary.cpp
index bc552b9c927..161dd3d3f60 100644
--- a/src/Databases/DatabaseOrdinary.cpp
+++ b/src/Databases/DatabaseOrdinary.cpp
@@ -469,7 +469,7 @@ void DatabaseOrdinary::alterTable(ContextPtr local_context, const StorageID & ta
         statement.data() + statement.size(),
         "in file " + table_metadata_path,
         0,
-        local_context->getSettingsRef().max_parser_depth);
+        local_context->getSettingsRef().max_parser_depth, local_context->getSettingsRef().max_parser_backtracks);
 
     applyMetadataChangesToCreateQuery(ast, metadata);
 
diff --git a/src/Databases/DatabaseReplicated.cpp b/src/Databases/DatabaseReplicated.cpp
index 9cf19a251f7..3b6a712510d 100644
--- a/src/Databases/DatabaseReplicated.cpp
+++ b/src/Databases/DatabaseReplicated.cpp
@@ -812,7 +812,8 @@ static UUID getTableUUIDIfReplicated(const String & metadata, ContextPtr context
     ParserCreateQuery parser;
     auto size = context->getSettingsRef().max_query_size;
     auto depth = context->getSettingsRef().max_parser_depth;
-    ASTPtr query = parseQuery(parser, metadata, size, depth);
+    auto backtracks = context->getSettingsRef().max_parser_backtracks;
+    ASTPtr query = parseQuery(parser, metadata, size, depth, backtracks);
     const ASTCreateQuery & create = query->as<const ASTCreateQuery &>();
     if (!create.storage || !create.storage->engine)
         return UUIDHelpers::Nil;
@@ -1234,7 +1235,7 @@ ASTPtr DatabaseReplicated::parseQueryFromMetadataInZooKeeper(const String & node
 {
     ParserCreateQuery parser;
     String description = "in ZooKeeper " + zookeeper_path + "/metadata/" + node_name;
-    auto ast = parseQuery(parser, query, description, 0, getContext()->getSettingsRef().max_parser_depth);
+    auto ast = parseQuery(parser, query, description, 0, getContext()->getSettingsRef().max_parser_depth, getContext()->getSettingsRef().max_parser_backtracks);
 
     auto & create = ast->as<ASTCreateQuery &>();
     if (create.uuid == UUIDHelpers::Nil || create.getTable() != TABLE_WITH_UUID_NAME_PLACEHOLDER || create.database)
@@ -1559,7 +1560,7 @@ DatabaseReplicated::getTablesForBackup(const FilterByNameFunction & filter, cons
     for (const auto & [table_name, metadata] : snapshot)
     {
         ParserCreateQuery parser;
-        auto create_table_query = parseQuery(parser, metadata, 0, getContext()->getSettingsRef().max_parser_depth);
+        auto create_table_query = parseQuery(parser, metadata, 0, getContext()->getSettingsRef().max_parser_depth, getContext()->getSettingsRef().max_parser_backtracks);
 
         auto & create = create_table_query->as<ASTCreateQuery &>();
         create.attach = false;
diff --git a/src/Databases/DatabaseS3.cpp b/src/Databases/DatabaseS3.cpp
index d2ca5a05ea4..159a5242dbe 100644
--- a/src/Databases/DatabaseS3.cpp
+++ b/src/Databases/DatabaseS3.cpp
@@ -191,7 +191,7 @@ ASTPtr DatabaseS3::getCreateDatabaseQuery() const
         creation_args += fmt::format(", '{}', '{}'", config.access_key_id.value(), config.secret_access_key.value());
 
     const String query = fmt::format("CREATE DATABASE {} ENGINE = S3({})", backQuoteIfNeed(getDatabaseName()), creation_args);
-    ASTPtr ast = parseQuery(parser, query.data(), query.data() + query.size(), "", 0, settings.max_parser_depth);
+    ASTPtr ast = parseQuery(parser, query.data(), query.data() + query.size(), "", 0, settings.max_parser_depth, settings.max_parser_backtracks);
 
     if (const auto database_comment = getDatabaseComment(); !database_comment.empty())
     {
diff --git a/src/Databases/DatabasesCommon.cpp b/src/Databases/DatabasesCommon.cpp
index 963cf0064df..f8d6ad69ba8 100644
--- a/src/Databases/DatabasesCommon.cpp
+++ b/src/Databases/DatabasesCommon.cpp
@@ -108,7 +108,8 @@ void applyMetadataChangesToCreateQuery(const ASTPtr & query, const StorageInMemo
 }
 
 
-ASTPtr getCreateQueryFromStorage(const StoragePtr & storage, const ASTPtr & ast_storage, bool only_ordinary, uint32_t max_parser_depth, bool throw_on_error)
+ASTPtr getCreateQueryFromStorage(const StoragePtr & storage, const ASTPtr & ast_storage, bool only_ordinary,
+    uint32_t max_parser_depth, uint32_t max_parser_backtracks, bool throw_on_error)
 {
     auto table_id = storage->getStorageID();
     auto metadata_ptr = storage->getInMemoryMetadataPtr();
@@ -148,7 +149,7 @@ ASTPtr getCreateQueryFromStorage(const StoragePtr & storage, const ASTPtr & ast_
                 Expected expected;
                 expected.max_parsed_pos = string_end;
                 Tokens tokens(type_name.c_str(), string_end);
-                IParser::Pos pos(tokens, max_parser_depth);
+                IParser::Pos pos(tokens, max_parser_depth, max_parser_backtracks);
                 ParserDataType parser;
                 if (!parser.parse(pos, ast_type, expected))
                 {
diff --git a/src/Databases/DatabasesCommon.h b/src/Databases/DatabasesCommon.h
index 4e9d967c11a..81a3c55a435 100644
--- a/src/Databases/DatabasesCommon.h
+++ b/src/Databases/DatabasesCommon.h
@@ -13,7 +13,8 @@ namespace DB
 {
 
 void applyMetadataChangesToCreateQuery(const ASTPtr & query, const StorageInMemoryMetadata & metadata);
-ASTPtr getCreateQueryFromStorage(const StoragePtr & storage, const ASTPtr & ast_storage, bool only_ordinary, uint32_t max_parser_depth, bool throw_on_error);
+ASTPtr getCreateQueryFromStorage(const StoragePtr & storage, const ASTPtr & ast_storage, bool only_ordinary,
+    uint32_t max_parser_depth, uint32_t max_parser_backtracks, bool throw_on_error);
 
 /// Cleans a CREATE QUERY from temporary flags like "IF NOT EXISTS", "OR REPLACE", "AS SELECT" (for non-views), etc.
 void cleanupObjectDefinitionFromTemporaryFlags(ASTCreateQuery & query);
diff --git a/src/Databases/MySQL/DatabaseMySQL.cpp b/src/Databases/MySQL/DatabaseMySQL.cpp
index 96a5c3a18ce..d9b0f7f9ac7 100644
--- a/src/Databases/MySQL/DatabaseMySQL.cpp
+++ b/src/Databases/MySQL/DatabaseMySQL.cpp
@@ -174,12 +174,14 @@ ASTPtr DatabaseMySQL::getCreateTableQueryImpl(const String & table_name, Context
         ast_storage->settings = nullptr;
     }
 
-    unsigned max_parser_depth = static_cast<unsigned>(getContext()->getSettingsRef().max_parser_depth);
-    auto create_table_query = DB::getCreateQueryFromStorage(storage,
-                                                            table_storage_define,
-                                                            true,
-                                                            max_parser_depth,
-                                                            throw_on_error);
+    const Settings & settings = getContext()->getSettingsRef();
+    auto create_table_query = DB::getCreateQueryFromStorage(
+        storage,
+        table_storage_define,
+        true,
+        static_cast<unsigned>(settings.max_parser_depth),
+        static_cast<unsigned>(settings.max_parser_backtracks),
+        throw_on_error);
     return create_table_query;
 }
 
diff --git a/src/Databases/MySQL/tryConvertStringLiterals.cpp b/src/Databases/MySQL/tryConvertStringLiterals.cpp
index ab392b301e8..ac65d510f67 100644
--- a/src/Databases/MySQL/tryConvertStringLiterals.cpp
+++ b/src/Databases/MySQL/tryConvertStringLiterals.cpp
@@ -61,7 +61,7 @@ static bool tryReadCharset(
 bool tryConvertStringLiterals(String & query)
 {
     Tokens tokens(query.data(), query.data() + query.size());
-    IParser::Pos pos(tokens, 0);
+    IParser::Pos pos(tokens, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
     Expected expected;
     String rewritten_query;
     rewritten_query.reserve(query.size());
diff --git a/src/Databases/MySQL/tryParseTableIDFromDDL.cpp b/src/Databases/MySQL/tryParseTableIDFromDDL.cpp
index a01eb311450..4fe0f44c767 100644
--- a/src/Databases/MySQL/tryParseTableIDFromDDL.cpp
+++ b/src/Databases/MySQL/tryParseTableIDFromDDL.cpp
@@ -10,7 +10,7 @@ StorageID tryParseTableIDFromDDL(const String & query, const String & default_da
 {
     bool is_ddl = false;
     Tokens tokens(query.data(), query.data() + query.size());
-    IParser::Pos pos(tokens, 0);
+    IParser::Pos pos(tokens, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
     Expected expected;
     if (ParserKeyword("CREATE TEMPORARY TABLE").ignore(pos, expected) || ParserKeyword("CREATE TABLE").ignore(pos, expected))
     {
diff --git a/src/Databases/MySQL/tryQuoteUnrecognizedTokens.cpp b/src/Databases/MySQL/tryQuoteUnrecognizedTokens.cpp
index c5a366698e6..9ecc81c693f 100644
--- a/src/Databases/MySQL/tryQuoteUnrecognizedTokens.cpp
+++ b/src/Databases/MySQL/tryQuoteUnrecognizedTokens.cpp
@@ -37,7 +37,7 @@ static void quoteLiteral(
 bool tryQuoteUnrecognizedTokens(String & query)
 {
     Tokens tokens(query.data(), query.data() + query.size());
-    IParser::Pos pos(tokens, 0);
+    IParser::Pos pos(tokens, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
     Expected expected;
     String rewritten_query;
     const char * copy_from = query.data();
diff --git a/src/Databases/SQLite/DatabaseSQLite.cpp b/src/Databases/SQLite/DatabaseSQLite.cpp
index b3d5288cdf7..b7a82fd9d0f 100644
--- a/src/Databases/SQLite/DatabaseSQLite.cpp
+++ b/src/Databases/SQLite/DatabaseSQLite.cpp
@@ -194,10 +194,10 @@ ASTPtr DatabaseSQLite::getCreateTableQueryImpl(const String & table_name, Contex
     /// Add table_name to engine arguments
     storage_engine_arguments->children.insert(storage_engine_arguments->children.begin() + 1, std::make_shared<ASTLiteral>(table_id.table_name));
 
-    unsigned max_parser_depth = static_cast<unsigned>(getContext()->getSettingsRef().max_parser_depth);
+    const Settings & settings = getContext()->getSettingsRef();
+
     auto create_table_query = DB::getCreateQueryFromStorage(storage, table_storage_define, true,
-                                                            max_parser_depth,
-                                                            throw_on_error);
+        static_cast<uint32_t>(settings.max_parser_depth), static_cast<uint32_t>(settings.max_parser_backtracks), throw_on_error);
 
     return create_table_query;
 }
diff --git a/src/Dictionaries/tests/gtest_dictionary_configuration.cpp b/src/Dictionaries/tests/gtest_dictionary_configuration.cpp
index 989ce5c8f18..08aad663a8c 100644
--- a/src/Dictionaries/tests/gtest_dictionary_configuration.cpp
+++ b/src/Dictionaries/tests/gtest_dictionary_configuration.cpp
@@ -48,7 +48,7 @@ TEST(ConvertDictionaryAST, SimpleDictConfiguration)
                    " COMMENT 'hello world!'";
 
     ParserCreateDictionaryQuery parser;
-    ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, 0, 0);
     ASTCreateQuery * create = ast->as<ASTCreateQuery>();
     DictionaryConfigurationPtr config = getDictionaryConfigurationFromAST(*create, getContext().context);
 
@@ -119,7 +119,7 @@ TEST(ConvertDictionaryAST, TrickyAttributes)
                    " SOURCE(CLICKHOUSE(HOST 'localhost'))";
 
     ParserCreateDictionaryQuery parser;
-    ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, 0, 0);
     ASTCreateQuery * create = ast->as<ASTCreateQuery>();
     DictionaryConfigurationPtr config = getDictionaryConfigurationFromAST(*create, getContext().context);
 
@@ -164,7 +164,7 @@ TEST(ConvertDictionaryAST, ComplexKeyAndLayoutWithParams)
                    " LIFETIME(MIN 1 MAX 10)";
 
     ParserCreateDictionaryQuery parser;
-    ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, 0, 0);
     ASTCreateQuery * create = ast->as<ASTCreateQuery>();
     DictionaryConfigurationPtr config = getDictionaryConfigurationFromAST(*create, getContext().context);
 
@@ -215,7 +215,7 @@ TEST(ConvertDictionaryAST, ComplexSource)
                    " RANGE(MIN second_column MAX third_column)";
 
     ParserCreateDictionaryQuery parser;
-    ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, 0, 0);
     ASTCreateQuery * create = ast->as<ASTCreateQuery>();
     DictionaryConfigurationPtr config = getDictionaryConfigurationFromAST(*create, getContext().context);
     /// source
diff --git a/src/Formats/SchemaInferenceUtils.cpp b/src/Formats/SchemaInferenceUtils.cpp
index cb574551d26..0bba7a1f424 100644
--- a/src/Formats/SchemaInferenceUtils.cpp
+++ b/src/Formats/SchemaInferenceUtils.cpp
@@ -1054,7 +1054,7 @@ namespace
     {
         if (depth > settings.max_parser_depth)
             throw Exception(ErrorCodes::TOO_DEEP_RECURSION,
-                            "Maximum parse depth ({}) exceeded. Consider raising max_parser_depth setting.", settings.max_parser_depth);
+                "Maximum parse depth ({}) exceeded. Consider raising max_parser_depth setting.", settings.max_parser_depth);
 
         assertChar('{', buf);
         skipWhitespaceIfAny(buf);
diff --git a/src/Functions/FunctionSQLJSON.h b/src/Functions/FunctionSQLJSON.h
index 3efa40df9be..37db514fd1f 100644
--- a/src/Functions/FunctionSQLJSON.h
+++ b/src/Functions/FunctionSQLJSON.h
@@ -123,7 +123,7 @@ public:
     class Executor
     {
     public:
-        static ColumnPtr run(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count, uint32_t parse_depth, const ContextPtr & context)
+        static ColumnPtr run(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count, uint32_t parse_depth, uint32_t parse_backtracks, const ContextPtr & context)
         {
             MutableColumnPtr to{result_type->createColumn()};
             to->reserve(input_rows_count);
@@ -161,7 +161,7 @@ public:
             /// Tokenize the query
             Tokens tokens(query.data(), query.data() + query.size());
             /// Max depth 0 indicates that depth is not limited
-            IParser::Pos token_iterator(tokens, parse_depth);
+            IParser::Pos token_iterator(tokens, parse_depth, parse_backtracks);
 
             /// Parse query and create AST tree
             Expected expected;
@@ -232,16 +232,17 @@ public:
         /// 3. Parser(Tokens, ASTPtr) -> complete AST
         /// 4. Execute functions: call getNextItem on generator and handle each item
         unsigned parse_depth = static_cast<unsigned>(getContext()->getSettingsRef().max_parser_depth);
+        unsigned parse_backtracks = static_cast<unsigned>(getContext()->getSettingsRef().max_parser_backtracks);
 #if USE_SIMDJSON
         if (getContext()->getSettingsRef().allow_simdjson)
             return FunctionSQLJSONHelpers::Executor<
                 Name,
                 Impl<SimdJSONParser, JSONStringSerializer<SimdJSONParser::Element, SimdJSONElementFormatter>>,
-                SimdJSONParser>::run(arguments, result_type, input_rows_count, parse_depth, getContext());
+                SimdJSONParser>::run(arguments, result_type, input_rows_count, parse_depth, parse_backtracks, getContext());
 #endif
         return FunctionSQLJSONHelpers::
             Executor<Name, Impl<DummyJSONParser, DefaultJSONStringSerializer<DummyJSONParser::Element>>, DummyJSONParser>::run(
-                arguments, result_type, input_rows_count, parse_depth, getContext());
+                arguments, result_type, input_rows_count, parse_depth, parse_backtracks, getContext());
     }
 };
 
diff --git a/src/Functions/UserDefined/UserDefinedSQLObjectsBackup.cpp b/src/Functions/UserDefined/UserDefinedSQLObjectsBackup.cpp
index 3ec5393fa6f..b7c7e5847bd 100644
--- a/src/Functions/UserDefined/UserDefinedSQLObjectsBackup.cpp
+++ b/src/Functions/UserDefined/UserDefinedSQLObjectsBackup.cpp
@@ -128,7 +128,7 @@ restoreUserDefinedSQLObjects(RestorerFromBackup & restorer, const String & data_
                     statement_def.data() + statement_def.size(),
                     "in file " + filepath + " from backup " + backup->getNameForLogging(),
                     0,
-                    context->getSettingsRef().max_parser_depth);
+                    context->getSettingsRef().max_parser_depth, context->getSettingsRef().max_parser_backtracks);
                 break;
             }
         }
diff --git a/src/Functions/UserDefined/UserDefinedSQLObjectsDiskStorage.cpp b/src/Functions/UserDefined/UserDefinedSQLObjectsDiskStorage.cpp
index 34946db7d9e..b083c540083 100644
--- a/src/Functions/UserDefined/UserDefinedSQLObjectsDiskStorage.cpp
+++ b/src/Functions/UserDefined/UserDefinedSQLObjectsDiskStorage.cpp
@@ -92,7 +92,8 @@ ASTPtr UserDefinedSQLObjectsDiskStorage::tryLoadObject(UserDefinedSQLObjectType
                     object_create_query.data() + object_create_query.size(),
                     "",
                     0,
-                    global_context->getSettingsRef().max_parser_depth);
+                    global_context->getSettingsRef().max_parser_depth,
+                    global_context->getSettingsRef().max_parser_backtracks);
                 return ast;
             }
         }
diff --git a/src/Functions/UserDefined/UserDefinedSQLObjectsZooKeeperStorage.cpp b/src/Functions/UserDefined/UserDefinedSQLObjectsZooKeeperStorage.cpp
index c43b223ffeb..4ec34c15efc 100644
--- a/src/Functions/UserDefined/UserDefinedSQLObjectsZooKeeperStorage.cpp
+++ b/src/Functions/UserDefined/UserDefinedSQLObjectsZooKeeperStorage.cpp
@@ -314,7 +314,8 @@ ASTPtr UserDefinedSQLObjectsZooKeeperStorage::parseObjectData(const String & obj
                 object_data.data() + object_data.size(),
                 "",
                 0,
-                global_context->getSettingsRef().max_parser_depth);
+                global_context->getSettingsRef().max_parser_depth,
+                global_context->getSettingsRef().max_parser_backtracks);
             return ast;
         }
     }
diff --git a/src/Functions/formatQuery.cpp b/src/Functions/formatQuery.cpp
index 92403d2e88e..d7addcc284e 100644
--- a/src/Functions/formatQuery.cpp
+++ b/src/Functions/formatQuery.cpp
@@ -17,6 +17,9 @@ namespace ErrorCodes
     extern const int ILLEGAL_COLUMN;
 }
 
+namespace
+{
+
 enum class OutputFormatting
 {
     SingleLine,
@@ -29,21 +32,16 @@ enum class ErrorHandling
     Null
 };
 
-template <OutputFormatting output_formatting, ErrorHandling error_handling, typename Name>
 class FunctionFormatQuery : public IFunction
 {
 public:
-    static constexpr auto name = Name::name;
-    static FunctionPtr create(ContextPtr context)
-    {
-        const auto & settings = context->getSettings();
-        return std::make_shared<FunctionFormatQuery>(settings.max_query_size, settings.max_parser_depth);
-    }
-
-    FunctionFormatQuery(size_t max_query_size_, size_t max_parser_depth_)
-        : max_query_size(max_query_size_)
-        , max_parser_depth(max_parser_depth_)
+    FunctionFormatQuery(ContextPtr context, String name_, OutputFormatting output_formatting_, ErrorHandling error_handling_)
+        : name(name_), output_formatting(output_formatting_), error_handling(error_handling_)
     {
+        const Settings & settings = context->getSettings();
+        max_query_size = settings.max_query_size;
+        max_parser_depth = settings.max_parser_depth;
+        max_parser_backtracks = settings.max_parser_backtracks;
     }
 
     String getName() const override { return name; }
@@ -59,7 +57,7 @@ public:
         validateFunctionArgumentTypes(*this, arguments, args);
 
         DataTypePtr string_type = std::make_shared<DataTypeString>();
-        if constexpr (error_handling == ErrorHandling::Null)
+        if (error_handling == ErrorHandling::Null)
             return std::make_shared<DataTypeNullable>(string_type);
         else
             return string_type;
@@ -70,7 +68,7 @@ public:
         const ColumnPtr col_query = arguments[0].column;
 
         ColumnUInt8::MutablePtr col_null_map;
-        if constexpr (error_handling == ErrorHandling::Null)
+        if (error_handling == ErrorHandling::Null)
             col_null_map = ColumnUInt8::create(input_rows_count, 0);
 
         if (const ColumnString * col_query_string = checkAndGetColumn<ColumnString>(col_query.get()))
@@ -78,7 +76,7 @@ public:
             auto col_res = ColumnString::create();
             formatVector(col_query_string->getChars(), col_query_string->getOffsets(), col_res->getChars(), col_res->getOffsets(), col_null_map);
 
-            if constexpr (error_handling == ErrorHandling::Null)
+            if (error_handling == ErrorHandling::Null)
                 return ColumnNullable::create(std::move(col_res), std::move(col_null_map));
             else
                 return col_res;
@@ -113,11 +111,11 @@ private:
 
             try
             {
-                ast = parseQuery(parser, begin, end, /*query_description*/ {}, max_query_size, max_parser_depth);
+                ast = parseQuery(parser, begin, end, /*query_description*/ {}, max_query_size, max_parser_depth, max_parser_backtracks);
             }
             catch (...)
             {
-                if constexpr (error_handling == ErrorHandling::Null)
+                if (error_handling == ErrorHandling::Null)
                 {
                     const size_t res_data_new_size = res_data_size + 1;
                     if (res_data_new_size > res_data.size())
@@ -135,7 +133,6 @@ private:
                 }
                 else
                 {
-                    static_assert(error_handling == ErrorHandling::Exception);
                     throw;
                 }
             }
@@ -160,92 +157,91 @@ private:
         res_data.resize(res_data_size);
     }
 
-    const size_t max_query_size;
-    const size_t max_parser_depth;
+    String name;
+    OutputFormatting output_formatting;
+    ErrorHandling error_handling;
+
+    size_t max_query_size;
+    size_t max_parser_depth;
+    size_t max_parser_backtracks;
 };
 
-struct NameFormatQuery
-{
-    static constexpr auto name = "formatQuery";
-};
-
-struct NameFormatQueryOrNull
-{
-    static constexpr auto name = "formatQueryOrNull";
-};
-
-struct NameFormatQuerySingleLine
-{
-    static constexpr auto name = "formatQuerySingleLine";
-};
-
-struct NameFormatQuerySingleLineOrNull
-{
-    static constexpr auto name = "formatQuerySingleLineOrNull";
-};
+}
 
 REGISTER_FUNCTION(formatQuery)
 {
-    factory.registerFunction<FunctionFormatQuery<OutputFormatting::MultiLine, ErrorHandling::Exception, NameFormatQuery>>(FunctionDocumentation{
-        .description = "Returns a formatted, possibly multi-line, version of the given SQL query. Throws in case of a parsing error.\n[example:multiline]",
-        .syntax = "formatQuery(query)",
-        .arguments = {{"query", "The SQL query to be formatted. [String](../../sql-reference/data-types/string.md)"}},
-        .returned_value = "The formatted query. [String](../../sql-reference/data-types/string.md).",
-        .examples{
-            {"multiline",
-             "SELECT formatQuery('select a,    b FRom tab WHERE a > 3 and  b < 3');",
-             "SELECT\n"
-             "    a,\n"
-             "    b\n"
-             "FROM tab\n"
-             "WHERE (a > 3) AND (b < 3)"}},
-        .categories{"Other"}});
+    factory.registerFunction(
+        "formatQuery",
+        [](ContextPtr context) { return std::make_shared<FunctionFormatQuery>(context, "formatQuery", OutputFormatting::MultiLine, ErrorHandling::Exception); },
+        FunctionDocumentation{
+            .description = "Returns a formatted, possibly multi-line, version of the given SQL query. Throws in case of a parsing error.\n[example:multiline]",
+            .syntax = "formatQuery(query)",
+            .arguments = {{"query", "The SQL query to be formatted. [String](../../sql-reference/data-types/string.md)"}},
+            .returned_value = "The formatted query. [String](../../sql-reference/data-types/string.md).",
+            .examples{
+                {"multiline",
+                 "SELECT formatQuery('select a,    b FRom tab WHERE a > 3 and  b < 3');",
+                 "SELECT\n"
+                 "    a,\n"
+                 "    b\n"
+                 "FROM tab\n"
+                 "WHERE (a > 3) AND (b < 3)"}},
+            .categories{"Other"}});
 }
 
 REGISTER_FUNCTION(formatQueryOrNull)
 {
-    factory.registerFunction<FunctionFormatQuery<OutputFormatting::MultiLine, ErrorHandling::Null, NameFormatQueryOrNull>>(FunctionDocumentation{
-        .description = "Returns a formatted, possibly multi-line, version of the given SQL query. Returns NULL in case of a parsing error.\n[example:multiline]",
-        .syntax = "formatQueryOrNull(query)",
-        .arguments = {{"query", "The SQL query to be formatted. [String](../../sql-reference/data-types/string.md)"}},
-        .returned_value = "The formatted query. [String](../../sql-reference/data-types/string.md).",
-        .examples{
-            {"multiline",
-             "SELECT formatQuery('select a,    b FRom tab WHERE a > 3 and  b < 3');",
-             "SELECT\n"
-             "    a,\n"
-             "    b\n"
-             "FROM tab\n"
-             "WHERE (a > 3) AND (b < 3)"}},
-        .categories{"Other"}});
+    factory.registerFunction(
+        "formatQueryOrNull",
+        [](ContextPtr context) { return std::make_shared<FunctionFormatQuery>(context, "formatQueryOrNull", OutputFormatting::MultiLine, ErrorHandling::Null); },
+        FunctionDocumentation{
+            .description = "Returns a formatted, possibly multi-line, version of the given SQL query. Returns NULL in case of a parsing error.\n[example:multiline]",
+            .syntax = "formatQueryOrNull(query)",
+            .arguments = {{"query", "The SQL query to be formatted. [String](../../sql-reference/data-types/string.md)"}},
+            .returned_value = "The formatted query. [String](../../sql-reference/data-types/string.md).",
+            .examples{
+                {"multiline",
+                 "SELECT formatQuery('select a,    b FRom tab WHERE a > 3 and  b < 3');",
+                 "SELECT\n"
+                 "    a,\n"
+                 "    b\n"
+                 "FROM tab\n"
+                 "WHERE (a > 3) AND (b < 3)"}},
+            .categories{"Other"}});
 }
 
 REGISTER_FUNCTION(formatQuerySingleLine)
 {
-    factory.registerFunction<FunctionFormatQuery<OutputFormatting::SingleLine, ErrorHandling::Exception, NameFormatQuerySingleLine>>(FunctionDocumentation{
-        .description = "Like formatQuery() but the returned formatted string contains no line breaks. Throws in case of a parsing error.\n[example:multiline]",
-        .syntax = "formatQuerySingleLine(query)",
-        .arguments = {{"query", "The SQL query to be formatted. [String](../../sql-reference/data-types/string.md)"}},
-        .returned_value = "The formatted query. [String](../../sql-reference/data-types/string.md).",
-        .examples{
-            {"multiline",
-             "SELECT formatQuerySingleLine('select a,    b FRom tab WHERE a > 3 and  b < 3');",
-             "SELECT a, b FROM tab WHERE (a > 3) AND (b < 3)"}},
-        .categories{"Other"}});
+    factory.registerFunction(
+        "formatQuerySingleLine",
+        [](ContextPtr context) { return std::make_shared<FunctionFormatQuery>(context, "formatQuerySingleLine", OutputFormatting::SingleLine, ErrorHandling::Exception); },
+        FunctionDocumentation{
+            .description = "Like formatQuery() but the returned formatted string contains no line breaks. Throws in case of a parsing error.\n[example:multiline]",
+            .syntax = "formatQuerySingleLine(query)",
+            .arguments = {{"query", "The SQL query to be formatted. [String](../../sql-reference/data-types/string.md)"}},
+            .returned_value = "The formatted query. [String](../../sql-reference/data-types/string.md).",
+            .examples{
+                {"multiline",
+                 "SELECT formatQuerySingleLine('select a,    b FRom tab WHERE a > 3 and  b < 3');",
+                 "SELECT a, b FROM tab WHERE (a > 3) AND (b < 3)"}},
+            .categories{"Other"}});
 }
 
 REGISTER_FUNCTION(formatQuerySingleLineOrNull)
 {
-    factory.registerFunction<FunctionFormatQuery<OutputFormatting::SingleLine, ErrorHandling::Null, NameFormatQuerySingleLineOrNull>>(FunctionDocumentation{
-        .description = "Like formatQuery() but the returned formatted string contains no line breaks. Returns NULL in case of a parsing error.\n[example:multiline]",
-        .syntax = "formatQuerySingleLineOrNull(query)",
-        .arguments = {{"query", "The SQL query to be formatted. [String](../../sql-reference/data-types/string.md)"}},
-        .returned_value = "The formatted query. [String](../../sql-reference/data-types/string.md).",
-        .examples{
-            {"multiline",
-             "SELECT formatQuerySingleLine('select a,    b FRom tab WHERE a > 3 and  b < 3');",
-             "SELECT a, b FROM tab WHERE (a > 3) AND (b < 3)"}},
-        .categories{"Other"}});
+    factory.registerFunction(
+        "formatQuerySingleLineOrNull",
+        [](ContextPtr context) { return std::make_shared<FunctionFormatQuery>(context, "formatQuerySingleLineOrNull", OutputFormatting::SingleLine, ErrorHandling::Null); },
+        FunctionDocumentation{
+            .description = "Like formatQuery() but the returned formatted string contains no line breaks. Returns NULL in case of a parsing error.\n[example:multiline]",
+            .syntax = "formatQuerySingleLineOrNull(query)",
+            .arguments = {{"query", "The SQL query to be formatted. [String](../../sql-reference/data-types/string.md)"}},
+            .returned_value = "The formatted query. [String](../../sql-reference/data-types/string.md).",
+            .examples{
+                {"multiline",
+                 "SELECT formatQuerySingleLine('select a,    b FRom tab WHERE a > 3 and  b < 3');",
+                 "SELECT a, b FROM tab WHERE (a > 3) AND (b < 3)"}},
+            .categories{"Other"}});
 }
 
 }
diff --git a/src/Interpreters/Access/InterpreterShowCreateAccessEntityQuery.cpp b/src/Interpreters/Access/InterpreterShowCreateAccessEntityQuery.cpp
index a55588baeaa..1147d74c146 100644
--- a/src/Interpreters/Access/InterpreterShowCreateAccessEntityQuery.cpp
+++ b/src/Interpreters/Access/InterpreterShowCreateAccessEntityQuery.cpp
@@ -206,7 +206,7 @@ namespace
             if (!filter.empty())
             {
                 ParserExpression parser;
-                ASTPtr expr = parseQuery(parser, filter, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+                ASTPtr expr = parseQuery(parser, filter, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
                 query->filters.emplace_back(type, std::move(expr));
             }
         }
diff --git a/src/Interpreters/AsynchronousMetricLog.cpp b/src/Interpreters/AsynchronousMetricLog.cpp
index 5cf7f951eec..dc67bd91550 100644
--- a/src/Interpreters/AsynchronousMetricLog.cpp
+++ b/src/Interpreters/AsynchronousMetricLog.cpp
@@ -21,31 +21,31 @@ ColumnsDescription AsynchronousMetricLogElement::getColumnsDescription()
         {
             "hostname",
             std::make_shared<DataTypeLowCardinality>(std::make_shared<DataTypeString>()),
-            parseQuery(codec_parser, "(ZSTD(1))", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH),
+            parseQuery(codec_parser, "(ZSTD(1))", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS),
             "Hostname of the server executing the query."
         },
         {
             "event_date",
             std::make_shared<DataTypeDate>(),
-            parseQuery(codec_parser, "(Delta(2), ZSTD(1))", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH),
+            parseQuery(codec_parser, "(Delta(2), ZSTD(1))", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS),
             "Event date."
         },
         {
             "event_time",
             std::make_shared<DataTypeDateTime>(),
-            parseQuery(codec_parser, "(Delta(4), ZSTD(1))", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH),
+            parseQuery(codec_parser, "(Delta(4), ZSTD(1))", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS),
             "Event time."
         },
         {
             "metric",
             std::make_shared<DataTypeLowCardinality>(std::make_shared<DataTypeString>()),
-            parseQuery(codec_parser, "(ZSTD(1))", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH),
+            parseQuery(codec_parser, "(ZSTD(1))", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS),
             "Metric name."
         },
         {
             "value",
             std::make_shared<DataTypeFloat64>(),
-            parseQuery(codec_parser, "(ZSTD(3))", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH),
+            parseQuery(codec_parser, "(ZSTD(3))", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS),
             "Metric value."
         }
     };
diff --git a/src/Interpreters/DDLTask.cpp b/src/Interpreters/DDLTask.cpp
index fe2baea6b4e..e10f3ecfbc9 100644
--- a/src/Interpreters/DDLTask.cpp
+++ b/src/Interpreters/DDLTask.cpp
@@ -154,7 +154,8 @@ void DDLLogEntry::parse(const String & data)
             rb >> "settings: " >> settings_str >> "\n";
             ParserSetQuery parser{true};
             constexpr UInt64 max_depth = 16;
-            ASTPtr settings_ast = parseQuery(parser, settings_str, Context::getGlobalContextInstance()->getSettingsRef().max_query_size, max_depth);
+            constexpr UInt64 max_backtracks = DBMS_DEFAULT_MAX_PARSER_BACKTRACKS;
+            ASTPtr settings_ast = parseQuery(parser, settings_str, Context::getGlobalContextInstance()->getSettingsRef().max_query_size, max_depth, max_backtracks);
             settings.emplace(std::move(settings_ast->as<ASTSetQuery>()->changes));
         }
     }
@@ -197,7 +198,7 @@ void DDLTaskBase::parseQueryFromEntry(ContextPtr context)
 
     ParserQuery parser_query(end, settings.allow_settings_after_format_in_insert);
     String description;
-    query = parseQuery(parser_query, begin, end, description, 0, settings.max_parser_depth);
+    query = parseQuery(parser_query, begin, end, description, 0, settings.max_parser_depth, settings.max_parser_backtracks);
 }
 
 void DDLTaskBase::formatRewrittenQuery(ContextPtr context)
diff --git a/src/Interpreters/IInterpreterUnionOrSelectQuery.cpp b/src/Interpreters/IInterpreterUnionOrSelectQuery.cpp
index 60110916760..fed29b410db 100644
--- a/src/Interpreters/IInterpreterUnionOrSelectQuery.cpp
+++ b/src/Interpreters/IInterpreterUnionOrSelectQuery.cpp
@@ -96,7 +96,7 @@ static ASTPtr parseAdditionalPostFilter(const Context & context)
     ParserExpression parser;
     return parseQuery(
                 parser, filter.data(), filter.data() + filter.size(),
-                "additional filter", settings.max_query_size, settings.max_parser_depth);
+                "additional filter", settings.max_query_size, settings.max_parser_depth, settings.max_parser_backtracks);
 }
 
 static ActionsDAGPtr makeAdditionalPostFilter(ASTPtr & ast, ContextPtr context, const Block & header)
diff --git a/src/Interpreters/InterpreterCreateQuery.cpp b/src/Interpreters/InterpreterCreateQuery.cpp
index edd7452c130..2a08e8458a4 100644
--- a/src/Interpreters/InterpreterCreateQuery.cpp
+++ b/src/Interpreters/InterpreterCreateQuery.cpp
@@ -381,7 +381,7 @@ ASTPtr InterpreterCreateQuery::formatColumns(const NamesAndTypesList & columns)
         String type_name = column.type->getName();
         const char * pos = type_name.data();
         const char * end = pos + type_name.size();
-        column_declaration->type = parseQuery(type_parser, pos, end, "data type", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+        column_declaration->type = parseQuery(type_parser, pos, end, "data type", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
         columns_list->children.emplace_back(column_declaration);
     }
 
@@ -401,7 +401,7 @@ ASTPtr InterpreterCreateQuery::formatColumns(const NamesAndTypesList & columns,
         String type_name = alias_column.type->getName();
         const char * type_pos = type_name.data();
         const char * type_end = type_pos + type_name.size();
-        column_declaration->type = parseQuery(type_parser, type_pos, type_end, "data type", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+        column_declaration->type = parseQuery(type_parser, type_pos, type_end, "data type", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
 
         column_declaration->default_specifier = "ALIAS";
 
@@ -409,7 +409,7 @@ ASTPtr InterpreterCreateQuery::formatColumns(const NamesAndTypesList & columns,
         const char * alias_pos = alias.data();
         const char * alias_end = alias_pos + alias.size();
         ParserExpression expression_parser;
-        column_declaration->default_expression = parseQuery(expression_parser, alias_pos, alias_end, "expression", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+        column_declaration->default_expression = parseQuery(expression_parser, alias_pos, alias_end, "expression", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
         column_declaration->children.push_back(column_declaration->default_expression);
 
         columns_list->children.emplace_back(column_declaration);
@@ -433,7 +433,7 @@ ASTPtr InterpreterCreateQuery::formatColumns(const ColumnsDescription & columns)
         String type_name = column.type->getName();
         const char * type_name_pos = type_name.data();
         const char * type_name_end = type_name_pos + type_name.size();
-        column_declaration->type = parseQuery(type_parser, type_name_pos, type_name_end, "data type", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+        column_declaration->type = parseQuery(type_parser, type_name_pos, type_name_end, "data type", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
 
         if (column.default_desc.expression)
         {
@@ -1852,10 +1852,12 @@ void InterpreterCreateQuery::addColumnsDescriptionToCreateQueryIfNecessary(ASTCr
 
     auto ast_storage = std::make_shared<ASTStorage>();
     unsigned max_parser_depth = static_cast<unsigned>(getContext()->getSettingsRef().max_parser_depth);
+    unsigned max_parser_backtracks = static_cast<unsigned>(getContext()->getSettingsRef().max_parser_backtracks);
     auto query_from_storage = DB::getCreateQueryFromStorage(storage,
                                                             ast_storage,
                                                             false,
                                                             max_parser_depth,
+                                                            max_parser_backtracks,
                                                             true);
     auto & create_query_from_storage = query_from_storage->as<ASTCreateQuery &>();
 
diff --git a/src/Interpreters/InterpreterDeleteQuery.cpp b/src/Interpreters/InterpreterDeleteQuery.cpp
index 97ae9649ae8..8fb0dabb5b5 100644
--- a/src/Interpreters/InterpreterDeleteQuery.cpp
+++ b/src/Interpreters/InterpreterDeleteQuery.cpp
@@ -97,7 +97,8 @@ BlockIO InterpreterDeleteQuery::execute()
             alter_query.data() + alter_query.size(),
             "ALTER query",
             0,
-            DBMS_DEFAULT_MAX_PARSER_DEPTH);
+            DBMS_DEFAULT_MAX_PARSER_DEPTH,
+            DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
 
         auto context = Context::createCopy(getContext());
         context->setSetting("mutations_sync", 2);   /// Lightweight delete is always synchronous
diff --git a/src/Interpreters/InterpreterKillQueryQuery.cpp b/src/Interpreters/InterpreterKillQueryQuery.cpp
index 86196270ed1..26dae6a1df3 100644
--- a/src/Interpreters/InterpreterKillQueryQuery.cpp
+++ b/src/Interpreters/InterpreterKillQueryQuery.cpp
@@ -281,7 +281,7 @@ BlockIO InterpreterKillQueryQuery::execute()
                     const auto with_round_bracket = alter_command.front() == '(';
                     ParserAlterCommand parser{with_round_bracket};
                     auto command_ast
-                        = parseQuery(parser, alter_command, 0, getContext()->getSettingsRef().max_parser_depth);
+                        = parseQuery(parser, alter_command, 0, getContext()->getSettingsRef().max_parser_depth, getContext()->getSettingsRef().max_parser_backtracks);
                     required_access_rights = InterpreterAlterQuery::getRequiredAccessForCommand(
                         command_ast->as<const ASTAlterCommand &>(), table_id.database_name, table_id.table_name);
                     if (!access->isGranted(required_access_rights))
diff --git a/src/Interpreters/InterpreterSelectQuery.cpp b/src/Interpreters/InterpreterSelectQuery.cpp
index a314492c5b0..07f4e94680c 100644
--- a/src/Interpreters/InterpreterSelectQuery.cpp
+++ b/src/Interpreters/InterpreterSelectQuery.cpp
@@ -160,7 +160,7 @@ FilterDAGInfoPtr generateFilterActions(
     {
         ParserExpression expr_parser;
         /// We should add back quotes around column name as it can contain dots.
-        expr_list->children.push_back(parseQuery(expr_parser, backQuoteIfNeed(column_str), 0, context->getSettingsRef().max_parser_depth));
+        expr_list->children.push_back(parseQuery(expr_parser, backQuoteIfNeed(column_str), 0, context->getSettingsRef().max_parser_depth, context->getSettingsRef().max_parser_backtracks));
     }
 
     select_ast->setExpression(ASTSelectQuery::Expression::TABLES, std::make_shared<ASTTablesInSelectQuery>());
@@ -331,7 +331,7 @@ ASTPtr parseAdditionalFilterConditionForTable(
             const auto & settings = context.getSettingsRef();
             return parseQuery(
                 parser, filter.data(), filter.data() + filter.size(),
-                "additional filter", settings.max_query_size, settings.max_parser_depth);
+                "additional filter", settings.max_query_size, settings.max_parser_depth, settings.max_parser_backtracks);
         }
     }
 
diff --git a/src/Interpreters/JoinToSubqueryTransformVisitor.cpp b/src/Interpreters/JoinToSubqueryTransformVisitor.cpp
index 6251a9604e1..5c4ae528fc1 100644
--- a/src/Interpreters/JoinToSubqueryTransformVisitor.cpp
+++ b/src/Interpreters/JoinToSubqueryTransformVisitor.cpp
@@ -43,7 +43,7 @@ ASTPtr makeSubqueryTemplate(const String & table_alias)
     String query_template = "(select * from _t)";
     if (!table_alias.empty())
         query_template += " as " + table_alias;
-    ASTPtr subquery_template = parseQuery(parser, query_template, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+    ASTPtr subquery_template = parseQuery(parser, query_template, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
     if (!subquery_template)
         throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot parse subquery template");
     return subquery_template;
diff --git a/src/Interpreters/MySQL/tests/gtest_create_rewritten.cpp b/src/Interpreters/MySQL/tests/gtest_create_rewritten.cpp
index 9f6e9b930fd..6d6077a0295 100644
--- a/src/Interpreters/MySQL/tests/gtest_create_rewritten.cpp
+++ b/src/Interpreters/MySQL/tests/gtest_create_rewritten.cpp
@@ -19,7 +19,7 @@ using namespace DB;
 static inline ASTPtr tryRewrittenCreateQuery(const String & query, ContextPtr context)
 {
     ParserExternalDDLQuery external_ddl_parser;
-    ASTPtr ast = parseQuery(external_ddl_parser, "EXTERNAL DDL FROM MySQL(test_database, test_database) " + query, 0, 0);
+    ASTPtr ast = parseQuery(external_ddl_parser, "EXTERNAL DDL FROM MySQL(test_database, test_database) " + query, 0, 0, 0);
 
     return MySQLInterpreter::InterpreterCreateImpl::getRewrittenQueries(
         *ast->as<ASTExternalDDLQuery>()->external_ddl->as<MySQLParser::ASTCreateQuery>(),
diff --git a/src/Interpreters/SystemLog.cpp b/src/Interpreters/SystemLog.cpp
index a74b5c67726..e4cbbb8f5f7 100644
--- a/src/Interpreters/SystemLog.cpp
+++ b/src/Interpreters/SystemLog.cpp
@@ -216,7 +216,7 @@ std::shared_ptr<TSystemLog> createSystemLog(
     /// Validate engine definition syntax to prevent some configuration errors.
     ParserStorageWithComment storage_parser;
     auto storage_ast = parseQuery(storage_parser, log_settings.engine.data(), log_settings.engine.data() + log_settings.engine.size(),
-            "Storage to create table for " + config_prefix, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+            "Storage to create table for " + config_prefix, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
     auto & storage_with_comment = storage_ast->as<StorageWithComment &>();
 
     /// Add comment to AST. So it will be saved when the table will be renamed.
@@ -647,7 +647,7 @@ ASTPtr SystemLog<LogElement>::getCreateTableQuery()
 
     ASTPtr storage_with_comment_ast = parseQuery(
         storage_parser, storage_def.data(), storage_def.data() + storage_def.size(),
-        "Storage to create table for " + LogElement::name(), 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+        "Storage to create table for " + LogElement::name(), 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
 
     StorageWithComment & storage_with_comment = storage_with_comment_ast->as<StorageWithComment &>();
 
diff --git a/src/Interpreters/executeQuery.cpp b/src/Interpreters/executeQuery.cpp
index 88021038ebb..7dd46534fdf 100644
--- a/src/Interpreters/executeQuery.cpp
+++ b/src/Interpreters/executeQuery.cpp
@@ -746,18 +746,18 @@ static std::tuple<ASTPtr, BlockIO> executeQueryImpl(
         {
             ParserKQLStatement parser(end, settings.allow_settings_after_format_in_insert);
             /// TODO: parser should fail early when max_query_size limit is reached.
-            ast = parseKQLQuery(parser, begin, end, "", max_query_size, settings.max_parser_depth);
+            ast = parseKQLQuery(parser, begin, end, "", max_query_size, settings.max_parser_depth, settings.max_parser_backtracks);
         }
         else if (settings.dialect == Dialect::prql && !internal)
         {
-            ParserPRQLQuery parser(max_query_size, settings.max_parser_depth);
-            ast = parseQuery(parser, begin, end, "", max_query_size, settings.max_parser_depth);
+            ParserPRQLQuery parser(max_query_size, settings.max_parser_depth, settings.max_parser_backtracks);
+            ast = parseQuery(parser, begin, end, "", max_query_size, settings.max_parser_depth, settings.max_parser_backtracks);
         }
         else
         {
             ParserQuery parser(end, settings.allow_settings_after_format_in_insert);
             /// TODO: parser should fail early when max_query_size limit is reached.
-            ast = parseQuery(parser, begin, end, "", max_query_size, settings.max_parser_depth);
+            ast = parseQuery(parser, begin, end, "", max_query_size, settings.max_parser_depth, settings.max_parser_backtracks);
 
 #ifndef NDEBUG
             /// Verify that AST formatting is consistent:
@@ -774,7 +774,7 @@ static std::tuple<ASTPtr, BlockIO> executeQueryImpl(
                 ast2 = parseQuery(parser,
                     formatted1.data(),
                     formatted1.data() + formatted1.size(),
-                    "", new_max_query_size, settings.max_parser_depth);
+                    "", new_max_query_size, settings.max_parser_depth, settings.max_parser_backtracks);
             }
             catch (const Exception & e)
             {
diff --git a/src/Interpreters/getCustomKeyFilterForParallelReplicas.cpp b/src/Interpreters/getCustomKeyFilterForParallelReplicas.cpp
index 1295a4d5a75..d78b6ab0c4d 100644
--- a/src/Interpreters/getCustomKeyFilterForParallelReplicas.cpp
+++ b/src/Interpreters/getCustomKeyFilterForParallelReplicas.cpp
@@ -122,7 +122,7 @@ ASTPtr parseCustomKeyForTable(const String & custom_key, const Context & context
     const auto & settings = context.getSettingsRef();
     return parseQuery(
         parser, custom_key.data(), custom_key.data() + custom_key.size(),
-        "parallel replicas custom key", settings.max_query_size, settings.max_parser_depth);
+        "parallel replicas custom key", settings.max_query_size, settings.max_parser_depth, settings.max_parser_backtracks);
 }
 
 }
diff --git a/src/Interpreters/loadMetadata.cpp b/src/Interpreters/loadMetadata.cpp
index 2723eb37350..226472175b3 100644
--- a/src/Interpreters/loadMetadata.cpp
+++ b/src/Interpreters/loadMetadata.cpp
@@ -55,9 +55,11 @@ static void executeCreateQuery(
     bool create,
     bool has_force_restore_data_flag)
 {
+    const Settings & settings = context->getSettingsRef();
     ParserCreateQuery parser;
     ASTPtr ast = parseQuery(
-        parser, query.data(), query.data() + query.size(), "in file " + file_name, 0, context->getSettingsRef().max_parser_depth);
+        parser, query.data(), query.data() + query.size(), "in file " + file_name,
+        0, settings.max_parser_depth, settings.max_parser_backtracks);
 
     auto & ast_create_query = ast->as<ASTCreateQuery &>();
     ast_create_query.setDatabase(database);
diff --git a/src/Interpreters/parseColumnsListForTableFunction.cpp b/src/Interpreters/parseColumnsListForTableFunction.cpp
index 78b72022a9a..30a41c090d5 100644
--- a/src/Interpreters/parseColumnsListForTableFunction.cpp
+++ b/src/Interpreters/parseColumnsListForTableFunction.cpp
@@ -115,7 +115,7 @@ ColumnsDescription parseColumnsListFromString(const std::string & structure, con
     ParserColumnDeclarationList parser(true, true);
     const Settings & settings = context->getSettingsRef();
 
-    ASTPtr columns_list_raw = parseQuery(parser, structure, "columns declaration list", settings.max_query_size, settings.max_parser_depth);
+    ASTPtr columns_list_raw = parseQuery(parser, structure, "columns declaration list", settings.max_query_size, settings.max_parser_depth, settings.max_parser_backtracks);
 
     auto * columns_list = dynamic_cast<ASTExpressionList *>(columns_list_raw.get());
     if (!columns_list)
@@ -136,7 +136,7 @@ bool tryParseColumnsListFromString(const std::string & structure, ColumnsDescrip
     const char * start = structure.data();
     const char * end = structure.data() + structure.size();
     ASTPtr columns_list_raw = tryParseQuery(
-        parser, start, end, error, false, "columns declaration list", false, settings.max_query_size, settings.max_parser_depth);
+        parser, start, end, error, false, "columns declaration list", false, settings.max_query_size, settings.max_parser_depth, settings.max_parser_backtracks, true);
     if (!columns_list_raw)
         return false;
 
diff --git a/src/Interpreters/tests/gtest_comparison_graph.cpp b/src/Interpreters/tests/gtest_comparison_graph.cpp
index 96a78241c8e..ac24a8de368 100644
--- a/src/Interpreters/tests/gtest_comparison_graph.cpp
+++ b/src/Interpreters/tests/gtest_comparison_graph.cpp
@@ -12,7 +12,7 @@ using namespace DB;
 static ComparisonGraph<ASTPtr> getGraph(const String & query)
 {
     ParserExpressionList parser(false);
-    ASTPtr ast = parseQuery(parser, query, 0, 0);
+    ASTPtr ast = parseQuery(parser, query, 0, 0, 0);
     return ComparisonGraph<ASTPtr>(ast->children);
 }
 
diff --git a/src/Interpreters/tests/gtest_cycle_aliases.cpp b/src/Interpreters/tests/gtest_cycle_aliases.cpp
index 2bdeac90f8f..5ff3fbe1c2d 100644
--- a/src/Interpreters/tests/gtest_cycle_aliases.cpp
+++ b/src/Interpreters/tests/gtest_cycle_aliases.cpp
@@ -14,10 +14,10 @@ TEST(QueryNormalizer, SimpleLoopAlias)
 {
     String query = "a as a";
     ParserExpressionList parser(false);
-    ASTPtr ast = parseQuery(parser, query, 0, 0);
+    ASTPtr ast = parseQuery(parser, query, 0, 0, 0);
 
     Aliases aliases;
-    aliases["a"] = parseQuery(parser, "a as a", 0, 0)->children[0];
+    aliases["a"] = parseQuery(parser, "a as a", 0, 0, 0)->children[0];
 
     Settings settings;
     QueryNormalizer::Data normalizer_data(aliases, {}, false, settings, false);
@@ -28,11 +28,11 @@ TEST(QueryNormalizer, SimpleCycleAlias)
 {
     String query = "a as b, b as a";
     ParserExpressionList parser(false);
-    ASTPtr ast = parseQuery(parser, query, 0, 0);
+    ASTPtr ast = parseQuery(parser, query, 0, 0, 0);
 
     Aliases aliases;
-    aliases["a"] = parseQuery(parser, "b as a", 0, 0)->children[0];
-    aliases["b"] = parseQuery(parser, "a as b", 0, 0)->children[0];
+    aliases["a"] = parseQuery(parser, "b as a", 0, 0, 0)->children[0];
+    aliases["b"] = parseQuery(parser, "a as b", 0, 0, 0)->children[0];
 
     Settings settings;
     QueryNormalizer::Data normalizer_data(aliases, {}, false, settings, true);
diff --git a/src/Interpreters/tests/gtest_table_overrides.cpp b/src/Interpreters/tests/gtest_table_overrides.cpp
index 779bc7a53a4..09aa2e1f37f 100644
--- a/src/Interpreters/tests/gtest_table_overrides.cpp
+++ b/src/Interpreters/tests/gtest_table_overrides.cpp
@@ -34,11 +34,11 @@ TEST_P(TableOverrideTest, applyOverrides)
     const auto & [database_query, table_query, expected_query] = GetParam();
     ParserCreateQuery parser;
     ASTPtr database_ast;
-    ASSERT_NO_THROW(database_ast = parseQuery(parser, database_query, 0, 0));
+    ASSERT_NO_THROW(database_ast = parseQuery(parser, database_query, 0, 0, 0));
     auto * database = database_ast->as<ASTCreateQuery>();
     ASSERT_NE(nullptr, database);
     ASTPtr table_ast;
-    ASSERT_NO_THROW(table_ast = parseQuery(parser, table_query, 0, 0));
+    ASSERT_NO_THROW(table_ast = parseQuery(parser, table_query, 0, 0, 0));
     auto * table = table_ast->as<ASTCreateQuery>();
     ASSERT_NE(nullptr, table);
     auto table_name = table->table->as<ASTIdentifier>()->name();
diff --git a/src/Parsers/ExpressionListParsers.cpp b/src/Parsers/ExpressionListParsers.cpp
index 6d267a7d215..1e7d0158878 100644
--- a/src/Parsers/ExpressionListParsers.cpp
+++ b/src/Parsers/ExpressionListParsers.cpp
@@ -1918,7 +1918,7 @@ public:
                 && string_literal->as<ASTLiteral &>().value.tryGet(literal))
             {
                 Tokens tokens(literal.data(), literal.data() + literal.size());
-                IParser::Pos token_pos(tokens, 0);
+                IParser::Pos token_pos(tokens, pos.max_depth, pos.max_backtracks);
                 Expected token_expected;
                 ASTPtr expr;
 
diff --git a/src/Parsers/IParser.cpp b/src/Parsers/IParser.cpp
new file mode 100644
index 00000000000..d1e9ace89b6
--- /dev/null
+++ b/src/Parsers/IParser.cpp
@@ -0,0 +1,33 @@
+#include <Parsers/IParser.h>
+#include <iostream>
+
+namespace DB
+{
+
+IParser::Pos & IParser::Pos::operator=(const IParser::Pos & rhs)
+{
+    depth = rhs.depth;
+    max_depth = rhs.max_depth;
+
+    if (rhs.backtracks > backtracks)
+        backtracks = rhs.backtracks;
+
+    max_backtracks = rhs.max_backtracks;
+
+    if (rhs < *this)
+    {
+        ++backtracks;
+        if (max_backtracks && backtracks > max_backtracks)
+            throw Exception(ErrorCodes::TOO_DEEP_RECURSION, "Maximum amount of backtracking ({}) exceeded in the parser. "
+                "Consider rising max_parser_backtracks parameter.", max_backtracks);
+    }
+
+    TokenIterator::operator=(rhs);
+
+    if (backtracks % 1000 == 0)
+        std::cerr << backtracks << "\n";
+
+    return *this;
+}
+
+}
diff --git a/src/Parsers/IParser.h b/src/Parsers/IParser.h
index 198ec0346ff..291f8ee7d44 100644
--- a/src/Parsers/IParser.h
+++ b/src/Parsers/IParser.h
@@ -62,11 +62,18 @@ public:
         uint32_t depth = 0;
         uint32_t max_depth = 0;
 
-        Pos(Tokens & tokens_, uint32_t max_depth_) : TokenIterator(tokens_), max_depth(max_depth_)
+        uint32_t backtracks = 0;
+        uint32_t max_backtracks = 0;
+
+        Pos(Tokens & tokens_, uint32_t max_depth_, uint32_t max_backtracks_)
+            : TokenIterator(tokens_), max_depth(max_depth_), max_backtracks(max_backtracks_)
         {
         }
 
-        Pos(TokenIterator token_iterator_, uint32_t max_depth_) : TokenIterator(token_iterator_), max_depth(max_depth_) { }
+        Pos(TokenIterator token_iterator_, uint32_t max_depth_, uint32_t max_backtracks_)
+            : TokenIterator(token_iterator_), max_depth(max_depth_), max_backtracks(max_backtracks_)
+        {
+        }
 
         ALWAYS_INLINE void increaseDepth()
         {
@@ -97,6 +104,10 @@ public:
                 throw Exception(ErrorCodes::LOGICAL_ERROR, "Logical error in parser: incorrect calculation of parse depth");
             --depth;
         }
+
+        Pos(const Pos & rhs) = default;
+
+        Pos & operator=(const Pos & rhs);
     };
 
     /** Get the text of this parser parses. */
diff --git a/src/Parsers/Kusto/KustoFunctions/IParserKQLFunction.cpp b/src/Parsers/Kusto/KustoFunctions/IParserKQLFunction.cpp
index 152c29e5941..1d77007a37c 100644
--- a/src/Parsers/Kusto/KustoFunctions/IParserKQLFunction.cpp
+++ b/src/Parsers/Kusto/KustoFunctions/IParserKQLFunction.cpp
@@ -279,13 +279,13 @@ String IParserKQLFunction::getKQLFunctionName(IParser::Pos & pos)
 }
 
 String IParserKQLFunction::kqlCallToExpression(
-    const std::string_view function_name, const std::initializer_list<const std::string_view> params, const uint32_t max_depth)
+    const std::string_view function_name, const std::initializer_list<const std::string_view> params, uint32_t max_depth, uint32_t max_backtracks)
 {
-    return kqlCallToExpression(function_name, std::span(params), max_depth);
+    return kqlCallToExpression(function_name, std::span(params), max_depth, max_backtracks);
 }
 
 String IParserKQLFunction::kqlCallToExpression(
-    const std::string_view function_name, const std::span<const std::string_view> params, const uint32_t max_depth)
+    const std::string_view function_name, const std::span<const std::string_view> params, uint32_t max_depth, uint32_t max_backtracks)
 {
     const auto params_str = std::accumulate(
         std::cbegin(params),
@@ -302,7 +302,7 @@ String IParserKQLFunction::kqlCallToExpression(
 
     const auto kql_call = std::format("{}({})", function_name, params_str);
     DB::Tokens call_tokens(kql_call.c_str(), kql_call.c_str() + kql_call.length());
-    DB::IParser::Pos tokens_pos(call_tokens, max_depth);
+    DB::IParser::Pos tokens_pos(call_tokens, max_depth, max_backtracks);
     return DB::IParserKQLFunction::getExpression(tokens_pos);
 }
 
diff --git a/src/Parsers/Kusto/KustoFunctions/IParserKQLFunction.h b/src/Parsers/Kusto/KustoFunctions/IParserKQLFunction.h
index 147436551f9..f5069e80745 100644
--- a/src/Parsers/Kusto/KustoFunctions/IParserKQLFunction.h
+++ b/src/Parsers/Kusto/KustoFunctions/IParserKQLFunction.h
@@ -77,8 +77,8 @@ public:
     static std::optional<String>
     getOptionalArgument(const String & function_name, DB::IParser::Pos & pos, ArgumentState argument_state = ArgumentState::Parsed);
     static String
-    kqlCallToExpression(std::string_view function_name, std::initializer_list<const std::string_view> params, uint32_t max_depth);
-    static String kqlCallToExpression(std::string_view function_name, std::span<const std::string_view> params, uint32_t max_depth);
+    kqlCallToExpression(std::string_view function_name, std::initializer_list<const std::string_view> params, uint32_t max_depth, uint32_t max_backtracks);
+    static String kqlCallToExpression(std::string_view function_name, std::span<const std::string_view> params, uint32_t max_depth, uint32_t max_backtracks);
     static String escapeSingleQuotes(const String & input);
 
 protected:
diff --git a/src/Parsers/Kusto/KustoFunctions/KQLCastingFunctions.cpp b/src/Parsers/Kusto/KustoFunctions/KQLCastingFunctions.cpp
index b0eec16f56f..87841e295ba 100644
--- a/src/Parsers/Kusto/KustoFunctions/KQLCastingFunctions.cpp
+++ b/src/Parsers/Kusto/KustoFunctions/KQLCastingFunctions.cpp
@@ -99,7 +99,7 @@ bool ToTimeSpan::convertImpl(String & out, IParser::Pos & pos)
         ++pos;
         try
         {
-            auto result = kqlCallToExpression("time", {arg}, pos.max_depth);
+            auto result = kqlCallToExpression("time", {arg}, pos.max_depth, pos.max_backtracks);
             out = std::format("{}", result);
         }
         catch (...)
diff --git a/src/Parsers/Kusto/KustoFunctions/KQLDynamicFunctions.cpp b/src/Parsers/Kusto/KustoFunctions/KQLDynamicFunctions.cpp
index 924ac9f6490..e90be363e4b 100644
--- a/src/Parsers/Kusto/KustoFunctions/KQLDynamicFunctions.cpp
+++ b/src/Parsers/Kusto/KustoFunctions/KQLDynamicFunctions.cpp
@@ -99,7 +99,7 @@ bool ArrayRotateRight::convertImpl(String & out, IParser::Pos & pos)
 
     const auto array = getArgument(function_name, pos, ArgumentState::Raw);
     const auto count = getArgument(function_name, pos, ArgumentState::Raw);
-    out = kqlCallToExpression("array_rotate_left", {array, "-1 * " + count}, pos.max_depth);
+    out = kqlCallToExpression("array_rotate_left", {array, "-1 * " + count}, pos.max_depth, pos.max_backtracks);
 
     return true;
 }
@@ -140,7 +140,7 @@ bool ArrayShiftRight::convertImpl(String & out, IParser::Pos & pos)
         "array_shift_left",
         fill ? std::initializer_list<std::string_view>{array, negated_count, *fill}
              : std::initializer_list<std::string_view>{array, negated_count},
-        pos.max_depth);
+        pos.max_depth, pos.max_backtracks);
 
     return true;
 }
@@ -233,8 +233,8 @@ bool JaccardIndex::convertImpl(String & out, IParser::Pos & pos)
     const auto rhs = getArgument(function_name, pos, ArgumentState::Raw);
     out = std::format(
         "divide(length({0}), length({1}))",
-        kqlCallToExpression("set_intersect", {lhs, rhs}, pos.max_depth),
-        kqlCallToExpression("set_union", {lhs, rhs}, pos.max_depth));
+        kqlCallToExpression("set_intersect", {lhs, rhs}, pos.max_depth, pos.max_backtracks),
+        kqlCallToExpression("set_union", {lhs, rhs}, pos.max_depth, pos.max_backtracks));
 
     return true;
 }
@@ -292,7 +292,7 @@ bool SetDifference::convertImpl(String & out, IParser::Pos & pos)
             while (auto next_array = getOptionalArgument(function_name, pos, ArgumentState::Raw))
                 arrays.push_back(*next_array);
 
-            return kqlCallToExpression("set_union", std::vector<std::string_view>(arrays.cbegin(), arrays.cend()), pos.max_depth);
+            return kqlCallToExpression("set_union", std::vector<std::string_view>(arrays.cbegin(), arrays.cend()), pos.max_depth, pos.max_backtracks);
         });
 
     out = std::format("arrayFilter(x -> not has({1}, x), arrayDistinct({0}))", lhs, rhs);
diff --git a/src/Parsers/Kusto/KustoFunctions/KQLIPFunctions.cpp b/src/Parsers/Kusto/KustoFunctions/KQLIPFunctions.cpp
index 6f853b16fbc..06566dc54ec 100644
--- a/src/Parsers/Kusto/KustoFunctions/KQLIPFunctions.cpp
+++ b/src/Parsers/Kusto/KustoFunctions/KQLIPFunctions.cpp
@@ -34,10 +34,10 @@ bool Ipv4Compare::convertImpl(String & out, IParser::Pos & pos)
         "sign(IPv4StringToNumOrNull(toString((tupleElement(IPv4CIDRToRange(assumeNotNull(lhs_ip_{5}), "
         "toUInt8(min2({4}, min2(assumeNotNull(lhs_mask_{5}), assumeNotNull(rhs_mask_{5})))) as mask_{5}), 1))))"
         "   - IPv4StringToNumOrNull(toString((tupleElement(IPv4CIDRToRange(assumeNotNull(rhs_ip_{5}), mask_{5}), 1))))))",
-        kqlCallToExpression("parse_ipv4", {lhs}, pos.max_depth),
-        kqlCallToExpression("ipv4_netmask_suffix", {lhs}, pos.max_depth),
-        kqlCallToExpression("parse_ipv4", {rhs}, pos.max_depth),
-        kqlCallToExpression("ipv4_netmask_suffix", {rhs}, pos.max_depth),
+        kqlCallToExpression("parse_ipv4", {lhs}, pos.max_depth, pos.max_backtracks),
+        kqlCallToExpression("ipv4_netmask_suffix", {lhs}, pos.max_depth, pos.max_backtracks),
+        kqlCallToExpression("parse_ipv4", {rhs}, pos.max_depth, pos.max_backtracks),
+        kqlCallToExpression("ipv4_netmask_suffix", {rhs}, pos.max_depth, pos.max_backtracks),
         mask ? *mask : "32",
         generateUniqueIdentifier());
     return true;
@@ -56,8 +56,8 @@ bool Ipv4IsInRange::convertImpl(String & out, IParser::Pos & pos)
         "or isNull({1} as range_start_ip_{3}) or isNull({2} as range_mask_{3}), null, "
         "bitXor(range_start_ip_{3}, bitAnd(ip_{3}, bitNot(toUInt32(intExp2(toInt32(32 - range_mask_{3})) - 1)))) = 0) ",
         ip_address,
-        kqlCallToExpression("parse_ipv4", {ip_range}, pos.max_depth),
-        kqlCallToExpression("ipv4_netmask_suffix", {ip_range}, pos.max_depth),
+        kqlCallToExpression("parse_ipv4", {ip_range}, pos.max_depth, pos.max_backtracks),
+        kqlCallToExpression("ipv4_netmask_suffix", {ip_range}, pos.max_depth, pos.max_backtracks),
         generateUniqueIdentifier());
     return true;
 }
@@ -71,7 +71,7 @@ bool Ipv4IsMatch::convertImpl(String & out, IParser::Pos & pos)
     const auto lhs = getArgument(function_name, pos, ArgumentState::Raw);
     const auto rhs = getArgument(function_name, pos, ArgumentState::Raw);
     const auto mask = getOptionalArgument(function_name, pos, ArgumentState::Raw);
-    out = std::format("equals({}, 0)", kqlCallToExpression("ipv4_compare", {lhs, rhs, mask ? *mask : "32"}, pos.max_depth));
+    out = std::format("equals({}, 0)", kqlCallToExpression("ipv4_compare", {lhs, rhs, mask ? *mask : "32"}, pos.max_depth, pos.max_backtracks));
     return true;
 }
 
@@ -196,7 +196,7 @@ bool Ipv6IsMatch::convertImpl(String & out, IParser::Pos & pos)
     const auto lhs = getArgument(function_name, pos, ArgumentState::Raw);
     const auto rhs = getArgument(function_name, pos, ArgumentState::Raw);
     const auto mask = getOptionalArgument(function_name, pos, ArgumentState::Raw);
-    out = std::format("equals({}, 0)", kqlCallToExpression("ipv6_compare", {lhs, rhs, mask ? *mask : "128"}, pos.max_depth));
+    out = std::format("equals({}, 0)", kqlCallToExpression("ipv6_compare", {lhs, rhs, mask ? *mask : "128"}, pos.max_depth, pos.max_backtracks));
     return true;
 }
 
@@ -228,9 +228,9 @@ bool ParseIpv6Mask::convertImpl(String & out, IParser::Pos & pos)
     const auto unique_identifier = generateUniqueIdentifier();
     out = std::format(
         "if(empty({0} as ipv4_{3}), {1}, {2})",
-        kqlCallToExpression("format_ipv4", {"trim_start('::', " + ip_address + ")", mask + " - 96"}, pos.max_depth),
-        kqlCallToExpression("parse_ipv6", {"strcat(tostring(parse_ipv6(" + ip_address + ")), '/', tostring(" + mask + "))"}, pos.max_depth),
-        kqlCallToExpression("parse_ipv6", {"ipv4_" + unique_identifier}, pos.max_depth),
+        kqlCallToExpression("format_ipv4", {"trim_start('::', " + ip_address + ")", mask + " - 96"}, pos.max_depth, pos.max_backtracks),
+        kqlCallToExpression("parse_ipv6", {"strcat(tostring(parse_ipv6(" + ip_address + ")), '/', tostring(" + mask + "))"}, pos.max_depth, pos.max_backtracks),
+        kqlCallToExpression("parse_ipv6", {"ipv4_" + unique_identifier}, pos.max_depth, pos.max_backtracks),
         unique_identifier);
     return true;
 }
@@ -247,9 +247,9 @@ bool FormatIpv4::convertImpl(String & out, IParser::Pos & pos)
         "ifNull(if(isNotNull(toUInt32OrNull(toString({0})) as param_as_uint32_{3}) and toTypeName({0}) = 'String' or ({1}) < 0 "
         "or isNull(ifNull(param_as_uint32_{3}, {2}) as ip_as_number_{3}), null, "
         "IPv4NumToString(bitAnd(ip_as_number_{3}, bitNot(toUInt32(intExp2(toInt32(32 - ({1}))) - 1))))), '')",
-        ParserKQLBase::getExprFromToken(ip_address, pos.max_depth),
+        ParserKQLBase::getExprFromToken(ip_address, pos.max_depth, pos.max_backtracks),
         mask ? *mask : "32",
-        kqlCallToExpression("parse_ipv4", {"tostring(" + ip_address + ")"}, pos.max_depth),
+        kqlCallToExpression("parse_ipv4", {"tostring(" + ip_address + ")"}, pos.max_depth, pos.max_backtracks),
         generateUniqueIdentifier());
     return true;
 }
@@ -266,10 +266,10 @@ bool FormatIpv4Mask::convertImpl(String & out, IParser::Pos & pos)
     out = std::format(
         "if(empty({1} as formatted_ip_{2}) or position(toTypeName({0}), 'Int') = 0 or not {0} between 0 and 32, '', "
         "concat(formatted_ip_{2}, '/', toString(toInt64(min2({0}, ifNull({3} as suffix_{2}, 32))))))",
-        ParserKQLBase::getExprFromToken(calculated_mask, pos.max_depth),
-        kqlCallToExpression("format_ipv4", {ip_address, calculated_mask}, pos.max_depth),
+        ParserKQLBase::getExprFromToken(calculated_mask, pos.max_depth, pos.max_backtracks),
+        kqlCallToExpression("format_ipv4", {ip_address, calculated_mask}, pos.max_depth, pos.max_backtracks),
         generateUniqueIdentifier(),
-        kqlCallToExpression("ipv4_netmask_suffix", {"tostring(" + ip_address + ")"}, pos.max_depth));
+        kqlCallToExpression("ipv4_netmask_suffix", {"tostring(" + ip_address + ")"}, pos.max_depth, pos.max_backtracks));
     return true;
 }
 }
diff --git a/src/Parsers/Kusto/KustoFunctions/KQLStringFunctions.cpp b/src/Parsers/Kusto/KustoFunctions/KQLStringFunctions.cpp
index 82cfa68b180..18c986c2191 100644
--- a/src/Parsers/Kusto/KustoFunctions/KQLStringFunctions.cpp
+++ b/src/Parsers/Kusto/KustoFunctions/KQLStringFunctions.cpp
@@ -442,7 +442,7 @@ bool ParseJSON::convertImpl(String & out, IParser::Pos & pos)
     {
         --pos;
         auto arg = getArgument(fn_name, pos);
-        auto result = kqlCallToExpression("dynamic", {arg}, pos.max_depth);
+        auto result = kqlCallToExpression("dynamic", {arg}, pos.max_depth, pos.max_backtracks);
         out = std::format("{}", result);
     }
     else
@@ -729,7 +729,7 @@ bool Trim::convertImpl(String & out, IParser::Pos & pos)
 
     const auto regex = getArgument(fn_name, pos, ArgumentState::Raw);
     const auto source = getArgument(fn_name, pos, ArgumentState::Raw);
-    out = kqlCallToExpression("trim_start", {regex, std::format("trim_end({0}, {1})", regex, source)}, pos.max_depth);
+    out = kqlCallToExpression("trim_start", {regex, std::format("trim_end({0}, {1})", regex, source)}, pos.max_depth, pos.max_backtracks);
 
     return true;
 }
diff --git a/src/Parsers/Kusto/ParserKQLDistinct.cpp b/src/Parsers/Kusto/ParserKQLDistinct.cpp
index 2de4d2c28e7..3ec823a61b5 100644
--- a/src/Parsers/Kusto/ParserKQLDistinct.cpp
+++ b/src/Parsers/Kusto/ParserKQLDistinct.cpp
@@ -12,7 +12,7 @@ bool ParserKQLDistinct::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
     expr = getExprFromToken(pos);
 
     Tokens tokens(expr.c_str(), expr.c_str() + expr.size());
-    IParser::Pos new_pos(tokens, pos.max_depth);
+    IParser::Pos new_pos(tokens, pos.max_depth, pos.max_backtracks);
 
     if (!ParserNotEmptyExpressionList(false).parse(new_pos, select_expression_list, expected))
         return false;
diff --git a/src/Parsers/Kusto/ParserKQLExtend.cpp b/src/Parsers/Kusto/ParserKQLExtend.cpp
index b37618f69fd..41ce296bd25 100644
--- a/src/Parsers/Kusto/ParserKQLExtend.cpp
+++ b/src/Parsers/Kusto/ParserKQLExtend.cpp
@@ -23,7 +23,7 @@ bool ParserKQLExtend ::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
     String except_str;
     String new_extend_str;
     Tokens ntokens(extend_expr.c_str(), extend_expr.c_str() + extend_expr.size());
-    IParser::Pos npos(ntokens, pos.max_depth);
+    IParser::Pos npos(ntokens, pos.max_depth, pos.max_backtracks);
 
     String alias;
 
@@ -77,7 +77,7 @@ bool ParserKQLExtend ::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 
     String expr = std::format("SELECT * {}, {} from prev", except_str, new_extend_str);
     Tokens tokens(expr.c_str(), expr.c_str() + expr.size());
-    IParser::Pos new_pos(tokens, pos.max_depth);
+    IParser::Pos new_pos(tokens, pos.max_depth, pos.max_backtracks);
 
     if (!ParserSelectQuery().parse(new_pos, select_query, expected))
         return false;
diff --git a/src/Parsers/Kusto/ParserKQLFilter.cpp b/src/Parsers/Kusto/ParserKQLFilter.cpp
index 74d8610ecd4..b060ce8d2c7 100644
--- a/src/Parsers/Kusto/ParserKQLFilter.cpp
+++ b/src/Parsers/Kusto/ParserKQLFilter.cpp
@@ -14,7 +14,7 @@ bool ParserKQLFilter::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
     ASTPtr where_expression;
 
     Tokens token_filter(expr.c_str(), expr.c_str() + expr.size());
-    IParser::Pos pos_filter(token_filter, pos.max_depth);
+    IParser::Pos pos_filter(token_filter, pos.max_depth, pos.max_backtracks);
     if (!ParserExpressionWithOptionalAlias(false).parse(pos_filter, where_expression, expected))
         return false;
 
diff --git a/src/Parsers/Kusto/ParserKQLLimit.cpp b/src/Parsers/Kusto/ParserKQLLimit.cpp
index 910f0e8e1a3..0eb460757b1 100644
--- a/src/Parsers/Kusto/ParserKQLLimit.cpp
+++ b/src/Parsers/Kusto/ParserKQLLimit.cpp
@@ -14,7 +14,7 @@ bool ParserKQLLimit::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
     auto expr = getExprFromToken(pos);
 
     Tokens tokens(expr.c_str(), expr.c_str() + expr.size());
-    IParser::Pos new_pos(tokens, pos.max_depth);
+    IParser::Pos new_pos(tokens, pos.max_depth, pos.max_backtracks);
 
     if (!ParserExpressionWithOptionalAlias(false).parse(new_pos, limit_length, expected))
         return false;
diff --git a/src/Parsers/Kusto/ParserKQLMVExpand.cpp b/src/Parsers/Kusto/ParserKQLMVExpand.cpp
index 7d242dffaf7..d174e9b5911 100644
--- a/src/Parsers/Kusto/ParserKQLMVExpand.cpp
+++ b/src/Parsers/Kusto/ParserKQLMVExpand.cpp
@@ -69,7 +69,7 @@ bool ParserKQLMVExpand::parseColumnArrayExprs(ColumnArrayExprs & column_array_ex
 
         auto add_columns = [&]
         {
-            column_array_expr = getExprFromToken(String(expr_begin_pos->begin, expr_end_pos->end), pos.max_depth);
+            column_array_expr = getExprFromToken(String(expr_begin_pos->begin, expr_end_pos->end), pos.max_depth, pos.max_backtracks);
 
             if (alias.empty())
             {
@@ -189,7 +189,7 @@ bool ParserKQLMVExpand::parserMVExpand(KQLMVExpand & kql_mv_expand, Pos & pos, E
     return true;
 }
 
-bool ParserKQLMVExpand::genQuery(KQLMVExpand & kql_mv_expand, ASTPtr & select_node, int32_t max_depth)
+bool ParserKQLMVExpand::genQuery(KQLMVExpand & kql_mv_expand, ASTPtr & select_node, uint32_t max_depth, uint32_t max_backtracks)
 {
     String expand_str;
     String cast_type_column_remove, cast_type_column_rename;
@@ -253,7 +253,7 @@ bool ParserKQLMVExpand::genQuery(KQLMVExpand & kql_mv_expand, ASTPtr & select_no
     if (cast_type_column_remove.empty())
     {
         query = std::format("Select {} {} From {} {}", columns, extra_columns, input, expand_str);
-        if (!parseSQLQueryByString(std::make_unique<ParserSelectQuery>(), query, sub_query_node, max_depth))
+        if (!parseSQLQueryByString(std::make_unique<ParserSelectQuery>(), query, sub_query_node, max_depth, max_backtracks))
             return false;
         if (!setSubQuerySource(sub_query_node, select_node, false, false))
             return false;
@@ -262,14 +262,14 @@ bool ParserKQLMVExpand::genQuery(KQLMVExpand & kql_mv_expand, ASTPtr & select_no
     else
     {
         query = std::format("(Select {} {} From {} {})", columns, extra_columns, input, expand_str);
-        if (!parseSQLQueryByString(std::make_unique<ParserTablesInSelectQuery>(), query, sub_query_node, max_depth))
+        if (!parseSQLQueryByString(std::make_unique<ParserTablesInSelectQuery>(), query, sub_query_node, max_depth, max_backtracks))
             return false;
         if (!setSubQuerySource(sub_query_node, select_node, true, false))
             return false;
         select_node = std::move(sub_query_node);
 
         auto rename_query = std::format("(Select * {}, {} From {})", cast_type_column_remove, cast_type_column_rename, "query");
-        if (!parseSQLQueryByString(std::make_unique<ParserTablesInSelectQuery>(), rename_query, sub_query_node, max_depth))
+        if (!parseSQLQueryByString(std::make_unique<ParserTablesInSelectQuery>(), rename_query, sub_query_node, max_depth, max_backtracks))
             return false;
         if (!setSubQuerySource(sub_query_node, select_node, true, true))
             return false;
@@ -277,7 +277,7 @@ bool ParserKQLMVExpand::genQuery(KQLMVExpand & kql_mv_expand, ASTPtr & select_no
         select_node = std::move(sub_query_node);
         query = std::format("Select * {}, {} from {}", cast_type_column_restore, cast_type_column_restore_name, "rename_query");
 
-        if (!parseSQLQueryByString(std::make_unique<ParserSelectQuery>(), query, sub_query_node, max_depth))
+        if (!parseSQLQueryByString(std::make_unique<ParserSelectQuery>(), query, sub_query_node, max_depth, max_backtracks))
             return false;
         sub_query_node->as<ASTSelectQuery>()->setExpression(ASTSelectQuery::Expression::TABLES, std::move(select_node));
         select_node = std::move(sub_query_node);
@@ -294,12 +294,12 @@ bool ParserKQLMVExpand::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
     KQLMVExpand kql_mv_expand;
     if (!parserMVExpand(kql_mv_expand, pos, expected))
         return false;
-    if (!genQuery(kql_mv_expand, node, pos.max_depth))
+    if (!genQuery(kql_mv_expand, node, pos.max_depth, pos.max_backtracks))
         return false;
 
     const String setting_str = "enable_unaligned_array_join = 1";
     Tokens token_settings(setting_str.c_str(), setting_str.c_str() + setting_str.size());
-    IParser::Pos pos_settings(token_settings, pos.max_depth);
+    IParser::Pos pos_settings(token_settings, pos.max_depth, pos.max_backtracks);
 
     if (!ParserSetQuery(true).parse(pos_settings, setting, expected))
         return false;
diff --git a/src/Parsers/Kusto/ParserKQLMVExpand.h b/src/Parsers/Kusto/ParserKQLMVExpand.h
index 61f206bb00d..068aee53f58 100644
--- a/src/Parsers/Kusto/ParserKQLMVExpand.h
+++ b/src/Parsers/Kusto/ParserKQLMVExpand.h
@@ -33,7 +33,7 @@ protected:
 
     static bool parseColumnArrayExprs(ColumnArrayExprs & column_array_exprs, Pos & pos, Expected & expected);
     static bool parserMVExpand(KQLMVExpand & kql_mv_expand, Pos & pos, Expected & expected);
-    static bool genQuery(KQLMVExpand & kql_mv_expand, ASTPtr & select_node, int32_t max_depth);
+    static bool genQuery(KQLMVExpand & kql_mv_expand, ASTPtr & select_node, uint32_t max_depth, uint32_t max_backtracks);
 
     const char * getName() const override { return "KQL mv-expand"; }
     bool parseImpl(Pos & pos, ASTPtr & node, Expected & expected) override;
diff --git a/src/Parsers/Kusto/ParserKQLMakeSeries.cpp b/src/Parsers/Kusto/ParserKQLMakeSeries.cpp
index e89423e2fc9..4759efc0025 100644
--- a/src/Parsers/Kusto/ParserKQLMakeSeries.cpp
+++ b/src/Parsers/Kusto/ParserKQLMakeSeries.cpp
@@ -142,7 +142,7 @@ bool ParserKQLMakeSeries ::parseFromToStepClause(FromToStepClause & from_to_step
         || ParserKQLDateTypeTimespan().parseConstKQLTimespan(from_to_step.step_str))
     {
         from_to_step.is_timespan = true;
-        from_to_step.step = std::stod(getExprFromToken(from_to_step.step_str, pos.max_depth));
+        from_to_step.step = std::stod(getExprFromToken(from_to_step.step_str, pos.max_depth, pos.max_backtracks));
     }
     else
         from_to_step.step = std::stod(from_to_step.step_str);
@@ -150,7 +150,7 @@ bool ParserKQLMakeSeries ::parseFromToStepClause(FromToStepClause & from_to_step
     return true;
 }
 
-bool ParserKQLMakeSeries ::makeSeries(KQLMakeSeries & kql_make_series, ASTPtr & select_node, const uint32_t & max_depth)
+bool ParserKQLMakeSeries ::makeSeries(KQLMakeSeries & kql_make_series, ASTPtr & select_node, uint32_t max_depth, uint32_t max_backtracks)
 {
     const uint64_t era_diff
         = 62135596800; // this magic number is the differicen is second form 0001-01-01 (Azure start time ) and 1970-01-01 (CH start time)
@@ -166,15 +166,15 @@ bool ParserKQLMakeSeries ::makeSeries(KQLMakeSeries & kql_make_series, ASTPtr &
     auto step = from_to_step.step;
 
     if (!kql_make_series.from_to_step.from_str.empty())
-        start_str = getExprFromToken(kql_make_series.from_to_step.from_str, max_depth);
+        start_str = getExprFromToken(kql_make_series.from_to_step.from_str, max_depth, max_backtracks);
 
     if (!kql_make_series.from_to_step.to_str.empty())
-        end_str = getExprFromToken(from_to_step.to_str, max_depth);
+        end_str = getExprFromToken(from_to_step.to_str, max_depth, max_backtracks);
 
     auto date_type_cast = [&](String & src)
     {
         Tokens tokens(src.c_str(), src.c_str() + src.size());
-        IParser::Pos pos(tokens, max_depth);
+        IParser::Pos pos(tokens, max_depth, max_backtracks);
         String res;
         while (isValidKQLPos(pos))
         {
@@ -201,7 +201,7 @@ bool ParserKQLMakeSeries ::makeSeries(KQLMakeSeries & kql_make_series, ASTPtr &
     {
         std::vector<String> group_expression_tokens;
         Tokens tokens(group_expression.c_str(), group_expression.c_str() + group_expression.size());
-        IParser::Pos pos(tokens, max_depth);
+        IParser::Pos pos(tokens, max_depth, max_backtracks);
         while (isValidKQLPos(pos))
         {
             if (String(pos->begin, pos->end) == "AS")
@@ -296,7 +296,7 @@ bool ParserKQLMakeSeries ::makeSeries(KQLMakeSeries & kql_make_series, ASTPtr &
 
     ASTPtr sub_query_node;
 
-    if (!ParserSimpleCHSubquery(select_node).parseByString(sub_sub_query, sub_query_node, max_depth))
+    if (!ParserSimpleCHSubquery(select_node).parseByString(sub_sub_query, sub_query_node, max_depth, max_backtracks))
         return false;
     select_node->as<ASTSelectQuery>()->setExpression(ASTSelectQuery::Expression::TABLES, std::move(sub_query_node));
 
@@ -351,7 +351,7 @@ bool ParserKQLMakeSeries ::makeSeries(KQLMakeSeries & kql_make_series, ASTPtr &
     else
         main_query = std::format("{},{}", group_expression_alias, final_axis_agg_alias_list);
 
-    if (!ParserSimpleCHSubquery(select_node).parseByString(sub_query, sub_query_node, max_depth))
+    if (!ParserSimpleCHSubquery(select_node).parseByString(sub_query, sub_query_node, max_depth, max_backtracks))
         return false;
     select_node->as<ASTSelectQuery>()->setExpression(ASTSelectQuery::Expression::TABLES, std::move(sub_query_node));
 
@@ -411,10 +411,10 @@ bool ParserKQLMakeSeries ::parseImpl(Pos & pos, ASTPtr & node, Expected & expect
             subquery_columns += ", " + column_str;
     }
 
-    makeSeries(kql_make_series, node, pos.max_depth);
+    makeSeries(kql_make_series, node, pos.max_depth, pos.max_backtracks);
 
     Tokens token_main_query(kql_make_series.main_query.c_str(), kql_make_series.main_query.c_str() + kql_make_series.main_query.size());
-    IParser::Pos pos_main_query(token_main_query, pos.max_depth);
+    IParser::Pos pos_main_query(token_main_query, pos.max_depth, pos.max_backtracks);
 
     if (!ParserNotEmptyExpressionList(true).parse(pos_main_query, select_expression_list, expected))
         return false;
diff --git a/src/Parsers/Kusto/ParserKQLMakeSeries.h b/src/Parsers/Kusto/ParserKQLMakeSeries.h
index ef7cc4976f6..6a32e76eff3 100644
--- a/src/Parsers/Kusto/ParserKQLMakeSeries.h
+++ b/src/Parsers/Kusto/ParserKQLMakeSeries.h
@@ -42,7 +42,7 @@ protected:
         String main_query;
     };
 
-    static bool makeSeries(KQLMakeSeries & kql_make_series, ASTPtr & select_node, const uint32_t & max_depth);
+    static bool makeSeries(KQLMakeSeries & kql_make_series, ASTPtr & select_node, uint32_t max_depth, uint32_t max_backtracks);
     static bool parseAggregationColumns(AggregationColumns & aggregation_columns, Pos & pos);
     static bool parseFromToStepClause(FromToStepClause & from_to_step, Pos & pos);
 
diff --git a/src/Parsers/Kusto/ParserKQLPrint.cpp b/src/Parsers/Kusto/ParserKQLPrint.cpp
index bd9980ea96d..37483439f14 100644
--- a/src/Parsers/Kusto/ParserKQLPrint.cpp
+++ b/src/Parsers/Kusto/ParserKQLPrint.cpp
@@ -10,7 +10,7 @@ bool ParserKQLPrint::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
     const String expr = getExprFromToken(pos);
 
     Tokens tokens(expr.c_str(), expr.c_str() + expr.size());
-    IParser::Pos new_pos(tokens, pos.max_depth);
+    IParser::Pos new_pos(tokens, pos.max_depth, pos.max_backtracks);
 
     if (!ParserNotEmptyExpressionList(true).parse(new_pos, select_expression_list, expected))
         return false;
diff --git a/src/Parsers/Kusto/ParserKQLProject.cpp b/src/Parsers/Kusto/ParserKQLProject.cpp
index fdc458b7707..eab9ee082c5 100644
--- a/src/Parsers/Kusto/ParserKQLProject.cpp
+++ b/src/Parsers/Kusto/ParserKQLProject.cpp
@@ -12,7 +12,7 @@ bool ParserKQLProject ::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
     expr = getExprFromToken(pos);
 
     Tokens tokens(expr.c_str(), expr.c_str() + expr.size());
-    IParser::Pos new_pos(tokens, pos.max_depth);
+    IParser::Pos new_pos(tokens, pos.max_depth, pos.max_backtracks);
 
     if (!ParserNotEmptyExpressionList(false).parse(new_pos, select_expression_list, expected))
         return false;
diff --git a/src/Parsers/Kusto/ParserKQLQuery.cpp b/src/Parsers/Kusto/ParserKQLQuery.cpp
index a54a2b0eda9..6fd9c95ec6f 100644
--- a/src/Parsers/Kusto/ParserKQLQuery.cpp
+++ b/src/Parsers/Kusto/ParserKQLQuery.cpp
@@ -33,20 +33,20 @@ namespace ErrorCodes
     extern const int SYNTAX_ERROR;
 }
 
-bool ParserKQLBase::parseByString(const String expr, ASTPtr & node, const uint32_t max_depth)
+bool ParserKQLBase::parseByString(String expr, ASTPtr & node, uint32_t max_depth, uint32_t max_backtracks)
 {
     Expected expected;
 
     Tokens tokens(expr.c_str(), expr.c_str() + expr.size());
-    IParser::Pos pos(tokens, max_depth);
+    IParser::Pos pos(tokens, max_depth, max_backtracks);
     return parse(pos, node, expected);
 }
 
-bool ParserKQLBase::parseSQLQueryByString(ParserPtr && parser, String & query, ASTPtr & select_node, int32_t max_depth)
+bool ParserKQLBase::parseSQLQueryByString(ParserPtr && parser, String & query, ASTPtr & select_node, uint32_t max_depth, uint32_t max_backtracks)
 {
     Expected expected;
     Tokens token_subquery(query.c_str(), query.c_str() + query.size());
-    IParser::Pos pos_subquery(token_subquery, max_depth);
+    IParser::Pos pos_subquery(token_subquery, max_depth, max_backtracks);
     if (!parser->parse(pos_subquery, select_node, expected))
         return false;
     return true;
@@ -121,10 +121,10 @@ bool ParserKQLBase::setSubQuerySource(ASTPtr & select_query, ASTPtr & source, bo
     return true;
 }
 
-String ParserKQLBase::getExprFromToken(const String & text, const uint32_t max_depth)
+String ParserKQLBase::getExprFromToken(const String & text, uint32_t max_depth, uint32_t max_backtracks)
 {
     Tokens tokens(text.c_str(), text.c_str() + text.size());
-    IParser::Pos pos(tokens, max_depth);
+    IParser::Pos pos(tokens, max_depth, max_backtracks);
 
     return getExprFromToken(pos);
 }
@@ -523,7 +523,7 @@ bool ParserKQLQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 
             String sub_query = std::format("({})", String(operation_pos.front().second->begin, last_pos->end));
             Tokens token_subquery(sub_query.c_str(), sub_query.c_str() + sub_query.size());
-            IParser::Pos pos_subquery(token_subquery, pos.max_depth);
+            IParser::Pos pos_subquery(token_subquery, pos.max_depth, pos.max_backtracks);
 
             if (!ParserKQLSubquery().parse(pos_subquery, tables, expected))
                 return false;
@@ -544,7 +544,7 @@ bool ParserKQLQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
             if (oprator)
             {
                 Tokens token_clause(op_calsue.c_str(), op_calsue.c_str() + op_calsue.size());
-                IParser::Pos pos_clause(token_clause, pos.max_depth);
+                IParser::Pos pos_clause(token_clause, pos.max_depth, pos.max_backtracks);
                 if (!oprator->parse(pos_clause, node, expected))
                     return false;
             }
@@ -577,7 +577,7 @@ bool ParserKQLQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
     {
         auto expr = String("*");
         Tokens tokens(expr.c_str(), expr.c_str() + expr.size());
-        IParser::Pos new_pos(tokens, pos.max_depth);
+        IParser::Pos new_pos(tokens, pos.max_depth, pos.max_backtracks);
         if (!std::make_unique<ParserKQLProject>()->parse(new_pos, node, expected))
             return false;
     }
diff --git a/src/Parsers/Kusto/ParserKQLQuery.h b/src/Parsers/Kusto/ParserKQLQuery.h
index a594f43ceec..e003ee3ee8b 100644
--- a/src/Parsers/Kusto/ParserKQLQuery.h
+++ b/src/Parsers/Kusto/ParserKQLQuery.h
@@ -9,11 +9,11 @@ class ParserKQLBase : public IParserBase
 {
 public:
     static String getExprFromToken(Pos & pos);
-    static String getExprFromToken(const String & text, uint32_t max_depth);
+    static String getExprFromToken(const String & text, uint32_t max_depth, uint32_t max_backtracks);
     static String getExprFromPipe(Pos & pos);
     static bool setSubQuerySource(ASTPtr & select_query, ASTPtr & source, bool dest_is_subquery, bool src_is_subquery);
-    static bool parseSQLQueryByString(ParserPtr && parser, String & query, ASTPtr & select_node, int32_t max_depth);
-    bool parseByString(String expr, ASTPtr & node, uint32_t max_depth);
+    static bool parseSQLQueryByString(ParserPtr && parser, String & query, ASTPtr & select_node, uint32_t max_depth, uint32_t max_backtracks);
+    bool parseByString(String expr, ASTPtr & node, uint32_t max_depth, uint32_t max_backtracks);
 };
 
 class ParserKQLQuery : public IParserBase
diff --git a/src/Parsers/Kusto/ParserKQLSort.cpp b/src/Parsers/Kusto/ParserKQLSort.cpp
index 7e5ac2b17e7..852ba50698d 100644
--- a/src/Parsers/Kusto/ParserKQLSort.cpp
+++ b/src/Parsers/Kusto/ParserKQLSort.cpp
@@ -19,7 +19,7 @@ bool ParserKQLSort::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
     auto expr = getExprFromToken(pos);
 
     Tokens tokens(expr.c_str(), expr.c_str() + expr.size());
-    IParser::Pos new_pos(tokens, pos.max_depth);
+    IParser::Pos new_pos(tokens, pos.max_depth, pos.max_backtracks);
 
     auto pos_backup = new_pos;
     if (!order_list.parse(pos_backup, order_expression_list, expected))
diff --git a/src/Parsers/Kusto/ParserKQLStatement.cpp b/src/Parsers/Kusto/ParserKQLStatement.cpp
index 668696fa9dc..fbf2110e664 100644
--- a/src/Parsers/Kusto/ParserKQLStatement.cpp
+++ b/src/Parsers/Kusto/ParserKQLStatement.cpp
@@ -95,7 +95,7 @@ bool ParserKQLTableFunction::parseImpl(Pos & pos, ASTPtr & node, Expected & expe
         }
         ++pos;
         Tokens token_kql(kql_statement.c_str(), kql_statement.c_str() + kql_statement.size());
-        IParser::Pos pos_kql(token_kql, pos.max_depth);
+        IParser::Pos pos_kql(token_kql, pos.max_depth, pos.max_backtracks);
 
         if (kql_p.parse(pos_kql, select, expected))
         {
diff --git a/src/Parsers/Kusto/ParserKQLSummarize.cpp b/src/Parsers/Kusto/ParserKQLSummarize.cpp
index a45717930bb..47d706d0b4b 100644
--- a/src/Parsers/Kusto/ParserKQLSummarize.cpp
+++ b/src/Parsers/Kusto/ParserKQLSummarize.cpp
@@ -192,10 +192,10 @@ bool ParserKQLSummarize::parseImpl(Pos & pos, ASTPtr & node, Expected & expected
             expr_columns = expr_columns + "," + expr_aggregation;
     }
 
-    String converted_columns = getExprFromToken(expr_columns, pos.max_depth);
+    String converted_columns = getExprFromToken(expr_columns, pos.max_depth, pos.max_backtracks);
 
     Tokens token_converted_columns(converted_columns.c_str(), converted_columns.c_str() + converted_columns.size());
-    IParser::Pos pos_converted_columns(token_converted_columns, pos.max_depth);
+    IParser::Pos pos_converted_columns(token_converted_columns, pos.max_depth, pos.max_backtracks);
 
     if (!ParserNotEmptyExpressionList(true).parse(pos_converted_columns, select_expression_list, expected))
         return false;
@@ -204,10 +204,10 @@ bool ParserKQLSummarize::parseImpl(Pos & pos, ASTPtr & node, Expected & expected
 
     if (groupby)
     {
-        String converted_groupby = getExprFromToken(expr_groupby, pos.max_depth);
+        String converted_groupby = getExprFromToken(expr_groupby, pos.max_depth, pos.max_backtracks);
 
         Tokens token_converted_groupby(converted_groupby.c_str(), converted_groupby.c_str() + converted_groupby.size());
-        IParser::Pos postoken_converted_groupby(token_converted_groupby, pos.max_depth);
+        IParser::Pos postoken_converted_groupby(token_converted_groupby, pos.max_depth, pos.max_backtracks);
 
         if (!ParserNotEmptyExpressionList(false).parse(postoken_converted_groupby, group_expression_list, expected))
             return false;
diff --git a/src/Parsers/Kusto/parseKQLQuery.cpp b/src/Parsers/Kusto/parseKQLQuery.cpp
index bcc04ef7001..34a009873f8 100644
--- a/src/Parsers/Kusto/parseKQLQuery.cpp
+++ b/src/Parsers/Kusto/parseKQLQuery.cpp
@@ -322,12 +322,13 @@ ASTPtr tryParseKQLQuery(
     bool allow_multi_statements,
     size_t max_query_size,
     size_t max_parser_depth,
+    size_t max_parser_backtracks,
     bool skip_insignificant)
 {
     const char * query_begin = _out_query_end;
     Tokens tokens(query_begin, all_queries_end, max_query_size, skip_insignificant);
     /// NOTE: consider use UInt32 for max_parser_depth setting.
-    IParser::Pos token_iterator(tokens, static_cast<uint32_t>(max_parser_depth));
+    IParser::Pos token_iterator(tokens, static_cast<uint32_t>(max_parser_depth), static_cast<uint32_t>(max_parser_backtracks));
 
     if (token_iterator->isEnd()
         || token_iterator->type == TokenType::Semicolon)
@@ -441,10 +442,11 @@ ASTPtr parseKQLQueryAndMovePosition(
     const std::string & query_description,
     bool allow_multi_statements,
     size_t max_query_size,
-    size_t max_parser_depth)
+    size_t max_parser_depth,
+    size_t max_parser_backtracks)
 {
     std::string error_message;
-    ASTPtr res = tryParseKQLQuery(parser, pos, end, error_message, false, query_description, allow_multi_statements, max_query_size, max_parser_depth);
+    ASTPtr res = tryParseKQLQuery(parser, pos, end, error_message, false, query_description, allow_multi_statements, max_query_size, max_parser_depth, max_parser_backtracks);
 
     if (res)
         return res;
@@ -458,9 +460,10 @@ ASTPtr parseKQLQuery(
     const char * end,
     const std::string & query_description,
     size_t max_query_size,
-    size_t max_parser_depth)
+    size_t max_parser_depth,
+    size_t max_parser_backtracks)
 {
-    return parseKQLQueryAndMovePosition(parser, begin, end, query_description, false, max_query_size, max_parser_depth);
+    return parseKQLQueryAndMovePosition(parser, begin, end, query_description, false, max_query_size, max_parser_depth, max_parser_backtracks);
 }
 
 ASTPtr parseKQLQuery(
@@ -468,18 +471,20 @@ ASTPtr parseKQLQuery(
     const std::string & query,
     const std::string & query_description,
     size_t max_query_size,
-    size_t max_parser_depth)
+    size_t max_parser_depth,
+    size_t max_parser_backtracks)
 {
-    return parseKQLQuery(parser, query.data(), query.data() + query.size(), query_description, max_query_size, max_parser_depth);
+    return parseKQLQuery(parser, query.data(), query.data() + query.size(), query_description, max_query_size, max_parser_depth, max_parser_backtracks);
 }
 
 ASTPtr parseKQLQuery(
     IParser & parser,
     const std::string & query,
     size_t max_query_size,
-    size_t max_parser_depth)
+    size_t max_parser_depth,
+    size_t max_parser_backtracks)
 {
-    return parseKQLQuery(parser, query.data(), query.data() + query.size(), parser.getName(), max_query_size, max_parser_depth);
+    return parseKQLQuery(parser, query.data(), query.data() + query.size(), parser.getName(), max_query_size, max_parser_depth, max_parser_backtracks);
 }
 
 }
diff --git a/src/Parsers/Kusto/parseKQLQuery.h b/src/Parsers/Kusto/parseKQLQuery.h
index fca017e70fe..9e52ba56307 100644
--- a/src/Parsers/Kusto/parseKQLQuery.h
+++ b/src/Parsers/Kusto/parseKQLQuery.h
@@ -3,6 +3,7 @@
 #include <Parsers/IAST_fwd.h>
 #include <Parsers/parseQuery.h>
 #include <IO/WriteBufferFromString.h>
+
 namespace DB
 {
 
@@ -10,10 +11,6 @@ namespace DB
   * Used in syntax error message.
   */
 
-}
-namespace DB
-{
-
 class IParser;
 
 /// Parse query or set 'out_error_message'.
@@ -24,11 +21,11 @@ ASTPtr tryParseKQLQuery(
     std::string & out_error_message,
     bool hilite,
     const std::string & description,
-    bool allow_multi_statements,    /// If false, check for non-space characters after semicolon and set error message if any.
-    size_t max_query_size,          /// If (end - pos) > max_query_size and query is longer than max_query_size then throws "Max query size exceeded".
-                                    /// Disabled if zero. Is used in order to check query size if buffer can contains data for INSERT query.
+    bool allow_multi_statements,
+    size_t max_query_size,
     size_t max_parser_depth,
-    bool skip_insignificant = true);  /// If true, lexer will skip all insignificant tokens (e.g. whitespaces)
+    size_t max_parser_backtracks,
+    bool skip_insignificant = true);
 
 
 /// Parse query or throw an exception with error message.
@@ -39,7 +36,8 @@ ASTPtr parseKQLQueryAndMovePosition(
     const std::string & description,
     bool allow_multi_statements,
     size_t max_query_size,
-    size_t max_parser_depth);
+    size_t max_parser_depth,
+    size_t max_parser_backtracks);
 
 ASTPtr parseKQLQuery(
     IParser & parser,
@@ -47,18 +45,22 @@ ASTPtr parseKQLQuery(
     const char * end,
     const std::string & description,
     size_t max_query_size,
-    size_t max_parser_depth);
+    size_t max_parser_depth,
+    size_t max_parser_backtracks);
 
 ASTPtr parseKQLQuery(
     IParser & parser,
     const std::string & query,
     const std::string & query_description,
     size_t max_query_size,
-    size_t max_parser_depth);
+    size_t max_parser_depth,
+    size_t max_parser_backtracks);
 
 ASTPtr parseKQLQuery(
     IParser & parser,
     const std::string & query,
     size_t max_query_size,
-    size_t max_parser_depth);
+    size_t max_parser_depth,
+    size_t max_parser_backtracks);
+
 }
diff --git a/src/Parsers/MySQL/tests/gtest_alter_command_parser.cpp b/src/Parsers/MySQL/tests/gtest_alter_command_parser.cpp
index d406cdbd3b9..4db96646e16 100644
--- a/src/Parsers/MySQL/tests/gtest_alter_command_parser.cpp
+++ b/src/Parsers/MySQL/tests/gtest_alter_command_parser.cpp
@@ -11,7 +11,7 @@ using namespace DB::MySQLParser;
 
 static inline ASTPtr tryParserQuery(IParser & parser, const String & query)
 {
-    return parseQuery(parser, query.data(), query.data() + query.size(), "", 0, 0);
+    return parseQuery(parser, query.data(), query.data() + query.size(), "", 0, 0, 0);
 }
 
 TEST(ParserAlterCommand, AddAlterCommand)
diff --git a/src/Parsers/MySQL/tests/gtest_alter_parser.cpp b/src/Parsers/MySQL/tests/gtest_alter_parser.cpp
index 4ebbe332710..2b12d7bdcf1 100644
--- a/src/Parsers/MySQL/tests/gtest_alter_parser.cpp
+++ b/src/Parsers/MySQL/tests/gtest_alter_parser.cpp
@@ -9,7 +9,7 @@ using namespace DB::MySQLParser;
 
 static inline ASTPtr tryParserQuery(IParser & parser, const String & query)
 {
-    return parseQuery(parser, query.data(), query.data() + query.size(), "", 0, 0);
+    return parseQuery(parser, query.data(), query.data() + query.size(), "", 0, 0, 0);
 }
 
 TEST(ParserAlterQuery, AlterQuery)
diff --git a/src/Parsers/MySQL/tests/gtest_column_parser.cpp b/src/Parsers/MySQL/tests/gtest_column_parser.cpp
index b1c7c778bea..21c37e4ee2e 100644
--- a/src/Parsers/MySQL/tests/gtest_column_parser.cpp
+++ b/src/Parsers/MySQL/tests/gtest_column_parser.cpp
@@ -17,7 +17,7 @@ TEST(ParserColumn, AllNonGeneratedColumnOption)
 
     String input = "col_01 VARCHAR(100) NOT NULL DEFAULT NULL AUTO_INCREMENT UNIQUE KEY PRIMARY KEY COMMENT 'column comment' COLLATE utf8 "
                    "COLUMN_FORMAT FIXED STORAGE MEMORY REFERENCES tbl_name (col_01) CHECK 1";
-    ASTPtr ast = parseQuery(p_column, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(p_column, input.data(), input.data() + input.size(), "", 0, 0, 0);
     EXPECT_EQ(ast->as<ASTDeclareColumn>()->name, "col_01");
     EXPECT_EQ(ast->as<ASTDeclareColumn>()->data_type->as<ASTFunction>()->name, "VARCHAR");
     EXPECT_EQ(ast->as<ASTDeclareColumn>()->data_type->as<ASTFunction>()->arguments->children[0]->as<ASTLiteral>()->value.safeGet<UInt64>(), 100);
@@ -42,7 +42,7 @@ TEST(ParserColumn, AllGeneratedColumnOption)
 
     String input = "col_01 VARCHAR(100) NULL UNIQUE KEY PRIMARY KEY COMMENT 'column comment' COLLATE utf8 "
                    "REFERENCES tbl_name (col_01) CHECK 1 GENERATED ALWAYS AS (1) STORED";
-    ASTPtr ast = parseQuery(p_column, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(p_column, input.data(), input.data() + input.size(), "", 0, 0, 0);
     EXPECT_EQ(ast->as<ASTDeclareColumn>()->name, "col_01");
     EXPECT_EQ(ast->as<ASTDeclareColumn>()->data_type->as<ASTFunction>()->name, "VARCHAR");
     EXPECT_EQ(ast->as<ASTDeclareColumn>()->data_type->as<ASTFunction>()->arguments->children[0]->as<ASTLiteral>()->value.safeGet<UInt64>(), 100);
diff --git a/src/Parsers/MySQL/tests/gtest_constraint_parser.cpp b/src/Parsers/MySQL/tests/gtest_constraint_parser.cpp
index 9c9124c9f58..a06f2ade24a 100644
--- a/src/Parsers/MySQL/tests/gtest_constraint_parser.cpp
+++ b/src/Parsers/MySQL/tests/gtest_constraint_parser.cpp
@@ -14,7 +14,7 @@ TEST(ParserConstraint, CheckConstraint)
     ParserDeclareConstraint p_constraint;
 
     String constraint_01 = "CONSTRAINT symbol_name CHECK col_01 = 1";
-    ASTPtr ast_constraint_01 = parseQuery(p_constraint, constraint_01.data(), constraint_01.data() + constraint_01.size(), "", 0, 0);
+    ASTPtr ast_constraint_01 = parseQuery(p_constraint, constraint_01.data(), constraint_01.data() + constraint_01.size(), "", 0, 0, 0);
     EXPECT_EQ(ast_constraint_01->as<ASTDeclareConstraint>()->constraint_name, "symbol_name");
     auto * check_expression_01 = ast_constraint_01->as<ASTDeclareConstraint>()->check_expression->as<ASTFunction>();
     EXPECT_EQ(check_expression_01->name, "equals");
@@ -22,7 +22,7 @@ TEST(ParserConstraint, CheckConstraint)
     EXPECT_EQ(check_expression_01->arguments->children[1]->as<ASTLiteral>()->value.safeGet<UInt64>(), 1);
 
     String constraint_02 = "CONSTRAINT CHECK col_01 = 1";
-    ASTPtr ast_constraint_02 = parseQuery(p_constraint, constraint_02.data(), constraint_02.data() + constraint_02.size(), "", 0, 0);
+    ASTPtr ast_constraint_02 = parseQuery(p_constraint, constraint_02.data(), constraint_02.data() + constraint_02.size(), "", 0, 0, 0);
     EXPECT_EQ(ast_constraint_02->as<ASTDeclareConstraint>()->constraint_name, "");
     auto * check_expression_02 = ast_constraint_02->as<ASTDeclareConstraint>()->check_expression->as<ASTFunction>();
     EXPECT_EQ(check_expression_02->name, "equals");
@@ -30,7 +30,7 @@ TEST(ParserConstraint, CheckConstraint)
     EXPECT_EQ(check_expression_02->arguments->children[1]->as<ASTLiteral>()->value.safeGet<UInt64>(), 1);
 
     String constraint_03 = "CHECK col_01 = 1";
-    ASTPtr ast_constraint_03 = parseQuery(p_constraint, constraint_03.data(), constraint_03.data() + constraint_03.size(), "", 0, 0);
+    ASTPtr ast_constraint_03 = parseQuery(p_constraint, constraint_03.data(), constraint_03.data() + constraint_03.size(), "", 0, 0, 0);
     EXPECT_EQ(ast_constraint_03->as<ASTDeclareConstraint>()->constraint_name, "");
     auto * check_expression_03 = ast_constraint_03->as<ASTDeclareConstraint>()->check_expression->as<ASTFunction>();
     EXPECT_EQ(check_expression_03->name, "equals");
@@ -38,7 +38,7 @@ TEST(ParserConstraint, CheckConstraint)
     EXPECT_EQ(check_expression_03->arguments->children[1]->as<ASTLiteral>()->value.safeGet<UInt64>(), 1);
 
     String constraint_04 = "CONSTRAINT CHECK col_01 = 1 ENFORCED";
-    ASTPtr ast_constraint_04 = parseQuery(p_constraint, constraint_04.data(), constraint_04.data() + constraint_04.size(), "", 0, 0);
+    ASTPtr ast_constraint_04 = parseQuery(p_constraint, constraint_04.data(), constraint_04.data() + constraint_04.size(), "", 0, 0, 0);
     EXPECT_TRUE(ast_constraint_04->as<ASTDeclareConstraint>()->enforced);
     EXPECT_EQ(ast_constraint_04->as<ASTDeclareConstraint>()->constraint_name, "");
     auto * check_expression_04 = ast_constraint_04->as<ASTDeclareConstraint>()->check_expression->as<ASTFunction>();
@@ -47,7 +47,7 @@ TEST(ParserConstraint, CheckConstraint)
     EXPECT_EQ(check_expression_04->arguments->children[1]->as<ASTLiteral>()->value.safeGet<UInt64>(), 1);
 
     String constraint_05 = "CONSTRAINT CHECK col_01 = 1 NOT ENFORCED";
-    ASTPtr ast_constraint_05 = parseQuery(p_constraint, constraint_05.data(), constraint_05.data() + constraint_05.size(), "", 0, 0);
+    ASTPtr ast_constraint_05 = parseQuery(p_constraint, constraint_05.data(), constraint_05.data() + constraint_05.size(), "", 0, 0, 0);
     EXPECT_FALSE(ast_constraint_05->as<ASTDeclareConstraint>()->enforced);
     EXPECT_EQ(ast_constraint_05->as<ASTDeclareConstraint>()->constraint_name, "");
     auto * check_expression_05 = ast_constraint_05->as<ASTDeclareConstraint>()->check_expression->as<ASTFunction>();
diff --git a/src/Parsers/MySQL/tests/gtest_create_parser.cpp b/src/Parsers/MySQL/tests/gtest_create_parser.cpp
index 2f65eb6e592..8512b88ffc1 100644
--- a/src/Parsers/MySQL/tests/gtest_create_parser.cpp
+++ b/src/Parsers/MySQL/tests/gtest_create_parser.cpp
@@ -15,16 +15,16 @@ TEST(CreateTableParser, LikeCreate)
 {
     ParserCreateQuery p_create_query;
     String like_create_01 = "CREATE TABLE IF NOT EXISTS table_name LIKE table_name_01";
-    parseQuery(p_create_query, like_create_01.data(), like_create_01.data() + like_create_01.size(), "", 0, 0);
+    parseQuery(p_create_query, like_create_01.data(), like_create_01.data() + like_create_01.size(), "", 0, 0, 0);
     String like_create_02 = "CREATE TABLE IF NOT EXISTS table_name (LIKE table_name_01)";
-    parseQuery(p_create_query, like_create_02.data(), like_create_02.data() + like_create_02.size(), "", 0, 0);
+    parseQuery(p_create_query, like_create_02.data(), like_create_02.data() + like_create_02.size(), "", 0, 0, 0);
 }
 
 TEST(CreateTableParser, SimpleCreate)
 {
     ParserCreateQuery p_create_query;
     String input = "CREATE TABLE IF NOT EXISTS table_name(col_01 VARCHAR(100), INDEX (col_01), CHECK 1) ENGINE INNODB PARTITION BY HASH(col_01)";
-    ASTPtr ast = parseQuery(p_create_query, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(p_create_query, input.data(), input.data() + input.size(), "", 0, 0, 0);
     EXPECT_TRUE(ast->as<ASTCreateQuery>()->if_not_exists);
     EXPECT_EQ(ast->as<ASTCreateQuery>()->columns_list->as<ASTCreateDefines>()->columns->children.size(), 1);
     EXPECT_EQ(ast->as<ASTCreateQuery>()->columns_list->as<ASTCreateDefines>()->indices->children.size(), 1);
@@ -37,7 +37,7 @@ TEST(CreateTableParser, SS)
 {
     ParserCreateQuery p_create_query;
     String input = "CREATE TABLE `test_table_1` (`a` int DEFAULT NULL, `b` int DEFAULT NULL) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_0900_ai_ci";
-    ASTPtr ast = parseQuery(p_create_query, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(p_create_query, input.data(), input.data() + input.size(), "", 0, 0, 0);
     WriteBufferFromOStream buf(std::cerr, 4096);
     ast->dumpTree(buf);
     buf.finalize();
diff --git a/src/Parsers/MySQL/tests/gtest_index_parser.cpp b/src/Parsers/MySQL/tests/gtest_index_parser.cpp
index a8be6787b2c..187bac3e090 100644
--- a/src/Parsers/MySQL/tests/gtest_index_parser.cpp
+++ b/src/Parsers/MySQL/tests/gtest_index_parser.cpp
@@ -14,7 +14,7 @@ TEST(ParserIndex, AllIndexOptions)
     String input = "INDEX (col_01, col_02(100), col_03 DESC) KEY_BLOCK_SIZE 3 USING HASH WITH PARSER parser_name COMMENT 'index comment' VISIBLE";
 
     ParserDeclareIndex p_index;
-    ASTPtr ast = parseQuery(p_index, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(p_index, input.data(), input.data() + input.size(), "", 0, 0, 0);
 
     ASTDeclareIndex * declare_index = ast->as<ASTDeclareIndex>();
     EXPECT_EQ(declare_index->index_columns->children[0]->as<ASTIdentifier>()->name(), "col_01");
@@ -33,7 +33,7 @@ TEST(ParserIndex, OptionalIndexOptions)
     String input = "INDEX (col_01, col_02(100), col_03 DESC) USING HASH INVISIBLE KEY_BLOCK_SIZE 3";
 
     ParserDeclareIndex p_index;
-    ASTPtr ast = parseQuery(p_index, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(p_index, input.data(), input.data() + input.size(), "", 0, 0, 0);
 
     ASTDeclareIndex * declare_index = ast->as<ASTDeclareIndex>();
     EXPECT_EQ(declare_index->index_columns->children[0]->as<ASTIdentifier>()->name(), "col_01");
@@ -50,28 +50,28 @@ TEST(ParserIndex, OrdinaryIndex)
 {
     ParserDeclareIndex p_index;
     String non_unique_index_01 = "KEY index_name USING HASH (col_01) INVISIBLE";
-    parseQuery(p_index, non_unique_index_01.data(), non_unique_index_01.data() + non_unique_index_01.size(), "", 0, 0);
+    parseQuery(p_index, non_unique_index_01.data(), non_unique_index_01.data() + non_unique_index_01.size(), "", 0, 0, 0);
 
     String non_unique_index_02 = "INDEX index_name USING HASH (col_01) INVISIBLE";
-    parseQuery(p_index, non_unique_index_02.data(), non_unique_index_02.data() + non_unique_index_02.size(), "", 0, 0);
+    parseQuery(p_index, non_unique_index_02.data(), non_unique_index_02.data() + non_unique_index_02.size(), "", 0, 0, 0);
 
     String fulltext_index_01 = "FULLTEXT index_name (col_01) INVISIBLE";
-    parseQuery(p_index, fulltext_index_01.data(), fulltext_index_01.data() + fulltext_index_01.size(), "", 0, 0);
+    parseQuery(p_index, fulltext_index_01.data(), fulltext_index_01.data() + fulltext_index_01.size(), "", 0, 0, 0);
 
     String fulltext_index_02 = "FULLTEXT INDEX index_name (col_01) INVISIBLE";
-    parseQuery(p_index, fulltext_index_02.data(), fulltext_index_02.data() + fulltext_index_02.size(), "", 0, 0);
+    parseQuery(p_index, fulltext_index_02.data(), fulltext_index_02.data() + fulltext_index_02.size(), "", 0, 0, 0);
 
     String fulltext_index_03 = "FULLTEXT KEY index_name (col_01) INVISIBLE";
-    parseQuery(p_index, fulltext_index_03.data(), fulltext_index_03.data() + fulltext_index_03.size(), "", 0, 0);
+    parseQuery(p_index, fulltext_index_03.data(), fulltext_index_03.data() + fulltext_index_03.size(), "", 0, 0, 0);
 
     String spatial_index_01 = "SPATIAL index_name (col_01) INVISIBLE";
-    parseQuery(p_index, spatial_index_01.data(), spatial_index_01.data() + spatial_index_01.size(), "", 0, 0);
+    parseQuery(p_index, spatial_index_01.data(), spatial_index_01.data() + spatial_index_01.size(), "", 0, 0, 0);
 
     String spatial_index_02 = "SPATIAL INDEX index_name (col_01) INVISIBLE";
-    parseQuery(p_index, spatial_index_02.data(), spatial_index_02.data() + spatial_index_02.size(), "", 0, 0);
+    parseQuery(p_index, spatial_index_02.data(), spatial_index_02.data() + spatial_index_02.size(), "", 0, 0, 0);
 
     String spatial_index_03 = "SPATIAL KEY index_name (col_01) INVISIBLE";
-    parseQuery(p_index, spatial_index_03.data(), spatial_index_03.data() + spatial_index_03.size(), "", 0, 0);
+    parseQuery(p_index, spatial_index_03.data(), spatial_index_03.data() + spatial_index_03.size(), "", 0, 0, 0);
 }
 
 TEST(ParserIndex, ConstraintIndex)
@@ -79,47 +79,47 @@ TEST(ParserIndex, ConstraintIndex)
     ParserDeclareIndex p_index;
 
     String primary_key_01 = "PRIMARY KEY (col_01) INVISIBLE";
-    parseQuery(p_index, primary_key_01.data(), primary_key_01.data() + primary_key_01.size(), "", 0, 0);
+    parseQuery(p_index, primary_key_01.data(), primary_key_01.data() + primary_key_01.size(), "", 0, 0, 0);
 
     String primary_key_02 = "PRIMARY KEY USING BTREE (col_01) INVISIBLE";
-    parseQuery(p_index, primary_key_02.data(), primary_key_02.data() + primary_key_02.size(), "", 0, 0);
+    parseQuery(p_index, primary_key_02.data(), primary_key_02.data() + primary_key_02.size(), "", 0, 0, 0);
 
     String primary_key_03 = "CONSTRAINT PRIMARY KEY USING BTREE (col_01) INVISIBLE";
-    parseQuery(p_index, primary_key_03.data(), primary_key_03.data() + primary_key_03.size(), "", 0, 0);
+    parseQuery(p_index, primary_key_03.data(), primary_key_03.data() + primary_key_03.size(), "", 0, 0, 0);
 
     String primary_key_04 = "CONSTRAINT index_name PRIMARY KEY USING BTREE (col_01) INVISIBLE";
-    parseQuery(p_index, primary_key_04.data(), primary_key_04.data() + primary_key_04.size(), "", 0, 0);
+    parseQuery(p_index, primary_key_04.data(), primary_key_04.data() + primary_key_04.size(), "", 0, 0, 0);
 
     String unique_key_01 = "UNIQUE (col_01) INVISIBLE";
-    parseQuery(p_index, unique_key_01.data(), unique_key_01.data() + unique_key_01.size(), "", 0, 0);
+    parseQuery(p_index, unique_key_01.data(), unique_key_01.data() + unique_key_01.size(), "", 0, 0, 0);
 
     String unique_key_02 = "UNIQUE INDEX (col_01) INVISIBLE";
-    parseQuery(p_index, unique_key_02.data(), unique_key_02.data() + unique_key_02.size(), "", 0, 0);
+    parseQuery(p_index, unique_key_02.data(), unique_key_02.data() + unique_key_02.size(), "", 0, 0, 0);
 
     String unique_key_03 = "UNIQUE KEY (col_01) INVISIBLE";
-    parseQuery(p_index, unique_key_03.data(), unique_key_03.data() + unique_key_03.size(), "", 0, 0);
+    parseQuery(p_index, unique_key_03.data(), unique_key_03.data() + unique_key_03.size(), "", 0, 0, 0);
 
     String unique_key_04 = "UNIQUE KEY index_name (col_01) INVISIBLE";
-    parseQuery(p_index, unique_key_04.data(), unique_key_04.data() + unique_key_04.size(), "", 0, 0);
+    parseQuery(p_index, unique_key_04.data(), unique_key_04.data() + unique_key_04.size(), "", 0, 0, 0);
 
     String unique_key_05 = "UNIQUE KEY index_name USING HASH (col_01) INVISIBLE";
-    parseQuery(p_index, unique_key_05.data(), unique_key_05.data() + unique_key_05.size(), "", 0, 0);
+    parseQuery(p_index, unique_key_05.data(), unique_key_05.data() + unique_key_05.size(), "", 0, 0, 0);
 
     String unique_key_06 = "CONSTRAINT UNIQUE KEY index_name USING HASH (col_01) INVISIBLE";
-    parseQuery(p_index, unique_key_06.data(), unique_key_06.data() + unique_key_06.size(), "", 0, 0);
+    parseQuery(p_index, unique_key_06.data(), unique_key_06.data() + unique_key_06.size(), "", 0, 0, 0);
 
     String unique_key_07 = "CONSTRAINT index_name UNIQUE KEY index_name_1 USING HASH (col_01) INVISIBLE";
-    parseQuery(p_index, unique_key_07.data(), unique_key_07.data() + unique_key_07.size(), "", 0, 0);
+    parseQuery(p_index, unique_key_07.data(), unique_key_07.data() + unique_key_07.size(), "", 0, 0, 0);
 
     String foreign_key_01 = "FOREIGN KEY (col_01) REFERENCES tbl_name (col_01)";
-    parseQuery(p_index, foreign_key_01.data(), foreign_key_01.data() + foreign_key_01.size(), "", 0, 0);
+    parseQuery(p_index, foreign_key_01.data(), foreign_key_01.data() + foreign_key_01.size(), "", 0, 0, 0);
 
     String foreign_key_02 = "FOREIGN KEY index_name (col_01) REFERENCES tbl_name (col_01)";
-    parseQuery(p_index, foreign_key_02.data(), foreign_key_02.data() + foreign_key_02.size(), "", 0, 0);
+    parseQuery(p_index, foreign_key_02.data(), foreign_key_02.data() + foreign_key_02.size(), "", 0, 0, 0);
 
     String foreign_key_03 = "CONSTRAINT FOREIGN KEY index_name (col_01) REFERENCES tbl_name (col_01)";
-    parseQuery(p_index, foreign_key_03.data(), foreign_key_03.data() + foreign_key_03.size(), "", 0, 0);
+    parseQuery(p_index, foreign_key_03.data(), foreign_key_03.data() + foreign_key_03.size(), "", 0, 0, 0);
 
     String foreign_key_04 = "CONSTRAINT index_name FOREIGN KEY index_name_01 (col_01) REFERENCES tbl_name (col_01)";
-    parseQuery(p_index, foreign_key_04.data(), foreign_key_04.data() + foreign_key_04.size(), "", 0, 0);
+    parseQuery(p_index, foreign_key_04.data(), foreign_key_04.data() + foreign_key_04.size(), "", 0, 0, 0);
 }
diff --git a/src/Parsers/MySQL/tests/gtest_partition_options_parser.cpp b/src/Parsers/MySQL/tests/gtest_partition_options_parser.cpp
index 01b757e5891..6ec8d73530e 100644
--- a/src/Parsers/MySQL/tests/gtest_partition_options_parser.cpp
+++ b/src/Parsers/MySQL/tests/gtest_partition_options_parser.cpp
@@ -14,14 +14,14 @@ TEST(ParserPartitionOptions, HashPatitionOptions)
     String hash_partition = "PARTITION BY HASH(col_01)";
 
     ParserDeclarePartitionOptions p_partition_options;
-    ASTPtr ast_01 = parseQuery(p_partition_options, hash_partition.data(), hash_partition.data() + hash_partition.size(), "", 0, 0);
+    ASTPtr ast_01 = parseQuery(p_partition_options, hash_partition.data(), hash_partition.data() + hash_partition.size(), "", 0, 0, 0);
 
     ASTDeclarePartitionOptions * declare_partition_options_01 = ast_01->as<ASTDeclarePartitionOptions>();
     EXPECT_EQ(declare_partition_options_01->partition_type, "hash");
     EXPECT_EQ(declare_partition_options_01->partition_expression->as<ASTIdentifier>()->name(), "col_01");
 
     String linear_hash_partition = "PARTITION BY LINEAR HASH(col_01)";
-    ASTPtr ast_02 = parseQuery(p_partition_options, linear_hash_partition.data(), linear_hash_partition.data() + linear_hash_partition.size(), "", 0, 0);
+    ASTPtr ast_02 = parseQuery(p_partition_options, linear_hash_partition.data(), linear_hash_partition.data() + linear_hash_partition.size(), "", 0, 0, 0);
 
     ASTDeclarePartitionOptions * declare_partition_options_02 = ast_02->as<ASTDeclarePartitionOptions>();
     EXPECT_EQ(declare_partition_options_02->partition_type, "linear_hash");
@@ -33,14 +33,14 @@ TEST(ParserPartitionOptions, KeyPatitionOptions)
     String key_partition = "PARTITION BY KEY(col_01)";
 
     ParserDeclarePartitionOptions p_partition_options;
-    ASTPtr ast_01 = parseQuery(p_partition_options, key_partition.data(), key_partition.data() + key_partition.size(), "", 0, 0);
+    ASTPtr ast_01 = parseQuery(p_partition_options, key_partition.data(), key_partition.data() + key_partition.size(), "", 0, 0, 0);
 
     ASTDeclarePartitionOptions * declare_partition_options_01 = ast_01->as<ASTDeclarePartitionOptions>();
     EXPECT_EQ(declare_partition_options_01->partition_type, "key");
     EXPECT_EQ(declare_partition_options_01->partition_expression->as<ASTIdentifier>()->name(), "col_01");
 
     String linear_key_partition = "PARTITION BY LINEAR KEY(col_01, col_02)";
-    ASTPtr ast_02 = parseQuery(p_partition_options, linear_key_partition.data(), linear_key_partition.data() + linear_key_partition.size(), "", 0, 0);
+    ASTPtr ast_02 = parseQuery(p_partition_options, linear_key_partition.data(), linear_key_partition.data() + linear_key_partition.size(), "", 0, 0, 0);
 
     ASTDeclarePartitionOptions * declare_partition_options_02 = ast_02->as<ASTDeclarePartitionOptions>();
     EXPECT_EQ(declare_partition_options_02->partition_type, "linear_key");
@@ -49,7 +49,7 @@ TEST(ParserPartitionOptions, KeyPatitionOptions)
     EXPECT_EQ(columns_list->children[1]->as<ASTIdentifier>()->name(), "col_02");
 
     String key_partition_with_algorithm = "PARTITION BY KEY ALGORITHM=1 (col_01)";
-    ASTPtr ast_03 = parseQuery(p_partition_options, key_partition_with_algorithm.data(), key_partition_with_algorithm.data() + key_partition_with_algorithm.size(), "", 0, 0);
+    ASTPtr ast_03 = parseQuery(p_partition_options, key_partition_with_algorithm.data(), key_partition_with_algorithm.data() + key_partition_with_algorithm.size(), "", 0, 0, 0);
 
     ASTDeclarePartitionOptions * declare_partition_options_03 = ast_03->as<ASTDeclarePartitionOptions>();
     EXPECT_EQ(declare_partition_options_03->partition_type, "key_1");
@@ -61,14 +61,14 @@ TEST(ParserPartitionOptions, RangePatitionOptions)
     String range_partition = "PARTITION BY RANGE(col_01)";
 
     ParserDeclarePartitionOptions p_partition_options;
-    ASTPtr ast_01 = parseQuery(p_partition_options, range_partition.data(), range_partition.data() + range_partition.size(), "", 0, 0);
+    ASTPtr ast_01 = parseQuery(p_partition_options, range_partition.data(), range_partition.data() + range_partition.size(), "", 0, 0, 0);
 
     ASTDeclarePartitionOptions * declare_partition_options_01 = ast_01->as<ASTDeclarePartitionOptions>();
     EXPECT_EQ(declare_partition_options_01->partition_type, "range");
     EXPECT_EQ(declare_partition_options_01->partition_expression->as<ASTIdentifier>()->name(), "col_01");
 
     String range_columns_partition = "PARTITION BY RANGE COLUMNS(col_01, col_02)";
-    ASTPtr ast_02 = parseQuery(p_partition_options, range_columns_partition.data(), range_columns_partition.data() + range_columns_partition.size(), "", 0, 0);
+    ASTPtr ast_02 = parseQuery(p_partition_options, range_columns_partition.data(), range_columns_partition.data() + range_columns_partition.size(), "", 0, 0, 0);
 
     ASTDeclarePartitionOptions * declare_partition_options_02 = ast_02->as<ASTDeclarePartitionOptions>();
     EXPECT_EQ(declare_partition_options_02->partition_type, "range");
@@ -82,14 +82,14 @@ TEST(ParserPartitionOptions, ListPatitionOptions)
     String range_partition = "PARTITION BY LIST(col_01)";
 
     ParserDeclarePartitionOptions p_partition_options;
-    ASTPtr ast_01 = parseQuery(p_partition_options, range_partition.data(), range_partition.data() + range_partition.size(), "", 0, 0);
+    ASTPtr ast_01 = parseQuery(p_partition_options, range_partition.data(), range_partition.data() + range_partition.size(), "", 0, 0, 0);
 
     ASTDeclarePartitionOptions * declare_partition_options_01 = ast_01->as<ASTDeclarePartitionOptions>();
     EXPECT_EQ(declare_partition_options_01->partition_type, "list");
     EXPECT_EQ(declare_partition_options_01->partition_expression->as<ASTIdentifier>()->name(), "col_01");
 
     String range_columns_partition = "PARTITION BY LIST COLUMNS(col_01, col_02)";
-    ASTPtr ast_02 = parseQuery(p_partition_options, range_columns_partition.data(), range_columns_partition.data() + range_columns_partition.size(), "", 0, 0);
+    ASTPtr ast_02 = parseQuery(p_partition_options, range_columns_partition.data(), range_columns_partition.data() + range_columns_partition.size(), "", 0, 0, 0);
 
     ASTDeclarePartitionOptions * declare_partition_options_02 = ast_02->as<ASTDeclarePartitionOptions>();
     EXPECT_EQ(declare_partition_options_02->partition_type, "list");
@@ -103,7 +103,7 @@ TEST(ParserPartitionOptions, PatitionNumberOptions)
     String numbers_partition = "PARTITION BY KEY(col_01) PARTITIONS 2";
 
     ParserDeclarePartitionOptions p_partition_options;
-    ASTPtr ast = parseQuery(p_partition_options, numbers_partition.data(), numbers_partition.data() + numbers_partition.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(p_partition_options, numbers_partition.data(), numbers_partition.data() + numbers_partition.size(), "", 0, 0, 0);
 
     ASTDeclarePartitionOptions * declare_partition_options = ast->as<ASTDeclarePartitionOptions>();
     EXPECT_EQ(declare_partition_options->partition_type, "key");
@@ -116,7 +116,7 @@ TEST(ParserPartitionOptions, PatitionWithSubpartitionOptions)
     String partition_with_subpartition = "PARTITION BY KEY(col_01) PARTITIONS 3 SUBPARTITION BY HASH(col_02) SUBPARTITIONS 4";
 
     ParserDeclarePartitionOptions p_partition_options;
-    ASTPtr ast = parseQuery(p_partition_options, partition_with_subpartition.data(), partition_with_subpartition.data() + partition_with_subpartition.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(p_partition_options, partition_with_subpartition.data(), partition_with_subpartition.data() + partition_with_subpartition.size(), "", 0, 0, 0);
 
     ASTDeclarePartitionOptions * declare_partition_options = ast->as<ASTDeclarePartitionOptions>();
     EXPECT_EQ(declare_partition_options->partition_type, "key");
@@ -134,7 +134,7 @@ TEST(ParserPartitionOptions, PatitionOptionsWithDeclarePartition)
     ParserDeclarePartitionOptions p_partition_options;
     ASTPtr ast = parseQuery(p_partition_options,
         partition_options_with_declare.data(),
-        partition_options_with_declare.data() + partition_options_with_declare.size(), "", 0, 0);
+        partition_options_with_declare.data() + partition_options_with_declare.size(), "", 0, 0, 0);
 
     ASTDeclarePartitionOptions * declare_partition_options = ast->as<ASTDeclarePartitionOptions>();
     EXPECT_EQ(declare_partition_options->partition_type, "key");
@@ -153,7 +153,7 @@ TEST(ParserPartitionOptions, PatitionOptionsWithDeclarePartitions)
     ParserDeclarePartitionOptions p_partition_options;
     ASTPtr ast = parseQuery(p_partition_options,
                             partition_options_with_declare.data(),
-                            partition_options_with_declare.data() + partition_options_with_declare.size(), "", 0, 0);
+                            partition_options_with_declare.data() + partition_options_with_declare.size(), "", 0, 0, 0);
 
     ASTDeclarePartitionOptions * declare_partition_options = ast->as<ASTDeclarePartitionOptions>();
     EXPECT_EQ(declare_partition_options->partition_type, "key");
diff --git a/src/Parsers/MySQL/tests/gtest_partition_parser.cpp b/src/Parsers/MySQL/tests/gtest_partition_parser.cpp
index 458c7acd553..07c7c03dbb7 100644
--- a/src/Parsers/MySQL/tests/gtest_partition_parser.cpp
+++ b/src/Parsers/MySQL/tests/gtest_partition_parser.cpp
@@ -17,7 +17,7 @@ TEST(ParserPartition, AllPatitionOptions)
                    " TABLESPACE table_space_name";
 
     ParserDeclarePartition p_partition;
-    ASTPtr ast = parseQuery(p_partition, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(p_partition, input.data(), input.data() + input.size(), "", 0, 0, 0);
 
     ASTDeclarePartition * declare_partition = ast->as<ASTDeclarePartition>();
     EXPECT_EQ(declare_partition->partition_name, "partition_name");
@@ -35,7 +35,7 @@ TEST(ParserPartition, OptionalPatitionOptions)
 {
     String input = "PARTITION partition_name STORAGE engine = engine_name max_rows 1000 min_rows 0 tablespace table_space_name";
     ParserDeclarePartition p_partition;
-    ASTPtr ast = parseQuery(p_partition, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(p_partition, input.data(), input.data() + input.size(), "", 0, 0, 0);
 
     ASTDeclarePartition * declare_partition = ast->as<ASTDeclarePartition>();
     EXPECT_EQ(declare_partition->partition_name, "partition_name");
@@ -50,7 +50,7 @@ TEST(ParserPartition, PatitionOptionsWithLessThan)
 {
     ParserDeclarePartition p_partition;
     String partition_01 = "PARTITION partition_01 VALUES LESS THAN (1991) STORAGE engine = engine_name";
-    ASTPtr ast_partition_01 = parseQuery(p_partition, partition_01.data(), partition_01.data() + partition_01.size(), "", 0, 0);
+    ASTPtr ast_partition_01 = parseQuery(p_partition, partition_01.data(), partition_01.data() + partition_01.size(), "", 0, 0, 0);
 
     ASTDeclarePartition * declare_partition_01 = ast_partition_01->as<ASTDeclarePartition>();
     EXPECT_EQ(declare_partition_01->partition_name, "partition_01");
@@ -59,7 +59,7 @@ TEST(ParserPartition, PatitionOptionsWithLessThan)
     EXPECT_EQ(declare_options_01->changes["engine"]->as<ASTIdentifier>()->name(), "engine_name");
 
     String partition_02 = "PARTITION partition_02 VALUES LESS THAN MAXVALUE STORAGE engine = engine_name";
-    ASTPtr ast_partition_02 = parseQuery(p_partition, partition_02.data(), partition_02.data() + partition_02.size(), "", 0, 0);
+    ASTPtr ast_partition_02 = parseQuery(p_partition, partition_02.data(), partition_02.data() + partition_02.size(), "", 0, 0, 0);
 
     ASTDeclarePartition * declare_partition_02 = ast_partition_02->as<ASTDeclarePartition>();
     EXPECT_EQ(declare_partition_02->partition_name, "partition_02");
@@ -68,7 +68,7 @@ TEST(ParserPartition, PatitionOptionsWithLessThan)
     EXPECT_EQ(declare_options_02->changes["engine"]->as<ASTIdentifier>()->name(), "engine_name");
 
     String partition_03 = "PARTITION partition_03 VALUES LESS THAN (50, MAXVALUE) STORAGE engine = engine_name";
-    ASTPtr ast_partition_03 = parseQuery(p_partition, partition_03.data(), partition_03.data() + partition_03.size(), "", 0, 0);
+    ASTPtr ast_partition_03 = parseQuery(p_partition, partition_03.data(), partition_03.data() + partition_03.size(), "", 0, 0, 0);
 
     ASTDeclarePartition * declare_partition_03 = ast_partition_03->as<ASTDeclarePartition>();
     EXPECT_EQ(declare_partition_03->partition_name, "partition_03");
@@ -79,7 +79,7 @@ TEST(ParserPartition, PatitionOptionsWithLessThan)
     EXPECT_EQ(declare_options_03->changes["engine"]->as<ASTIdentifier>()->name(), "engine_name");
 
     String partition_04 = "PARTITION partition_04 VALUES LESS THAN (MAXVALUE, MAXVALUE) STORAGE engine = engine_name";
-    ASTPtr ast_partition_04 = parseQuery(p_partition, partition_04.data(), partition_04.data() + partition_04.size(), "", 0, 0);
+    ASTPtr ast_partition_04 = parseQuery(p_partition, partition_04.data(), partition_04.data() + partition_04.size(), "", 0, 0, 0);
 
     ASTDeclarePartition * declare_partition_04 = ast_partition_04->as<ASTDeclarePartition>();
     EXPECT_EQ(declare_partition_04->partition_name, "partition_04");
@@ -94,7 +94,7 @@ TEST(ParserPartition, PatitionOptionsWithInExpression)
 {
     ParserDeclarePartition p_partition;
     String partition_01 = "PARTITION partition_01 VALUES IN (NULL, 1991, MAXVALUE) STORAGE engine = engine_name";
-    ASTPtr ast_partition_01 = parseQuery(p_partition, partition_01.data(), partition_01.data() + partition_01.size(), "", 0, 0);
+    ASTPtr ast_partition_01 = parseQuery(p_partition, partition_01.data(), partition_01.data() + partition_01.size(), "", 0, 0, 0);
 
     ASTDeclarePartition * declare_partition_01 = ast_partition_01->as<ASTDeclarePartition>();
     EXPECT_EQ(declare_partition_01->partition_name, "partition_01");
@@ -106,7 +106,7 @@ TEST(ParserPartition, PatitionOptionsWithInExpression)
     EXPECT_EQ(declare_options_01->changes["engine"]->as<ASTIdentifier>()->name(), "engine_name");
 
     String partition_02 = "PARTITION partition_02 VALUES IN ((NULL, 1991), (1991, NULL), (MAXVALUE, MAXVALUE)) STORAGE engine = engine_name";
-    ASTPtr ast_partition_02 = parseQuery(p_partition, partition_02.data(), partition_02.data() + partition_02.size(), "", 0, 0);
+    ASTPtr ast_partition_02 = parseQuery(p_partition, partition_02.data(), partition_02.data() + partition_02.size(), "", 0, 0, 0);
 
     ASTDeclarePartition * declare_partition_02 = ast_partition_02->as<ASTDeclarePartition>();
     EXPECT_EQ(declare_partition_02->partition_name, "partition_02");
@@ -132,18 +132,17 @@ TEST(ParserPartition, PatitionOptionsWithSubpartitions)
 {
     ParserDeclarePartition p_partition;
     String partition_01 = "PARTITION partition_01 VALUES IN (NULL, 1991, MAXVALUE) STORAGE engine = engine_name (SUBPARTITION s_p01)";
-    ASTPtr ast_partition_01 = parseQuery(p_partition, partition_01.data(), partition_01.data() + partition_01.size(), "", 0, 0);
+    ASTPtr ast_partition_01 = parseQuery(p_partition, partition_01.data(), partition_01.data() + partition_01.size(), "", 0, 0, 0);
 
     ASTDeclarePartition * declare_partition_01 = ast_partition_01->as<ASTDeclarePartition>();
     EXPECT_EQ(declare_partition_01->partition_name, "partition_01");
     EXPECT_TRUE(declare_partition_01->subpartitions->as<ASTExpressionList>()->children[0]->as<ASTDeclareSubPartition>());
 
     String partition_02 = "PARTITION partition_02 VALUES IN (NULL, 1991, MAXVALUE) STORAGE engine = engine_name (SUBPARTITION s_p01, SUBPARTITION s_p02)";
-    ASTPtr ast_partition_02 = parseQuery(p_partition, partition_02.data(), partition_02.data() + partition_02.size(), "", 0, 0);
+    ASTPtr ast_partition_02 = parseQuery(p_partition, partition_02.data(), partition_02.data() + partition_02.size(), "", 0, 0, 0);
 
     ASTDeclarePartition * declare_partition_02 = ast_partition_02->as<ASTDeclarePartition>();
     EXPECT_EQ(declare_partition_02->partition_name, "partition_02");
     EXPECT_TRUE(declare_partition_02->subpartitions->as<ASTExpressionList>()->children[0]->as<ASTDeclareSubPartition>());
     EXPECT_TRUE(declare_partition_02->subpartitions->as<ASTExpressionList>()->children[1]->as<ASTDeclareSubPartition>());
 }
-
diff --git a/src/Parsers/MySQL/tests/gtest_reference_parser.cpp b/src/Parsers/MySQL/tests/gtest_reference_parser.cpp
index 7447f16fc7c..d5b3c9b596d 100644
--- a/src/Parsers/MySQL/tests/gtest_reference_parser.cpp
+++ b/src/Parsers/MySQL/tests/gtest_reference_parser.cpp
@@ -12,12 +12,12 @@ TEST(ParserReference, SimpleReference)
     ParserDeclareReference p_reference;
 
     String reference_01 = "REFERENCES table_name (ref_col_01)";
-    ASTPtr ast_reference_01 = parseQuery(p_reference, reference_01.data(), reference_01.data() + reference_01.size(), "", 0, 0);
+    ASTPtr ast_reference_01 = parseQuery(p_reference, reference_01.data(), reference_01.data() + reference_01.size(), "", 0, 0, 0);
     EXPECT_EQ(ast_reference_01->as<ASTDeclareReference>()->reference_table_name, "table_name");
     EXPECT_EQ(ast_reference_01->as<ASTDeclareReference>()->reference_expression->as<ASTIdentifier>()->name(), "ref_col_01");
 
     String reference_02 = "REFERENCES table_name (ref_col_01, ref_col_02)";
-    ASTPtr ast_reference_02 = parseQuery(p_reference, reference_02.data(), reference_02.data() + reference_02.size(), "", 0, 0);
+    ASTPtr ast_reference_02 = parseQuery(p_reference, reference_02.data(), reference_02.data() + reference_02.size(), "", 0, 0, 0);
     EXPECT_EQ(ast_reference_02->as<ASTDeclareReference>()->reference_table_name, "table_name");
     ASTPtr arguments = ast_reference_02->as<ASTDeclareReference>()->reference_expression->as<ASTFunction>()->arguments;
     EXPECT_EQ(arguments->children[0]->as<ASTIdentifier>()->name(), "ref_col_01");
@@ -28,19 +28,19 @@ TEST(ParserReference, ReferenceDifferenceKind)
 {
     ParserDeclareReference p_reference;
     String reference_01 = "REFERENCES table_name (ref_col_01) MATCH FULL";
-    ASTPtr ast_reference_01 = parseQuery(p_reference, reference_01.data(), reference_01.data() + reference_01.size(), "", 0, 0);
+    ASTPtr ast_reference_01 = parseQuery(p_reference, reference_01.data(), reference_01.data() + reference_01.size(), "", 0, 0, 0);
     EXPECT_EQ(ast_reference_01->as<ASTDeclareReference>()->reference_table_name, "table_name");
     EXPECT_EQ(ast_reference_01->as<ASTDeclareReference>()->reference_expression->as<ASTIdentifier>()->name(), "ref_col_01");
     EXPECT_EQ(ast_reference_01->as<ASTDeclareReference>()->kind, ASTDeclareReference::MATCH_FULL);
 
     String reference_02 = "REFERENCES table_name (ref_col_01) MATCH PARTIAL";
-    ASTPtr ast_reference_02 = parseQuery(p_reference, reference_02.data(), reference_02.data() + reference_02.size(), "", 0, 0);
+    ASTPtr ast_reference_02 = parseQuery(p_reference, reference_02.data(), reference_02.data() + reference_02.size(), "", 0, 0, 0);
     EXPECT_EQ(ast_reference_02->as<ASTDeclareReference>()->reference_table_name, "table_name");
     EXPECT_EQ(ast_reference_02->as<ASTDeclareReference>()->reference_expression->as<ASTIdentifier>()->name(), "ref_col_01");
     EXPECT_EQ(ast_reference_02->as<ASTDeclareReference>()->kind, ASTDeclareReference::MATCH_PARTIAL);
 
     String reference_03 = "REFERENCES table_name (ref_col_01) MATCH SIMPLE";
-    ASTPtr ast_reference_03 = parseQuery(p_reference, reference_03.data(), reference_03.data() + reference_03.size(), "", 0, 0);
+    ASTPtr ast_reference_03 = parseQuery(p_reference, reference_03.data(), reference_03.data() + reference_03.size(), "", 0, 0, 0);
     EXPECT_EQ(ast_reference_03->as<ASTDeclareReference>()->reference_table_name, "table_name");
     EXPECT_EQ(ast_reference_03->as<ASTDeclareReference>()->reference_expression->as<ASTIdentifier>()->name(), "ref_col_01");
     EXPECT_EQ(ast_reference_03->as<ASTDeclareReference>()->kind, ASTDeclareReference::MATCH_SIMPLE);
@@ -50,7 +50,7 @@ TEST(ParserReference, ReferenceDifferenceOption)
 {
     ParserDeclareReference p_reference;
     String reference_01 = "REFERENCES table_name (ref_col_01) MATCH FULL ON DELETE RESTRICT ON UPDATE RESTRICT";
-    ASTPtr ast_reference_01 = parseQuery(p_reference, reference_01.data(), reference_01.data() + reference_01.size(), "", 0, 0);
+    ASTPtr ast_reference_01 = parseQuery(p_reference, reference_01.data(), reference_01.data() + reference_01.size(), "", 0, 0, 0);
     EXPECT_EQ(ast_reference_01->as<ASTDeclareReference>()->reference_table_name, "table_name");
     EXPECT_EQ(ast_reference_01->as<ASTDeclareReference>()->reference_expression->as<ASTIdentifier>()->name(), "ref_col_01");
     EXPECT_EQ(ast_reference_01->as<ASTDeclareReference>()->kind, ASTDeclareReference::MATCH_FULL);
@@ -58,7 +58,7 @@ TEST(ParserReference, ReferenceDifferenceOption)
     EXPECT_EQ(ast_reference_01->as<ASTDeclareReference>()->on_update_option, ASTDeclareReference::RESTRICT);
 
     String reference_02 = "REFERENCES table_name (ref_col_01) MATCH FULL ON DELETE CASCADE ON UPDATE CASCADE";
-    ASTPtr ast_reference_02 = parseQuery(p_reference, reference_02.data(), reference_02.data() + reference_02.size(), "", 0, 0);
+    ASTPtr ast_reference_02 = parseQuery(p_reference, reference_02.data(), reference_02.data() + reference_02.size(), "", 0, 0, 0);
     EXPECT_EQ(ast_reference_02->as<ASTDeclareReference>()->reference_table_name, "table_name");
     EXPECT_EQ(ast_reference_02->as<ASTDeclareReference>()->reference_expression->as<ASTIdentifier>()->name(), "ref_col_01");
     EXPECT_EQ(ast_reference_02->as<ASTDeclareReference>()->kind, ASTDeclareReference::MATCH_FULL);
@@ -66,7 +66,7 @@ TEST(ParserReference, ReferenceDifferenceOption)
     EXPECT_EQ(ast_reference_02->as<ASTDeclareReference>()->on_update_option, ASTDeclareReference::CASCADE);
 
     String reference_03 = "REFERENCES table_name (ref_col_01) MATCH FULL ON DELETE SET NULL ON UPDATE SET NULL";
-    ASTPtr ast_reference_03 = parseQuery(p_reference, reference_03.data(), reference_03.data() + reference_03.size(), "", 0, 0);
+    ASTPtr ast_reference_03 = parseQuery(p_reference, reference_03.data(), reference_03.data() + reference_03.size(), "", 0, 0, 0);
     EXPECT_EQ(ast_reference_03->as<ASTDeclareReference>()->reference_table_name, "table_name");
     EXPECT_EQ(ast_reference_03->as<ASTDeclareReference>()->reference_expression->as<ASTIdentifier>()->name(), "ref_col_01");
     EXPECT_EQ(ast_reference_03->as<ASTDeclareReference>()->kind, ASTDeclareReference::MATCH_FULL);
@@ -74,7 +74,7 @@ TEST(ParserReference, ReferenceDifferenceOption)
     EXPECT_EQ(ast_reference_03->as<ASTDeclareReference>()->on_update_option, ASTDeclareReference::SET_NULL);
 
     String reference_04 = "REFERENCES table_name (ref_col_01) MATCH FULL ON UPDATE NO ACTION ON DELETE NO ACTION";
-    ASTPtr ast_reference_04 = parseQuery(p_reference, reference_04.data(), reference_04.data() + reference_04.size(), "", 0, 0);
+    ASTPtr ast_reference_04 = parseQuery(p_reference, reference_04.data(), reference_04.data() + reference_04.size(), "", 0, 0, 0);
     EXPECT_EQ(ast_reference_04->as<ASTDeclareReference>()->reference_table_name, "table_name");
     EXPECT_EQ(ast_reference_04->as<ASTDeclareReference>()->reference_expression->as<ASTIdentifier>()->name(), "ref_col_01");
     EXPECT_EQ(ast_reference_04->as<ASTDeclareReference>()->kind, ASTDeclareReference::MATCH_FULL);
@@ -82,11 +82,10 @@ TEST(ParserReference, ReferenceDifferenceOption)
     EXPECT_EQ(ast_reference_04->as<ASTDeclareReference>()->on_update_option, ASTDeclareReference::NO_ACTION);
 
     String reference_05 = "REFERENCES table_name (ref_col_01) MATCH FULL ON UPDATE SET DEFAULT ON DELETE SET DEFAULT";
-    ASTPtr ast_reference_05 = parseQuery(p_reference, reference_05.data(), reference_05.data() + reference_05.size(), "", 0, 0);
+    ASTPtr ast_reference_05 = parseQuery(p_reference, reference_05.data(), reference_05.data() + reference_05.size(), "", 0, 0, 0);
     EXPECT_EQ(ast_reference_05->as<ASTDeclareReference>()->reference_table_name, "table_name");
     EXPECT_EQ(ast_reference_05->as<ASTDeclareReference>()->reference_expression->as<ASTIdentifier>()->name(), "ref_col_01");
     EXPECT_EQ(ast_reference_05->as<ASTDeclareReference>()->kind, ASTDeclareReference::MATCH_FULL);
     EXPECT_EQ(ast_reference_05->as<ASTDeclareReference>()->on_delete_option, ASTDeclareReference::SET_DEFAULT);
     EXPECT_EQ(ast_reference_05->as<ASTDeclareReference>()->on_update_option, ASTDeclareReference::SET_DEFAULT);
 }
-
diff --git a/src/Parsers/MySQL/tests/gtest_subpartition_parser.cpp b/src/Parsers/MySQL/tests/gtest_subpartition_parser.cpp
index b375f73c55c..1876cd1d028 100644
--- a/src/Parsers/MySQL/tests/gtest_subpartition_parser.cpp
+++ b/src/Parsers/MySQL/tests/gtest_subpartition_parser.cpp
@@ -14,7 +14,7 @@ TEST(ParserSubpartition, AllSubpatitionOptions)
                    " DATA DIRECTORY 'data_directory' INDEX DIRECTORY 'index_directory' max_rows 1000 MIN_ROWs 0"
                    " TABLESPACE table_space_name";
     MySQLParser::ParserDeclareSubPartition p_subpartition;
-    ASTPtr ast = parseQuery(p_subpartition, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(p_subpartition, input.data(), input.data() + input.size(), "", 0, 0, 0);
 
     ASTDeclareSubPartition * declare_subpartition = ast->as<ASTDeclareSubPartition>();
     EXPECT_EQ(declare_subpartition->logical_name, "subpartition_name");
@@ -32,7 +32,7 @@ TEST(ParserSubpartition, OptionalSubpatitionOptions)
 {
     String input = "SUBPARTITION subpartition_name STORAGE engine = engine_name max_rows 1000 min_rows 0 tablespace table_space_name";
     MySQLParser::ParserDeclareSubPartition p_subpartition;
-    ASTPtr ast = parseQuery(p_subpartition, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(p_subpartition, input.data(), input.data() + input.size(), "", 0, 0, 0);
 
     ASTDeclareSubPartition * declare_subpartition = ast->as<ASTDeclareSubPartition>();
     EXPECT_EQ(declare_subpartition->logical_name, "subpartition_name");
@@ -42,4 +42,3 @@ TEST(ParserSubpartition, OptionalSubpatitionOptions)
     EXPECT_EQ(declare_options->changes["max_rows"]->as<ASTLiteral>()->value.safeGet<UInt64>(), 1000);
     EXPECT_EQ(declare_options->changes["tablespace"]->as<ASTIdentifier>()->name(), "table_space_name");
 }
-
diff --git a/src/Parsers/MySQL/tests/gtest_table_options_parser.cpp b/src/Parsers/MySQL/tests/gtest_table_options_parser.cpp
index 42b9279c96d..a84da7cb9d5 100644
--- a/src/Parsers/MySQL/tests/gtest_table_options_parser.cpp
+++ b/src/Parsers/MySQL/tests/gtest_table_options_parser.cpp
@@ -18,7 +18,7 @@ TEST(ParserTableOptions, AllSubpatitionOptions)
                    " STATS_PERSISTENT DEFAULT STATS_SAMPLE_PAGES 3 TABLESPACE tablespace_name STORAGE MEMORY UNION (table_01, table_02)";
 
     ParserDeclareTableOptions p_table_options;
-    ASTPtr ast = parseQuery(p_table_options, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(p_table_options, input.data(), input.data() + input.size(), "", 0, 0, 0);
 
     ASTDeclareOptions * declare_options = ast->as<ASTDeclareOptions>();
     EXPECT_EQ(declare_options->changes["auto_increment"]->as<ASTLiteral>()->value.safeGet<UInt64>(), 1);
@@ -56,7 +56,7 @@ TEST(ParserTableOptions, OptionalTableOptions)
 {
     String input = "STATS_AUTO_RECALC DEFAULT AUTO_INCREMENt = 1 ";
     ParserDeclareTableOptions p_table_options;
-    ASTPtr ast = parseQuery(p_table_options, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(p_table_options, input.data(), input.data() + input.size(), "", 0, 0, 0);
 
     ASTDeclareOptions * declare_options = ast->as<ASTDeclareOptions>();
     EXPECT_EQ(declare_options->changes["auto_increment"]->as<ASTLiteral>()->value.safeGet<UInt64>(), 1);
diff --git a/src/Parsers/PRQL/ParserPRQLQuery.cpp b/src/Parsers/PRQL/ParserPRQLQuery.cpp
index b3733b727dc..fb1796714cb 100644
--- a/src/Parsers/PRQL/ParserPRQLQuery.cpp
+++ b/src/Parsers/PRQL/ParserPRQLQuery.cpp
@@ -69,7 +69,9 @@ bool ParserPRQLQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
         "",
         false,
         max_query_size,
-        max_parser_depth);
+        max_parser_depth,
+        max_parser_backtracks,
+        true);
 
     if (!node)
         throw Exception(
diff --git a/src/Parsers/PRQL/ParserPRQLQuery.h b/src/Parsers/PRQL/ParserPRQLQuery.h
index 4fc450df6b6..88bf97f69d1 100644
--- a/src/Parsers/PRQL/ParserPRQLQuery.h
+++ b/src/Parsers/PRQL/ParserPRQLQuery.h
@@ -13,9 +13,10 @@ private:
     // These fields are not used when PRQL is disabled at build time.
     [[maybe_unused]] size_t max_query_size;
     [[maybe_unused]] size_t max_parser_depth;
+    [[maybe_unused]] size_t max_parser_backtracks;
 
 public:
-    ParserPRQLQuery(size_t max_query_size_, size_t max_parser_depth_) : max_query_size{max_query_size_}, max_parser_depth{max_parser_depth_}
+    ParserPRQLQuery(size_t max_query_size_, size_t max_parser_depth_, size_t max_parser_backtracks_) : max_query_size(max_query_size_), max_parser_depth(max_parser_depth_), max_parser_backtracks(max_parser_backtracks_)
     {
     }
 
diff --git a/src/Parsers/ParserAlterQuery.cpp b/src/Parsers/ParserAlterQuery.cpp
index b1cc7622e00..4bc95e67afb 100644
--- a/src/Parsers/ParserAlterQuery.cpp
+++ b/src/Parsers/ParserAlterQuery.cpp
@@ -865,7 +865,8 @@ bool ParserAlterCommand::parseImpl(Pos & pos, ASTPtr & node, Expected & expected
             else if (s_modify_sql_security.ignore(pos, expected))
             {
                 /// This is a hack so we can reuse parser from create and don't have to write `MODIFY SQL SECURITY SQL SECURITY INVOKER`
-                pos -= 2;
+                --pos;
+                --pos;
                 if (!sql_security_p.parse(pos, command_sql_security, expected))
                     return false;
                 command->type = ASTAlterCommand::MODIFY_SQL_SECURITY;
diff --git a/src/Parsers/ParserCreateQuery.h b/src/Parsers/ParserCreateQuery.h
index 440a8bc1dc7..30bce57f9d9 100644
--- a/src/Parsers/ParserCreateQuery.h
+++ b/src/Parsers/ParserCreateQuery.h
@@ -286,7 +286,7 @@ bool IParserColumnDeclaration<NameParser>::parseImpl(Pos & pos, ASTPtr & node, E
         {
             const String type_int("INT");
             Tokens tokens(type_int.data(), type_int.data() + type_int.size());
-            Pos tmp_pos(tokens, 0);
+            Pos tmp_pos(tokens, pos.max_depth, pos.max_backtracks);
             Expected tmp_expected;
             ParserDataType().parse(tmp_pos, type, tmp_expected);
         }
diff --git a/src/Parsers/QueryParameterVisitor.cpp b/src/Parsers/QueryParameterVisitor.cpp
index b8679cc3b96..9afd9a8615c 100644
--- a/src/Parsers/QueryParameterVisitor.cpp
+++ b/src/Parsers/QueryParameterVisitor.cpp
@@ -43,7 +43,7 @@ NameSet analyzeReceiveQueryParams(const std::string & query)
     const char * query_end = query.data() + query.size();
 
     ParserQuery parser(query_end);
-    ASTPtr extract_query_ast = parseQuery(parser, query_begin, query_end, "analyzeReceiveQueryParams", 0, 0);
+    ASTPtr extract_query_ast = parseQuery(parser, query_begin, query_end, "analyzeReceiveQueryParams", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
     QueryParameterVisitor(query_params).visit(extract_query_ast);
 
     NameSet query_param_names;
diff --git a/src/Parsers/TokenIterator.h b/src/Parsers/TokenIterator.h
index 8cb59aa12e2..192f2f55e6a 100644
--- a/src/Parsers/TokenIterator.h
+++ b/src/Parsers/TokenIterator.h
@@ -62,18 +62,6 @@ public:
         return *this;
     }
 
-    ALWAYS_INLINE TokenIterator & operator-=(int value)
-    {
-        index -= value;
-        return *this;
-    }
-
-    ALWAYS_INLINE TokenIterator & operator+=(int value)
-    {
-        index += value;
-        return *this;
-    }
-
     ALWAYS_INLINE bool operator<(const TokenIterator & rhs) const { return index < rhs.index; }
     ALWAYS_INLINE bool operator<=(const TokenIterator & rhs) const { return index <= rhs.index; }
     ALWAYS_INLINE bool operator==(const TokenIterator & rhs) const { return index == rhs.index; }
diff --git a/src/Parsers/examples/create_parser.cpp b/src/Parsers/examples/create_parser.cpp
index c241b353b4f..b628c79435c 100644
--- a/src/Parsers/examples/create_parser.cpp
+++ b/src/Parsers/examples/create_parser.cpp
@@ -13,7 +13,7 @@ int main(int, char **)
 
     std::string input = "CREATE TABLE hits (URL String, UserAgentMinor2 FixedString(2), EventTime DateTime) ENGINE = Log";
     ParserCreateQuery parser;
-    ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, 0, 0);
 
     WriteBufferFromOStream out(std::cerr, 4096);
     formatAST(*ast, out);
diff --git a/src/Parsers/examples/select_parser.cpp b/src/Parsers/examples/select_parser.cpp
index 15295170c6b..3ed358121f6 100644
--- a/src/Parsers/examples/select_parser.cpp
+++ b/src/Parsers/examples/select_parser.cpp
@@ -23,7 +23,7 @@ try
         " FORMAT TabSeparated";
 
     ParserQueryWithOutput parser(input.data() + input.size());
-    ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, 0, 0);
 
     std::cout << "Success." << std::endl;
     WriteBufferFromOStream out(std::cerr, 4096);
diff --git a/src/Parsers/fuzzers/select_parser_fuzzer.cpp b/src/Parsers/fuzzers/select_parser_fuzzer.cpp
index ae490ed4e56..aed83853c33 100644
--- a/src/Parsers/fuzzers/select_parser_fuzzer.cpp
+++ b/src/Parsers/fuzzers/select_parser_fuzzer.cpp
@@ -15,7 +15,8 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t * data, size_t size)
         DB::ParserQueryWithOutput parser(input.data() + input.size());
 
         const UInt64 max_parser_depth = 1000;
-        DB::ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, max_parser_depth);
+        const UInt64 max_parser_backtracks = 1000000;
+        DB::ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, max_parser_depth, max_parser_backtracks);
 
         const UInt64 max_ast_depth = 1000;
         ast->checkDepth(max_ast_depth);
diff --git a/src/Parsers/parseQuery.cpp b/src/Parsers/parseQuery.cpp
index 8f9977c0b8d..7aad0b010a5 100644
--- a/src/Parsers/parseQuery.cpp
+++ b/src/Parsers/parseQuery.cpp
@@ -234,12 +234,13 @@ ASTPtr tryParseQuery(
     bool allow_multi_statements,
     size_t max_query_size,
     size_t max_parser_depth,
+    size_t max_parser_backtracks,
     bool skip_insignificant)
 {
     const char * query_begin = _out_query_end;
     Tokens tokens(query_begin, all_queries_end, max_query_size, skip_insignificant);
     /// NOTE: consider use UInt32 for max_parser_depth setting.
-    IParser::Pos token_iterator(tokens, static_cast<uint32_t>(max_parser_depth));
+    IParser::Pos token_iterator(tokens, static_cast<uint32_t>(max_parser_depth), static_cast<uint32_t>(max_parser_backtracks));
 
     if (token_iterator->isEnd()
         || token_iterator->type == TokenType::Semicolon)
@@ -356,10 +357,13 @@ ASTPtr parseQueryAndMovePosition(
     const std::string & query_description,
     bool allow_multi_statements,
     size_t max_query_size,
-    size_t max_parser_depth)
+    size_t max_parser_depth,
+    size_t max_parser_backtracks)
 {
     std::string error_message;
-    ASTPtr res = tryParseQuery(parser, pos, end, error_message, false, query_description, allow_multi_statements, max_query_size, max_parser_depth);
+    ASTPtr res = tryParseQuery(
+        parser, pos, end, error_message, false, query_description, allow_multi_statements,
+        max_query_size, max_parser_depth, max_parser_backtracks, true);
 
     if (res)
         return res;
@@ -374,9 +378,10 @@ ASTPtr parseQuery(
     const char * end,
     const std::string & query_description,
     size_t max_query_size,
-    size_t max_parser_depth)
+    size_t max_parser_depth,
+    size_t max_parser_backtracks)
 {
-    return parseQueryAndMovePosition(parser, begin, end, query_description, false, max_query_size, max_parser_depth);
+    return parseQueryAndMovePosition(parser, begin, end, query_description, false, max_query_size, max_parser_depth, max_parser_backtracks);
 }
 
 
@@ -385,9 +390,10 @@ ASTPtr parseQuery(
     const std::string & query,
     const std::string & query_description,
     size_t max_query_size,
-    size_t max_parser_depth)
+    size_t max_parser_depth,
+    size_t max_parser_backtracks)
 {
-    return parseQuery(parser, query.data(), query.data() + query.size(), query_description, max_query_size, max_parser_depth);
+    return parseQuery(parser, query.data(), query.data() + query.size(), query_description, max_query_size, max_parser_depth, max_parser_backtracks);
 }
 
 
@@ -395,9 +401,10 @@ ASTPtr parseQuery(
     IParser & parser,
     const std::string & query,
     size_t max_query_size,
-    size_t max_parser_depth)
+    size_t max_parser_depth,
+    size_t max_parser_backtracks)
 {
-    return parseQuery(parser, query.data(), query.data() + query.size(), parser.getName(), max_query_size, max_parser_depth);
+    return parseQuery(parser, query.data(), query.data() + query.size(), parser.getName(), max_query_size, max_parser_depth, max_parser_backtracks);
 }
 
 
@@ -406,6 +413,7 @@ std::pair<const char *, bool> splitMultipartQuery(
     std::vector<std::string> & queries_list,
     size_t max_query_size,
     size_t max_parser_depth,
+    size_t max_parser_backtracks,
     bool allow_settings_after_format_in_insert)
 {
     ASTPtr ast;
@@ -422,7 +430,7 @@ std::pair<const char *, bool> splitMultipartQuery(
     {
         begin = pos;
 
-        ast = parseQueryAndMovePosition(parser, pos, end, "", true, max_query_size, max_parser_depth);
+        ast = parseQueryAndMovePosition(parser, pos, end, "", true, max_query_size, max_parser_depth, max_parser_backtracks);
 
         auto * insert = ast->as<ASTInsertQuery>();
 
diff --git a/src/Parsers/parseQuery.h b/src/Parsers/parseQuery.h
index a087f145d2c..93c1a465267 100644
--- a/src/Parsers/parseQuery.h
+++ b/src/Parsers/parseQuery.h
@@ -19,7 +19,8 @@ ASTPtr tryParseQuery(
     size_t max_query_size,          /// If (end - pos) > max_query_size and query is longer than max_query_size then throws "Max query size exceeded".
                                     /// Disabled if zero. Is used in order to check query size if buffer can contains data for INSERT query.
     size_t max_parser_depth,
-    bool skip_insignificant = true);  /// If true, lexer will skip all insignificant tokens (e.g. whitespaces)
+    size_t max_parser_backtracks,
+    bool skip_insignificant);  /// If true, lexer will skip all insignificant tokens (e.g. whitespaces)
 
 
 /// Parse query or throw an exception with error message.
@@ -30,7 +31,8 @@ ASTPtr parseQueryAndMovePosition(
     const std::string & description,
     bool allow_multi_statements,
     size_t max_query_size,
-    size_t max_parser_depth);
+    size_t max_parser_depth,
+    size_t max_parser_backtracks);
 
 ASTPtr parseQuery(
     IParser & parser,
@@ -38,20 +40,23 @@ ASTPtr parseQuery(
     const char * end,
     const std::string & description,
     size_t max_query_size,
-    size_t max_parser_depth);
+    size_t max_parser_depth,
+    size_t max_parser_backtracks);
 
 ASTPtr parseQuery(
     IParser & parser,
     const std::string & query,
     const std::string & query_description,
     size_t max_query_size,
-    size_t max_parser_depth);
+    size_t max_parser_depth,
+    size_t max_parser_backtracks);
 
 ASTPtr parseQuery(
     IParser & parser,
     const std::string & query,
     size_t max_query_size,
-    size_t max_parser_depth);
+    size_t max_parser_depth,
+    size_t max_parser_backtracks);
 
 
 /** Split queries separated by ; on to list of single queries
@@ -63,6 +68,7 @@ std::pair<const char *, bool> splitMultipartQuery(
     std::vector<std::string> & queries_list,
     size_t max_query_size,
     size_t max_parser_depth,
+    size_t max_parser_backtracks,
     bool allow_settings_after_format_in_insert);
 
 }
diff --git a/src/Parsers/tests/gtest_Parser.cpp b/src/Parsers/tests/gtest_Parser.cpp
index 19947cd38cc..f0abc68f966 100644
--- a/src/Parsers/tests/gtest_Parser.cpp
+++ b/src/Parsers/tests/gtest_Parser.cpp
@@ -1,7 +1,4 @@
-#include <IO/WriteBufferFromOStream.h>
-#include <Interpreters/applyTableOverride.h>
 #include <Parsers/ASTCreateQuery.h>
-#include <Parsers/ASTFunction.h>
 #include <Parsers/ASTIdentifier.h>
 #include <Parsers/Access/ASTCreateUserQuery.h>
 #include <Parsers/Access/ParserCreateUserQuery.h>
@@ -10,7 +7,6 @@
 #include <Parsers/ParserCreateQuery.h>
 #include <Parsers/ParserOptimizeQuery.h>
 #include <Parsers/ParserRenameQuery.h>
-#include <Parsers/ParserQueryWithOutput.h>
 #include <Parsers/ParserAttachAccessEntity.h>
 #include <Parsers/formatAST.h>
 #include <Parsers/parseQuery.h>
@@ -54,12 +50,12 @@ TEST_P(ParserTest, parseQuery)
     {
         if (std::string(expected_ast).starts_with("throws"))
         {
-            EXPECT_THROW(parseQuery(*parser, input_text.begin(), input_text.end(), 0, 0), DB::Exception);
+            EXPECT_THROW(parseQuery(*parser, input_text.begin(), input_text.end(), 0, 0, 0), DB::Exception);
         }
         else
         {
             ASTPtr ast;
-            ASSERT_NO_THROW(ast = parseQuery(*parser, input_text.begin(), input_text.end(), 0, 0));
+            ASSERT_NO_THROW(ast = parseQuery(*parser, input_text.begin(), input_text.end(), 0, 0, 0));
             if (std::string("CREATE USER or ALTER USER query") != parser->getName()
                     && std::string("ATTACH access entity query") != parser->getName())
             {
@@ -106,7 +102,7 @@ TEST_P(ParserTest, parseQuery)
     }
     else
     {
-        ASSERT_THROW(parseQuery(*parser, input_text.begin(), input_text.end(), 0, 0), DB::Exception);
+        ASSERT_THROW(parseQuery(*parser, input_text.begin(), input_text.end(), 0, 0, 0), DB::Exception);
     }
 }
 
@@ -649,12 +645,13 @@ INSTANTIATE_TEST_SUITE_P(ParserKQLQuery, ParserKQLTest,
 
 static constexpr size_t kDummyMaxQuerySize = 256 * 1024;
 static constexpr size_t kDummyMaxParserDepth = 256;
+static constexpr size_t kDummyMaxParserBacktracks = 1000000;
 
 INSTANTIATE_TEST_SUITE_P(
     ParserPRQL,
     ParserTest,
     ::testing::Combine(
-        ::testing::Values(std::make_shared<ParserPRQLQuery>(kDummyMaxQuerySize, kDummyMaxParserDepth)),
+        ::testing::Values(std::make_shared<ParserPRQLQuery>(kDummyMaxQuerySize, kDummyMaxParserDepth, kDummyMaxParserBacktracks)),
         ::testing::ValuesIn(std::initializer_list<ParserTestCase>{
             {
                 "from albums\ngroup {author_id} (\n  aggregate {first_published = min published}\n)\njoin a=author side:left (==author_id)\njoin p=purchases side:right (==author_id)\ngroup {a.id, p.purchase_id} (\n  aggregate {avg_sell = min first_published}\n)",
diff --git a/src/Parsers/tests/gtest_common.cpp b/src/Parsers/tests/gtest_common.cpp
index 52d3ceb47e2..8ff9400d8a2 100644
--- a/src/Parsers/tests/gtest_common.cpp
+++ b/src/Parsers/tests/gtest_common.cpp
@@ -28,7 +28,7 @@ TEST_P(ParserRegexTest, parseQuery)
     ASSERT_TRUE(expected_ast);
 
     DB::ASTPtr ast;
-    ASSERT_NO_THROW(ast = parseQuery(*parser, input_text.begin(), input_text.end(), 0, 0));
+    ASSERT_NO_THROW(ast = parseQuery(*parser, input_text.begin(), input_text.end(), 0, 0, 0));
     DB::WriteBufferFromOwnString buf;
     formatAST(*ast->clone(), buf, false, false);
     EXPECT_THAT(buf.str(), ::testing::MatchesRegex(expected_ast));
@@ -45,12 +45,12 @@ TEST_P(ParserKQLTest, parseKQLQuery)
     {
         if (std::string(expected_ast).starts_with("throws"))
         {
-            EXPECT_THROW(parseKQLQuery(*parser, input_text.begin(), input_text.end(), 0, 0), DB::Exception);
+            EXPECT_THROW(parseKQLQuery(*parser, input_text.begin(), input_text.end(), 0, 0, 0), DB::Exception);
         }
         else
         {
             DB::ASTPtr ast;
-            ASSERT_NO_THROW(ast = parseKQLQuery(*parser, input_text.begin(), input_text.end(), 0, 0));
+            ASSERT_NO_THROW(ast = parseKQLQuery(*parser, input_text.begin(), input_text.end(), 0, 0, 0));
             if (std::string("CREATE USER or ALTER USER query") != parser->getName()
                     && std::string("ATTACH access entity query") != parser->getName())
             {
@@ -78,6 +78,6 @@ TEST_P(ParserKQLTest, parseKQLQuery)
     }
     else
     {
-        ASSERT_THROW(parseKQLQuery(*parser, input_text.begin(), input_text.end(), 0, 0), DB::Exception);
+        ASSERT_THROW(parseKQLQuery(*parser, input_text.begin(), input_text.end(), 0, 0, 0), DB::Exception);
     }
 }
diff --git a/src/Parsers/tests/gtest_dictionary_parser.cpp b/src/Parsers/tests/gtest_dictionary_parser.cpp
index c0a975f7a38..a1ba46125a7 100644
--- a/src/Parsers/tests/gtest_dictionary_parser.cpp
+++ b/src/Parsers/tests/gtest_dictionary_parser.cpp
@@ -40,7 +40,7 @@ TEST(ParserDictionaryDDL, SimpleDictionary)
                    " RANGE(MIN second_column MAX third_column)";
 
     ParserCreateDictionaryQuery parser;
-    ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, 0, 0);
     ASTCreateQuery * create = ast->as<ASTCreateQuery>();
     EXPECT_EQ(create->getTable(), "dict1");
     EXPECT_EQ(create->getDatabase(), "test");
@@ -136,7 +136,7 @@ TEST(ParserDictionaryDDL, AttributesWithMultipleProperties)
                    " SOURCE(CLICKHOUSE(HOST 'localhost'))";
 
     ParserCreateDictionaryQuery parser;
-    ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, 0, 0);
     ASTCreateQuery * create = ast->as<ASTCreateQuery>();
     EXPECT_EQ(create->getTable(), "dict2");
     EXPECT_EQ(create->getDatabase(), "");
@@ -183,7 +183,7 @@ TEST(ParserDictionaryDDL, CustomAttributePropertiesOrder)
                    " LIFETIME(300)";
 
     ParserCreateDictionaryQuery parser;
-    ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, 0, 0);
     ASTCreateQuery * create = ast->as<ASTCreateQuery>();
 
     /// test attributes
@@ -238,7 +238,7 @@ TEST(ParserDictionaryDDL, NestedSource)
                    " RANGE(MIN second_column MAX third_column)";
 
     ParserCreateDictionaryQuery parser;
-    ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, 0, 0);
     ASTCreateQuery * create = ast->as<ASTCreateQuery>();
     EXPECT_EQ(create->getTable(), "dict4");
     EXPECT_EQ(create->getDatabase(), "");
@@ -286,7 +286,7 @@ TEST(ParserDictionaryDDL, Formatting)
                    " RANGE(MIN second_column MAX third_column)";
 
     ParserCreateDictionaryQuery parser;
-    ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, 0);
+    ASTPtr ast = parseQuery(parser, input.data(), input.data() + input.size(), "", 0, 0, 0);
     ASTCreateQuery * create = ast->as<ASTCreateQuery>();
     auto str = serializeAST(*create);
     EXPECT_EQ(str, "CREATE DICTIONARY test.dict5 (`key_column1` UInt64 DEFAULT 1 HIERARCHICAL INJECTIVE, `key_column2` String DEFAULT '', `second_column` UInt8 EXPRESSION intDiv(50, rand() % 1000), `third_column` UInt8) PRIMARY KEY key_column1, key_column2 SOURCE(MYSQL(HOST 'localhost' PORT 9000 USER 'default' REPLICA (HOST '127.0.0.1' PRIORITY 1) PASSWORD '')) LIFETIME(MIN 1 MAX 10) LAYOUT(CACHE(SIZE_IN_CELLS 50)) RANGE(MIN second_column MAX third_column)");
@@ -297,7 +297,7 @@ TEST(ParserDictionaryDDL, ParseDropQuery)
     String input1 = "DROP DICTIONARY test.dict1";
 
     ParserDropQuery parser;
-    ASTPtr ast1 = parseQuery(parser, input1.data(), input1.data() + input1.size(), "", 0, 0);
+    ASTPtr ast1 = parseQuery(parser, input1.data(), input1.data() + input1.size(), "", 0, 0, 0);
     ASTDropQuery * drop1 = ast1->as<ASTDropQuery>();
 
     EXPECT_TRUE(drop1->is_dictionary);
@@ -308,7 +308,7 @@ TEST(ParserDictionaryDDL, ParseDropQuery)
 
     String input2 = "DROP DICTIONARY IF EXISTS dict2";
 
-    ASTPtr ast2 = parseQuery(parser, input2.data(), input2.data() + input2.size(), "", 0, 0);
+    ASTPtr ast2 = parseQuery(parser, input2.data(), input2.data() + input2.size(), "", 0, 0, 0);
     ASTDropQuery * drop2 = ast2->as<ASTDropQuery>();
 
     EXPECT_TRUE(drop2->is_dictionary);
@@ -323,7 +323,7 @@ TEST(ParserDictionaryDDL, ParsePropertiesQueries)
     String input1 = "SHOW CREATE DICTIONARY test.dict1";
 
     ParserTablePropertiesQuery parser;
-    ASTPtr ast1 = parseQuery(parser, input1.data(), input1.data() + input1.size(), "", 0, 0);
+    ASTPtr ast1 = parseQuery(parser, input1.data(), input1.data() + input1.size(), "", 0, 0, 0);
     ASTShowCreateDictionaryQuery * show1 = ast1->as<ASTShowCreateDictionaryQuery>();
 
     EXPECT_EQ(show1->getTable(), "dict1");
@@ -332,7 +332,7 @@ TEST(ParserDictionaryDDL, ParsePropertiesQueries)
 
     String input2 = "EXISTS DICTIONARY dict2";
 
-    ASTPtr ast2 = parseQuery(parser, input2.data(), input2.data() + input2.size(), "", 0, 0);
+    ASTPtr ast2 = parseQuery(parser, input2.data(), input2.data() + input2.size(), "", 0, 0, 0);
     ASTExistsDictionaryQuery * show2 = ast2->as<ASTExistsDictionaryQuery>();
 
     EXPECT_EQ(show2->getTable(), "dict2");
diff --git a/src/Parsers/tests/gtest_format_hiliting.cpp b/src/Parsers/tests/gtest_format_hiliting.cpp
index e87b093db9d..00e8197af1f 100644
--- a/src/Parsers/tests/gtest_format_hiliting.cpp
+++ b/src/Parsers/tests/gtest_format_hiliting.cpp
@@ -50,7 +50,7 @@ void compare(const String & expected, const String & query)
 {
     using namespace DB;
     ParserQuery parser(query.data() + query.size());
-    ASTPtr ast = parseQuery(parser, query, 0, 0);
+    ASTPtr ast = parseQuery(parser, query, 0, 0, 0);
 
     WriteBufferFromOwnString write_buffer;
     IAST::FormatSettings settings(write_buffer, true, true);
diff --git a/src/Planner/PlannerJoinTree.cpp b/src/Planner/PlannerJoinTree.cpp
index 7b3fb0c5c91..8ca8f0f258b 100644
--- a/src/Planner/PlannerJoinTree.cpp
+++ b/src/Planner/PlannerJoinTree.cpp
@@ -538,7 +538,7 @@ FilterDAGInfo buildAdditionalFiltersIfNeeded(const StoragePtr & storage,
             ParserExpression parser;
             additional_filter_ast = parseQuery(
                 parser, filter.data(), filter.data() + filter.size(),
-                "additional filter", settings.max_query_size, settings.max_parser_depth);
+                "additional filter", settings.max_query_size, settings.max_parser_depth, settings.max_parser_backtracks);
             break;
         }
     }
diff --git a/src/Planner/Utils.cpp b/src/Planner/Utils.cpp
index bd0b831ee58..a04f9f502e2 100644
--- a/src/Planner/Utils.cpp
+++ b/src/Planner/Utils.cpp
@@ -523,7 +523,7 @@ ASTPtr parseAdditionalResultFilter(const Settings & settings)
     ParserExpression parser;
     auto additional_result_filter_ast = parseQuery(
                 parser, additional_result_filter.data(), additional_result_filter.data() + additional_result_filter.size(),
-                "additional result filter", settings.max_query_size, settings.max_parser_depth);
+                "additional result filter", settings.max_query_size, settings.max_parser_depth, settings.max_parser_backtracks);
     return additional_result_filter_ast;
 }
 
diff --git a/src/Processors/Formats/Impl/ConstantExpressionTemplate.cpp b/src/Processors/Formats/Impl/ConstantExpressionTemplate.cpp
index f91f7cf536b..9d056b42101 100644
--- a/src/Processors/Formats/Impl/ConstantExpressionTemplate.cpp
+++ b/src/Processors/Formats/Impl/ConstantExpressionTemplate.cpp
@@ -537,7 +537,7 @@ bool ConstantExpressionTemplate::parseLiteralAndAssertType(
         ParserArrayOfLiterals parser_array;
         ParserTupleOfLiterals parser_tuple;
 
-        IParser::Pos iterator(token_iterator, static_cast<unsigned>(settings.max_parser_depth));
+        IParser::Pos iterator(token_iterator, static_cast<unsigned>(settings.max_parser_depth), static_cast<unsigned>(settings.max_parser_backtracks));
         while (iterator->begin < istr.position())
             ++iterator;
         Expected expected;
diff --git a/src/Processors/Formats/Impl/MySQLDumpRowInputFormat.cpp b/src/Processors/Formats/Impl/MySQLDumpRowInputFormat.cpp
index 9c7f095e661..67bdd1cf877 100644
--- a/src/Processors/Formats/Impl/MySQLDumpRowInputFormat.cpp
+++ b/src/Processors/Formats/Impl/MySQLDumpRowInputFormat.cpp
@@ -274,7 +274,8 @@ static bool tryToExtractStructureFromCreateQuery(ReadBuffer & in, NamesAndTypesL
     String error;
     const char * start = create_query_str.data();
     const char * end = create_query_str.data() + create_query_str.size();
-    ASTPtr query = tryParseQuery(parser, start, end, error, false, "MySQL create query", false, DBMS_DEFAULT_MAX_QUERY_SIZE, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+    ASTPtr query = tryParseQuery(parser, start, end, error, false, "MySQL create query", false,
+        DBMS_DEFAULT_MAX_QUERY_SIZE, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS, true);
     if (!query)
         return false;
 
diff --git a/src/Processors/Formats/Impl/ValuesBlockInputFormat.cpp b/src/Processors/Formats/Impl/ValuesBlockInputFormat.cpp
index 8659dcd2318..353de76eea8 100644
--- a/src/Processors/Formats/Impl/ValuesBlockInputFormat.cpp
+++ b/src/Processors/Formats/Impl/ValuesBlockInputFormat.cpp
@@ -194,7 +194,7 @@ void ValuesBlockInputFormat::readUntilTheEndOfRowAndReTokenize(size_t current_co
     auto * row_end = buf->position();
     buf->rollbackToCheckpoint();
     tokens.emplace(buf->position(), row_end);
-    token_iterator.emplace(*tokens, static_cast<unsigned>(context->getSettingsRef().max_parser_depth));
+    token_iterator.emplace(*tokens, static_cast<unsigned>(context->getSettingsRef().max_parser_depth), static_cast<unsigned>(context->getSettingsRef().max_parser_backtracks));
     auto const & first = (*token_iterator).get();
     if (first.isError() || first.isEnd())
     {
@@ -418,7 +418,7 @@ bool ValuesBlockInputFormat::parseExpression(IColumn & column, size_t column_idx
     {
         Expected expected;
         /// Keep a copy to the start of the column tokens to use if later if necessary
-        ti_start = IParser::Pos(*token_iterator, static_cast<unsigned>(settings.max_parser_depth));
+        ti_start = IParser::Pos(*token_iterator, static_cast<unsigned>(settings.max_parser_depth), static_cast<unsigned>(settings.max_parser_backtracks));
 
         parsed = parser.parse(*token_iterator, ast, expected);
 
diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp
index 21e3cfcceab..fb92be6eed9 100644
--- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp
+++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp
@@ -1343,7 +1343,7 @@ static void buildIndexes(
     {
         const auto & indices = settings.ignore_data_skipping_indices.toString();
         Tokens tokens(indices.data(), indices.data() + indices.size(), settings.max_query_size);
-        IParser::Pos pos(tokens, static_cast<unsigned>(settings.max_parser_depth));
+        IParser::Pos pos(tokens, static_cast<unsigned>(settings.max_parser_depth), static_cast<unsigned>(settings.max_parser_backtracks));
         Expected expected;
 
         /// Use an unordered list rather than string vector
diff --git a/src/Server/GRPCServer.cpp b/src/Server/GRPCServer.cpp
index 15765f99b4b..f21991e8d58 100644
--- a/src/Server/GRPCServer.cpp
+++ b/src/Server/GRPCServer.cpp
@@ -885,7 +885,7 @@ namespace
         const char * begin = query_text.data();
         const char * end = begin + query_text.size();
         ParserQuery parser(end, settings.allow_settings_after_format_in_insert);
-        ast = parseQuery(parser, begin, end, "", settings.max_query_size, settings.max_parser_depth);
+        ast = parseQuery(parser, begin, end, "", settings.max_query_size, settings.max_parser_depth, settings.max_parser_backtracks);
 
         /// Choose input format.
         insert_query = ast->as<ASTInsertQuery>();
diff --git a/src/Server/PostgreSQLHandler.cpp b/src/Server/PostgreSQLHandler.cpp
index c62dc8109ea..83e06628185 100644
--- a/src/Server/PostgreSQLHandler.cpp
+++ b/src/Server/PostgreSQLHandler.cpp
@@ -284,6 +284,7 @@ void PostgreSQLHandler::processQuery()
         auto parse_res = splitMultipartQuery(query->query, queries,
             settings.max_query_size,
             settings.max_parser_depth,
+            settings.max_parser_backtracks,
             settings.allow_settings_after_format_in_insert);
         if (!parse_res.second)
             throw Exception(ErrorCodes::SYNTAX_ERROR, "Cannot parse and execute the following part of query: {}", String(parse_res.first));
diff --git a/src/Storages/ColumnsDescription.cpp b/src/Storages/ColumnsDescription.cpp
index e08dac3a332..16b89f24243 100644
--- a/src/Storages/ColumnsDescription.cpp
+++ b/src/Storages/ColumnsDescription.cpp
@@ -145,7 +145,7 @@ void ColumnDescription::readText(ReadBuffer & buf)
         readEscapedStringUntilEOL(modifiers, buf);
 
         ParserColumnDeclaration column_parser(/* require type */ true);
-        ASTPtr ast = parseQuery(column_parser, "x T " + modifiers, "column parser", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+        ASTPtr ast = parseQuery(column_parser, "x T " + modifiers, "column parser", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
 
         if (auto * col_ast = ast->as<ASTColumnDeclaration>())
         {
@@ -211,7 +211,7 @@ void ColumnsDescription::setAliases(NamesAndAliases aliases)
         const char * alias_expression_pos = alias.expression.data();
         const char * alias_expression_end = alias_expression_pos + alias.expression.size();
         ParserExpression expression_parser;
-        description.default_desc.expression = parseQuery(expression_parser, alias_expression_pos, alias_expression_end, "expression", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+        description.default_desc.expression = parseQuery(expression_parser, alias_expression_pos, alias_expression_end, "expression", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
 
         add(std::move(description));
     }
diff --git a/src/Storages/ConstraintsDescription.cpp b/src/Storages/ConstraintsDescription.cpp
index 219c3fd0c97..d492de2c2b2 100644
--- a/src/Storages/ConstraintsDescription.cpp
+++ b/src/Storages/ConstraintsDescription.cpp
@@ -45,7 +45,7 @@ ConstraintsDescription ConstraintsDescription::parse(const String & str)
 
     ConstraintsDescription res;
     ParserConstraintDeclarationList parser;
-    ASTPtr list = parseQuery(parser, str, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+    ASTPtr list = parseQuery(parser, str, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
 
     for (const auto & constraint : list->children)
         res.constraints.push_back(constraint);
diff --git a/src/Storages/IndicesDescription.cpp b/src/Storages/IndicesDescription.cpp
index c723fa4225c..14555dca63b 100644
--- a/src/Storages/IndicesDescription.cpp
+++ b/src/Storages/IndicesDescription.cpp
@@ -173,7 +173,7 @@ IndicesDescription IndicesDescription::parse(const String & str, const ColumnsDe
         return result;
 
     ParserIndexDeclarationList parser;
-    ASTPtr list = parseQuery(parser, str, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+    ASTPtr list = parseQuery(parser, str, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
 
     for (const auto & index : list->children)
         result.emplace_back(IndexDescription::getIndexFromAST(index, columns, context));
diff --git a/src/Storages/KeyDescription.cpp b/src/Storages/KeyDescription.cpp
index c407cef627d..d63b40e2b11 100644
--- a/src/Storages/KeyDescription.cpp
+++ b/src/Storages/KeyDescription.cpp
@@ -171,7 +171,7 @@ KeyDescription KeyDescription::parse(const String & str, const ColumnsDescriptio
         return result;
 
     ParserExpression parser;
-    ASTPtr ast = parseQuery(parser, "(" + str + ")", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+    ASTPtr ast = parseQuery(parser, "(" + str + ")", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
     FunctionNameNormalizer().visit(ast.get());
 
     return getKeyFromAST(ast, columns, context);
diff --git a/src/Storages/MergeTree/IMergeTreeDataPart.cpp b/src/Storages/MergeTree/IMergeTreeDataPart.cpp
index 984d06e6a61..023202019e4 100644
--- a/src/Storages/MergeTree/IMergeTreeDataPart.cpp
+++ b/src/Storages/MergeTree/IMergeTreeDataPart.cpp
@@ -957,7 +957,7 @@ void IMergeTreeDataPart::loadDefaultCompressionCodec()
         try
         {
             ParserCodec codec_parser;
-            auto codec_ast = parseQuery(codec_parser, codec_line.data() + buf.getPosition(), codec_line.data() + codec_line.length(), "codec parser", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+            auto codec_ast = parseQuery(codec_parser, codec_line.data() + buf.getPosition(), codec_line.data() + codec_line.length(), "codec parser", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
             default_codec = CompressionCodecFactory::instance().get(codec_ast, {});
         }
         catch (const DB::Exception & ex)
diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.cpp
index 1721fd15b8d..fe45d0bee54 100644
--- a/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.cpp
+++ b/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.cpp
@@ -12,7 +12,7 @@ namespace ErrorCodes
 static CompressionCodecPtr getMarksCompressionCodec(const String & marks_compression_codec)
 {
     ParserCodec codec_parser;
-    auto ast = parseQuery(codec_parser, "(" + Poco::toUpper(marks_compression_codec) + ")", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+    auto ast = parseQuery(codec_parser, "(" + Poco::toUpper(marks_compression_codec) + ")", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
     return CompressionCodecFactory::instance().get(ast, nullptr);
 }
 
diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp
index fd83d2ebfe9..a31da5bc4fe 100644
--- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp
+++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp
@@ -243,7 +243,7 @@ void MergeTreeDataPartWriterOnDisk::initPrimaryIndex()
         if (compress_primary_key)
         {
             ParserCodec codec_parser;
-            auto ast = parseQuery(codec_parser, "(" + Poco::toUpper(settings.primary_key_compression_codec) + ")", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+            auto ast = parseQuery(codec_parser, "(" + Poco::toUpper(settings.primary_key_compression_codec) + ")", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
             CompressionCodecPtr primary_key_compression_codec = CompressionCodecFactory::instance().get(ast, nullptr);
             index_compressor_stream = std::make_unique<CompressedWriteBuffer>(*index_file_hashing_stream, primary_key_compression_codec, settings.primary_key_compress_block_size);
             index_source_hashing_stream = std::make_unique<HashingWriteBuffer>(*index_compressor_stream);
@@ -268,7 +268,7 @@ void MergeTreeDataPartWriterOnDisk::initStatistics()
 void MergeTreeDataPartWriterOnDisk::initSkipIndices()
 {
     ParserCodec codec_parser;
-    auto ast = parseQuery(codec_parser, "(" + Poco::toUpper(settings.marks_compression_codec) + ")", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+    auto ast = parseQuery(codec_parser, "(" + Poco::toUpper(settings.marks_compression_codec) + ")", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
     CompressionCodecPtr marks_compression_codec = CompressionCodecFactory::instance().get(ast, nullptr);
 
     for (const auto & skip_index : skip_indices)
diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp
index d79590ded21..6a3b08d4d65 100644
--- a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp
+++ b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp
@@ -135,7 +135,7 @@ void MergeTreeDataPartWriterWide::addStreams(
             compression_codec = CompressionCodecFactory::instance().get(effective_codec_desc, nullptr, default_codec, true);
 
         ParserCodec codec_parser;
-        auto ast = parseQuery(codec_parser, "(" + Poco::toUpper(settings.marks_compression_codec) + ")", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+        auto ast = parseQuery(codec_parser, "(" + Poco::toUpper(settings.marks_compression_codec) + ")", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
         CompressionCodecPtr marks_compression_codec = CompressionCodecFactory::instance().get(ast, nullptr);
 
         const auto column_desc = metadata_snapshot->columns.tryGetColumnDescription(GetColumnsOptions(GetColumnsOptions::AllPhysical), column.getNameInStorage());
diff --git a/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp b/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp
index ef679b61a79..6471f510291 100644
--- a/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp
+++ b/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp
@@ -606,7 +606,7 @@ RangesInDataParts MergeTreeDataSelectExecutor::filterPartsByPrimaryKeyAndSkipInd
         Strings forced_indices;
         {
             Tokens tokens(indices.data(), indices.data() + indices.size(), settings.max_query_size);
-            IParser::Pos pos(tokens, static_cast<unsigned>(settings.max_parser_depth));
+            IParser::Pos pos(tokens, static_cast<unsigned>(settings.max_parser_depth), static_cast<unsigned>(settings.max_parser_backtracks));
             Expected expected;
             if (!parseIdentifiersOrStringLiterals(pos, expected, forced_indices))
                 throw Exception(ErrorCodes::CANNOT_PARSE_TEXT, "Cannot parse force_data_skipping_indices ('{}')", indices);
diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeTableMetadata.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeTableMetadata.cpp
index 41188891118..0ca7a4d74d9 100644
--- a/src/Storages/MergeTree/ReplicatedMergeTreeTableMetadata.cpp
+++ b/src/Storages/MergeTree/ReplicatedMergeTreeTableMetadata.cpp
@@ -432,7 +432,7 @@ StorageInMemoryMetadata ReplicatedMergeTreeTableMetadata::Diff::getNewMetadata(c
         auto parse_key_expr = [] (const String & key_expr)
         {
             ParserNotEmptyExpressionList parser(false);
-            auto new_sorting_key_expr_list = parseQuery(parser, key_expr, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+            auto new_sorting_key_expr_list = parseQuery(parser, key_expr, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
 
             ASTPtr order_by_ast;
             if (new_sorting_key_expr_list->children.size() == 1)
@@ -489,7 +489,7 @@ StorageInMemoryMetadata ReplicatedMergeTreeTableMetadata::Diff::getNewMetadata(c
             if (!new_ttl_table.empty())
             {
                 ParserTTLExpressionList parser;
-                auto ttl_for_table_ast = parseQuery(parser, new_ttl_table, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+                auto ttl_for_table_ast = parseQuery(parser, new_ttl_table, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
                 new_metadata.table_ttl = TTLTableDescription::getTTLForTableFromAST(
                     ttl_for_table_ast, new_metadata.columns, context, new_metadata.primary_key, true /* allow_suspicious; because it is replication */);
             }
diff --git a/src/Storages/MutationCommands.cpp b/src/Storages/MutationCommands.cpp
index f6ec277c270..aaf5c1b5d87 100644
--- a/src/Storages/MutationCommands.cpp
+++ b/src/Storages/MutationCommands.cpp
@@ -228,7 +228,7 @@ void MutationCommands::readText(ReadBuffer & in)
 
     ParserAlterCommandList p_alter_commands;
     auto commands_ast = parseQuery(
-        p_alter_commands, commands_str.data(), commands_str.data() + commands_str.length(), "mutation commands list", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+        p_alter_commands, commands_str.data(), commands_str.data() + commands_str.length(), "mutation commands list", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
 
     for (const auto & child : commands_ast->children)
     {
diff --git a/src/Storages/PostgreSQL/StorageMaterializedPostgreSQL.cpp b/src/Storages/PostgreSQL/StorageMaterializedPostgreSQL.cpp
index 64d329f74b2..f686fbda664 100644
--- a/src/Storages/PostgreSQL/StorageMaterializedPostgreSQL.cpp
+++ b/src/Storages/PostgreSQL/StorageMaterializedPostgreSQL.cpp
@@ -479,7 +479,7 @@ ASTPtr StorageMaterializedPostgreSQL::getCreateNestedTableQuery(
                     ASTPtr result;
 
                     Tokens tokens(attr.attr_def.data(), attr.attr_def.data() + attr.attr_def.size());
-                    IParser::Pos pos(tokens, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+                    IParser::Pos pos(tokens, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
                     if (!expr_parser.parse(pos, result, expected))
                     {
                         throw Exception(ErrorCodes::BAD_ARGUMENTS, "Failed to parse default expression: {}", attr.attr_def);
diff --git a/src/Storages/ProjectionsDescription.cpp b/src/Storages/ProjectionsDescription.cpp
index 08ebe3a10d0..0bcbedee41a 100644
--- a/src/Storages/ProjectionsDescription.cpp
+++ b/src/Storages/ProjectionsDescription.cpp
@@ -341,7 +341,7 @@ ProjectionsDescription ProjectionsDescription::parse(const String & str, const C
         return result;
 
     ParserProjectionDeclarationList parser;
-    ASTPtr list = parseQuery(parser, str, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+    ASTPtr list = parseQuery(parser, str, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
 
     for (const auto & projection_ast : list->children)
     {
diff --git a/src/Storages/System/StorageSystemDDLWorkerQueue.cpp b/src/Storages/System/StorageSystemDDLWorkerQueue.cpp
index ac5dd6c05d0..635686780a0 100644
--- a/src/Storages/System/StorageSystemDDLWorkerQueue.cpp
+++ b/src/Storages/System/StorageSystemDDLWorkerQueue.cpp
@@ -78,7 +78,8 @@ static String clusterNameFromDDLQuery(ContextPtr context, const DDLTask & task)
     ParserQuery parser_query(end, settings.allow_settings_after_format_in_insert);
     ASTPtr query = parseQuery(parser_query, begin, end, description,
                               settings.max_query_size,
-                              settings.max_parser_depth);
+                              settings.max_parser_depth,
+                              settings.max_parser_backtracks);
 
     String cluster_name;
     if (const auto * query_on_cluster = dynamic_cast<const ASTQueryWithOnCluster *>(query.get()))
diff --git a/src/Storages/System/attachInformationSchemaTables.cpp b/src/Storages/System/attachInformationSchemaTables.cpp
index 3482867bbf7..5afdd7a02ac 100644
--- a/src/Storages/System/attachInformationSchemaTables.cpp
+++ b/src/Storages/System/attachInformationSchemaTables.cpp
@@ -478,7 +478,7 @@ static void createInformationSchemaView(ContextMutablePtr context, IDatabase & d
         ParserCreateQuery parser;
         ASTPtr ast = parseQuery(parser, query.data(), query.data() + query.size(),
                                 "Attach query from embedded resource " + metadata_resource_name,
-                                DBMS_DEFAULT_MAX_QUERY_SIZE, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+                                DBMS_DEFAULT_MAX_QUERY_SIZE, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
 
         auto & ast_create = ast->as<ASTCreateQuery &>();
         assert(view_name == ast_create.getTable());
diff --git a/src/Storages/TTLDescription.cpp b/src/Storages/TTLDescription.cpp
index a675afbdc26..3d1ce76dff1 100644
--- a/src/Storages/TTLDescription.cpp
+++ b/src/Storages/TTLDescription.cpp
@@ -425,7 +425,7 @@ TTLTableDescription TTLTableDescription::parse(const String & str, const Columns
         return result;
 
     ParserTTLExpressionList parser;
-    ASTPtr ast = parseQuery(parser, str, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH);
+    ASTPtr ast = parseQuery(parser, str, 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
     FunctionNameNormalizer().visit(ast.get());
 
     return getTTLForTableFromAST(ast, columns, context, primary_key, context->getSettingsRef().allow_suspicious_ttl_expressions);
diff --git a/src/Storages/getStructureOfRemoteTable.cpp b/src/Storages/getStructureOfRemoteTable.cpp
index 32266f20923..c545367b63d 100644
--- a/src/Storages/getStructureOfRemoteTable.cpp
+++ b/src/Storages/getStructureOfRemoteTable.cpp
@@ -32,6 +32,7 @@ ColumnsDescription getStructureOfRemoteTableInShard(
     const ASTPtr & table_func_ptr)
 {
     String query;
+    const Settings & settings = context->getSettingsRef();
 
     if (table_func_ptr)
     {
@@ -110,7 +111,8 @@ ColumnsDescription getStructureOfRemoteTableInShard(
                 column.default_desc.kind = columnDefaultKindFromString(kind_name);
                 String expr_str = (*default_expr)[i].get<const String &>();
                 column.default_desc.expression = parseQuery(
-                    expr_parser, expr_str.data(), expr_str.data() + expr_str.size(), "default expression", 0, context->getSettingsRef().max_parser_depth);
+                    expr_parser, expr_str.data(), expr_str.data() + expr_str.size(), "default expression",
+                    0, settings.max_parser_depth, settings.max_parser_backtracks);
             }
 
             res.add(column);
diff --git a/src/Storages/tests/gtest_transform_query_for_external_database.cpp b/src/Storages/tests/gtest_transform_query_for_external_database.cpp
index 475cf5a4eae..7e2d393c3d1 100644
--- a/src/Storages/tests/gtest_transform_query_for_external_database.cpp
+++ b/src/Storages/tests/gtest_transform_query_for_external_database.cpp
@@ -118,7 +118,7 @@ static void checkOld(
     const std::string & expected)
 {
     ParserSelectQuery parser;
-    ASTPtr ast = parseQuery(parser, query, 1000, 1000);
+    ASTPtr ast = parseQuery(parser, query, 1000, 1000, 1000000);
     SelectQueryInfo query_info;
     SelectQueryOptions select_options;
     query_info.syntax_analyzer_result
@@ -161,7 +161,7 @@ static void checkNewAnalyzer(
     const std::string & expected)
 {
     ParserSelectQuery parser;
-    ASTPtr ast = parseQuery(parser, query, 1000, 1000);
+    ASTPtr ast = parseQuery(parser, query, 1000, 1000, 1000000);
 
     SelectQueryOptions select_query_options;
     auto query_tree = buildQueryTree(ast, state.context);
diff --git a/src/TableFunctions/Hive/TableFunctionHive.cpp b/src/TableFunctions/Hive/TableFunctionHive.cpp
index e840d5fc8be..80494dbe5a8 100644
--- a/src/TableFunctions/Hive/TableFunctionHive.cpp
+++ b/src/TableFunctions/Hive/TableFunctionHive.cpp
@@ -17,6 +17,7 @@
 #include <TableFunctions/TableFunctionFactory.h>
 #include <Interpreters/parseColumnsListForTableFunction.h>
 
+
 namespace DB
 {
 
@@ -99,7 +100,8 @@ StoragePtr TableFunctionHive::executeImpl(
         "(" + partition_by_def + ")",
         "partition by declaration list",
         settings.max_query_size,
-        settings.max_parser_depth);
+        settings.max_parser_depth,
+        settings.max_parser_backtracks);
     StoragePtr storage;
     storage = std::make_shared<StorageHive>(
         hive_metastore_url,
diff --git a/src/TableFunctions/TableFunctionExplain.cpp b/src/TableFunctions/TableFunctionExplain.cpp
index 400fc81e6d4..8607597fa67 100644
--- a/src/TableFunctions/TableFunctionExplain.cpp
+++ b/src/TableFunctions/TableFunctionExplain.cpp
@@ -63,7 +63,7 @@ std::vector<size_t> TableFunctionExplain::skipAnalysisForArguments(const QueryTr
     return {};
 }
 
-void TableFunctionExplain::parseArguments(const ASTPtr & ast_function, ContextPtr /*context*/)
+void TableFunctionExplain::parseArguments(const ASTPtr & ast_function, ContextPtr context)
 {
     const auto * function = ast_function->as<ASTFunction>();
     if (!function || !function->arguments)
@@ -94,12 +94,12 @@ void TableFunctionExplain::parseArguments(const ASTPtr & ast_function, ContextPt
     const auto & settings_str = settings_arg->value.get<String>();
     if (!settings_str.empty())
     {
-        constexpr UInt64 max_size = 4096;
-        constexpr UInt64 max_depth = 16;
+        const Settings & settings = context->getSettingsRef();
 
         /// parse_only_internals_ = true - we don't want to parse `SET` keyword
         ParserSetQuery settings_parser(/* parse_only_internals_ = */ true);
-        ASTPtr settings_ast = parseQuery(settings_parser, settings_str, max_size, max_depth);
+        ASTPtr settings_ast = parseQuery(settings_parser, settings_str,
+            settings.max_query_size, settings.max_parser_depth, settings.max_parser_backtracks);
         explain_query->setSettings(std::move(settings_ast));
     }
 

From f0c9fe6bc90798d3d4e402817e14c83b831663f1 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 17 Mar 2024 19:54:20 +0100
Subject: [PATCH 421/443] Limit backtracking in parser

---
 src/Parsers/IParser.cpp | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/src/Parsers/IParser.cpp b/src/Parsers/IParser.cpp
index d1e9ace89b6..5679fba1a0c 100644
--- a/src/Parsers/IParser.cpp
+++ b/src/Parsers/IParser.cpp
@@ -24,9 +24,6 @@ IParser::Pos & IParser::Pos::operator=(const IParser::Pos & rhs)
 
     TokenIterator::operator=(rhs);
 
-    if (backtracks % 1000 == 0)
-        std::cerr << backtracks << "\n";
-
     return *this;
 }
 

From 65d091cc65362f9f86e1efa5a51001a67c11b03c Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 17 Mar 2024 20:16:23 +0100
Subject: [PATCH 422/443] Limit backtracking in parser

---
 src/Functions/getFuzzerData.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Functions/getFuzzerData.cpp b/src/Functions/getFuzzerData.cpp
index 5c536477401..a6f8dd1de2c 100644
--- a/src/Functions/getFuzzerData.cpp
+++ b/src/Functions/getFuzzerData.cpp
@@ -41,7 +41,7 @@ public:
         return DataTypeString().createColumnConst(input_rows_count, fuzz_data);
     }
 
-    static void update(const String & fuzz_data_)
+    [[maybe_unused]] static void update(const String & fuzz_data_)
     {
         fuzz_data = fuzz_data_;
     }

From 13797b9712994edfe37d5bcfbca9d19d174ea95b Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 17 Mar 2024 20:21:07 +0100
Subject: [PATCH 423/443] Fix style

---
 src/Common/ErrorCodes.cpp | 1 +
 src/Parsers/IParser.cpp   | 7 ++++++-
 2 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/src/Common/ErrorCodes.cpp b/src/Common/ErrorCodes.cpp
index 9f2572cbfc6..0f01036aa06 100644
--- a/src/Common/ErrorCodes.cpp
+++ b/src/Common/ErrorCodes.cpp
@@ -584,6 +584,7 @@
     M(703, INVALID_IDENTIFIER) \
     M(704, QUERY_CACHE_USED_WITH_NONDETERMINISTIC_FUNCTIONS) \
     M(705, TABLE_NOT_EMPTY) \
+    M(706, TOO_SLOW_PARSING) \
     \
     M(900, DISTRIBUTED_CACHE_ERROR) \
     M(901, CANNOT_USE_DISTRIBUTED_CACHE) \
diff --git a/src/Parsers/IParser.cpp b/src/Parsers/IParser.cpp
index 5679fba1a0c..41981a4bb8a 100644
--- a/src/Parsers/IParser.cpp
+++ b/src/Parsers/IParser.cpp
@@ -4,6 +4,11 @@
 namespace DB
 {
 
+namespace ErrorCodes
+{
+    extern const int TOO_SLOW_PARSING;
+}
+
 IParser::Pos & IParser::Pos::operator=(const IParser::Pos & rhs)
 {
     depth = rhs.depth;
@@ -18,7 +23,7 @@ IParser::Pos & IParser::Pos::operator=(const IParser::Pos & rhs)
     {
         ++backtracks;
         if (max_backtracks && backtracks > max_backtracks)
-            throw Exception(ErrorCodes::TOO_DEEP_RECURSION, "Maximum amount of backtracking ({}) exceeded in the parser. "
+            throw Exception(ErrorCodes::TOO_SLOW_PARSING, "Maximum amount of backtracking ({}) exceeded in the parser. "
                 "Consider rising max_parser_backtracks parameter.", max_backtracks);
     }
 

From 95cfba9439b1c298fec390521510af32b6a5d66e Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 17 Mar 2024 20:26:43 +0100
Subject: [PATCH 424/443] Add a test

---
 src/Common/ErrorCodes.cpp                                | 9 ++++-----
 .../0_stateless/03012_parser_backtracking.reference      | 1 +
 tests/queries/0_stateless/03012_parser_backtracking.sh   | 7 +++++++
 3 files changed, 12 insertions(+), 5 deletions(-)
 create mode 100644 tests/queries/0_stateless/03012_parser_backtracking.reference
 create mode 100755 tests/queries/0_stateless/03012_parser_backtracking.sh

diff --git a/src/Common/ErrorCodes.cpp b/src/Common/ErrorCodes.cpp
index 0f01036aa06..75ba9cff81e 100644
--- a/src/Common/ErrorCodes.cpp
+++ b/src/Common/ErrorCodes.cpp
@@ -584,11 +584,6 @@
     M(703, INVALID_IDENTIFIER) \
     M(704, QUERY_CACHE_USED_WITH_NONDETERMINISTIC_FUNCTIONS) \
     M(705, TABLE_NOT_EMPTY) \
-    M(706, TOO_SLOW_PARSING) \
-    \
-    M(900, DISTRIBUTED_CACHE_ERROR) \
-    M(901, CANNOT_USE_DISTRIBUTED_CACHE) \
-    \
     M(706, LIBSSH_ERROR) \
     M(707, GCP_ERROR) \
     M(708, ILLEGAL_STATISTIC) \
@@ -600,6 +595,10 @@
     M(715, CANNOT_DETECT_FORMAT) \
     M(716, CANNOT_FORGET_PARTITION) \
     M(717, EXPERIMENTAL_FEATURE_ERROR) \
+    M(718, TOO_SLOW_PARSING) \
+    \
+    M(900, DISTRIBUTED_CACHE_ERROR) \
+    M(901, CANNOT_USE_DISTRIBUTED_CACHE) \
     \
     M(999, KEEPER_EXCEPTION) \
     M(1000, POCO_EXCEPTION) \
diff --git a/tests/queries/0_stateless/03012_parser_backtracking.reference b/tests/queries/0_stateless/03012_parser_backtracking.reference
new file mode 100644
index 00000000000..84727754516
--- /dev/null
+++ b/tests/queries/0_stateless/03012_parser_backtracking.reference
@@ -0,0 +1 @@
+TOO_SLOW_PARSING
diff --git a/tests/queries/0_stateless/03012_parser_backtracking.sh b/tests/queries/0_stateless/03012_parser_backtracking.sh
new file mode 100755
index 00000000000..889753fb048
--- /dev/null
+++ b/tests/queries/0_stateless/03012_parser_backtracking.sh
@@ -0,0 +1,7 @@
+#!/usr/bin/env bash
+
+CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
+# shellcheck source=../shell_config.sh
+. "$CUR_DIR"/../shell_config.sh
+
+$CLICKHOUSE_LOCAL --query "SELECT((((((((((SELECT(((((((((SELECT((((((((((SELECT(((((((((SELECT((((((((((SELECT(((((((((SELECT 1+)))))))))))))))))))))))))))))))))))))))))))))))))))))))))" 2>&1 | grep -o -F 'TOO_SLOW_PARSING'

From 58a2f780d9f224dcc5463860e986dbd9deb094b8 Mon Sep 17 00:00:00 2001
From: San <santrancisco@users.noreply.github.com>
Date: Mon, 18 Mar 2024 11:30:36 +1100
Subject: [PATCH 425/443] Update jdbc.md

We will no longer support jdbc bridge
---
 docs/en/sql-reference/table-functions/jdbc.md | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/docs/en/sql-reference/table-functions/jdbc.md b/docs/en/sql-reference/table-functions/jdbc.md
index fbc917c1e1a..a566c168794 100644
--- a/docs/en/sql-reference/table-functions/jdbc.md
+++ b/docs/en/sql-reference/table-functions/jdbc.md
@@ -6,6 +6,10 @@ sidebar_label: jdbc
 
 # jdbc
 
+:::note
+clickhouse-jdbc-bridge contains experimental codes and is no longer supported. It may contain reliability issues and security vulnerabilities. Use it at your own risk. ClickHouse recommend using built-in table functions in ClickHouse which provide a better alternative for ad-hoc querying scenarios (Postgres, MySQL, MongoDB, etc).
+:::
+
 `jdbc(datasource, schema, table)` - returns table that is connected via JDBC driver.
 
 This table function requires separate [clickhouse-jdbc-bridge](https://github.com/ClickHouse/clickhouse-jdbc-bridge) program to be running.

From 0a8b7dd9cdaf2a5df6d18811a33698c6edc4db1c Mon Sep 17 00:00:00 2001
From: San <santrancisco@users.noreply.github.com>
Date: Mon, 18 Mar 2024 11:36:39 +1100
Subject: [PATCH 426/443] Update jdbc.md

---
 docs/en/engines/table-engines/integrations/jdbc.md | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/docs/en/engines/table-engines/integrations/jdbc.md b/docs/en/engines/table-engines/integrations/jdbc.md
index a4a1e2a31ae..16ed01fecb5 100644
--- a/docs/en/engines/table-engines/integrations/jdbc.md
+++ b/docs/en/engines/table-engines/integrations/jdbc.md
@@ -6,6 +6,11 @@ sidebar_label: JDBC
 
 # JDBC
 
+:::note
+clickhouse-jdbc-bridge contains experimental codes and is no longer supported. It may contain reliability issues and security vulnerabilities. Use it at your own risk. 
+ClickHouse recommend using built-in table functions in ClickHouse which provide a better alternative for ad-hoc querying scenarios (Postgres, MySQL, MongoDB, etc).
+:::
+
 Allows ClickHouse to connect to external databases via [JDBC](https://en.wikipedia.org/wiki/Java_Database_Connectivity).
 
 To implement the JDBC connection, ClickHouse uses the separate program [clickhouse-jdbc-bridge](https://github.com/ClickHouse/clickhouse-jdbc-bridge) that should run as a daemon.

From 973b2781282e8e6fb7900cf404abe94142e649ff Mon Sep 17 00:00:00 2001
From: San <santrancisco@users.noreply.github.com>
Date: Mon, 18 Mar 2024 11:37:21 +1100
Subject: [PATCH 427/443] Update jdbc.md

---
 docs/en/sql-reference/table-functions/jdbc.md | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/docs/en/sql-reference/table-functions/jdbc.md b/docs/en/sql-reference/table-functions/jdbc.md
index a566c168794..6b801344a83 100644
--- a/docs/en/sql-reference/table-functions/jdbc.md
+++ b/docs/en/sql-reference/table-functions/jdbc.md
@@ -7,7 +7,8 @@ sidebar_label: jdbc
 # jdbc
 
 :::note
-clickhouse-jdbc-bridge contains experimental codes and is no longer supported. It may contain reliability issues and security vulnerabilities. Use it at your own risk. ClickHouse recommend using built-in table functions in ClickHouse which provide a better alternative for ad-hoc querying scenarios (Postgres, MySQL, MongoDB, etc).
+clickhouse-jdbc-bridge contains experimental codes and is no longer supported. It may contain reliability issues and security vulnerabilities. Use it at your own risk. 
+ClickHouse recommend using built-in table functions in ClickHouse which provide a better alternative for ad-hoc querying scenarios (Postgres, MySQL, MongoDB, etc).
 :::
 
 `jdbc(datasource, schema, table)` - returns table that is connected via JDBC driver.

From c17536c3b31ca48cdd1771266d8385628de622e5 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Thu, 14 Mar 2024 12:16:33 +0000
Subject: [PATCH 428/443] Enable clang-tidy in headers

---
 .clang-tidy                                   |  6 +-
 .../Poco/Net/HTTPRequestHandlerFactory.h      |  1 -
 contrib/libmetrohash/src/metrohash64.cpp      | 64 +++++++++----------
 contrib/libmetrohash/src/metrohash64.h        | 10 +--
 src/Access/AccessBackup.cpp                   |  4 +-
 src/Access/AccessBackup.h                     |  2 +-
 src/Access/ContextAccessParams.cpp            |  2 +-
 src/Access/ContextAccessParams.h              |  2 +-
 src/Access/GSSAcceptor.h                      |  2 +-
 src/Access/LDAPAccessStorage.cpp              |  2 +-
 src/Access/LDAPAccessStorage.h                | 22 +++----
 .../AggregateFunctionFlameGraph.cpp           |  2 -
 .../AggregateFunctionMLMethod.h               |  6 +-
 .../AggregateFunctionStatisticsSimple.h       |  5 --
 .../AggregateFunctionUniqCombined.h           |  5 --
 .../Combinators/AggregateFunctionNull.h       |  2 +-
 src/AggregateFunctions/DDSketch/Store.h       |  2 +-
 src/AggregateFunctions/QuantileExact.h        |  8 +--
 .../ReservoirSamplerDeterministic.h           |  2 +-
 src/AggregateFunctions/SingleValueData.h      |  2 +-
 src/AggregateFunctions/UniqExactSet.h         |  4 +-
 src/Analyzer/Passes/QueryAnalysisPass.h       |  2 +-
 src/Backups/BackupSettings.h                  |  2 +-
 src/Backups/IBackup.h                         |  8 +--
 src/Backups/RestorerFromBackup.cpp            |  8 +--
 src/Client/HedgedConnectionsFactory.h         |  2 +-
 src/Client/TestHint.h                         |  6 +-
 src/Columns/ColumnUnique.h                    |  2 +-
 src/Columns/ReverseIndex.h                    |  2 +-
 src/Common/AllocatorWithMemoryTracking.h      |  6 +-
 src/Common/Arena.h                            |  6 +-
 src/Common/FrequencyHolder.h                  |  1 -
 src/Common/HTTPConnectionPool.h               | 10 +--
 src/Common/HashTable/HashTable.h              |  5 +-
 src/Common/HostResolvePool.h                  | 15 +++--
 src/Common/MortonUtils.h                      |  4 +-
 src/Common/NetException.h                     |  2 +-
 src/Common/PODArray.h                         |  4 ++
 src/Common/ProxyConfiguration.h               |  2 +-
 .../Scheduler/Nodes/DynamicResourceManager.h  |  1 -
 src/Common/Scheduler/Nodes/FairPolicy.h       |  1 -
 src/Common/Scheduler/SchedulerRoot.h          |  2 -
 src/Common/ZooKeeper/ZooKeeperImpl.h          |  2 +-
 src/Common/ZooKeeper/ZooKeeperRetries.h       |  2 +-
 src/Common/memory.h                           |  2 +-
 src/Common/mysqlxx/mysqlxx/Transaction.h      |  2 +-
 src/Common/mysqlxx/mysqlxx/Types.h            |  2 +-
 src/Common/threadPoolCallbackRunner.h         |  2 +-
 src/Compression/CompressionCodecDeflateQpl.h  |  2 +-
 .../WriteBufferFromNuraftBuffer.h             |  2 +-
 src/Core/Field.h                              |  2 +
 src/Core/MySQL/MySQLReplication.h             |  2 +-
 src/Core/Settings.h                           |  2 -
 src/Core/tests/gtest_charset_conv.cpp         |  2 -
 src/Daemon/BaseDaemon.h                       |  4 +-
 src/DataTypes/NumberTraits.h                  |  2 +
 .../SerializationCustomSimpleText.h           |  2 +-
 .../Serializations/SerializationSparse.h      |  2 +-
 .../SerializationVariantElement.cpp           |  4 +-
 .../SerializationVariantElement.h             |  4 +-
 src/Databases/DatabasesOverlay.h              |  1 -
 src/Dictionaries/CacheDictionary.cpp          |  2 +-
 src/Dictionaries/CacheDictionary.h            |  2 +-
 src/Dictionaries/CacheDictionaryStorage.h     |  4 +-
 src/Dictionaries/ICacheDictionaryStorage.h    |  4 +-
 src/Dictionaries/SSDCacheDictionaryStorage.h  |  4 +-
 src/Disks/IDiskTransaction.h                  |  4 +-
 src/Disks/IO/IOUringReader.h                  |  2 +-
 src/Disks/IVolume.h                           |  2 +-
 .../ObjectStorages/DiskObjectStorage.cpp      |  2 +-
 .../ObjectStorages/ObjectStorageIterator.h    |  2 +-
 src/Disks/ObjectStorages/S3/diskSettings.h    |  8 ---
 src/Functions/CountSubstringsImpl.h           |  2 +-
 src/Functions/CustomWeekTransforms.h          |  2 +-
 src/Functions/FunctionConstantBase.h          |  5 +-
 src/Functions/FunctionsAES.h                  |  4 +-
 src/Functions/FunctionsBitmap.h               |  4 +-
 src/Functions/FunctionsComparison.h           |  2 +-
 src/Functions/FunctionsExternalDictionaries.h |  4 +-
 src/Functions/FunctionsTimeWindow.h           |  4 ++
 src/Functions/GatherUtils/Selectors.h         | 14 ++--
 src/Functions/GatherUtils/Sources.h           |  4 ++
 src/Functions/GregorianDate.h                 |  4 +-
 src/Functions/HasSubsequenceImpl.h            |  4 --
 .../JSONPath/Generator/GeneratorJSONPath.h    |  2 +-
 src/Functions/LowerUpperUTF8Impl.h            |  2 +-
 src/Functions/URL/domain.h                    |  4 +-
 src/Functions/UniqTheta/FunctionsUniqTheta.h  |  4 +-
 .../UserDefinedSQLObjectsBackup.cpp           |  2 +-
 src/Functions/geometryConverters.h            |  5 --
 src/Functions/stl.hpp                         | 58 ++++++++---------
 src/IO/Archives/IArchiveWriter.h              |  4 +-
 src/IO/BrotliWriteBuffer.h                    |  4 +-
 src/IO/Bzip2WriteBuffer.h                     |  4 +-
 src/IO/ConnectionTimeouts.h                   |  2 +
 src/IO/HTTPCommon.h                           |  2 +-
 src/IO/LZMADeflatingWriteBuffer.h             |  4 +-
 src/IO/Lz4DeflatingWriteBuffer.h              |  4 +-
 src/IO/ReadBufferFromFileBase.h               |  2 +-
 src/IO/ReadSettings.h                         |  1 -
 src/IO/ReadWriteBufferFromHTTP.h              |  4 +-
 src/IO/S3/PocoHTTPClient.h                    |  2 +-
 src/IO/S3/tests/TestPocoHTTPServer.h          | 10 ++-
 src/IO/SeekableReadBuffer.h                   |  2 +-
 src/IO/WriteBufferFromArena.h                 |  2 +-
 src/IO/ZlibDeflatingWriteBuffer.h             |  8 +--
 src/IO/ZstdDeflatingWriteBuffer.h             |  4 +-
 src/IO/readDecimalText.h                      |  2 +-
 src/Interpreters/AddDefaultDatabaseVisitor.h  |  2 +-
 .../AggregateFunctionOfGroupByKeysVisitor.h   |  4 +-
 src/Interpreters/AggregationCommon.h          |  2 +-
 src/Interpreters/ArrayJoinedColumnsVisitor.h  | 10 +--
 src/Interpreters/AsynchronousInsertQueue.h    |  2 +-
 src/Interpreters/BackupLog.h                  |  2 +-
 src/Interpreters/Context.h                    |  2 -
 src/Interpreters/DirectJoin.h                 | 22 +++----
 src/Interpreters/GetAggregatesVisitor.h       |  4 +-
 src/Interpreters/GroupByFunctionKeysVisitor.h |  4 +-
 src/Interpreters/HashJoin.h                   |  2 +
 src/Interpreters/MonotonicityCheckVisitor.h   |  2 +-
 .../RedundantFunctionsInOrderByVisitor.h      |  2 +-
 src/Interpreters/SessionLog.h                 |  2 +-
 src/Interpreters/SetVariants.h                |  4 +-
 .../TranslateQualifiedNamesVisitor.h          |  2 +-
 src/Interpreters/TreeCNFConverter.h           |  2 +-
 src/Interpreters/joinDispatch.h               |  3 +-
 src/Loggers/OwnFormattingChannel.h            |  4 +-
 .../Executors/PullingAsyncPipelineExecutor.h  |  1 -
 src/Processors/Formats/IRowOutputFormat.h     |  6 +-
 .../Formats/Impl/ArrowBufferedStreams.cpp     |  2 +-
 .../Formats/Impl/AvroRowInputFormat.h         |  2 +-
 .../Formats/Impl/AvroRowOutputFormat.h        |  8 +--
 .../Formats/Impl/ConstantExpressionTemplate.h |  1 -
 .../Impl/ParallelFormattingOutputFormat.h     |  2 +-
 .../Formats/Impl/ParquetBlockOutputFormat.h   |  4 +-
 .../Impl/TabSeparatedRowOutputFormat.h        |  6 +-
 .../Formats/RowInputFormatWithNamesAndTypes.h |  2 +-
 .../Algorithms/AggregatingSortedAlgorithm.h   |  2 +-
 .../Merges/Algorithms/IMergingAlgorithm.h     |  2 +-
 src/Processors/Merges/Algorithms/RowRef.h     |  2 +-
 .../Transforms/MergeJoinTransform.h           |  6 +-
 .../Transforms/PasteJoinTransform.h           |  6 +-
 src/QueryPipeline/QueryPipelineBuilder.h      |  2 +-
 src/Server/GRPCServer.cpp                     |  2 -
 src/Server/HTTP/HTTPServerResponse.h          |  6 +-
 src/Server/HTTPHandler.h                      |  2 +-
 src/Storages/DataLakes/S3MetadataReader.h     |  4 +-
 src/Storages/FileLog/FileLogSource.h          |  2 +-
 src/Storages/Hive/HiveFile.h                  |  2 +-
 src/Storages/IStorage.cpp                     |  2 +-
 src/Storages/Kafka/KafkaConsumer.h            |  2 +-
 src/Storages/Kafka/parseSyslogLevel.cpp       |  2 +-
 src/Storages/Kafka/parseSyslogLevel.h         |  2 +-
 src/Storages/MergeTree/ColumnSizeEstimator.h  |  4 +-
 src/Storages/MergeTree/IMergeTreeDataPart.h   |  1 -
 src/Storages/MergeTree/LeaderElection.h       |  2 +-
 src/Storages/MergeTree/MarkRange.h            |  2 +-
 src/Storages/MergeTree/MergeTreeData.cpp      |  4 +-
 src/Storages/MergeTree/MergeTreeData.h        |  2 +-
 .../MergeTree/MergeTreeMutationStatus.h       |  2 +
 src/Storages/MergeTree/RPNBuilder.h           |  2 +-
 src/Storages/MergeTree/RangesInDataPart.h     |  2 +-
 .../ReplicatedMergeTreeQuorumAddedParts.h     |  2 +-
 src/Storages/MergeTree/checkDataPart.cpp      |  2 +-
 src/Storages/MergeTree/checkDataPart.h        |  2 +-
 src/Storages/RabbitMQ/RabbitMQProducer.cpp    |  6 +-
 src/Storages/RabbitMQ/RabbitMQProducer.h      |  6 +-
 src/Storages/S3Queue/StorageS3Queue.h         |  5 --
 src/Storages/StorageInMemoryMetadata.h        |  4 +-
 src/Storages/StorageMongoDBSocketFactory.h    |  2 +-
 src/Storages/StorageS3.h                      |  5 --
 src/Storages/StorageView.cpp                  |  2 +-
 src/Storages/StorageView.h                    |  4 +-
 src/Storages/System/StorageSystemPartsBase.h  |  2 +-
 src/Storages/System/StorageSystemTimeZones.h  |  1 -
 175 files changed, 349 insertions(+), 388 deletions(-)

diff --git a/.clang-tidy b/.clang-tidy
index 4aeb38ca409..e7900cbe2aa 100644
--- a/.clang-tidy
+++ b/.clang-tidy
@@ -5,9 +5,7 @@
 # a) the new check is not controversial (this includes many checks in readability-* and google-*) or
 # b) too noisy (checks with > 100 new warnings are considered noisy, this includes e.g. cppcoreguidelines-*).
 
-# TODO Let clang-tidy check headers in further directories
-#      --> HeaderFilterRegex: '^.*/(src|base|programs|utils)/.*(h|hpp)$'
-HeaderFilterRegex: '^.*/(base|programs|utils)/.*(h|hpp)$'
+HeaderFilterRegex: '^.*/(base|src|programs|utils)/.*(h|hpp)$'
 
 Checks: [
   '*',
@@ -37,6 +35,8 @@ Checks: [
   '-cert-oop54-cpp',
   '-cert-oop57-cpp',
 
+  '-clang-analyzer-optin.performance.Padding',
+
   '-clang-analyzer-unix.Malloc',
 
   '-cppcoreguidelines-*', # impractical in a codebase as large as ClickHouse, also slow
diff --git a/base/poco/Net/include/Poco/Net/HTTPRequestHandlerFactory.h b/base/poco/Net/include/Poco/Net/HTTPRequestHandlerFactory.h
index 029d936769b..9bc35b7ff70 100644
--- a/base/poco/Net/include/Poco/Net/HTTPRequestHandlerFactory.h
+++ b/base/poco/Net/include/Poco/Net/HTTPRequestHandlerFactory.h
@@ -30,7 +30,6 @@ namespace Net
 
 
     class HTTPServerRequest;
-    class HTTPServerResponse;
     class HTTPRequestHandler;
 
 
diff --git a/contrib/libmetrohash/src/metrohash64.cpp b/contrib/libmetrohash/src/metrohash64.cpp
index 7b5ec7f1a42..6ff64027292 100644
--- a/contrib/libmetrohash/src/metrohash64.cpp
+++ b/contrib/libmetrohash/src/metrohash64.cpp
@@ -26,13 +26,13 @@ const uint8_t MetroHash64::test_seed_1[8] =   { 0x3B, 0x0D, 0x48, 0x1C, 0xF4, 0x
 
 
 
-MetroHash64::MetroHash64(const uint64_t seed)
+MetroHash64::MetroHash64(uint64_t seed)
 {
     Initialize(seed);
 }
 
 
-void MetroHash64::Initialize(const uint64_t seed)
+void MetroHash64::Initialize(uint64_t seed)
 {
     vseed = (static_cast<uint64_t>(seed) + k2) * k0;
 
@@ -47,7 +47,7 @@ void MetroHash64::Initialize(const uint64_t seed)
 }
 
 
-void MetroHash64::Update(const uint8_t * const buffer, const uint64_t length)
+void MetroHash64::Update(const uint8_t * const buffer, uint64_t length)
 {
     const uint8_t * ptr = reinterpret_cast<const uint8_t*>(buffer);
     const uint8_t * const end = ptr + length;
@@ -62,7 +62,7 @@ void MetroHash64::Update(const uint8_t * const buffer, const uint64_t length)
         memcpy(input.b + (bytes % 32), ptr, static_cast<size_t>(fill));
         ptr   += fill;
         bytes += fill;
-        
+
         // input buffer is still partially filled
         if ((bytes % 32) != 0) return;
 
@@ -72,7 +72,7 @@ void MetroHash64::Update(const uint8_t * const buffer, const uint64_t length)
         state.v[2] += read_u64(&input.b[16]) * k2; state.v[2] = rotate_right(state.v[2],29) + state.v[0];
         state.v[3] += read_u64(&input.b[24]) * k3; state.v[3] = rotate_right(state.v[3],29) + state.v[1];
     }
-    
+
     // bulk update
     bytes += static_cast<uint64_t>(end - ptr);
     while (ptr <= (end - 32))
@@ -83,14 +83,14 @@ void MetroHash64::Update(const uint8_t * const buffer, const uint64_t length)
         state.v[2] += read_u64(ptr) * k2; ptr += 8; state.v[2] = rotate_right(state.v[2],29) + state.v[0];
         state.v[3] += read_u64(ptr) * k3; ptr += 8; state.v[3] = rotate_right(state.v[3],29) + state.v[1];
     }
-    
+
     // store remaining bytes in input buffer
     if (ptr < end)
         memcpy(input.b, ptr, static_cast<size_t>(end - ptr));
 }
 
 
-void MetroHash64::Finalize(uint8_t * const hash)
+void MetroHash64::Finalize(uint8_t * hash)
 {
     // finalize bulk loop, if used
     if (bytes >= 32)
@@ -102,11 +102,11 @@ void MetroHash64::Finalize(uint8_t * const hash)
 
         state.v[0] = vseed + (state.v[0] ^ state.v[1]);
     }
-    
+
     // process any bytes remaining in the input buffer
     const uint8_t * ptr = reinterpret_cast<const uint8_t*>(input.b);
     const uint8_t * const end = ptr + (bytes % 32);
-    
+
     if ((end - ptr) >= 16)
     {
         state.v[1]  = state.v[0] + (read_u64(ptr) * k2); ptr += 8; state.v[1] = rotate_right(state.v[1],29) * k3;
@@ -139,7 +139,7 @@ void MetroHash64::Finalize(uint8_t * const hash)
         state.v[0] += read_u8 (ptr) * k3;
         state.v[0] ^= rotate_right(state.v[0], 37) * k1;
     }
-    
+
     state.v[0] ^= rotate_right(state.v[0], 28);
     state.v[0] *= k0;
     state.v[0] ^= rotate_right(state.v[0], 29);
@@ -152,7 +152,7 @@ void MetroHash64::Finalize(uint8_t * const hash)
 }
 
 
-void MetroHash64::Hash(const uint8_t * buffer, const uint64_t length, uint8_t * const hash, const uint64_t seed)
+void MetroHash64::Hash(const uint8_t * buffer, uint64_t length, uint8_t * const hash, uint64_t seed)
 {
     const uint8_t * ptr = reinterpret_cast<const uint8_t*>(buffer);
     const uint8_t * const end = ptr + length;
@@ -238,7 +238,7 @@ bool MetroHash64::ImplementationVerified()
 
     // verify incremental implementation
     MetroHash64 metro;
-    
+
     metro.Initialize(0);
     metro.Update(reinterpret_cast<const uint8_t *>(MetroHash64::test_string), strlen(MetroHash64::test_string));
     metro.Finalize(hash);
@@ -262,9 +262,9 @@ void metrohash64_1(const uint8_t * key, uint64_t len, uint32_t seed, uint8_t * o
 
     const uint8_t * ptr = reinterpret_cast<const uint8_t*>(key);
     const uint8_t * const end = ptr + len;
-    
+
     uint64_t hash = ((static_cast<uint64_t>(seed) + k2) * k0) + len;
-    
+
     if (len >= 32)
     {
         uint64_t v[4];
@@ -272,7 +272,7 @@ void metrohash64_1(const uint8_t * key, uint64_t len, uint32_t seed, uint8_t * o
         v[1] = hash;
         v[2] = hash;
         v[3] = hash;
-        
+
         do
         {
             v[0] += read_u64(ptr) * k0; ptr += 8; v[0] = rotate_right(v[0],29) + v[2];
@@ -288,7 +288,7 @@ void metrohash64_1(const uint8_t * key, uint64_t len, uint32_t seed, uint8_t * o
         v[1] ^= rotate_right(((v[1] + v[3]) * k1) + v[2], 33) * k0;
         hash += v[0] ^ v[1];
     }
-    
+
     if ((end - ptr) >= 16)
     {
         uint64_t v0 = hash + (read_u64(ptr) * k0); ptr += 8; v0 = rotate_right(v0,33) * k1;
@@ -297,32 +297,32 @@ void metrohash64_1(const uint8_t * key, uint64_t len, uint32_t seed, uint8_t * o
         v1 ^= rotate_right(v1 * k3, 35) + v0;
         hash += v1;
     }
-    
+
     if ((end - ptr) >= 8)
     {
         hash += read_u64(ptr) * k3; ptr += 8;
         hash ^= rotate_right(hash, 33) * k1;
-        
+
     }
-    
+
     if ((end - ptr) >= 4)
     {
         hash += read_u32(ptr) * k3; ptr += 4;
         hash ^= rotate_right(hash, 15) * k1;
     }
-    
+
     if ((end - ptr) >= 2)
     {
         hash += read_u16(ptr) * k3; ptr += 2;
         hash ^= rotate_right(hash, 13) * k1;
     }
-    
+
     if ((end - ptr) >= 1)
     {
         hash += read_u8 (ptr) * k3;
         hash ^= rotate_right(hash, 25) * k1;
     }
-    
+
     hash ^= rotate_right(hash, 33);
     hash *= k0;
     hash ^= rotate_right(hash, 33);
@@ -336,13 +336,13 @@ void metrohash64_2(const uint8_t * key, uint64_t len, uint32_t seed, uint8_t * o
     static const uint64_t k0 = 0xD6D018F5;
     static const uint64_t k1 = 0xA2AA033B;
     static const uint64_t k2 = 0x62992FC1;
-    static const uint64_t k3 = 0x30BC5B29; 
+    static const uint64_t k3 = 0x30BC5B29;
 
     const uint8_t * ptr = reinterpret_cast<const uint8_t*>(key);
     const uint8_t * const end = ptr + len;
-    
+
     uint64_t hash = ((static_cast<uint64_t>(seed) + k2) * k0) + len;
-    
+
     if (len >= 32)
     {
         uint64_t v[4];
@@ -350,7 +350,7 @@ void metrohash64_2(const uint8_t * key, uint64_t len, uint32_t seed, uint8_t * o
         v[1] = hash;
         v[2] = hash;
         v[3] = hash;
-        
+
         do
         {
             v[0] += read_u64(ptr) * k0; ptr += 8; v[0] = rotate_right(v[0],29) + v[2];
@@ -366,7 +366,7 @@ void metrohash64_2(const uint8_t * key, uint64_t len, uint32_t seed, uint8_t * o
         v[1] ^= rotate_right(((v[1] + v[3]) * k1) + v[2], 30) * k0;
         hash += v[0] ^ v[1];
     }
-    
+
     if ((end - ptr) >= 16)
     {
         uint64_t v0 = hash + (read_u64(ptr) * k2); ptr += 8; v0 = rotate_right(v0,29) * k3;
@@ -375,31 +375,31 @@ void metrohash64_2(const uint8_t * key, uint64_t len, uint32_t seed, uint8_t * o
         v1 ^= rotate_right(v1 * k3, 34) + v0;
         hash += v1;
     }
-    
+
     if ((end - ptr) >= 8)
     {
         hash += read_u64(ptr) * k3; ptr += 8;
         hash ^= rotate_right(hash, 36) * k1;
     }
-    
+
     if ((end - ptr) >= 4)
     {
         hash += read_u32(ptr) * k3; ptr += 4;
         hash ^= rotate_right(hash, 15) * k1;
     }
-    
+
     if ((end - ptr) >= 2)
     {
         hash += read_u16(ptr) * k3; ptr += 2;
         hash ^= rotate_right(hash, 15) * k1;
     }
-    
+
     if ((end - ptr) >= 1)
     {
         hash += read_u8 (ptr) * k3;
         hash ^= rotate_right(hash, 23) * k1;
     }
-    
+
     hash ^= rotate_right(hash, 28);
     hash *= k0;
     hash ^= rotate_right(hash, 29);
diff --git a/contrib/libmetrohash/src/metrohash64.h b/contrib/libmetrohash/src/metrohash64.h
index 911e54e6863..7003a1848be 100644
--- a/contrib/libmetrohash/src/metrohash64.h
+++ b/contrib/libmetrohash/src/metrohash64.h
@@ -25,24 +25,24 @@ public:
     static const uint32_t bits = 64;
 
     // Constructor initializes the same as Initialize()
-    explicit MetroHash64(const uint64_t seed=0);
+    explicit MetroHash64(uint64_t seed=0);
 
     // Initializes internal state for new hash with optional seed
-    void Initialize(const uint64_t seed=0);
+    void Initialize(uint64_t seed=0);
 
     // Update the hash state with a string of bytes. If the length
     // is sufficiently long, the implementation switches to a bulk
     // hashing algorithm directly on the argument buffer for speed.
-    void Update(const uint8_t * buffer, const uint64_t length);
+    void Update(const uint8_t * buffer, uint64_t length);
 
     // Constructs the final hash and writes it to the argument buffer.
     // After a hash is finalized, this instance must be Initialized()-ed
     // again or the behavior of Update() and Finalize() is undefined.
-    void Finalize(uint8_t * const hash);
+    void Finalize(uint8_t * hash);
 
     // A non-incremental function implementation. This can be significantly
     // faster than the incremental implementation for some usage patterns.
-    static void Hash(const uint8_t * buffer, const uint64_t length, uint8_t * const hash, const uint64_t seed=0);
+    static void Hash(const uint8_t * buffer, uint64_t length, uint8_t * hash, uint64_t seed=0);
 
     // Does implementation correctly execute test vectors?
     static bool ImplementationVerified();
diff --git a/src/Access/AccessBackup.cpp b/src/Access/AccessBackup.cpp
index 800a54e69b3..ba89899dd8f 100644
--- a/src/Access/AccessBackup.cpp
+++ b/src/Access/AccessBackup.cpp
@@ -297,7 +297,7 @@ namespace
 
 
 std::pair<String, BackupEntryPtr> makeBackupEntryForAccess(
-    const std::vector<std::pair<UUID, AccessEntityPtr>> access_entities,
+    const std::vector<std::pair<UUID, AccessEntityPtr>> & access_entities,
     const String & data_path_in_backup,
     size_t counter,
     const AccessControl & access_control)
@@ -326,7 +326,7 @@ void AccessRestorerFromBackup::addDataPath(const String & data_path)
         return;
 
     fs::path data_path_in_backup_fs = data_path;
-    Strings filenames = backup->listFiles(data_path);
+    Strings filenames = backup->listFiles(data_path, /*recursive*/ false);
     if (filenames.empty())
         return;
 
diff --git a/src/Access/AccessBackup.h b/src/Access/AccessBackup.h
index 99f22b6c6a8..1aaac589ccb 100644
--- a/src/Access/AccessBackup.h
+++ b/src/Access/AccessBackup.h
@@ -21,7 +21,7 @@ struct RestoreSettings;
 
 /// Makes a backup of access entities of a specified type.
 std::pair<String, BackupEntryPtr> makeBackupEntryForAccess(
-    const std::vector<std::pair<UUID, AccessEntityPtr>> access_entities,
+    const std::vector<std::pair<UUID, AccessEntityPtr>> & access_entities,
     const String & data_path_in_backup,
     size_t counter,
     const AccessControl & access_control);
diff --git a/src/Access/ContextAccessParams.cpp b/src/Access/ContextAccessParams.cpp
index ec839a37b1a..07374e9bbf5 100644
--- a/src/Access/ContextAccessParams.cpp
+++ b/src/Access/ContextAccessParams.cpp
@@ -7,7 +7,7 @@ namespace DB
 {
 
 ContextAccessParams::ContextAccessParams(
-    const std::optional<UUID> user_id_,
+    std::optional<UUID> user_id_,
     bool full_access_,
     bool use_default_roles_,
     const std::shared_ptr<const std::vector<UUID>> & current_roles_,
diff --git a/src/Access/ContextAccessParams.h b/src/Access/ContextAccessParams.h
index 8b68fa44ed4..07503a3af6d 100644
--- a/src/Access/ContextAccessParams.h
+++ b/src/Access/ContextAccessParams.h
@@ -15,7 +15,7 @@ class ContextAccessParams
 {
 public:
     ContextAccessParams(
-        const std::optional<UUID> user_id_,
+        std::optional<UUID> user_id_,
         bool full_access_,
         bool use_default_roles_,
         const std::shared_ptr<const std::vector<UUID>> & current_roles_,
diff --git a/src/Access/GSSAcceptor.h b/src/Access/GSSAcceptor.h
index 8d490fb47ae..225e03056ac 100644
--- a/src/Access/GSSAcceptor.h
+++ b/src/Access/GSSAcceptor.h
@@ -34,7 +34,7 @@ public:
     };
 
     explicit GSSAcceptorContext(const Params & params_);
-    virtual ~GSSAcceptorContext() override;
+    ~GSSAcceptorContext() override;
 
     GSSAcceptorContext(const GSSAcceptorContext &) = delete;
     GSSAcceptorContext(GSSAcceptorContext &&) = delete;
diff --git a/src/Access/LDAPAccessStorage.cpp b/src/Access/LDAPAccessStorage.cpp
index c271cd39845..b1f10cd2d49 100644
--- a/src/Access/LDAPAccessStorage.cpp
+++ b/src/Access/LDAPAccessStorage.cpp
@@ -204,7 +204,7 @@ void LDAPAccessStorage::assignRolesNoLock(User & user, const LDAPClient::SearchR
 }
 
 
-void LDAPAccessStorage::assignRolesNoLock(User & user, const LDAPClient::SearchResultsList & external_roles, const std::size_t external_roles_hash) const
+void LDAPAccessStorage::assignRolesNoLock(User & user, const LDAPClient::SearchResultsList & external_roles, std::size_t external_roles_hash) const
 {
     const auto & user_name = user.getName();
     auto & granted_roles = user.granted_roles;
diff --git a/src/Access/LDAPAccessStorage.h b/src/Access/LDAPAccessStorage.h
index b01f58c724a..d5da64e4111 100644
--- a/src/Access/LDAPAccessStorage.h
+++ b/src/Access/LDAPAccessStorage.h
@@ -33,29 +33,29 @@ public:
     static constexpr char STORAGE_TYPE[] = "ldap";
 
     explicit LDAPAccessStorage(const String & storage_name_, AccessControl & access_control_, const Poco::Util::AbstractConfiguration & config, const String & prefix);
-    virtual ~LDAPAccessStorage() override = default;
+    ~LDAPAccessStorage() override = default;
 
     String getLDAPServerName() const;
 
     // IAccessStorage implementations.
-    virtual const char * getStorageType() const override;
-    virtual String getStorageParamsJSON() const override;
-    virtual bool isReadOnly() const override { return true; }
-    virtual bool exists(const UUID & id) const override;
+    const char * getStorageType() const override;
+    String getStorageParamsJSON() const override;
+    bool isReadOnly() const override { return true; }
+    bool exists(const UUID & id) const override;
 
 private: // IAccessStorage implementations.
-    virtual std::optional<UUID> findImpl(AccessEntityType type, const String & name) const override;
-    virtual std::vector<UUID> findAllImpl(AccessEntityType type) const override;
-    virtual AccessEntityPtr readImpl(const UUID & id, bool throw_if_not_exists) const override;
-    virtual std::optional<std::pair<String, AccessEntityType>> readNameWithTypeImpl(const UUID & id, bool throw_if_not_exists) const override;
-    virtual std::optional<AuthResult> authenticateImpl(const Credentials & credentials, const Poco::Net::IPAddress & address, const ExternalAuthenticators & external_authenticators, bool throw_if_user_not_exists, bool allow_no_password, bool allow_plaintext_password) const override;
+    std::optional<UUID> findImpl(AccessEntityType type, const String & name) const override;
+    std::vector<UUID> findAllImpl(AccessEntityType type) const override;
+    AccessEntityPtr readImpl(const UUID & id, bool throw_if_not_exists) const override;
+    std::optional<std::pair<String, AccessEntityType>> readNameWithTypeImpl(const UUID & id, bool throw_if_not_exists) const override;
+    std::optional<AuthResult> authenticateImpl(const Credentials & credentials, const Poco::Net::IPAddress & address, const ExternalAuthenticators & external_authenticators, bool throw_if_user_not_exists, bool allow_no_password, bool allow_plaintext_password) const override;
 
     void setConfiguration(const Poco::Util::AbstractConfiguration & config, const String & prefix);
     void processRoleChange(const UUID & id, const AccessEntityPtr & entity);
 
     void applyRoleChangeNoLock(bool grant, const UUID & role_id, const String & role_name);
     void assignRolesNoLock(User & user, const LDAPClient::SearchResultsList & external_roles) const;
-    void assignRolesNoLock(User & user, const LDAPClient::SearchResultsList & external_roles, const std::size_t external_roles_hash) const;
+    void assignRolesNoLock(User & user, const LDAPClient::SearchResultsList & external_roles, std::size_t external_roles_hash) const;
     void updateAssignedRolesNoLock(const UUID & id, const String & user_name, const LDAPClient::SearchResultsList & external_roles) const;
     std::set<String> mapExternalRolesNoLock(const LDAPClient::SearchResultsList & external_roles) const;
     bool areLDAPCredentialsValidNoLock(const User & user, const Credentials & credentials,
diff --git a/src/AggregateFunctions/AggregateFunctionFlameGraph.cpp b/src/AggregateFunctions/AggregateFunctionFlameGraph.cpp
index f6aafd9a9b1..f3d99046036 100644
--- a/src/AggregateFunctions/AggregateFunctionFlameGraph.cpp
+++ b/src/AggregateFunctions/AggregateFunctionFlameGraph.cpp
@@ -252,7 +252,6 @@ void dumpFlameGraph(
     fillColumn(chars, offsets, out.str());
 }
 
-// NOLINTBEGIN(clang-analyzer-optin.performance.Padding)
 struct AggregateFunctionFlameGraphData
 {
     struct Entry
@@ -469,7 +468,6 @@ struct AggregateFunctionFlameGraphData
         DB::dumpFlameGraph(tree.dump(max_depth, min_bytes), chars, offsets);
     }
 };
-// NOLINTEND(clang-analyzer-optin.performance.Padding)
 
 /// Aggregate function which builds a flamegraph using the list of stacktraces.
 /// The output is an array of strings which can be used by flamegraph.pl util.
diff --git a/src/AggregateFunctions/AggregateFunctionMLMethod.h b/src/AggregateFunctions/AggregateFunctionMLMethod.h
index fd50fe4b28c..cfb8550d829 100644
--- a/src/AggregateFunctions/AggregateFunctionMLMethod.h
+++ b/src/AggregateFunctions/AggregateFunctionMLMethod.h
@@ -157,7 +157,7 @@ public:
 
     void update(UInt64 batch_size, std::vector<Float64> & weights, Float64 & bias, Float64 learning_rate, const std::vector<Float64> & batch_gradient) override;
 
-    virtual void merge(const IWeightsUpdater & rhs, Float64 frac, Float64 rhs_frac) override;
+    void merge(const IWeightsUpdater & rhs, Float64 frac, Float64 rhs_frac) override;
 
     void write(WriteBuffer & buf) const override;
 
@@ -189,7 +189,7 @@ public:
 
     void update(UInt64 batch_size, std::vector<Float64> & weights, Float64 & bias, Float64 learning_rate, const std::vector<Float64> & batch_gradient) override;
 
-    virtual void merge(const IWeightsUpdater & rhs, Float64 frac, Float64 rhs_frac) override;
+    void merge(const IWeightsUpdater & rhs, Float64 frac, Float64 rhs_frac) override;
 
     void write(WriteBuffer & buf) const override;
 
@@ -226,7 +226,7 @@ public:
 
     void update(UInt64 batch_size, std::vector<Float64> & weights, Float64 & bias, Float64 learning_rate, const std::vector<Float64> & batch_gradient) override;
 
-    virtual void merge(const IWeightsUpdater & rhs, Float64 frac, Float64 rhs_frac) override;
+    void merge(const IWeightsUpdater & rhs, Float64 frac, Float64 rhs_frac) override;
 
     void write(WriteBuffer & buf) const override;
 
diff --git a/src/AggregateFunctions/AggregateFunctionStatisticsSimple.h b/src/AggregateFunctions/AggregateFunctionStatisticsSimple.h
index f9a60c0c0b4..cb20d54b2a8 100644
--- a/src/AggregateFunctions/AggregateFunctionStatisticsSimple.h
+++ b/src/AggregateFunctions/AggregateFunctionStatisticsSimple.h
@@ -234,9 +234,6 @@ namespace ErrorCodes
     extern const int ILLEGAL_TYPE_OF_ARGUMENT;
 }
 
-namespace
-{
-
 template <template <typename> typename FunctionTemplate, StatisticsFunctionKind kind>
 AggregateFunctionPtr createAggregateFunctionStatisticsUnary(
     const std::string & name, const DataTypes & argument_types, const Array & parameters, const Settings *)
@@ -273,5 +270,3 @@ AggregateFunctionPtr createAggregateFunctionStatisticsBinary(
 }
 
 }
-
-}
diff --git a/src/AggregateFunctions/AggregateFunctionUniqCombined.h b/src/AggregateFunctions/AggregateFunctionUniqCombined.h
index 10774442610..6d5486a3b09 100644
--- a/src/AggregateFunctions/AggregateFunctionUniqCombined.h
+++ b/src/AggregateFunctions/AggregateFunctionUniqCombined.h
@@ -42,9 +42,6 @@ struct UniqCombinedHashTableGrower : public HashTableGrowerWithPrecalculation<>
     void increaseSize() { increaseSizeDegree(1); }
 };
 
-namespace
-{
-
 template <typename T, UInt8 K, typename HashValueType>
 struct AggregateFunctionUniqCombinedData
 {
@@ -268,8 +265,6 @@ AggregateFunctionPtr createAggregateFunctionWithK(const DataTypes & argument_typ
         return std::make_shared<typename WithK<K, HashValueType>::template AggregateFunctionVariadic<false, false>>(argument_types, params);
 }
 
-}
-
 template <UInt8 K>
 AggregateFunctionPtr createAggregateFunctionWithHashType(bool use_64_bit_hash, const DataTypes & argument_types, const Array & params)
 {
diff --git a/src/AggregateFunctions/Combinators/AggregateFunctionNull.h b/src/AggregateFunctions/Combinators/AggregateFunctionNull.h
index ba72f960852..72ab3cf5acb 100644
--- a/src/AggregateFunctions/Combinators/AggregateFunctionNull.h
+++ b/src/AggregateFunctions/Combinators/AggregateFunctionNull.h
@@ -521,7 +521,7 @@ public:
             }
         }
 
-        chassert(nullable_filters.size() > 0);
+        chassert(!nullable_filters.empty());
         bool found_one = false;
         if (nullable_filters.size() == 1)
         {
diff --git a/src/AggregateFunctions/DDSketch/Store.h b/src/AggregateFunctions/DDSketch/Store.h
index 91eaeacedfe..428b2a6c9b8 100644
--- a/src/AggregateFunctions/DDSketch/Store.h
+++ b/src/AggregateFunctions/DDSketch/Store.h
@@ -38,7 +38,7 @@ public:
         offset = other->offset;
     }
 
-    int length()
+    int length() const
     {
         return static_cast<int>(bins.size());
     }
diff --git a/src/AggregateFunctions/QuantileExact.h b/src/AggregateFunctions/QuantileExact.h
index a92d1979bab..94e321083a7 100644
--- a/src/AggregateFunctions/QuantileExact.h
+++ b/src/AggregateFunctions/QuantileExact.h
@@ -150,7 +150,7 @@ struct QuantileExactExclusive : public QuantileExact<Value>
                 return static_cast<Float64>(*std::min_element(array.begin(), array.end()));
 
             ::nth_element(array.begin(), array.begin() + n - 1, array.end());
-            auto nth_elem = std::min_element(array.begin() + n, array.end());
+            auto * nth_elem = std::min_element(array.begin() + n, array.end());
 
             return static_cast<Float64>(array[n - 1]) + (h - n) * static_cast<Float64>(*nth_elem - array[n - 1]);
         }
@@ -179,7 +179,7 @@ struct QuantileExactExclusive : public QuantileExact<Value>
                 else
                 {
                     ::nth_element(array.begin() + prev_n, array.begin() + n - 1, array.end());
-                    auto nth_elem = std::min_element(array.begin() + n, array.end());
+                    auto * nth_elem = std::min_element(array.begin() + n, array.end());
 
                     result[indices[i]] = static_cast<Float64>(array[n - 1]) + (h - n) * static_cast<Float64>(*nth_elem - array[n - 1]);
                     prev_n = n - 1;
@@ -214,7 +214,7 @@ struct QuantileExactInclusive : public QuantileExact<Value>
             else if (n < 1)
                 return static_cast<Float64>(*std::min_element(array.begin(), array.end()));
             ::nth_element(array.begin(), array.begin() + n - 1, array.end());
-            auto nth_elem = std::min_element(array.begin() + n, array.end());
+            auto * nth_elem = std::min_element(array.begin() + n, array.end());
 
             return static_cast<Float64>(array[n - 1]) + (h - n) * static_cast<Float64>(*nth_elem - array[n - 1]);
         }
@@ -241,7 +241,7 @@ struct QuantileExactInclusive : public QuantileExact<Value>
                 else
                 {
                     ::nth_element(array.begin() + prev_n, array.begin() + n - 1, array.end());
-                    auto nth_elem = std::min_element(array.begin() + n, array.end());
+                    auto * nth_elem = std::min_element(array.begin() + n, array.end());
 
                     result[indices[i]] = static_cast<Float64>(array[n - 1]) + (h - n) * (static_cast<Float64>(*nth_elem) - array[n - 1]);
                     prev_n = n - 1;
diff --git a/src/AggregateFunctions/ReservoirSamplerDeterministic.h b/src/AggregateFunctions/ReservoirSamplerDeterministic.h
index daed0b98ca3..b7032a01884 100644
--- a/src/AggregateFunctions/ReservoirSamplerDeterministic.h
+++ b/src/AggregateFunctions/ReservoirSamplerDeterministic.h
@@ -191,7 +191,7 @@ public:
             /// TODO: After implementation of "versioning aggregate function state",
             /// change the serialization format.
             Element elem;
-            memset(&elem, 0, sizeof(elem));
+            memset(&elem, 0, sizeof(elem)); /// NOLINT(bugprone-undefined-memory-manipulation)
             elem = samples[i];
 
             DB::transformEndianness<std::endian::little>(elem);
diff --git a/src/AggregateFunctions/SingleValueData.h b/src/AggregateFunctions/SingleValueData.h
index 2ba98f8a9c3..cde13b128e5 100644
--- a/src/AggregateFunctions/SingleValueData.h
+++ b/src/AggregateFunctions/SingleValueData.h
@@ -23,7 +23,7 @@ struct SingleValueDataBase
     /// For example argMin holds 1 of these (for the result), while keeping a template for the value
     static constexpr UInt32 MAX_STORAGE_SIZE = 64;
 
-    virtual ~SingleValueDataBase() { }
+    virtual ~SingleValueDataBase() = default;
     virtual bool has() const = 0;
     virtual void insertResultInto(IColumn &) const = 0;
     virtual void write(WriteBuffer &, const ISerialization &) const = 0;
diff --git a/src/AggregateFunctions/UniqExactSet.h b/src/AggregateFunctions/UniqExactSet.h
index 2a790690f51..131c59b9ed6 100644
--- a/src/AggregateFunctions/UniqExactSet.h
+++ b/src/AggregateFunctions/UniqExactSet.h
@@ -39,8 +39,8 @@ public:
     /// This method will convert all the SingleLevelSet to TwoLevelSet in parallel if the hashsets are not all singlelevel or not all twolevel.
     static void parallelizeMergePrepare(const std::vector<UniqExactSet *> & data_vec, ThreadPool & thread_pool)
     {
-        unsigned long single_level_set_num = 0;
-        unsigned long all_single_hash_size = 0;
+        UInt64 single_level_set_num = 0;
+        UInt64 all_single_hash_size = 0;
 
         for (auto ele : data_vec)
         {
diff --git a/src/Analyzer/Passes/QueryAnalysisPass.h b/src/Analyzer/Passes/QueryAnalysisPass.h
index 8c746833eee..e0a42a530b0 100644
--- a/src/Analyzer/Passes/QueryAnalysisPass.h
+++ b/src/Analyzer/Passes/QueryAnalysisPass.h
@@ -77,7 +77,7 @@ public:
       * Available expression columns are extracted from table expression.
       * Table expression node must have query, union, table, table function type.
       */
-    QueryAnalysisPass(QueryTreeNodePtr table_expression_, bool only_analyze_ = false);
+    explicit QueryAnalysisPass(QueryTreeNodePtr table_expression_, bool only_analyze_ = false);
 
     String getName() override
     {
diff --git a/src/Backups/BackupSettings.h b/src/Backups/BackupSettings.h
index f26b992b348..10181ea464a 100644
--- a/src/Backups/BackupSettings.h
+++ b/src/Backups/BackupSettings.h
@@ -26,7 +26,7 @@ struct BackupSettings
     String password;
 
     /// S3 storage class.
-    String s3_storage_class = "";
+    String s3_storage_class;
 
     /// If this is set to true then only create queries will be written to backup,
     /// without the data of tables.
diff --git a/src/Backups/IBackup.h b/src/Backups/IBackup.h
index 5663f1ed3e8..6f366a58103 100644
--- a/src/Backups/IBackup.h
+++ b/src/Backups/IBackup.h
@@ -80,7 +80,7 @@ public:
 
     /// Returns names of entries stored in a specified directory in the backup.
     /// If `directory` is empty or '/' the functions returns entries in the backup's root.
-    virtual Strings listFiles(const String & directory, bool recursive = false) const = 0;
+    virtual Strings listFiles(const String & directory, bool recursive) const = 0;
 
     /// Checks if a specified directory contains any files.
     /// The function returns the same as `!listFiles(directory).empty()`.
@@ -108,11 +108,9 @@ public:
     virtual std::unique_ptr<SeekableReadBuffer> readFile(const SizeAndChecksum & size_and_checksum) const = 0;
 
     /// Copies a file from the backup to a specified destination disk. Returns the number of bytes written.
-    virtual size_t copyFileToDisk(const String & file_name, DiskPtr destination_disk, const String & destination_path,
-                                  WriteMode write_mode = WriteMode::Rewrite) const = 0;
+    virtual size_t copyFileToDisk(const String & file_name, DiskPtr destination_disk, const String & destination_path, WriteMode write_mode) const = 0;
 
-    virtual size_t copyFileToDisk(const SizeAndChecksum & size_and_checksum, DiskPtr destination_disk, const String & destination_path,
-                                  WriteMode write_mode = WriteMode::Rewrite) const = 0;
+    virtual size_t copyFileToDisk(const SizeAndChecksum & size_and_checksum, DiskPtr destination_disk, const String & destination_path, WriteMode write_mode) const = 0;
 
     /// Puts a new entry to the backup.
     virtual void writeFile(const BackupFileInfo & file_info, BackupEntryPtr entry) = 0;
diff --git a/src/Backups/RestorerFromBackup.cpp b/src/Backups/RestorerFromBackup.cpp
index e20e8eb66c6..ed1d5b8a103 100644
--- a/src/Backups/RestorerFromBackup.cpp
+++ b/src/Backups/RestorerFromBackup.cpp
@@ -273,7 +273,7 @@ void RestorerFromBackup::findRootPathsInBackup()
     root_paths_in_backup.push_back(root_path);
 
     /// Add shard-related part to the root path.
-    Strings shards_in_backup = backup->listFiles(root_path / "shards");
+    Strings shards_in_backup = backup->listFiles(root_path / "shards", /*recursive*/ false);
     if (shards_in_backup.empty())
     {
         if (restore_settings.shard_num_in_backup > 1)
@@ -295,7 +295,7 @@ void RestorerFromBackup::findRootPathsInBackup()
     }
 
     /// Add replica-related part to the root path.
-    Strings replicas_in_backup = backup->listFiles(root_path / "replicas");
+    Strings replicas_in_backup = backup->listFiles(root_path / "replicas", /*recursive*/ false);
     if (replicas_in_backup.empty())
     {
         if (restore_settings.replica_num_in_backup > 1)
@@ -514,7 +514,7 @@ void RestorerFromBackup::findDatabaseInBackupImpl(const String & database_name_i
         if (!metadata_path && !try_metadata_path.empty() && backup->fileExists(try_metadata_path))
             metadata_path = try_metadata_path;
 
-        Strings file_names = backup->listFiles(try_tables_metadata_path);
+        Strings file_names = backup->listFiles(try_tables_metadata_path, /*recursive*/ false);
         for (const String & file_name : file_names)
         {
             if (!file_name.ends_with(".sql"))
@@ -575,7 +575,7 @@ void RestorerFromBackup::findEverythingInBackup(const std::set<String> & except_
 
     for (const auto & root_path_in_backup : root_paths_in_backup)
     {
-        Strings file_names = backup->listFiles(root_path_in_backup / "metadata");
+        Strings file_names = backup->listFiles(root_path_in_backup / "metadata", /*recursive*/ false);
         for (String & file_name : file_names)
         {
             if (file_name.ends_with(".sql"))
diff --git a/src/Client/HedgedConnectionsFactory.h b/src/Client/HedgedConnectionsFactory.h
index ce7b553acdd..2b7ec3f3fe5 100644
--- a/src/Client/HedgedConnectionsFactory.h
+++ b/src/Client/HedgedConnectionsFactory.h
@@ -159,7 +159,7 @@ private:
     size_t requested_connections_count = 0;
 
     const size_t max_parallel_replicas = 0;
-    const bool skip_unavailable_shards = 0;
+    const bool skip_unavailable_shards = false;
 };
 
 }
diff --git a/src/Client/TestHint.h b/src/Client/TestHint.h
index 7858b7f04a4..eaf854be5df 100644
--- a/src/Client/TestHint.h
+++ b/src/Client/TestHint.h
@@ -77,12 +77,12 @@ private:
     {
         if (actual_server_error && std::find(server_errors.begin(), server_errors.end(), actual_server_error) == server_errors.end())
             return false;
-        if (!actual_server_error && server_errors.size())
+        if (!actual_server_error && !server_errors.empty())
             return false;
 
         if (actual_client_error && std::find(client_errors.begin(), client_errors.end(), actual_client_error) == client_errors.end())
             return false;
-        if (!actual_client_error && client_errors.size())
+        if (!actual_client_error && !client_errors.empty())
             return false;
 
         return true;
@@ -90,7 +90,7 @@ private:
 
     bool lostExpectedError(int actual_server_error, int actual_client_error) const
     {
-        return (server_errors.size() && !actual_server_error) || (client_errors.size() && !actual_client_error);
+        return (!server_errors.empty() && !actual_server_error) || (!client_errors.empty() && !actual_client_error);
     }
 };
 
diff --git a/src/Columns/ColumnUnique.h b/src/Columns/ColumnUnique.h
index 76bbbbacdbf..a8873140817 100644
--- a/src/Columns/ColumnUnique.h
+++ b/src/Columns/ColumnUnique.h
@@ -671,7 +671,7 @@ IColumnUnique::IndexesWithOverflow ColumnUnique<ColumnType>::uniqueInsertRangeWi
     size_t max_dictionary_size)
 {
     auto overflowed_keys = column_holder->cloneEmpty();
-    auto overflowed_keys_ptr = typeid_cast<ColumnType *>(overflowed_keys.get());
+    auto * overflowed_keys_ptr = typeid_cast<ColumnType *>(overflowed_keys.get());
     if (!overflowed_keys_ptr)
         throw Exception(ErrorCodes::LOGICAL_ERROR, "Invalid keys type for ColumnUnique.");
 
diff --git a/src/Columns/ReverseIndex.h b/src/Columns/ReverseIndex.h
index 2e59cea2c22..e2c422a0b3b 100644
--- a/src/Columns/ReverseIndex.h
+++ b/src/Columns/ReverseIndex.h
@@ -86,7 +86,7 @@ struct ReverseIndexHashTableCell
     {
         /// Careful: apparently this uses SFINAE to redefine isZero for all types
         /// except the IndexType, for which the default ZeroTraits::isZero is used.
-        static_assert(!std::is_same_v<typename std::decay<T>::type, typename std::decay<IndexType>::type>);
+        static_assert(!std::is_same_v<typename std::decay_t<T>, typename std::decay_t<IndexType>>);
         return false;
     }
 
diff --git a/src/Common/AllocatorWithMemoryTracking.h b/src/Common/AllocatorWithMemoryTracking.h
index b43870e05b2..52bfe1cc3fa 100644
--- a/src/Common/AllocatorWithMemoryTracking.h
+++ b/src/Common/AllocatorWithMemoryTracking.h
@@ -26,10 +26,10 @@ struct AllocatorWithMemoryTracking
 
     [[nodiscard]] T * allocate(size_t n)
     {
-        if (n > std::numeric_limits<size_t>::max() / sizeof(T))
+        if (n > std::numeric_limits<size_t>::max() / sizeof(T)) /// NOLINT(bugprone-sizeof-expression)
             throw std::bad_alloc();
 
-        size_t bytes = n * sizeof(T);
+        size_t bytes = n * sizeof(T); /// NOLINT(bugprone-sizeof-expression)
         auto trace = CurrentMemoryTracker::alloc(bytes);
 
         T * p = static_cast<T *>(malloc(bytes));
@@ -43,7 +43,7 @@ struct AllocatorWithMemoryTracking
 
     void deallocate(T * p, size_t n) noexcept
     {
-        size_t bytes = n * sizeof(T);
+        size_t bytes = n * sizeof(T); /// NOLINT(bugprone-sizeof-expression)
 
         free(p);
         auto trace = CurrentMemoryTracker::free(bytes);
diff --git a/src/Common/Arena.h b/src/Common/Arena.h
index ba5b9ea9205..db0e4883553 100644
--- a/src/Common/Arena.h
+++ b/src/Common/Arena.h
@@ -49,7 +49,7 @@ private:
 
         MemoryChunk() = default;
 
-        void swap(MemoryChunk & other)
+        void swap(MemoryChunk & other) noexcept
         {
             std::swap(begin, other.begin);
             std::swap(pos, other.pos);
@@ -57,12 +57,12 @@ private:
             prev.swap(other.prev);
         }
 
-        MemoryChunk(MemoryChunk && other)
+        MemoryChunk(MemoryChunk && other) noexcept
         {
             *this = std::move(other);
         }
 
-        MemoryChunk & operator=(MemoryChunk && other)
+        MemoryChunk & operator=(MemoryChunk && other) noexcept
         {
             swap(other);
             return *this;
diff --git a/src/Common/FrequencyHolder.h b/src/Common/FrequencyHolder.h
index 73675ed9814..64207dc5423 100644
--- a/src/Common/FrequencyHolder.h
+++ b/src/Common/FrequencyHolder.h
@@ -47,7 +47,6 @@ public:
         HashMap<UInt16, Float64> map;
     };
 
-public:
     using Map = HashMap<StringRef, Float64>;
     using Container = std::vector<Language>;
 
diff --git a/src/Common/HTTPConnectionPool.h b/src/Common/HTTPConnectionPool.h
index 7fd0847f5a7..a883a18ab8e 100644
--- a/src/Common/HTTPConnectionPool.h
+++ b/src/Common/HTTPConnectionPool.h
@@ -44,11 +44,12 @@ public:
     virtual const Metrics & getMetrics() const = 0;
     virtual ~IHTTPConnectionPoolForEndpoint() = default;
 
+    IHTTPConnectionPoolForEndpoint(const IHTTPConnectionPoolForEndpoint &) = delete;
+    IHTTPConnectionPoolForEndpoint & operator=(const IHTTPConnectionPoolForEndpoint &) = delete;
+
 protected:
     IHTTPConnectionPoolForEndpoint() = default;
 
-    IHTTPConnectionPoolForEndpoint(const IHTTPConnectionPoolForEndpoint &) = delete;
-    IHTTPConnectionPoolForEndpoint & operator=(const IHTTPConnectionPoolForEndpoint &) = delete;
 };
 
 enum class HTTPConnectionGroupType
@@ -70,11 +71,12 @@ public:
         static constexpr size_t warning_step = 100;
     };
 
-private:
-    HTTPConnectionPools();
     HTTPConnectionPools(const HTTPConnectionPools &) = delete;
     HTTPConnectionPools & operator=(const HTTPConnectionPools &) = delete;
 
+private:
+    HTTPConnectionPools();
+
 public:
     static HTTPConnectionPools & instance();
 
diff --git a/src/Common/HashTable/HashTable.h b/src/Common/HashTable/HashTable.h
index f23c4ca15dd..9050b7ef6d7 100644
--- a/src/Common/HashTable/HashTable.h
+++ b/src/Common/HashTable/HashTable.h
@@ -92,7 +92,8 @@ inline bool bitEquals(T && a, T && b)
     using RealT = std::decay_t<T>;
 
     if constexpr (std::is_floating_point_v<RealT>)
-        return 0 == memcmp(&a, &b, sizeof(RealT));  /// Note that memcmp with constant size is compiler builtin.
+        /// Note that memcmp with constant size is compiler builtin.
+        return 0 == memcmp(&a, &b, sizeof(RealT)); /// NOLINT
     else
         return a == b;
 }
@@ -644,7 +645,7 @@ protected:
 
         /// Copy to a new location and zero the old one.
         x.setHash(hash_value);
-        memcpy(static_cast<void*>(&buf[place_value]), &x, sizeof(x));
+        memcpy(static_cast<void*>(&buf[place_value]), &x, sizeof(x)); /// NOLINT(bugprone-undefined-memory-manipulation)
         x.setZero();
 
         /// Then the elements that previously were in collision with this can move to the old place.
diff --git a/src/Common/HostResolvePool.h b/src/Common/HostResolvePool.h
index 4f127f05253..ec4594bc8dd 100644
--- a/src/Common/HostResolvePool.h
+++ b/src/Common/HostResolvePool.h
@@ -67,8 +67,8 @@ public:
     class Entry
     {
     public:
-        explicit Entry(Entry && entry) = default;
-        explicit Entry(Entry & entry) = delete;
+        Entry(Entry && entry) = default;
+        Entry(Entry & entry) = delete;
 
         // no access as r-value
         const String * operator->() && = delete;
@@ -89,7 +89,7 @@ public:
 
         Entry(HostResolver & pool_, Poco::Net::IPAddress address_)
             : pool(pool_.getWeakFromThis())
-            , address(std::move(address_))
+            , address(address_)
             , resolved_host(address.toString())
         { }
 
@@ -126,14 +126,14 @@ protected:
     struct Record
     {
         Record(Poco::Net::IPAddress address_, Poco::Timestamp resolve_time_)
-            : address(std::move(address_))
+            : address(address_)
             , resolve_time(resolve_time_)
         {}
 
-        explicit Record(Record && rec) = default;
+        Record(Record && rec) = default;
         Record& operator=(Record && s) = default;
 
-        explicit Record(const Record & rec) = default;
+        Record(const Record & rec) = default;
         Record& operator=(const Record & s) = default;
 
         Poco::Net::IPAddress address;
@@ -198,10 +198,11 @@ class HostResolversPool
 {
 private:
     HostResolversPool() = default;
+
+public:
     HostResolversPool(const HostResolversPool &) = delete;
     HostResolversPool & operator=(const HostResolversPool &) = delete;
 
-public:
     static HostResolversPool & instance();
 
     void dropCache();
diff --git a/src/Common/MortonUtils.h b/src/Common/MortonUtils.h
index 591621f7a47..c55b36c6354 100644
--- a/src/Common/MortonUtils.h
+++ b/src/Common/MortonUtils.h
@@ -27,7 +27,7 @@
   */
 
 
-namespace
+namespace impl
 {
     /// After the most significant bit 1, set all subsequent less significant bits to 1 as well.
     inline UInt64 toMask(UInt64 n)
@@ -85,7 +85,7 @@ void intervalBinaryPartition(UInt64 first, UInt64 last, F && callback)
     /// split = 15:   00001111
 
     UInt64 diff = first ^ last;
-    UInt64 mask = toMask(diff) >> 1;
+    UInt64 mask = impl::toMask(diff) >> 1;
 
     /// The current interval represents a whole range with fixed prefix.
     if ((first & mask) == 0 && (last & mask) == mask)
diff --git a/src/Common/NetException.h b/src/Common/NetException.h
index 0ec3a7c423f..09c51949501 100644
--- a/src/Common/NetException.h
+++ b/src/Common/NetException.h
@@ -29,7 +29,7 @@ public:
     }
 
     NetException * clone() const override { return new NetException(*this); }
-    void rethrow() const override { throw *this; }
+    void rethrow() const override { throw *this; } /// NOLINT(cert-err60-cpp)
 
 private:
     const char * name() const noexcept override { return "DB::NetException"; }
diff --git a/src/Common/PODArray.h b/src/Common/PODArray.h
index af863e01fb2..02281eb776a 100644
--- a/src/Common/PODArray.h
+++ b/src/Common/PODArray.h
@@ -301,6 +301,8 @@ public:
     }
 };
 
+/// NOLINTBEGIN(bugprone-sizeof-expression)
+
 template <typename T, size_t initial_bytes, typename TAllocator, size_t pad_right_, size_t pad_left_>
 class PODArray : public PODArrayBase<sizeof(T), initial_bytes, TAllocator, pad_right_, pad_left_>
 {
@@ -755,6 +757,8 @@ public:
     }
 };
 
+/// NOLINTEND(bugprone-sizeof-expression)
+
 template <typename T, size_t initial_bytes, typename TAllocator, size_t pad_right_, size_t pad_left_>
 void swap(PODArray<T, initial_bytes, TAllocator, pad_right_, pad_left_> & lhs, PODArray<T, initial_bytes, TAllocator, pad_right_, pad_left_> & rhs) /// NOLINT
 {
diff --git a/src/Common/ProxyConfiguration.h b/src/Common/ProxyConfiguration.h
index 11a09cb5924..289ae2b6813 100644
--- a/src/Common/ProxyConfiguration.h
+++ b/src/Common/ProxyConfiguration.h
@@ -50,7 +50,7 @@ struct ProxyConfiguration
     bool tunneling = false;
     Protocol original_request_protocol = Protocol::HTTP;
 
-    bool isEmpty() const { return host.size() == 0; }
+    bool isEmpty() const { return host.empty(); }
 };
 
 }
diff --git a/src/Common/Scheduler/Nodes/DynamicResourceManager.h b/src/Common/Scheduler/Nodes/DynamicResourceManager.h
index ff736693664..4b0a3a48b61 100644
--- a/src/Common/Scheduler/Nodes/DynamicResourceManager.h
+++ b/src/Common/Scheduler/Nodes/DynamicResourceManager.h
@@ -85,7 +85,6 @@ private:
         StatePtr state; // hold state to avoid ResourceLink invalidation due to resource deregistration from SchedulerRoot
     };
 
-private:
     SchedulerRoot scheduler;
     std::mutex mutex;
     StatePtr state;
diff --git a/src/Common/Scheduler/Nodes/FairPolicy.h b/src/Common/Scheduler/Nodes/FairPolicy.h
index ce2bf729a04..0a4e55c253b 100644
--- a/src/Common/Scheduler/Nodes/FairPolicy.h
+++ b/src/Common/Scheduler/Nodes/FairPolicy.h
@@ -248,7 +248,6 @@ private:
             parent->activateChild(this);
     }
 
-private:
     /// Beginning of `items` vector is heap of active children: [0; `heap_size`).
     /// Next go inactive children in unsorted order.
     /// NOTE: we have to track vruntime of inactive children for max-min fairness.
diff --git a/src/Common/Scheduler/SchedulerRoot.h b/src/Common/Scheduler/SchedulerRoot.h
index ab3f702a422..7af42fdbbea 100644
--- a/src/Common/Scheduler/SchedulerRoot.h
+++ b/src/Common/Scheduler/SchedulerRoot.h
@@ -231,7 +231,6 @@ private:
         value->next = nullptr;
     }
 
-private:
     void schedulerThread()
     {
         while (!stop_flag.load())
@@ -253,7 +252,6 @@ private:
         request->execute();
     }
 
-private:
     TResource * current = nullptr; // round-robin pointer
     std::unordered_map<ISchedulerNode *, TResource> children; // resources by pointer
     std::atomic<bool> stop_flag = false;
diff --git a/src/Common/ZooKeeper/ZooKeeperImpl.h b/src/Common/ZooKeeper/ZooKeeperImpl.h
index d089ab7cc04..b5ab4f04b8e 100644
--- a/src/Common/ZooKeeper/ZooKeeperImpl.h
+++ b/src/Common/ZooKeeper/ZooKeeperImpl.h
@@ -291,7 +291,7 @@ private:
     class ThreadReference
     {
     public:
-        const ThreadReference & operator = (ThreadFromGlobalPool && thread_)
+        ThreadReference & operator = (ThreadFromGlobalPool && thread_)
         {
             std::lock_guard<std::mutex> l(lock);
             thread = std::move(thread_);
diff --git a/src/Common/ZooKeeper/ZooKeeperRetries.h b/src/Common/ZooKeeper/ZooKeeperRetries.h
index d8d234ba913..39782d08b33 100644
--- a/src/Common/ZooKeeper/ZooKeeperRetries.h
+++ b/src/Common/ZooKeeper/ZooKeeperRetries.h
@@ -120,7 +120,7 @@ public:
 
         iteration_succeeded = false;
         user_error.code = code;
-        user_error.message = std::move(message);
+        user_error.message = message;
         user_error.exception = exception;
         keeper_error = KeeperError{};
     }
diff --git a/src/Common/memory.h b/src/Common/memory.h
index a4c9d46e21b..a828ba7a38e 100644
--- a/src/Common/memory.h
+++ b/src/Common/memory.h
@@ -215,7 +215,7 @@ inline ALWAYS_INLINE size_t untrackMemory(void * ptr [[maybe_unused]], Allocatio
 #endif
         trace = CurrentMemoryTracker::free(actual_size);
     }
-    catch (...)
+    catch (...) /// NOLINT(bugprone-empty-catch)
     {
     }
 
diff --git a/src/Common/mysqlxx/mysqlxx/Transaction.h b/src/Common/mysqlxx/mysqlxx/Transaction.h
index 38c175d69d5..4f06042eb06 100644
--- a/src/Common/mysqlxx/mysqlxx/Transaction.h
+++ b/src/Common/mysqlxx/mysqlxx/Transaction.h
@@ -27,7 +27,7 @@ public:
             if (!finished)
                 rollback();
         }
-        catch (...)
+        catch (...) /// NOLINT(bugprone-empty-catch)
         {
         }
     }
diff --git a/src/Common/mysqlxx/mysqlxx/Types.h b/src/Common/mysqlxx/mysqlxx/Types.h
index 883eb41989e..b58ed20d740 100644
--- a/src/Common/mysqlxx/mysqlxx/Types.h
+++ b/src/Common/mysqlxx/mysqlxx/Types.h
@@ -36,7 +36,7 @@ using Int64 = int64_t;
 using UInt32 = uint32_t;
 using Int32 = int32_t;
 
-using MYSQL_LENGTH = unsigned long;
+using MYSQL_LENGTH = unsigned long; /// NOLINT
 using MYSQL_LENGTHS = MYSQL_LENGTH *;
 using MYSQL_FIELDS = MYSQL_FIELD *;
 
diff --git a/src/Common/threadPoolCallbackRunner.h b/src/Common/threadPoolCallbackRunner.h
index 2b943110273..6f7892ae4bb 100644
--- a/src/Common/threadPoolCallbackRunner.h
+++ b/src/Common/threadPoolCallbackRunner.h
@@ -57,7 +57,7 @@ template <typename Result, typename T>
 std::future<Result> scheduleFromThreadPool(T && task, ThreadPool & pool, const std::string & thread_name, Priority priority = {})
 {
     auto schedule = threadPoolCallbackRunner<Result, T>(pool, thread_name);
-    return schedule(std::move(task), priority);
+    return schedule(std::move(task), priority); /// NOLINT
 }
 
 }
diff --git a/src/Compression/CompressionCodecDeflateQpl.h b/src/Compression/CompressionCodecDeflateQpl.h
index 42d9bb69c31..86fd9051bd8 100644
--- a/src/Compression/CompressionCodecDeflateQpl.h
+++ b/src/Compression/CompressionCodecDeflateQpl.h
@@ -26,7 +26,7 @@ public:
 
     qpl_job * acquireJob(UInt32 & job_id);
     void releaseJob(UInt32 job_id);
-    const bool & isJobPoolReady() { return job_pool_ready; }
+    const bool & isJobPoolReady() const { return job_pool_ready; }
 
 private:
     bool tryLockJob(UInt32 index);
diff --git a/src/Coordination/WriteBufferFromNuraftBuffer.h b/src/Coordination/WriteBufferFromNuraftBuffer.h
index c9ca1e2a227..61b8632631a 100644
--- a/src/Coordination/WriteBufferFromNuraftBuffer.h
+++ b/src/Coordination/WriteBufferFromNuraftBuffer.h
@@ -16,7 +16,7 @@ public:
     ~WriteBufferFromNuraftBuffer() override;
 
 private:
-    void finalizeImpl() override final;
+    void finalizeImpl() final;
 
     void nextImpl() override;
 
diff --git a/src/Core/Field.h b/src/Core/Field.h
index 43f175cb476..aed5fab2106 100644
--- a/src/Core/Field.h
+++ b/src/Core/Field.h
@@ -40,6 +40,7 @@ using FieldVector = std::vector<Field, AllocatorWithMemoryTracking<Field>>;
 /// construct a Field of Array or a Tuple type. An alternative approach would be
 /// to construct both of these types from FieldVector, and have the caller
 /// specify the desired Field type explicitly.
+/// NOLINTBEGIN(modernize-type-traits)
 #define DEFINE_FIELD_VECTOR(X) \
 struct X : public FieldVector \
 { \
@@ -48,6 +49,7 @@ struct X : public FieldVector \
 
 DEFINE_FIELD_VECTOR(Array);
 DEFINE_FIELD_VECTOR(Tuple);
+/// NOLINTEND(modernize-type-traits)
 
 /// An array with the following structure: [(key1, value1), (key2, value2), ...]
 DEFINE_FIELD_VECTOR(Map); /// TODO: use map instead of vector.
diff --git a/src/Core/MySQL/MySQLReplication.h b/src/Core/MySQL/MySQLReplication.h
index 2e2023ff47f..3387d952711 100644
--- a/src/Core/MySQL/MySQLReplication.h
+++ b/src/Core/MySQL/MySQLReplication.h
@@ -546,7 +546,7 @@ namespace MySQLReplication
         virtual void setGTIDSets(GTIDSets sets) = 0;
         virtual void setChecksumSignatureLength(size_t checksum_signature_length_) = 0;
 
-        virtual ~IFlavor() override = default;
+        ~IFlavor() override = default;
     };
 
     class MySQLFlavor : public IFlavor
diff --git a/src/Core/Settings.h b/src/Core/Settings.h
index e6adb00137f..b5e519fb8b4 100644
--- a/src/Core/Settings.h
+++ b/src/Core/Settings.h
@@ -1194,7 +1194,6 @@ class IColumn;
     FORMAT_FACTORY_SETTINGS(M, ALIAS)  \
     OBSOLETE_FORMAT_SETTINGS(M, ALIAS) \
 
-/// NOLINTNEXTLINE(clang-analyzer-optin.performance.Padding)
 DECLARE_SETTINGS_TRAITS_ALLOW_CUSTOM_SETTINGS(SettingsTraits, LIST_OF_SETTINGS)
 
 
@@ -1239,7 +1238,6 @@ private:
 /*
  * User-specified file format settings for File and URL engines.
  */
-/// NOLINTNEXTLINE(clang-analyzer-optin.performance.Padding)
 DECLARE_SETTINGS_TRAITS(FormatFactorySettingsTraits, LIST_OF_ALL_FORMAT_SETTINGS)
 
 struct FormatFactorySettings : public BaseSettings<FormatFactorySettingsTraits>
diff --git a/src/Core/tests/gtest_charset_conv.cpp b/src/Core/tests/gtest_charset_conv.cpp
index 9b501b5f8fc..073b0dd74b4 100644
--- a/src/Core/tests/gtest_charset_conv.cpp
+++ b/src/Core/tests/gtest_charset_conv.cpp
@@ -5,14 +5,12 @@
 namespace DB
 {
 
-// NOLINTBEGIN(clang-analyzer-optin.performance.Padding)
 struct CheckResult
 {
     Int32 id;
     String name;
     bool need_convert;
 };
-// NOLINTEND(clang-analyzer-optin.performance.Padding)
 
 TEST(CharsetTest, CharsetTest)
 {
diff --git a/src/Daemon/BaseDaemon.h b/src/Daemon/BaseDaemon.h
index 952cf61d8e0..a0f47c44460 100644
--- a/src/Daemon/BaseDaemon.h
+++ b/src/Daemon/BaseDaemon.h
@@ -103,7 +103,7 @@ public:
 
     GraphiteWriter * getGraphiteWriter(const std::string & config_name = DEFAULT_GRAPHITE_CONFIG_NAME)
     {
-        if (graphite_writers.count(config_name))
+        if (graphite_writers.contains(config_name))
             return graphite_writers[config_name].get();
         return nullptr;
     }
@@ -183,7 +183,7 @@ std::optional<std::reference_wrapper<Daemon>> BaseDaemon::tryGetInstance()
     {
         ptr = dynamic_cast<Daemon *>(&Poco::Util::Application::instance());
     }
-    catch (const Poco::NullPointerException &)
+    catch (const Poco::NullPointerException &) /// NOLINT(bugprone-empty-catch)
     {
         /// if daemon doesn't exist than instance() throw NullPointerException
     }
diff --git a/src/DataTypes/NumberTraits.h b/src/DataTypes/NumberTraits.h
index cf283d3358c..59a64017af3 100644
--- a/src/DataTypes/NumberTraits.h
+++ b/src/DataTypes/NumberTraits.h
@@ -217,11 +217,13 @@ template <typename A> struct ToInteger
 
 // CLICKHOUSE-29. The same depth, different signs
 // NOTE: This case is applied for 64-bit integers only (for backward compatibility), but could be used for any-bit integers
+/// NOLINTBEGIN(misc-redundant-expression)
 template <typename A, typename B>
 constexpr bool LeastGreatestSpecialCase =
     std::is_integral_v<A> && std::is_integral_v<B>
     && (8 == sizeof(A) && sizeof(A) == sizeof(B))
     && (is_signed_v<A> ^ is_signed_v<B>);
+/// NOLINTEND(misc-redundant-expression)
 
 template <typename A, typename B>
 using ResultOfLeast = std::conditional_t<LeastGreatestSpecialCase<A, B>,
diff --git a/src/DataTypes/Serializations/SerializationCustomSimpleText.h b/src/DataTypes/Serializations/SerializationCustomSimpleText.h
index c80a57e234c..f7aea8697ff 100644
--- a/src/DataTypes/Serializations/SerializationCustomSimpleText.h
+++ b/src/DataTypes/Serializations/SerializationCustomSimpleText.h
@@ -18,7 +18,7 @@ public:
     explicit SerializationCustomSimpleText(const SerializationPtr & nested_);
 
     // Methods that subclasses must override in order to get full serialization/deserialization support.
-    virtual void serializeText(const IColumn & column, size_t row_num, WriteBuffer & ostr, const FormatSettings &) const override = 0;
+    void serializeText(const IColumn & column, size_t row_num, WriteBuffer & ostr, const FormatSettings &) const override = 0;
     /// whole = true means that buffer contains only one value, so we should read until EOF.
     /// It's needed to check if there is garbage after parsed field.
     virtual void deserializeText(IColumn & column, ReadBuffer & istr, const FormatSettings &, bool whole) const = 0;
diff --git a/src/DataTypes/Serializations/SerializationSparse.h b/src/DataTypes/Serializations/SerializationSparse.h
index 2d31fba2509..b1ed7b613f0 100644
--- a/src/DataTypes/Serializations/SerializationSparse.h
+++ b/src/DataTypes/Serializations/SerializationSparse.h
@@ -27,7 +27,7 @@ public:
 
     Kind getKind() const override { return Kind::SPARSE; }
 
-    virtual void enumerateStreams(
+    void enumerateStreams(
         EnumerateStreamsSettings & settings,
         const StreamCallback & callback,
         const SubstreamData & data) const override;
diff --git a/src/DataTypes/Serializations/SerializationVariantElement.cpp b/src/DataTypes/Serializations/SerializationVariantElement.cpp
index 053f8d22d5a..7d4487fe6da 100644
--- a/src/DataTypes/Serializations/SerializationVariantElement.cpp
+++ b/src/DataTypes/Serializations/SerializationVariantElement.cpp
@@ -205,8 +205,8 @@ void SerializationVariantElement::removeVariantFromPath(DB::ISerialization::Subs
 SerializationVariantElement::VariantSubcolumnCreator::VariantSubcolumnCreator(
     const ColumnPtr & local_discriminators_,
     const String & variant_element_name_,
-    const ColumnVariant::Discriminator global_variant_discriminator_,
-    const ColumnVariant::Discriminator local_variant_discriminator_)
+    ColumnVariant::Discriminator global_variant_discriminator_,
+    ColumnVariant::Discriminator local_variant_discriminator_)
     : local_discriminators(local_discriminators_)
     , variant_element_name(variant_element_name_)
     , global_variant_discriminator(global_variant_discriminator_)
diff --git a/src/DataTypes/Serializations/SerializationVariantElement.h b/src/DataTypes/Serializations/SerializationVariantElement.h
index c343c219cf3..aafecf43d39 100644
--- a/src/DataTypes/Serializations/SerializationVariantElement.h
+++ b/src/DataTypes/Serializations/SerializationVariantElement.h
@@ -75,8 +75,8 @@ private:
         VariantSubcolumnCreator(
             const ColumnPtr & local_discriminators_,
             const String & variant_element_name_,
-            const ColumnVariant::Discriminator global_variant_discriminator_,
-            const ColumnVariant::Discriminator local_variant_discriminator_);
+            ColumnVariant::Discriminator global_variant_discriminator_,
+            ColumnVariant::Discriminator local_variant_discriminator_);
 
         DataTypePtr create(const DataTypePtr & prev) const override;
         ColumnPtr create(const ColumnPtr & prev) const override;
diff --git a/src/Databases/DatabasesOverlay.h b/src/Databases/DatabasesOverlay.h
index b58df506f70..859063a4ac4 100644
--- a/src/Databases/DatabasesOverlay.h
+++ b/src/Databases/DatabasesOverlay.h
@@ -23,7 +23,6 @@ public:
 
     String getEngineName() const override { return "Overlay"; }
 
-public:
     bool isTableExist(const String & table_name, ContextPtr context) const override;
 
     StoragePtr tryGetTable(const String & table_name, ContextPtr context) const override;
diff --git a/src/Dictionaries/CacheDictionary.cpp b/src/Dictionaries/CacheDictionary.cpp
index 935658719db..fbb04eab3e4 100644
--- a/src/Dictionaries/CacheDictionary.cpp
+++ b/src/Dictionaries/CacheDictionary.cpp
@@ -308,7 +308,7 @@ ColumnUInt8::Ptr CacheDictionary<dictionary_key_type>::hasKeys(const Columns & k
         /// Write lock on storage
         const ProfilingScopedWriteRWLock write_lock{rw_lock, ProfileEvents::DictCacheLockWriteNs};
 
-        result_of_fetch_from_storage = cache_storage_ptr->fetchColumnsForKeys(keys, request);
+        result_of_fetch_from_storage = cache_storage_ptr->fetchColumnsForKeys(keys, request, /*default_mask*/ nullptr);
     }
 
     size_t found_keys_size = result_of_fetch_from_storage.found_keys_size;
diff --git a/src/Dictionaries/CacheDictionary.h b/src/Dictionaries/CacheDictionary.h
index c02fb91c60e..a23f64e3e3f 100644
--- a/src/Dictionaries/CacheDictionary.h
+++ b/src/Dictionaries/CacheDictionary.h
@@ -171,7 +171,7 @@ private:
         const PaddedPODArray<KeyState> & key_index_to_fetched_columns_from_storage_result,
         const MutableColumns & fetched_columns_during_update,
         const HashMap<KeyType, size_t> & found_keys_to_fetched_columns_during_update_index,
-        IColumn::Filter * const default_mask = nullptr) const;
+        IColumn::Filter * default_mask = nullptr) const;
 
     void update(CacheDictionaryUpdateUnitPtr<dictionary_key_type> update_unit_ptr);
 
diff --git a/src/Dictionaries/CacheDictionaryStorage.h b/src/Dictionaries/CacheDictionaryStorage.h
index 4cbbaa070d0..01217c58e31 100644
--- a/src/Dictionaries/CacheDictionaryStorage.h
+++ b/src/Dictionaries/CacheDictionaryStorage.h
@@ -73,7 +73,7 @@ public:
     SimpleKeysStorageFetchResult fetchColumnsForKeys(
         const PaddedPODArray<UInt64> & keys,
         const DictionaryStorageFetchRequest & fetch_request,
-        IColumn::Filter * const default_mask = nullptr) override
+        IColumn::Filter * const default_mask) override
     {
         if constexpr (dictionary_key_type == DictionaryKeyType::Simple)
             return fetchColumnsForKeysImpl<SimpleKeysStorageFetchResult>(keys, fetch_request, default_mask);
@@ -110,7 +110,7 @@ public:
     ComplexKeysStorageFetchResult fetchColumnsForKeys(
         const PaddedPODArray<StringRef> & keys,
         const DictionaryStorageFetchRequest & column_fetch_requests,
-        IColumn::Filter * const default_mask = nullptr) override
+        IColumn::Filter * const default_mask) override
     {
         if constexpr (dictionary_key_type == DictionaryKeyType::Complex)
             return fetchColumnsForKeysImpl<ComplexKeysStorageFetchResult>(keys, column_fetch_requests, default_mask);
diff --git a/src/Dictionaries/ICacheDictionaryStorage.h b/src/Dictionaries/ICacheDictionaryStorage.h
index 047d67135a6..dcd7434946f 100644
--- a/src/Dictionaries/ICacheDictionaryStorage.h
+++ b/src/Dictionaries/ICacheDictionaryStorage.h
@@ -82,7 +82,7 @@ public:
     virtual SimpleKeysStorageFetchResult fetchColumnsForKeys(
         const PaddedPODArray<UInt64> & keys,
         const DictionaryStorageFetchRequest & fetch_request,
-        IColumn::Filter * default_mask = nullptr) = 0;
+        IColumn::Filter * default_mask) = 0;
 
     /// Fetch columns for keys, this method is not write thread safe
     virtual void insertColumnsForKeys(const PaddedPODArray<UInt64> & keys, Columns columns) = 0;
@@ -100,7 +100,7 @@ public:
     virtual ComplexKeysStorageFetchResult fetchColumnsForKeys(
         const PaddedPODArray<StringRef> & keys,
         const DictionaryStorageFetchRequest & column_fetch_requests,
-        IColumn::Filter * default_mask = nullptr) = 0;
+        IColumn::Filter * default_mask) = 0;
 
     /// Fetch columns for keys, this method is not write thread safe
     virtual void insertColumnsForKeys(const PaddedPODArray<StringRef> & keys, Columns columns) = 0;
diff --git a/src/Dictionaries/SSDCacheDictionaryStorage.h b/src/Dictionaries/SSDCacheDictionaryStorage.h
index acebea676a5..73c96e0aedc 100644
--- a/src/Dictionaries/SSDCacheDictionaryStorage.h
+++ b/src/Dictionaries/SSDCacheDictionaryStorage.h
@@ -870,7 +870,7 @@ public:
     SimpleKeysStorageFetchResult fetchColumnsForKeys(
         const PaddedPODArray<UInt64> & keys,
         const DictionaryStorageFetchRequest & fetch_request,
-        IColumn::Filter * const default_mask = nullptr) override
+        IColumn::Filter * const default_mask) override
     {
         if constexpr (dictionary_key_type == DictionaryKeyType::Simple)
             return fetchColumnsForKeysImpl<SimpleKeysStorageFetchResult>(keys, fetch_request, default_mask);
@@ -907,7 +907,7 @@ public:
     ComplexKeysStorageFetchResult fetchColumnsForKeys(
         const PaddedPODArray<StringRef> & keys,
         const DictionaryStorageFetchRequest & fetch_request,
-        IColumn::Filter * const default_mask = nullptr) override
+        IColumn::Filter * const default_mask) override
     {
         if constexpr (dictionary_key_type == DictionaryKeyType::Complex)
             return fetchColumnsForKeysImpl<ComplexKeysStorageFetchResult>(keys, fetch_request, default_mask);
diff --git a/src/Disks/IDiskTransaction.h b/src/Disks/IDiskTransaction.h
index 975c41cb70b..7df1b71eb2b 100644
--- a/src/Disks/IDiskTransaction.h
+++ b/src/Disks/IDiskTransaction.h
@@ -62,8 +62,8 @@ public:
     virtual void copyFile(
         const std::string & from_file_path,
         const std::string & to_file_path,
-        const ReadSettings & read_settings = {},
-        const WriteSettings & write_settings = {}) = 0;
+        const ReadSettings & read_settings,
+        const WriteSettings & write_settings) = 0;
 
     /// Open the file for write and return WriteBufferFromFileBase object.
     virtual std::unique_ptr<WriteBufferFromFileBase> writeFile( /// NOLINT
diff --git a/src/Disks/IO/IOUringReader.h b/src/Disks/IO/IOUringReader.h
index a29781171e5..89e71e4b215 100644
--- a/src/Disks/IO/IOUringReader.h
+++ b/src/Disks/IO/IOUringReader.h
@@ -78,7 +78,7 @@ private:
 public:
     explicit IOUringReader(uint32_t entries_);
 
-    inline bool isSupported() { return is_supported; }
+    bool isSupported() const { return is_supported; }
     std::future<Result> submit(Request request) override;
     Result execute(Request /* request */) override { throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Method `execute` not implemented for IOUringReader"); }
 
diff --git a/src/Disks/IVolume.h b/src/Disks/IVolume.h
index bc6706a1829..2da896a3514 100644
--- a/src/Disks/IVolume.h
+++ b/src/Disks/IVolume.h
@@ -64,7 +64,7 @@ public:
         DiskSelectorPtr disk_selector
     );
 
-    virtual ReservationPtr reserve(UInt64 bytes) override = 0;
+    ReservationPtr reserve(UInt64 bytes) override = 0;
 
     /// This is a volume.
     bool isVolume() const override { return true; }
diff --git a/src/Disks/ObjectStorages/DiskObjectStorage.cpp b/src/Disks/ObjectStorages/DiskObjectStorage.cpp
index 16183ec20c1..696a0ef7416 100644
--- a/src/Disks/ObjectStorages/DiskObjectStorage.cpp
+++ b/src/Disks/ObjectStorages/DiskObjectStorage.cpp
@@ -204,7 +204,7 @@ void DiskObjectStorage::copyFile( /// NOLINT
             /// It may use s3-server-side copy
             auto & to_disk_object_storage = dynamic_cast<DiskObjectStorage &>(to_disk);
             auto transaction = createObjectStorageTransactionToAnotherDisk(to_disk_object_storage);
-            transaction->copyFile(from_file_path, to_file_path);
+            transaction->copyFile(from_file_path, to_file_path, /*read_settings*/ {}, /*write_settings*/ {});
             transaction->commit();
     }
     else
diff --git a/src/Disks/ObjectStorages/ObjectStorageIterator.h b/src/Disks/ObjectStorages/ObjectStorageIterator.h
index 841b0ea6664..9af2593579a 100644
--- a/src/Disks/ObjectStorages/ObjectStorageIterator.h
+++ b/src/Disks/ObjectStorages/ObjectStorageIterator.h
@@ -54,7 +54,7 @@ public:
         return batch;
     }
 
-    virtual std::optional<RelativePathsWithMetadata> getCurrrentBatchAndScheduleNext() override
+    std::optional<RelativePathsWithMetadata> getCurrrentBatchAndScheduleNext() override
     {
         return std::nullopt;
     }
diff --git a/src/Disks/ObjectStorages/S3/diskSettings.h b/src/Disks/ObjectStorages/S3/diskSettings.h
index 83bf7b179ef..e461daa99e2 100644
--- a/src/Disks/ObjectStorages/S3/diskSettings.h
+++ b/src/Disks/ObjectStorages/S3/diskSettings.h
@@ -9,14 +9,6 @@
 
 #include <IO/S3/Client.h>
 
-namespace Aws
-{
-namespace S3
-{
-class Client;
-}
-}
-
 namespace DB
 {
 
diff --git a/src/Functions/CountSubstringsImpl.h b/src/Functions/CountSubstringsImpl.h
index 8ba9ee99de8..9ff3e4e1f2a 100644
--- a/src/Functions/CountSubstringsImpl.h
+++ b/src/Functions/CountSubstringsImpl.h
@@ -47,7 +47,7 @@ struct CountSubstringsImpl
         const UInt8 * pos = begin;
 
         /// FIXME: suboptimal
-        memset(&res[0], 0, res.size() * sizeof(res[0]));
+        memset(&res[0], 0, res.size() * sizeof(res[0])); /// NOLINT(readability-container-data-pointer)
 
         if (needle.empty())
             return; // Return all zeros
diff --git a/src/Functions/CustomWeekTransforms.h b/src/Functions/CustomWeekTransforms.h
index b8d25cc30b6..75fb3c32f16 100644
--- a/src/Functions/CustomWeekTransforms.h
+++ b/src/Functions/CustomWeekTransforms.h
@@ -58,7 +58,7 @@ struct CustomWeekTransformImpl
     template <typename Transform>
     static ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/, Transform transform = {})
     {
-        const auto op = WeekTransformer<typename FromDataType::FieldType, typename ToDataType::FieldType, Transform, is_extended_result>{std::move(transform)};
+        const auto op = WeekTransformer<typename FromDataType::FieldType, typename ToDataType::FieldType, Transform, is_extended_result>{transform};
 
         static constexpr UInt8 default_week_mode = 0;
         UInt8 week_mode = default_week_mode;
diff --git a/src/Functions/FunctionConstantBase.h b/src/Functions/FunctionConstantBase.h
index ad969268713..9b999af38b6 100644
--- a/src/Functions/FunctionConstantBase.h
+++ b/src/Functions/FunctionConstantBase.h
@@ -13,8 +13,9 @@ class FunctionConstantBase : public IFunction
 {
 public:
     template <typename U>
-    explicit FunctionConstantBase(U && constant_value_, bool is_distributed_ = false)
-        : constant_value(static_cast<T>(std::forward<U>(constant_value_))), is_distributed(is_distributed_)
+    explicit FunctionConstantBase(const U & constant_value_, bool is_distributed_ = false)
+        : constant_value(static_cast<T>(constant_value_))
+        , is_distributed(is_distributed_)
     {
     }
 
diff --git a/src/Functions/FunctionsAES.h b/src/Functions/FunctionsAES.h
index a03f0b602b9..3bb264dfaec 100644
--- a/src/Functions/FunctionsAES.h
+++ b/src/Functions/FunctionsAES.h
@@ -96,7 +96,7 @@ inline void validateCipherMode(const EVP_CIPHER * evp_cipher)
 {
     if constexpr (compatibility_mode == CompatibilityMode::MySQL)
     {
-        switch (EVP_CIPHER_mode(evp_cipher))
+        switch (EVP_CIPHER_mode(evp_cipher)) /// NOLINT(bugprone-switch-missing-default-case)
         {
             case EVP_CIPH_ECB_MODE: [[fallthrough]];
             case EVP_CIPH_CBC_MODE: [[fallthrough]];
@@ -107,7 +107,7 @@ inline void validateCipherMode(const EVP_CIPHER * evp_cipher)
     }
     else if constexpr (compatibility_mode == CompatibilityMode::OpenSSL)
     {
-        switch (EVP_CIPHER_mode(evp_cipher))
+        switch (EVP_CIPHER_mode(evp_cipher)) /// NOLINT(bugprone-switch-missing-default-case)
         {
             case EVP_CIPH_ECB_MODE: [[fallthrough]];
             case EVP_CIPH_CBC_MODE: [[fallthrough]];
diff --git a/src/Functions/FunctionsBitmap.h b/src/Functions/FunctionsBitmap.h
index 801f2ccd012..22d46fa7728 100644
--- a/src/Functions/FunctionsBitmap.h
+++ b/src/Functions/FunctionsBitmap.h
@@ -519,12 +519,12 @@ public:
                                        "but one of them has type {}.", getName(), arguments[i + 1]->getName());
 
             if (!array_type)
-                throw exception;
+                throw exception; /// NOLINT
 
             auto nested_type = array_type->getNestedType();
             WhichDataType which(nested_type);
             if (!(which.isUInt8() || which.isUInt16() || which.isUInt32() || which.isUInt64()))
-                throw exception;
+                throw exception; /// NOLINT
         }
         return arguments[0];
     }
diff --git a/src/Functions/FunctionsComparison.h b/src/Functions/FunctionsComparison.h
index 3be675f39b3..57aebc11da0 100644
--- a/src/Functions/FunctionsComparison.h
+++ b/src/Functions/FunctionsComparison.h
@@ -283,7 +283,7 @@ struct StringComparisonImpl
         size_t size = a_data.size();
 
         for (size_t i = 0, j = 0; i < size; i += 16, ++j)
-            c[j] = Op::apply(memcmp16(&a_data[i], &b_data[0]), 0);
+            c[j] = Op::apply(memcmp16(&a_data[i], &b_data[0]), 0); /// NOLINT(readability-container-data-pointer)
     }
 
     static void NO_INLINE fixed_string_vector_fixed_string_vector( /// NOLINT
diff --git a/src/Functions/FunctionsExternalDictionaries.h b/src/Functions/FunctionsExternalDictionaries.h
index 011772baab9..d3317e2dfcf 100644
--- a/src/Functions/FunctionsExternalDictionaries.h
+++ b/src/Functions/FunctionsExternalDictionaries.h
@@ -393,8 +393,8 @@ public:
                 throw Exception(ErrorCodes::UNSUPPORTED_METHOD, "Function {} does not support nullable keys", getName());
 
             // Wrap all the attribute types in Array()
-            for (auto it = attribute_types.begin(); it != attribute_types.end(); ++it)
-                *it = std::make_shared<DataTypeArray>(*it);
+            for (auto & attr_type : attribute_types)
+                attr_type = std::make_shared<DataTypeArray>(attr_type);
         }
         if (attribute_types.size() > 1)
         {
diff --git a/src/Functions/FunctionsTimeWindow.h b/src/Functions/FunctionsTimeWindow.h
index 658963cb69c..5145ae79377 100644
--- a/src/Functions/FunctionsTimeWindow.h
+++ b/src/Functions/FunctionsTimeWindow.h
@@ -73,6 +73,8 @@ struct ToStartOfTransform;
     TRANSFORM_TIME(Second)
 #undef TRANSFORM_TIME
 
+/// NOLINTBEGIN(bugprone-macro-parentheses)
+
 #define TRANSFORM_SUBSECONDS(INTERVAL_KIND, DEF_SCALE) \
 template<> \
     struct ToStartOfTransform<IntervalKind::Kind::INTERVAL_KIND> \
@@ -156,6 +158,8 @@ template <> \
     ADD_SUBSECONDS(Nanosecond, 9)
 #undef ADD_SUBSECONDS
 
+/// NOLINTEND(bugprone-macro-parentheses)
+
 template <TimeWindowFunctionName type>
 struct TimeWindowImpl
 {
diff --git a/src/Functions/GatherUtils/Selectors.h b/src/Functions/GatherUtils/Selectors.h
index 00367727a39..1f933c13ba0 100644
--- a/src/Functions/GatherUtils/Selectors.h
+++ b/src/Functions/GatherUtils/Selectors.h
@@ -126,19 +126,19 @@ struct ArrayAndValueSourceSelectorBySink : public ArraySinkSelector<ArrayAndValu
     template <typename Sink, typename ... Args>
     static void selectImpl(Sink && sink, IArraySource & array_source, IValueSource & value_source, Args && ... args)
     {
-        using SynkType = typename std::decay<Sink>::type;
-        using ArraySource = typename SynkType::CompatibleArraySource;
-        using ValueSource = typename SynkType::CompatibleValueSource;
+        using SinkType = typename std::decay_t<Sink>;
+        using ArraySource = typename SinkType::CompatibleArraySource;
+        using ValueSource = typename SinkType::CompatibleValueSource;
 
         auto check_type = [] (auto source_ptr)
         {
             if (source_ptr == nullptr)
                 throw Exception(ErrorCodes::LOGICAL_ERROR, "{} expected {} or {} or {} or {} but got {}",
                                 demangle(typeid(Base).name()),
-                                demangle(typeid(typename SynkType::CompatibleArraySource).name()),
-                                demangle(typeid(ConstSource<typename SynkType::CompatibleArraySource>).name()),
-                                demangle(typeid(typename SynkType::CompatibleValueSource).name()),
-                                demangle(typeid(ConstSource<typename SynkType::CompatibleValueSource>).name()),
+                                demangle(typeid(typename SinkType::CompatibleArraySource).name()),
+                                demangle(typeid(ConstSource<typename SinkType::CompatibleArraySource>).name()),
+                                demangle(typeid(typename SinkType::CompatibleValueSource).name()),
+                                demangle(typeid(ConstSource<typename SinkType::CompatibleValueSource>).name()),
                                 demangle(typeid(*source_ptr).name()));
         };
         auto check_type_and_call_concat = [& sink, & check_type, & args ...] (auto array_source_ptr, auto value_source_ptr)
diff --git a/src/Functions/GatherUtils/Sources.h b/src/Functions/GatherUtils/Sources.h
index 80fb9ce3900..01b63338559 100644
--- a/src/Functions/GatherUtils/Sources.h
+++ b/src/Functions/GatherUtils/Sources.h
@@ -144,6 +144,8 @@ struct NumericArraySource : public ArraySourceImpl<NumericArraySource<T>>
 #pragma clang diagnostic ignored "-Wsuggest-override"
 #pragma clang diagnostic ignored "-Wsuggest-destructor-override"
 
+/// NOLINTBEGIN(hicpp-use-override)
+
 template <typename Base>
 struct ConstSource : public Base
 {
@@ -231,6 +233,8 @@ struct ConstSource : public Base
     }
 };
 
+/// NOLINTEND(hicpp-use-override)
+
 #pragma clang diagnostic pop
 
 struct StringSource
diff --git a/src/Functions/GregorianDate.h b/src/Functions/GregorianDate.h
index 2528223443e..fb00e4276b6 100644
--- a/src/Functions/GregorianDate.h
+++ b/src/Functions/GregorianDate.h
@@ -13,7 +13,7 @@ class WriteBuffer;
 class GregorianDate
 {
 public:
-    GregorianDate() {}
+    GregorianDate() = default;
 
     void init(ReadBuffer & in);
     bool tryInit(ReadBuffer & in);
@@ -84,7 +84,7 @@ private:
 class OrdinalDate
 {
 public:
-    OrdinalDate() {}
+    OrdinalDate() = default;
 
     void init(int32_t year, uint16_t day_of_year);
     bool tryInit(int32_t year, uint16_t day_of_year);
diff --git a/src/Functions/HasSubsequenceImpl.h b/src/Functions/HasSubsequenceImpl.h
index 17955746aa2..630a2afb720 100644
--- a/src/Functions/HasSubsequenceImpl.h
+++ b/src/Functions/HasSubsequenceImpl.h
@@ -14,8 +14,6 @@ namespace ErrorCodes
     extern const int ILLEGAL_TYPE_OF_ARGUMENT;
     extern const int ILLEGAL_COLUMN;
 }
-namespace
-{
 
 using namespace GatherUtils;
 
@@ -154,5 +152,3 @@ private:
 };
 
 }
-
-}
diff --git a/src/Functions/JSONPath/Generator/GeneratorJSONPath.h b/src/Functions/JSONPath/Generator/GeneratorJSONPath.h
index de62be98d31..1016d776be0 100644
--- a/src/Functions/JSONPath/Generator/GeneratorJSONPath.h
+++ b/src/Functions/JSONPath/Generator/GeneratorJSONPath.h
@@ -35,7 +35,7 @@ public:
         }
         const auto * query = path->jsonpath_query;
 
-        for (auto child_ast : query->children)
+        for (const auto & child_ast : query->children)
         {
             if (typeid_cast<ASTJSONPathRoot *>(child_ast.get()))
             {
diff --git a/src/Functions/LowerUpperUTF8Impl.h b/src/Functions/LowerUpperUTF8Impl.h
index 460f75f9bde..7ca98166576 100644
--- a/src/Functions/LowerUpperUTF8Impl.h
+++ b/src/Functions/LowerUpperUTF8Impl.h
@@ -174,7 +174,7 @@ private:
 
     static void array(const UInt8 * src, const UInt8 * src_end, const ColumnString::Offsets & offsets, UInt8 * dst)
     {
-        auto offset_it = offsets.begin();
+        const auto * offset_it = offsets.begin();
         const UInt8 * begin = src;
 
 #ifdef __SSE2__
diff --git a/src/Functions/URL/domain.h b/src/Functions/URL/domain.h
index f38f106e9a2..87f5aeffda7 100644
--- a/src/Functions/URL/domain.h
+++ b/src/Functions/URL/domain.h
@@ -88,7 +88,7 @@ exloop: if ((scheme_end - pos) > 2 && *pos == ':' && *(pos + 1) == '/' && *(pos
     const auto * start_of_host = pos;
     for (; pos < end; ++pos)
     {
-        switch (*pos)
+        switch (*pos) /// NOLINT(bugprone-switch-missing-default-case)
         {
         case '.':
             if (has_open_bracket)
@@ -220,7 +220,7 @@ exloop: if ((scheme_end - pos) > 2 && *pos == ':' && *(pos + 1) == '/' && *(pos
     const auto * start_of_host = pos;
     for (; pos < end; ++pos)
     {
-        switch (*pos)
+        switch (*pos) /// NOLINT(bugprone-switch-missing-default-case)
         {
         case '.':
             dot_pos = pos;
diff --git a/src/Functions/UniqTheta/FunctionsUniqTheta.h b/src/Functions/UniqTheta/FunctionsUniqTheta.h
index 2d616841c7f..c312dcc09f5 100644
--- a/src/Functions/UniqTheta/FunctionsUniqTheta.h
+++ b/src/Functions/UniqTheta/FunctionsUniqTheta.h
@@ -137,8 +137,8 @@ namespace DB
 
             for (size_t i = 0; i < input_rows_count; ++i)
             {
-                const AggregateDataPtr data_ptr_0 = is_column_const[0] ? container0[0] : container0[i];
-                const AggregateDataPtr data_ptr_1 = is_column_const[1] ? container1[0] : container1[i];
+                AggregateDataPtr data_ptr_0 = is_column_const[0] ? container0[0] : container0[i];
+                AggregateDataPtr data_ptr_1 = is_column_const[1] ? container1[0] : container1[i];
 
                 col_to->insertFrom(data_ptr_0);
                 AggregateFunctionUniqThetaData & sketch_data_1 = *reinterpret_cast<AggregateFunctionUniqThetaData *>(col_to->getData()[i]);
diff --git a/src/Functions/UserDefined/UserDefinedSQLObjectsBackup.cpp b/src/Functions/UserDefined/UserDefinedSQLObjectsBackup.cpp
index b7c7e5847bd..c15958f81c9 100644
--- a/src/Functions/UserDefined/UserDefinedSQLObjectsBackup.cpp
+++ b/src/Functions/UserDefined/UserDefinedSQLObjectsBackup.cpp
@@ -88,7 +88,7 @@ restoreUserDefinedSQLObjects(RestorerFromBackup & restorer, const String & data_
     auto backup = restorer.getBackup();
     fs::path data_path_in_backup_fs{data_path_in_backup};
 
-    Strings filenames = backup->listFiles(data_path_in_backup);
+    Strings filenames = backup->listFiles(data_path_in_backup, /*recursive*/ false);
     if (filenames.empty())
         return {}; /// Nothing to restore.
 
diff --git a/src/Functions/geometryConverters.h b/src/Functions/geometryConverters.h
index dba984b4184..97162fa9dd0 100644
--- a/src/Functions/geometryConverters.h
+++ b/src/Functions/geometryConverters.h
@@ -28,9 +28,6 @@ namespace ErrorCodes
     extern const int ILLEGAL_TYPE_OF_ARGUMENT;
 }
 
-namespace
-{
-
 template <typename Point>
 using Ring = boost::geometry::model::ring<Point>;
 
@@ -374,5 +371,3 @@ static void callOnTwoGeometryDataTypes(DataTypePtr left_type, DataTypePtr right_
 }
 
 }
-
-}
diff --git a/src/Functions/stl.hpp b/src/Functions/stl.hpp
index fbfc52f959c..9d4fa30e2e3 100644
--- a/src/Functions/stl.hpp
+++ b/src/Functions/stl.hpp
@@ -1,4 +1,4 @@
-// Dump of https://github.com/ankane/stl-cpp/blob/3b1b3a3e9335cda26c8b0797d8b8d24ac8e350ad/include/stl.hpp. 
+// Dump of https://github.com/ankane/stl-cpp/blob/3b1b3a3e9335cda26c8b0797d8b8d24ac8e350ad/include/stl.hpp.
 // Added to ClickHouse source code and not referenced as a submodule because its easier maintain and modify/customize.
 
 /*!
@@ -24,12 +24,12 @@
 
 namespace stl {
 
-bool est(const float* y, size_t n, size_t len, int ideg, float xs, float* ys, size_t nleft, size_t nright, float* w, bool userw, const float* rw) {
+inline bool est(const float* y, size_t n, size_t len, int ideg, float xs, float* ys, size_t nleft, size_t nright, float* w, bool userw, const float* rw) {
     auto range = ((float) n) - 1.0;
     auto h = std::max(xs - ((float) nleft), ((float) nright) - xs);
 
     if (len > n) {
-        h += (float) ((len - n) / 2);
+        h += (float) ((len - n) / 2); /// NOLINT(bugprone-integer-division)
     }
 
     auto h9 = 0.999 * h;
@@ -89,7 +89,7 @@ bool est(const float* y, size_t n, size_t len, int ideg, float xs, float* ys, si
     }
 }
 
-void ess(const float* y, size_t n, size_t len, int ideg, size_t njump, bool userw, const float* rw, float* ys, float* res) {
+inline void ess(const float* y, size_t n, size_t len, int ideg, size_t njump, bool userw, const float* rw, float* ys, float* res) {
     if (n < 2) {
         ys[0] = y[0];
         return;
@@ -165,7 +165,7 @@ void ess(const float* y, size_t n, size_t len, int ideg, size_t njump, bool user
     }
 }
 
-void ma(const float* x, size_t n, size_t len, float* ave) {
+inline void ma(const float* x, size_t n, size_t len, float* ave) {
     auto newn = n - len + 1;
     auto flen = (float) len;
     auto v = 0.0;
@@ -189,13 +189,13 @@ void ma(const float* x, size_t n, size_t len, float* ave) {
     }
 }
 
-void fts(const float* x, size_t n, size_t np, float* trend, float* work) {
+inline void fts(const float* x, size_t n, size_t np, float* trend, float* work) {
     ma(x, n, np, trend);
     ma(trend, n - np + 1, np, work);
     ma(work, n - 2 * np + 2, 3, trend);
 }
 
-void rwts(const float* y, size_t n, const float* fit, float* rw) {
+inline void rwts(const float* y, size_t n, const float* fit, float* rw) {
     for (size_t i = 0; i < n; i++) {
         rw[i] = fabs(y[i] - fit[i]);
     }
@@ -222,7 +222,7 @@ void rwts(const float* y, size_t n, const float* fit, float* rw) {
     }
 }
 
-void ss(const float* y, size_t n, size_t np, size_t ns, int isdeg, size_t nsjump, bool userw, float* rw, float* season, float* work1, float* work2, float* work3, float* work4) {
+inline void ss(const float* y, size_t n, size_t np, size_t ns, int isdeg, size_t nsjump, bool userw, float* rw, float* season, float* work1, float* work2, float* work3, float* work4) { /// NOLINT(readability-non-const-parameter)
     for (size_t j = 1; j <= np; j++) {
         size_t k = (n - j) / np + 1;
 
@@ -253,7 +253,7 @@ void ss(const float* y, size_t n, size_t np, size_t ns, int isdeg, size_t nsjump
     }
 }
 
-void onestp(const float* y, size_t n, size_t np, size_t ns, size_t nt, size_t nl, int isdeg, int itdeg, int ildeg, size_t nsjump, size_t ntjump, size_t nljump, size_t ni, bool userw, float* rw, float* season, float* trend, float* work1, float* work2, float* work3, float* work4, float* work5) {
+inline void onestp(const float* y, size_t n, size_t np, size_t ns, size_t nt, size_t nl, int isdeg, int itdeg, int ildeg, size_t nsjump, size_t ntjump, size_t nljump, size_t ni, bool userw, float* rw, float* season, float* trend, float* work1, float* work2, float* work3, float* work4, float* work5) {
     for (size_t j = 0; j < ni; j++) {
         for (size_t i = 0; i < n; i++) {
             work1[i] = y[i] - trend[i];
@@ -272,7 +272,7 @@ void onestp(const float* y, size_t n, size_t np, size_t ns, size_t nt, size_t nl
     }
 }
 
-void stl(const float* y, size_t n, size_t np, size_t ns, size_t nt, size_t nl, int isdeg, int itdeg, int ildeg, size_t nsjump, size_t ntjump, size_t nljump, size_t ni, size_t no, float* rw, float* season, float* trend) {
+inline void stl(const float* y, size_t n, size_t np, size_t ns, size_t nt, size_t nl, int isdeg, int itdeg, int ildeg, size_t nsjump, size_t ntjump, size_t nljump, size_t ni, size_t no, float* rw, float* season, float* trend) {
     if (ns < 3) {
         throw std::invalid_argument("seasonal_length must be at least 3");
     }
@@ -335,7 +335,7 @@ void stl(const float* y, size_t n, size_t np, size_t ns, size_t nt, size_t nl, i
     }
 }
 
-float var(const std::vector<float>& series) {
+inline float var(const std::vector<float>& series) {
     auto mean = std::accumulate(series.begin(), series.end(), 0.0) / series.size();
     std::vector<float> tmp;
     tmp.reserve(series.size());
@@ -345,7 +345,7 @@ float var(const std::vector<float>& series) {
     return std::accumulate(tmp.begin(), tmp.end(), 0.0) / (series.size() - 1);
 }
 
-float strength(const std::vector<float>& component, const std::vector<float>& remainder) {
+inline float strength(const std::vector<float>& component, const std::vector<float>& remainder) {
     std::vector<float> sr;
     sr.reserve(remainder.size());
     for (size_t i = 0; i < remainder.size(); i++) {
@@ -361,11 +361,11 @@ public:
     std::vector<float> remainder;
     std::vector<float> weights;
 
-    inline float seasonal_strength() {
+    float seasonal_strength() const {
         return strength(seasonal, remainder);
     }
 
-    inline float trend_strength() {
+    float trend_strength() const {
         return strength(trend, remainder);
     }
 };
@@ -385,62 +385,62 @@ class StlParams {
     bool robust_ = false;
 
 public:
-    inline StlParams seasonal_length(size_t ns) {
+    StlParams seasonal_length(size_t ns) {
         this->ns_ = ns;
         return *this;
     }
 
-    inline StlParams trend_length(size_t nt) {
+    StlParams trend_length(size_t nt) {
         this->nt_ = nt;
         return *this;
     }
 
-    inline StlParams low_pass_length(size_t nl) {
+    StlParams low_pass_length(size_t nl) {
         this->nl_ = nl;
         return *this;
     }
 
-    inline StlParams seasonal_degree(int isdeg) {
+    StlParams seasonal_degree(int isdeg) {
         this->isdeg_ = isdeg;
         return *this;
     }
 
-    inline StlParams trend_degree(int itdeg) {
+    StlParams trend_degree(int itdeg) {
         this->itdeg_ = itdeg;
         return *this;
     }
 
-    inline StlParams low_pass_degree(int ildeg) {
+    StlParams low_pass_degree(int ildeg) {
         this->ildeg_ = ildeg;
         return *this;
     }
 
-    inline StlParams seasonal_jump(size_t nsjump) {
+    StlParams seasonal_jump(size_t nsjump) {
         this->nsjump_ = nsjump;
         return *this;
     }
 
-    inline StlParams trend_jump(size_t ntjump) {
+    StlParams trend_jump(size_t ntjump) {
         this->ntjump_ = ntjump;
         return *this;
     }
 
-    inline StlParams low_pass_jump(size_t nljump) {
+    StlParams low_pass_jump(size_t nljump) {
         this->nljump_ = nljump;
         return *this;
     }
 
-    inline StlParams inner_loops(bool ni) {
+    StlParams inner_loops(bool ni) {
         this->ni_ = ni;
         return *this;
     }
 
-    inline StlParams outer_loops(bool no) {
+    StlParams outer_loops(bool no) {
         this->no_ = no;
         return *this;
     }
 
-    inline StlParams robust(bool robust) {
+    StlParams robust(bool robust) {
         this->robust_ = robust;
         return *this;
     }
@@ -449,11 +449,11 @@ public:
     StlResult fit(const std::vector<float>& y, size_t np);
 };
 
-StlParams params() {
+inline StlParams params() {
     return StlParams();
 }
 
-StlResult StlParams::fit(const float* y, size_t n, size_t np) {
+inline StlResult StlParams::fit(const float* y, size_t n, size_t np) {
     if (n < 2 * np) {
         throw std::invalid_argument("series has less than two periods");
     }
@@ -506,7 +506,7 @@ StlResult StlParams::fit(const float* y, size_t n, size_t np) {
     return res;
 }
 
-StlResult StlParams::fit(const std::vector<float>& y, size_t np) {
+inline StlResult StlParams::fit(const std::vector<float>& y, size_t np) {
     return StlParams::fit(y.data(), y.size(), np);
 }
 
diff --git a/src/IO/Archives/IArchiveWriter.h b/src/IO/Archives/IArchiveWriter.h
index c8f0c609da1..a3b06e6c4be 100644
--- a/src/IO/Archives/IArchiveWriter.h
+++ b/src/IO/Archives/IArchiveWriter.h
@@ -36,10 +36,10 @@ public:
 
     /// Sets compression method and level.
     /// Changing them will affect next file in the archive.
-    virtual void setCompression(const String & /* compression_method */, int /* compression_level */ = kDefaultCompressionLevel) {}
+    virtual void setCompression(const String & /*compression_method*/, int /*compression_level*/) {}
 
     /// Sets password. If the password is not empty it will enable encryption in the archive.
-    virtual void setPassword(const String & /* password */) {}
+    virtual void setPassword(const String & /*password*/) {}
 };
 
 }
diff --git a/src/IO/BrotliWriteBuffer.h b/src/IO/BrotliWriteBuffer.h
index d4cda7b270c..c68743ca13f 100644
--- a/src/IO/BrotliWriteBuffer.h
+++ b/src/IO/BrotliWriteBuffer.h
@@ -21,10 +21,10 @@ public:
         WriteBufferT && out_,
         int compression_level,
         size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE,
-        char * existing_memory = nullptr,
+        char * existing_memory = nullptr, /// NOLINT(readability-non-const-parameter)
         size_t alignment = 0,
         bool compress_empty_ = true)
-    : WriteBufferWithOwnMemoryDecorator(std::move(out_), buf_size, existing_memory, alignment)
+    : WriteBufferWithOwnMemoryDecorator(std::move(out_), buf_size, existing_memory, alignment) /// NOLINT(bugprone-move-forwarding-reference)
     , brotli(std::make_unique<BrotliStateWrapper>())
     , in_available(0)
     , in_data(nullptr)
diff --git a/src/IO/Bzip2WriteBuffer.h b/src/IO/Bzip2WriteBuffer.h
index a685d4efa75..7808f8987cc 100644
--- a/src/IO/Bzip2WriteBuffer.h
+++ b/src/IO/Bzip2WriteBuffer.h
@@ -25,10 +25,10 @@ public:
         WriteBufferT && out_,
         int compression_level,
         size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE,
-        char * existing_memory = nullptr,
+        char * existing_memory = nullptr, /// NOLINT(readability-non-const-parameter)
         size_t alignment = 0,
         bool compress_empty_ = true)
-    : WriteBufferWithOwnMemoryDecorator(std::move(out_), buf_size, existing_memory, alignment)
+    : WriteBufferWithOwnMemoryDecorator(std::move(out_), buf_size, existing_memory, alignment) /// NOLINT(bugprone-move-forwarding-reference)
     , compress_empty(compress_empty_)
     {
         memset(&stream, 0, sizeof(stream));
diff --git a/src/IO/ConnectionTimeouts.h b/src/IO/ConnectionTimeouts.h
index 2b2ab0e7ab8..49305f42d85 100644
--- a/src/IO/ConnectionTimeouts.h
+++ b/src/IO/ConnectionTimeouts.h
@@ -71,6 +71,7 @@ APPLY_FOR_ALL_CONNECTION_TIMEOUT_MEMBERS(DECLARE_BUILDER_FOR_MEMBER)
     ConnectionTimeouts & withConnectionTimeout(Poco::Timespan span);
 };
 
+/// NOLINTBEGIN(bugprone-macro-parentheses)
 #define DEFINE_BUILDER_FOR_MEMBER(member, setter_func) \
     inline ConnectionTimeouts & ConnectionTimeouts::setter_func(size_t seconds) \
     { \
@@ -83,6 +84,7 @@ APPLY_FOR_ALL_CONNECTION_TIMEOUT_MEMBERS(DECLARE_BUILDER_FOR_MEMBER)
     } \
 
     APPLY_FOR_ALL_CONNECTION_TIMEOUT_MEMBERS(DEFINE_BUILDER_FOR_MEMBER)
+/// NOLINTEND(bugprone-macro-parentheses)
 
 #undef DEFINE_BUILDER_FOR_MEMBER
 
diff --git a/src/IO/HTTPCommon.h b/src/IO/HTTPCommon.h
index e27269e2559..63dffcf6878 100644
--- a/src/IO/HTTPCommon.h
+++ b/src/IO/HTTPCommon.h
@@ -34,7 +34,7 @@ public:
     {}
 
     HTTPException * clone() const override { return new HTTPException(*this); }
-    void rethrow() const override { throw *this; }
+    void rethrow() const override { throw *this; } /// NOLINT(cert-err60-cpp)
 
     Poco::Net::HTTPResponse::HTTPStatus getHTTPStatus() const { return http_status; }
 
diff --git a/src/IO/LZMADeflatingWriteBuffer.h b/src/IO/LZMADeflatingWriteBuffer.h
index 797b85cd400..a3bcad7603c 100644
--- a/src/IO/LZMADeflatingWriteBuffer.h
+++ b/src/IO/LZMADeflatingWriteBuffer.h
@@ -19,10 +19,10 @@ public:
         WriteBufferT && out_,
         int compression_level,
         size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE,
-        char * existing_memory = nullptr,
+        char * existing_memory = nullptr, /// NOLINT(readability-non-const-parameter)
         size_t alignment = 0,
         bool compress_empty_ = true)
-    : WriteBufferWithOwnMemoryDecorator(std::move(out_), buf_size, existing_memory, alignment), compress_empty(compress_empty_)
+    : WriteBufferWithOwnMemoryDecorator(std::move(out_), buf_size, existing_memory, alignment), compress_empty(compress_empty_) /// NOLINT(bugprone-move-forwarding-reference)
     {
         initialize(compression_level);
     }
diff --git a/src/IO/Lz4DeflatingWriteBuffer.h b/src/IO/Lz4DeflatingWriteBuffer.h
index b37d61fa732..6efd092afd0 100644
--- a/src/IO/Lz4DeflatingWriteBuffer.h
+++ b/src/IO/Lz4DeflatingWriteBuffer.h
@@ -19,10 +19,10 @@ public:
         WriteBufferT && out_,
         int compression_level,
         size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE,
-        char * existing_memory = nullptr,
+        char * existing_memory = nullptr, /// NOLINT(readability-non-const-parameter)
         size_t alignment = 0,
         bool compress_empty_ = true)
-    : WriteBufferWithOwnMemoryDecorator(std::move(out_), buf_size, existing_memory, alignment)
+    : WriteBufferWithOwnMemoryDecorator(std::move(out_), buf_size, existing_memory, alignment) /// NOLINT(bugprone-move-forwarding-reference)
     , tmp_memory(buf_size)
     , compress_empty(compress_empty_)
     {
diff --git a/src/IO/ReadBufferFromFileBase.h b/src/IO/ReadBufferFromFileBase.h
index 296edf9c689..9870d8bbe43 100644
--- a/src/IO/ReadBufferFromFileBase.h
+++ b/src/IO/ReadBufferFromFileBase.h
@@ -58,7 +58,7 @@ public:
     /// I.e. it can be read using open() or mmap(). If this buffer is a "view" into a subrange of the
     /// file, *out_view_offset is set to the start of that subrange, i.e. the difference between actual
     /// file offset and what getPosition() returns.
-    virtual bool isRegularLocalFile(size_t * /* out_view_offset */ = nullptr) { return false; }
+    virtual bool isRegularLocalFile(size_t * /*out_view_offsee*/) { return false; }
 
 protected:
     std::optional<size_t> file_size;
diff --git a/src/IO/ReadSettings.h b/src/IO/ReadSettings.h
index c937f77708c..6a0cac35878 100644
--- a/src/IO/ReadSettings.h
+++ b/src/IO/ReadSettings.h
@@ -63,7 +63,6 @@ enum class RemoteFSReadMethod
 class MMappedFileCache;
 class PageCache;
 
-/// NOLINTNEXTLINE(clang-analyzer-optin.performance.Padding)
 struct ReadSettings
 {
     /// Method to use reading from local filesystem.
diff --git a/src/IO/ReadWriteBufferFromHTTP.h b/src/IO/ReadWriteBufferFromHTTP.h
index 1b7437ea0c6..224aac809a4 100644
--- a/src/IO/ReadWriteBufferFromHTTP.h
+++ b/src/IO/ReadWriteBufferFromHTTP.h
@@ -210,10 +210,11 @@ class BuilderRWBufferFromHTTP
     bool delay_initialization = true;
 
 public:
-    BuilderRWBufferFromHTTP(Poco::URI uri_)
+    explicit BuilderRWBufferFromHTTP(Poco::URI uri_)
         : uri(uri_)
     {}
 
+/// NOLINTBEGIN(bugprone-macro-parentheses)
 #define setterMember(name, member) \
     BuilderRWBufferFromHTTP & name(decltype(BuilderRWBufferFromHTTP::member) arg_##member) \
     { \
@@ -235,6 +236,7 @@ public:
     setterMember(withDelayInit, delay_initialization)
     setterMember(withSkipNotFound, http_skip_not_found_url)
 #undef setterMember
+/// NOLINTEND(bugprone-macro-parentheses)
 
     ReadWriteBufferFromHTTPPtr create(const Poco::Net::HTTPBasicCredentials & credentials_)
     {
diff --git a/src/IO/S3/PocoHTTPClient.h b/src/IO/S3/PocoHTTPClient.h
index a93a4dfbaf7..ebbddbb2c7e 100644
--- a/src/IO/S3/PocoHTTPClient.h
+++ b/src/IO/S3/PocoHTTPClient.h
@@ -95,7 +95,7 @@ public:
 
     void SetResponseBody(std::string & response_body) /// NOLINT
     {
-        auto stream = Aws::New<std::stringstream>("http result buf", response_body); // STYLE_CHECK_ALLOW_STD_STRING_STREAM
+        auto * stream = Aws::New<std::stringstream>("http result buf", response_body); // STYLE_CHECK_ALLOW_STD_STRING_STREAM
         stream->exceptions(std::ios::failbit);
         body_stream = Aws::Utils::Stream::ResponseStream(std::move(stream));
     }
diff --git a/src/IO/S3/tests/TestPocoHTTPServer.h b/src/IO/S3/tests/TestPocoHTTPServer.h
index a26e811d646..35ceb5e1632 100644
--- a/src/IO/S3/tests/TestPocoHTTPServer.h
+++ b/src/IO/S3/tests/TestPocoHTTPServer.h
@@ -22,7 +22,7 @@ class MockRequestHandler : public Poco::Net::HTTPRequestHandler
 
 public:
     explicit MockRequestHandler(Poco::Net::MessageHeader & last_request_header_)
-    : Poco::Net::HTTPRequestHandler(), last_request_header(last_request_header_)
+    : last_request_header(last_request_header_)
     {
     }
 
@@ -38,20 +38,18 @@ class HTTPRequestHandlerFactory : public Poco::Net::HTTPRequestHandlerFactory
 {
     Poco::Net::MessageHeader & last_request_header;
 
-    virtual Poco::Net::HTTPRequestHandler * createRequestHandler(const Poco::Net::HTTPServerRequest &) override
+    Poco::Net::HTTPRequestHandler * createRequestHandler(const Poco::Net::HTTPServerRequest &) override
     {
         return new MockRequestHandler(last_request_header);
     }
 
 public:
     explicit HTTPRequestHandlerFactory(Poco::Net::MessageHeader & last_request_header_)
-    : Poco::Net::HTTPRequestHandlerFactory(), last_request_header(last_request_header_)
+    : last_request_header(last_request_header_)
     {
     }
 
-    virtual ~HTTPRequestHandlerFactory() override
-    {
-    }
+    ~HTTPRequestHandlerFactory() override = default;
 };
 
 class TestPocoHTTPServer
diff --git a/src/IO/SeekableReadBuffer.h b/src/IO/SeekableReadBuffer.h
index 798833e1a9b..808efe9e6d0 100644
--- a/src/IO/SeekableReadBuffer.h
+++ b/src/IO/SeekableReadBuffer.h
@@ -82,7 +82,7 @@ public:
     ///    (e.g. next() or supportsReadAt()).
     ///  * Performance: there's no buffering. Each readBigAt() call typically translates into actual
     ///    IO operation (e.g. HTTP request). Don't use it for small adjacent reads.
-    virtual size_t readBigAt(char * /*to*/, size_t /*n*/, size_t /*offset*/, const std::function<bool(size_t m)> & /*progress_callback*/ = nullptr) const
+    virtual size_t readBigAt(char * /*to*/, size_t /*n*/, size_t /*offset*/, const std::function<bool(size_t m)> & /*progress_callback*/) const
         { throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Method readBigAt() not implemented"); }
 
     /// Checks if readBigAt() is allowed. May be slow, may throw (e.g. it may do an HTTP request or an fstat).
diff --git a/src/IO/WriteBufferFromArena.h b/src/IO/WriteBufferFromArena.h
index 8e9276496b5..f901f5bff77 100644
--- a/src/IO/WriteBufferFromArena.h
+++ b/src/IO/WriteBufferFromArena.h
@@ -66,7 +66,7 @@ private:
 
     /// it is super strange,
     /// but addition next call changes the data in serializeValueIntoArena result
-    virtual void finalizeImpl() override { /* no op */ }
+    void finalizeImpl() override { /* no op */ }
 };
 
 }
diff --git a/src/IO/ZlibDeflatingWriteBuffer.h b/src/IO/ZlibDeflatingWriteBuffer.h
index f01c41c7d13..35760b3e246 100644
--- a/src/IO/ZlibDeflatingWriteBuffer.h
+++ b/src/IO/ZlibDeflatingWriteBuffer.h
@@ -27,10 +27,10 @@ public:
             CompressionMethod compression_method,
             int compression_level,
             size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE,
-            char * existing_memory = nullptr,
+            char * existing_memory = nullptr, /// NOLINT(readability-non-const-parameter)
             size_t alignment = 0,
             bool compress_empty_ = true)
-    : WriteBufferWithOwnMemoryDecorator(std::move(out_), buf_size, existing_memory, alignment), compress_empty(compress_empty_)
+    : WriteBufferWithOwnMemoryDecorator(std::move(out_), buf_size, existing_memory, alignment), compress_empty(compress_empty_) /// NOLINT(bugprone-move-forwarding-reference)
     {
         zstr.zalloc = nullptr;
         zstr.zfree = nullptr;
@@ -60,8 +60,8 @@ private:
     /// Flush all pending data and write zlib footer to the underlying buffer.
     /// After the first call to this function, subsequent calls will have no effect and
     /// an attempt to write to this buffer will result in exception.
-    virtual void finalizeBefore() override;
-    virtual void finalizeAfter() override;
+    void finalizeBefore() override;
+    void finalizeAfter() override;
 
     z_stream zstr;
     bool compress_empty = true;
diff --git a/src/IO/ZstdDeflatingWriteBuffer.h b/src/IO/ZstdDeflatingWriteBuffer.h
index 15c3869062f..9d1c677e725 100644
--- a/src/IO/ZstdDeflatingWriteBuffer.h
+++ b/src/IO/ZstdDeflatingWriteBuffer.h
@@ -20,10 +20,10 @@ public:
         int compression_level,
         int window_log = 0,
         size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE,
-        char * existing_memory = nullptr,
+        char * existing_memory = nullptr, /// NOLINT(readability-non-const-parameter)
         size_t alignment = 0,
         bool compress_empty_ = true)
-    : WriteBufferWithOwnMemoryDecorator(std::move(out_), buf_size, existing_memory, alignment), compress_empty(compress_empty_)
+    : WriteBufferWithOwnMemoryDecorator(std::move(out_), buf_size, existing_memory, alignment), compress_empty(compress_empty_) /// NOLINT(bugprone-move-forwarding-reference)
     {
         initialize(compression_level, window_log);
     }
diff --git a/src/IO/readDecimalText.h b/src/IO/readDecimalText.h
index 8b4405ee2e9..f3b08065696 100644
--- a/src/IO/readDecimalText.h
+++ b/src/IO/readDecimalText.h
@@ -38,7 +38,7 @@ inline bool readDigits(ReadBuffer & buf, T & x, uint32_t & digits, int32_t & exp
         return false;
     }
 
-    switch (*buf.position())
+    switch (*buf.position()) /// NOLINT(bugprone-switch-missing-default-case)
     {
         case '-':
             sign = -1;
diff --git a/src/Interpreters/AddDefaultDatabaseVisitor.h b/src/Interpreters/AddDefaultDatabaseVisitor.h
index e36f22e8ba1..356bffa75e9 100644
--- a/src/Interpreters/AddDefaultDatabaseVisitor.h
+++ b/src/Interpreters/AddDefaultDatabaseVisitor.h
@@ -163,7 +163,7 @@ private:
         if (identifier.compound())
             return;
         /// There is temporary table with such name, should not be rewritten.
-        if (external_tables.count(identifier.shortName()))
+        if (external_tables.contains(identifier.shortName()))
             return;
 
         auto qualified_identifier = std::make_shared<ASTTableIdentifier>(database_name, identifier.name());
diff --git a/src/Interpreters/AggregateFunctionOfGroupByKeysVisitor.h b/src/Interpreters/AggregateFunctionOfGroupByKeysVisitor.h
index ab01cc6a0b6..d8fae7e80ad 100644
--- a/src/Interpreters/AggregateFunctionOfGroupByKeysVisitor.h
+++ b/src/Interpreters/AggregateFunctionOfGroupByKeysVisitor.h
@@ -37,7 +37,7 @@ struct KeepAggregateFunctionMatcher
             return;
         }
 
-        if (!data.group_by_keys.count(function_node.getColumnName()))
+        if (!data.group_by_keys.contains(function_node.getColumnName()))
         {
             Visitor(data).visit(function_node.arguments);
         }
@@ -46,7 +46,7 @@ struct KeepAggregateFunctionMatcher
     static void visit(ASTIdentifier & ident, Data & data)
     {
         /// if variable of a function is not in GROUP BY keys, this function should not be deleted
-        if (!data.group_by_keys.count(ident.getColumnName()))
+        if (!data.group_by_keys.contains(ident.getColumnName()))
             data.keep_aggregator = true;
     }
 
diff --git a/src/Interpreters/AggregationCommon.h b/src/Interpreters/AggregationCommon.h
index ced29c2f862..ab078d1c5e5 100644
--- a/src/Interpreters/AggregationCommon.h
+++ b/src/Interpreters/AggregationCommon.h
@@ -94,7 +94,7 @@ void fillFixedBatch(size_t keys_size, const ColumnRawPtrs & key_columns, const S
             if constexpr (std::endian::native == std::endian::big)
                 offset_to = sizeof(Key) - sizeof(T) - offset;
             T * dest = reinterpret_cast<T *>(reinterpret_cast<char *>(out.data()) + offset_to);
-            fillFixedBatch<T, sizeof(Key) / sizeof(T)>(num_rows, reinterpret_cast<const T *>(source), dest);
+            fillFixedBatch<T, sizeof(Key) / sizeof(T)>(num_rows, reinterpret_cast<const T *>(source), dest); /// NOLINT(bugprone-sizeof-expression)
             offset += sizeof(T);
         }
     }
diff --git a/src/Interpreters/ArrayJoinedColumnsVisitor.h b/src/Interpreters/ArrayJoinedColumnsVisitor.h
index f16751c4561..86db342073a 100644
--- a/src/Interpreters/ArrayJoinedColumnsVisitor.h
+++ b/src/Interpreters/ArrayJoinedColumnsVisitor.h
@@ -75,7 +75,7 @@ private:
             if (nested_table_alias == nested_table_name && !ast->as<ASTIdentifier>())
                 throw Exception(ErrorCodes::ALIAS_REQUIRED, "No alias for non-trivial value in ARRAY JOIN: {}", nested_table_name);
 
-            if (data.array_join_alias_to_name.count(nested_table_alias) || data.aliases.count(nested_table_alias))
+            if (data.array_join_alias_to_name.contains(nested_table_alias) || data.aliases.contains(nested_table_alias))
                 throw Exception(ErrorCodes::MULTIPLE_EXPRESSIONS_FOR_ALIAS, "Duplicate alias in ARRAY JOIN: {}", nested_table_alias);
 
             data.array_join_alias_to_name[nested_table_alias] = nested_table_name;
@@ -100,18 +100,18 @@ private:
 
         auto split = Nested::splitName(node.name());  /// ParsedParams, Key1
 
-        if (array_join_alias_to_name.count(node.name()))
+        if (array_join_alias_to_name.contains(node.name()))
         {
             /// ARRAY JOIN was written with an array column. Example: SELECT K1 FROM ... ARRAY JOIN ParsedParams.Key1 AS K1
             array_join_result_to_source[node.name()] = array_join_alias_to_name[node.name()];    /// K1 -> ParsedParams.Key1
         }
-        else if (array_join_alias_to_name.count(split.first) && !split.second.empty())
+        else if (array_join_alias_to_name.contains(split.first) && !split.second.empty())
         {
             /// ARRAY JOIN was written with a nested table. Example: SELECT PP.KEY1 FROM ... ARRAY JOIN ParsedParams AS PP
             array_join_result_to_source[node.name()]    /// PP.Key1 -> ParsedParams.Key1
                 = Nested::concatenateName(array_join_alias_to_name[split.first], split.second);
         }
-        else if (array_join_name_to_alias.count(node.name()))
+        else if (array_join_name_to_alias.contains(node.name()))
         {
             /** Example: SELECT ParsedParams.Key1 FROM ... ARRAY JOIN ParsedParams.Key1 AS PP.Key1.
             * That is, the query uses the original array, replicated by itself.
@@ -119,7 +119,7 @@ private:
             array_join_result_to_source[    /// PP.Key1 -> ParsedParams.Key1
                 array_join_name_to_alias[node.name()]] = node.name();
         }
-        else if (array_join_name_to_alias.count(split.first) && !split.second.empty())
+        else if (array_join_name_to_alias.contains(split.first) && !split.second.empty())
         {
             /** Example: SELECT ParsedParams.Key1 FROM ... ARRAY JOIN ParsedParams AS PP.
             */
diff --git a/src/Interpreters/AsynchronousInsertQueue.h b/src/Interpreters/AsynchronousInsertQueue.h
index 5076701d0b0..9d3c38d0a4c 100644
--- a/src/Interpreters/AsynchronousInsertQueue.h
+++ b/src/Interpreters/AsynchronousInsertQueue.h
@@ -282,7 +282,7 @@ private:
 public:
     auto getQueueLocked(size_t shard_num) const
     {
-        auto & shard = queue_shards[shard_num];
+        const auto & shard = queue_shards[shard_num];
         std::unique_lock lock(shard.mutex);
         return std::make_pair(std::ref(shard.queue), std::move(lock));
     }
diff --git a/src/Interpreters/BackupLog.h b/src/Interpreters/BackupLog.h
index ee6a4c214ee..e17b9b69cfd 100644
--- a/src/Interpreters/BackupLog.h
+++ b/src/Interpreters/BackupLog.h
@@ -23,7 +23,7 @@ struct BackupLogElement
 
     std::chrono::system_clock::time_point event_time{};
     Decimal64 event_time_usec{};
-    BackupOperationInfo info{};
+    BackupOperationInfo info{}; /// NOLINT(bugprone-throw-keyword-missing)
 
     static std::string name() { return "BackupLog"; }
     static ColumnsDescription getColumnsDescription();
diff --git a/src/Interpreters/Context.h b/src/Interpreters/Context.h
index 43df8d6adf2..dbd9bbd1ff8 100644
--- a/src/Interpreters/Context.h
+++ b/src/Interpreters/Context.h
@@ -129,7 +129,6 @@ using ActionLocksManagerPtr = std::shared_ptr<ActionLocksManager>;
 class ShellCommand;
 class ICompressionCodec;
 class AccessControl;
-class Credentials;
 class GSSAcceptorContext;
 struct SettingsConstraintsAndProfileIDs;
 class SettingsProfileElements;
@@ -164,7 +163,6 @@ using OrdinaryBackgroundExecutorPtr = std::shared_ptr<OrdinaryBackgroundExecutor
 struct PartUUIDs;
 using PartUUIDsPtr = std::shared_ptr<PartUUIDs>;
 class KeeperDispatcher;
-class Session;
 struct WriteSettings;
 
 class IInputFormat;
diff --git a/src/Interpreters/DirectJoin.h b/src/Interpreters/DirectJoin.h
index ef8d12a1b8f..019e32c3130 100644
--- a/src/Interpreters/DirectJoin.h
+++ b/src/Interpreters/DirectJoin.h
@@ -31,28 +31,24 @@ public:
         const Block & right_sample_block_with_storage_column_names_);
 
     std::string getName() const override { return "DirectKeyValueJoin"; }
-    virtual const TableJoin & getTableJoin() const override { return *table_join; }
+    const TableJoin & getTableJoin() const override { return *table_join; }
 
-    virtual bool addBlockToJoin(const Block &, bool) override;
-    virtual void checkTypesOfKeys(const Block &) const override;
+    bool addBlockToJoin(const Block &, bool) override;
+    void checkTypesOfKeys(const Block &) const override;
 
     /// Join the block with data from left hand of JOIN to the right hand data (that was previously built by calls to addBlockToJoin).
     /// Could be called from different threads in parallel.
-    virtual void joinBlock(Block & block, std::shared_ptr<ExtraBlock> &) override;
+    void joinBlock(Block & block, std::shared_ptr<ExtraBlock> &) override;
 
-    virtual size_t getTotalRowCount() const override { return 0; }
+    size_t getTotalRowCount() const override { return 0; }
 
-    virtual size_t getTotalByteCount() const override { return 0; }
+    size_t getTotalByteCount() const override { return 0; }
 
-    virtual bool alwaysReturnsEmptySet() const override { return false; }
+    bool alwaysReturnsEmptySet() const override { return false; }
 
-    virtual bool isFilled() const override { return true; }
+    bool isFilled() const override { return true; }
 
-    virtual IBlocksStreamPtr
-    getNonJoinedBlocks(const Block &, const Block &, UInt64) const override
-    {
-        return nullptr;
-    }
+    IBlocksStreamPtr getNonJoinedBlocks(const Block &, const Block &, UInt64) const override { return nullptr; }
 
 private:
     std::shared_ptr<TableJoin> table_join;
diff --git a/src/Interpreters/GetAggregatesVisitor.h b/src/Interpreters/GetAggregatesVisitor.h
index 863c4f81a1b..e1d22e4de82 100644
--- a/src/Interpreters/GetAggregatesVisitor.h
+++ b/src/Interpreters/GetAggregatesVisitor.h
@@ -68,7 +68,7 @@ private:
                                 node.getColumnName(), String(data.assert_no_aggregates));
 
             String column_name = node.getColumnName();
-            if (data.uniq_names.count(column_name))
+            if (data.uniq_names.contains(column_name))
                 return;
 
             data.uniq_names.insert(column_name);
@@ -81,7 +81,7 @@ private:
                                 node.getColumnName(), String(data.assert_no_windows));
 
             String column_name = node.getColumnName();
-            if (data.uniq_names.count(column_name))
+            if (data.uniq_names.contains(column_name))
                 return;
 
             data.uniq_names.insert(column_name);
diff --git a/src/Interpreters/GroupByFunctionKeysVisitor.h b/src/Interpreters/GroupByFunctionKeysVisitor.h
index b1d40320a45..9b082291bfb 100644
--- a/src/Interpreters/GroupByFunctionKeysVisitor.h
+++ b/src/Interpreters/GroupByFunctionKeysVisitor.h
@@ -38,7 +38,7 @@ struct KeepFunctionMatcher
             return;
         }
 
-        if (!data.key_names_to_keep.count(function_node->getColumnName()))
+        if (!data.key_names_to_keep.contains(function_node->getColumnName()))
         {
             Visitor(data).visit(function_node->arguments);
         }
@@ -46,7 +46,7 @@ struct KeepFunctionMatcher
 
     static void visit(ASTIdentifier * ident, Data & data)
     {
-        if (!data.key_names_to_keep.count(ident->shortName()))
+        if (!data.key_names_to_keep.contains(ident->shortName()))
         {
             /// if variable of a function is not in GROUP BY keys, this function should not be deleted
             data.keep_key = true;
diff --git a/src/Interpreters/HashJoin.h b/src/Interpreters/HashJoin.h
index 29bb9070009..3e11517edad 100644
--- a/src/Interpreters/HashJoin.h
+++ b/src/Interpreters/HashJoin.h
@@ -253,6 +253,7 @@ public:
     template <typename Mapped>
     struct MapsTemplate
     {
+/// NOLINTBEGIN(bugprone-macro-parentheses)
         using MappedType = Mapped;
         std::unique_ptr<FixedHashMap<UInt8, Mapped>>                  key8;
         std::unique_ptr<FixedHashMap<UInt16, Mapped>>                 key16;
@@ -341,6 +342,7 @@ public:
 
             UNREACHABLE();
         }
+/// NOLINTEND(bugprone-macro-parentheses)
     };
 
     using MapsOne = MapsTemplate<RowRef>;
diff --git a/src/Interpreters/MonotonicityCheckVisitor.h b/src/Interpreters/MonotonicityCheckVisitor.h
index cc386825024..e1df45ed36b 100644
--- a/src/Interpreters/MonotonicityCheckVisitor.h
+++ b/src/Interpreters/MonotonicityCheckVisitor.h
@@ -41,7 +41,7 @@ public:
             /// if GROUP BY contains the same function ORDER BY shouldn't be optimized
             const auto hash = ast_function.getTreeHash(/*ignore_aliases=*/ true);
             const auto key = toString(hash);
-            if (group_by_function_hashes.count(key))
+            if (group_by_function_hashes.contains(key))
                 return false;
 
             /// if ORDER BY contains aggregate function or window functions, it
diff --git a/src/Interpreters/RedundantFunctionsInOrderByVisitor.h b/src/Interpreters/RedundantFunctionsInOrderByVisitor.h
index 60c9fcf2a24..516e6f0a9d4 100644
--- a/src/Interpreters/RedundantFunctionsInOrderByVisitor.h
+++ b/src/Interpreters/RedundantFunctionsInOrderByVisitor.h
@@ -61,7 +61,7 @@ public:
 
             /// Allow known identifiers: they are present in ORDER BY before current item
             if (auto * identifier = arg->as<ASTIdentifier>())
-                if (data.keys.count(getIdentifierName(identifier)))
+                if (data.keys.contains(getIdentifierName(identifier)))
                     continue;
 
             /// Reject erase others
diff --git a/src/Interpreters/SessionLog.h b/src/Interpreters/SessionLog.h
index 74e72c54bcc..5bacb9677c0 100644
--- a/src/Interpreters/SessionLog.h
+++ b/src/Interpreters/SessionLog.h
@@ -37,7 +37,7 @@ struct SessionLogElement
     SessionLogElement(const UUID & auth_id_, Type type_);
     SessionLogElement(const SessionLogElement &) = default;
     SessionLogElement & operator=(const SessionLogElement &) = default;
-    SessionLogElement(SessionLogElement &&) = default;
+    SessionLogElement(SessionLogElement &&) = default; /// NOLINT(performance-noexcept-move-constructor,hicpp-noexcept-move)
     SessionLogElement & operator=(SessionLogElement &&) = default;
 
     UUID auth_id;
diff --git a/src/Interpreters/SetVariants.h b/src/Interpreters/SetVariants.h
index 71187c9f109..ff527102080 100644
--- a/src/Interpreters/SetVariants.h
+++ b/src/Interpreters/SetVariants.h
@@ -80,7 +80,7 @@ protected:
 
         for (const auto & col : key_columns)
         {
-            if (auto * nullable = checkAndGetColumn<ColumnNullable>(*col))
+            if (const auto * nullable = checkAndGetColumn<ColumnNullable>(*col))
             {
                 actual_columns.push_back(&nullable->getNestedColumn());
                 null_maps.push_back(&nullable->getNullMapColumn());
@@ -96,7 +96,7 @@ protected:
     /// Return the columns which actually contain the values of the keys.
     /// For a given key column, if it is nullable, we return its nested
     /// column. Otherwise we return the key column itself.
-    inline const ColumnRawPtrs & getActualColumns() const
+    const ColumnRawPtrs & getActualColumns() const
     {
         return actual_columns;
     }
diff --git a/src/Interpreters/TranslateQualifiedNamesVisitor.h b/src/Interpreters/TranslateQualifiedNamesVisitor.h
index 73e45fc7ea0..00c85d08873 100644
--- a/src/Interpreters/TranslateQualifiedNamesVisitor.h
+++ b/src/Interpreters/TranslateQualifiedNamesVisitor.h
@@ -35,7 +35,7 @@ public:
             , has_columns(has_columns_)
         {}
 
-        bool hasColumn(const String & name) const { return source_columns.count(name); }
+        bool hasColumn(const String & name) const { return source_columns.contains(name); }
         bool hasTable() const { return !tables.empty(); }
         bool processAsterisks() const { return hasTable() && has_columns; }
         bool unknownColumn(size_t table_pos, const ASTIdentifier & identifier) const;
diff --git a/src/Interpreters/TreeCNFConverter.h b/src/Interpreters/TreeCNFConverter.h
index 3cc8b05d6c6..8258412f1a6 100644
--- a/src/Interpreters/TreeCNFConverter.h
+++ b/src/Interpreters/TreeCNFConverter.h
@@ -58,7 +58,7 @@ public:
         for (const auto & or_group : statements)
         {
             OrGroup filtered_group;
-            for (auto ast : or_group)
+            for (const auto & ast : or_group)
             {
                 if (predicate_is_unknown(ast))
                     filtered_group.insert(ast);
diff --git a/src/Interpreters/joinDispatch.h b/src/Interpreters/joinDispatch.h
index ca2fcd27cda..324d6a4d434 100644
--- a/src/Interpreters/joinDispatch.h
+++ b/src/Interpreters/joinDispatch.h
@@ -113,10 +113,9 @@ inline bool joinDispatch(JoinKind kind, JoinStrictness strictness, std::vector<c
         {
             using MapType = typename MapGetter<KINDS[i], STRICTNESSES[j]>::Map;
             std::vector<const MapType *> v;
+            v.resize(mapsv.size());
             for (const auto & el : mapsv)
-            {
                 v.push_back(&std::get<MapType>(*el));
-            }
 
             func(
                 std::integral_constant<JoinKind, KINDS[i]>(),
diff --git a/src/Loggers/OwnFormattingChannel.h b/src/Loggers/OwnFormattingChannel.h
index da7cde315f9..00300fe99ff 100644
--- a/src/Loggers/OwnFormattingChannel.h
+++ b/src/Loggers/OwnFormattingChannel.h
@@ -16,11 +16,11 @@ class OwnFormattingChannel : public Poco::Channel, public ExtendedLogChannel
 public:
     explicit OwnFormattingChannel(
         Poco::AutoPtr<OwnPatternFormatter> pFormatter_ = nullptr, Poco::AutoPtr<Poco::Channel> pChannel_ = nullptr)
-        : pFormatter(std::move(pFormatter_)), pChannel(std::move(pChannel_)), priority(Poco::Message::PRIO_TRACE)
+        : pFormatter(pFormatter_), pChannel(pChannel_), priority(Poco::Message::PRIO_TRACE)
     {
     }
 
-    void setChannel(Poco::AutoPtr<Poco::Channel> pChannel_) { pChannel = std::move(pChannel_); }
+    void setChannel(Poco::AutoPtr<Poco::Channel> pChannel_) { pChannel = pChannel_; }
 
     void setLevel(Poco::Message::Priority priority_) { priority = priority_; }
 
diff --git a/src/Processors/Executors/PullingAsyncPipelineExecutor.h b/src/Processors/Executors/PullingAsyncPipelineExecutor.h
index 361bcc0155c..618a7e790ab 100644
--- a/src/Processors/Executors/PullingAsyncPipelineExecutor.h
+++ b/src/Processors/Executors/PullingAsyncPipelineExecutor.h
@@ -58,7 +58,6 @@ private:
 
     void cancelWithExceptionHandling(CancelFunc && cancel_func);
 
-private:
     QueryPipeline & pipeline;
     std::shared_ptr<LazyOutputFormat> lazy_format;
     std::unique_ptr<Data> data;
diff --git a/src/Processors/Formats/IRowOutputFormat.h b/src/Processors/Formats/IRowOutputFormat.h
index 3a648f00eba..6083e1a6bc8 100644
--- a/src/Processors/Formats/IRowOutputFormat.h
+++ b/src/Processors/Formats/IRowOutputFormat.h
@@ -50,13 +50,13 @@ protected:
     virtual void writeFieldDelimiter() {}       /// delimiter between values
     virtual void writeRowStartDelimiter() {}    /// delimiter before each row
     virtual void writeRowEndDelimiter() {}      /// delimiter after each row
-    virtual void writePrefix() override {}      /// delimiter before resultset
-    virtual void writeSuffix() override {}      /// delimiter after resultset
+    void writePrefix() override {}      /// delimiter before resultset
+    void writeSuffix() override {}      /// delimiter after resultset
     virtual void writeBeforeTotals() {}
     virtual void writeAfterTotals() {}
     virtual void writeBeforeExtremes() {}
     virtual void writeAfterExtremes() {}
-    virtual void finalizeImpl() override {}  /// Write something after resultset, totals end extremes.
+    void finalizeImpl() override {}  /// Write something after resultset, totals end extremes.
 
     bool haveWrittenData() { return !first_row || getRowsReadBefore() != 0; }
 
diff --git a/src/Processors/Formats/Impl/ArrowBufferedStreams.cpp b/src/Processors/Formats/Impl/ArrowBufferedStreams.cpp
index 83d7a8b7bbd..c6e951ada6a 100644
--- a/src/Processors/Formats/Impl/ArrowBufferedStreams.cpp
+++ b/src/Processors/Formats/Impl/ArrowBufferedStreams.cpp
@@ -154,7 +154,7 @@ arrow::Result<int64_t> RandomAccessFileFromRandomAccessReadBuffer::GetSize()
 
 arrow::Result<int64_t> RandomAccessFileFromRandomAccessReadBuffer::ReadAt(int64_t position, int64_t nbytes, void* out)
 {
-    return in.readBigAt(reinterpret_cast<char*>(out), nbytes, position);
+    return in.readBigAt(reinterpret_cast<char*>(out), nbytes, position, nullptr);
 }
 
 arrow::Result<std::shared_ptr<arrow::Buffer>> RandomAccessFileFromRandomAccessReadBuffer::ReadAt(int64_t position, int64_t nbytes)
diff --git a/src/Processors/Formats/Impl/AvroRowInputFormat.h b/src/Processors/Formats/Impl/AvroRowInputFormat.h
index a8364df6e80..78dfe65fa5e 100644
--- a/src/Processors/Formats/Impl/AvroRowInputFormat.h
+++ b/src/Processors/Formats/Impl/AvroRowInputFormat.h
@@ -183,7 +183,7 @@ public:
     class SchemaRegistry;
 
 private:
-    virtual bool readRow(MutableColumns & columns, RowReadExtension & ext) override;
+    bool readRow(MutableColumns & columns, RowReadExtension & ext) override;
     void readPrefix() override;
 
     bool allowSyncAfterError() const override { return true; }
diff --git a/src/Processors/Formats/Impl/AvroRowOutputFormat.h b/src/Processors/Formats/Impl/AvroRowOutputFormat.h
index c526936b383..5d308d78966 100644
--- a/src/Processors/Formats/Impl/AvroRowOutputFormat.h
+++ b/src/Processors/Formats/Impl/AvroRowOutputFormat.h
@@ -48,16 +48,16 @@ class AvroRowOutputFormat final : public IRowOutputFormat
 {
 public:
     AvroRowOutputFormat(WriteBuffer & out_, const Block & header_, const FormatSettings & settings_);
-    virtual ~AvroRowOutputFormat() override;
+    ~AvroRowOutputFormat() override;
 
     String getName() const override { return "AvroRowOutputFormat"; }
 
 private:
     void write(const Columns & columns, size_t row_num) override;
     void writeField(const IColumn &, const ISerialization &, size_t) override {}
-    virtual void writePrefix() override;
-    virtual void finalizeImpl() override;
-    virtual void resetFormatterImpl() override;
+    void writePrefix() override;
+    void finalizeImpl() override;
+    void resetFormatterImpl() override;
 
     void createFileWriter();
 
diff --git a/src/Processors/Formats/Impl/ConstantExpressionTemplate.h b/src/Processors/Formats/Impl/ConstantExpressionTemplate.h
index 71d0d0f7134..11165d5c449 100644
--- a/src/Processors/Formats/Impl/ConstantExpressionTemplate.h
+++ b/src/Processors/Formats/Impl/ConstantExpressionTemplate.h
@@ -92,7 +92,6 @@ private:
     bool parseLiteralAndAssertType(
         ReadBuffer & istr, const TokenIterator & token_iterator, const IDataType * type, size_t column_idx, const Settings & settings);
 
-private:
     TemplateStructurePtr structure;
     MutableColumns columns;
 
diff --git a/src/Processors/Formats/Impl/ParallelFormattingOutputFormat.h b/src/Processors/Formats/Impl/ParallelFormattingOutputFormat.h
index 50a73646359..b04c1b0c16e 100644
--- a/src/Processors/Formats/Impl/ParallelFormattingOutputFormat.h
+++ b/src/Processors/Formats/Impl/ParallelFormattingOutputFormat.h
@@ -154,7 +154,7 @@ public:
     void setException(const String & exception_message_) override { exception_message = exception_message_; }
 
 private:
-    void consume(Chunk chunk) override final
+    void consume(Chunk chunk) final
     {
         addChunk(std::move(chunk), ProcessingUnitType::PLAIN, /*can_throw_exception*/ true);
     }
diff --git a/src/Processors/Formats/Impl/ParquetBlockOutputFormat.h b/src/Processors/Formats/Impl/ParquetBlockOutputFormat.h
index 3db90f4299b..422bae5c315 100644
--- a/src/Processors/Formats/Impl/ParquetBlockOutputFormat.h
+++ b/src/Processors/Formats/Impl/ParquetBlockOutputFormat.h
@@ -48,10 +48,10 @@ private:
             set(b);
         }
 
-        MemoryToken(MemoryToken && t)
+        MemoryToken(MemoryToken && t) /// NOLINT
           : parent(std::exchange(t.parent, nullptr)), bytes(std::exchange(t.bytes, 0)) {}
 
-        MemoryToken & operator=(MemoryToken && t)
+        MemoryToken & operator=(MemoryToken && t) /// NOLINT
         {
             parent = std::exchange(t.parent, nullptr);
             bytes = std::exchange(t.bytes, 0);
diff --git a/src/Processors/Formats/Impl/TabSeparatedRowOutputFormat.h b/src/Processors/Formats/Impl/TabSeparatedRowOutputFormat.h
index 9facba2052a..a514323873e 100644
--- a/src/Processors/Formats/Impl/TabSeparatedRowOutputFormat.h
+++ b/src/Processors/Formats/Impl/TabSeparatedRowOutputFormat.h
@@ -34,14 +34,14 @@ public:
 
 protected:
     void writeField(const IColumn & column, const ISerialization & serialization, size_t row_num) override;
-    void writeFieldDelimiter() override final;
+    void writeFieldDelimiter() final;
     void writeRowEndDelimiter() override;
 
     bool supportTotals() const override { return true; }
     bool supportExtremes() const override { return true; }
 
-    void writeBeforeTotals() override final;
-    void writeBeforeExtremes() override final;
+    void writeBeforeTotals() final;
+    void writeBeforeExtremes() final;
 
     void writePrefix() override;
     void writeLine(const std::vector<String> & values);
diff --git a/src/Processors/Formats/RowInputFormatWithNamesAndTypes.h b/src/Processors/Formats/RowInputFormatWithNamesAndTypes.h
index d0457990714..b7d9507151e 100644
--- a/src/Processors/Formats/RowInputFormatWithNamesAndTypes.h
+++ b/src/Processors/Formats/RowInputFormatWithNamesAndTypes.h
@@ -175,7 +175,7 @@ public:
     void transformTypesIfNeeded(DataTypePtr & type, DataTypePtr & new_type) override;
 
 protected:
-    virtual std::optional<DataTypes> readRowAndGetDataTypes() override;
+    std::optional<DataTypes> readRowAndGetDataTypes() override;
 
     virtual std::optional<DataTypes> readRowAndGetDataTypesImpl()
     {
diff --git a/src/Processors/Merges/Algorithms/AggregatingSortedAlgorithm.h b/src/Processors/Merges/Algorithms/AggregatingSortedAlgorithm.h
index 72a1f4871d8..aa221573151 100644
--- a/src/Processors/Merges/Algorithms/AggregatingSortedAlgorithm.h
+++ b/src/Processors/Merges/Algorithms/AggregatingSortedAlgorithm.h
@@ -47,7 +47,7 @@ public:
         bool created = false;
 
         SimpleAggregateDescription(
-            AggregateFunctionPtr function_, const size_t column_number_,
+            AggregateFunctionPtr function_, size_t column_number_,
             DataTypePtr nested_type_, DataTypePtr real_type_);
 
         void createState();
diff --git a/src/Processors/Merges/Algorithms/IMergingAlgorithm.h b/src/Processors/Merges/Algorithms/IMergingAlgorithm.h
index d55ceb78af4..6e352c3f104 100644
--- a/src/Processors/Merges/Algorithms/IMergingAlgorithm.h
+++ b/src/Processors/Merges/Algorithms/IMergingAlgorithm.h
@@ -31,7 +31,7 @@ public:
 
         IColumn::Permutation * permutation = nullptr;
 
-        void swap(Input & other)
+        void swap(Input & other) noexcept
         {
             chunk.swap(other.chunk);
             std::swap(skip_last_row, other.skip_last_row);
diff --git a/src/Processors/Merges/Algorithms/RowRef.h b/src/Processors/Merges/Algorithms/RowRef.h
index ee64224d44d..af379ec59f6 100644
--- a/src/Processors/Merges/Algorithms/RowRef.h
+++ b/src/Processors/Merges/Algorithms/RowRef.h
@@ -181,7 +181,7 @@ struct RowRefWithOwnedChunk
 
     UInt64 source_stream_index = 0;
 
-    void swap(RowRefWithOwnedChunk & other)
+    void swap(RowRefWithOwnedChunk & other) /// NOLINT(performance-noexcept-swap)
     {
         owned_chunk.swap(other.owned_chunk);
         std::swap(all_columns, other.all_columns);
diff --git a/src/Processors/Transforms/MergeJoinTransform.h b/src/Processors/Transforms/MergeJoinTransform.h
index 793de00db40..959550067f7 100644
--- a/src/Processors/Transforms/MergeJoinTransform.h
+++ b/src/Processors/Transforms/MergeJoinTransform.h
@@ -230,9 +230,9 @@ public:
     explicit MergeJoinAlgorithm(JoinPtr table_join, const Blocks & input_headers, size_t max_block_size_);
 
     const char * getName() const override { return "MergeJoinAlgorithm"; }
-    virtual void initialize(Inputs inputs) override;
-    virtual void consume(Input & input, size_t source_num) override;
-    virtual Status merge() override;
+    void initialize(Inputs inputs) override;
+    void consume(Input & input, size_t source_num) override;
+    Status merge() override;
 
     void logElapsed(double seconds);
 
diff --git a/src/Processors/Transforms/PasteJoinTransform.h b/src/Processors/Transforms/PasteJoinTransform.h
index 04cb5486cd5..6a7e65ee27c 100644
--- a/src/Processors/Transforms/PasteJoinTransform.h
+++ b/src/Processors/Transforms/PasteJoinTransform.h
@@ -32,9 +32,9 @@ public:
     explicit PasteJoinAlgorithm(JoinPtr table_join, const Blocks & input_headers, size_t max_block_size_);
 
     const char * getName() const override { return "PasteJoinAlgorithm"; }
-    virtual void initialize(Inputs inputs) override;
-    virtual void consume(Input & input, size_t source_num) override;
-    virtual Status merge() override;
+    void initialize(Inputs inputs) override;
+    void consume(Input & input, size_t source_num) override;
+    Status merge() override;
 
     void logElapsed(double seconds);
 
diff --git a/src/QueryPipeline/QueryPipelineBuilder.h b/src/QueryPipeline/QueryPipelineBuilder.h
index 5d273df7068..f0b2ead687e 100644
--- a/src/QueryPipeline/QueryPipelineBuilder.h
+++ b/src/QueryPipeline/QueryPipelineBuilder.h
@@ -188,7 +188,7 @@ public:
         concurrency_control = concurrency_control_;
     }
 
-    bool getConcurrencyControl()
+    bool getConcurrencyControl() const
     {
         return concurrency_control;
     }
diff --git a/src/Server/GRPCServer.cpp b/src/Server/GRPCServer.cpp
index f21991e8d58..dfae812c620 100644
--- a/src/Server/GRPCServer.cpp
+++ b/src/Server/GRPCServer.cpp
@@ -618,7 +618,6 @@ namespace
 
 
     /// Handles a connection after a responder is started (i.e. after getting a new call).
-// NOLINTBEGIN(clang-analyzer-optin.performance.Padding)
     class Call
     {
     public:
@@ -724,7 +723,6 @@ namespace
 
         ThreadFromGlobalPool call_thread;
     };
-// NOLINTEND(clang-analyzer-optin.performance.Padding)
 
     Call::Call(CallType call_type_, std::unique_ptr<BaseResponder> responder_, IServer & iserver_, LoggerRawPtr log_)
         : call_type(call_type_), responder(std::move(responder_)), iserver(iserver_), log(log_)
diff --git a/src/Server/HTTP/HTTPServerResponse.h b/src/Server/HTTP/HTTPServerResponse.h
index 6c5be008bf8..8edb785e7c5 100644
--- a/src/Server/HTTP/HTTPServerResponse.h
+++ b/src/Server/HTTP/HTTPServerResponse.h
@@ -90,7 +90,7 @@ public:
         resizeIfNeeded(buf_size);
     }
 
-    bool isChunked()
+    bool isChunked() const
     {
         return chunked;
     }
@@ -103,7 +103,7 @@ public:
         resizeIfNeeded(length);
     }
 
-    size_t isFixedLength()
+    size_t isFixedLength() const
     {
         return chunked ? 0 : fixed_length;
     }
@@ -116,7 +116,7 @@ public:
         resizeIfNeeded(buf_size);
     }
 
-    bool isPlain()
+    bool isPlain() const
     {
         return !(isChunked() || isFixedLength());
     }
diff --git a/src/Server/HTTPHandler.h b/src/Server/HTTPHandler.h
index 0e30b466694..39cef0e7112 100644
--- a/src/Server/HTTPHandler.h
+++ b/src/Server/HTTPHandler.h
@@ -32,7 +32,7 @@ class HTTPHandler : public HTTPRequestHandler
 {
 public:
     HTTPHandler(IServer & server_, const std::string & name, const std::optional<String> & content_type_override_);
-    virtual ~HTTPHandler() override;
+    ~HTTPHandler() override;
 
     void handleRequest(HTTPServerRequest & request, HTTPServerResponse & response, const ProfileEvents::Event & write_event) override;
 
diff --git a/src/Storages/DataLakes/S3MetadataReader.h b/src/Storages/DataLakes/S3MetadataReader.h
index cae7dd1fa3d..c29a66b3813 100644
--- a/src/Storages/DataLakes/S3MetadataReader.h
+++ b/src/Storages/DataLakes/S3MetadataReader.h
@@ -6,11 +6,11 @@
 
 #include <Storages/StorageS3.h>
 
-class ReadBuffer;
-
 namespace DB
 {
 
+class ReadBuffer;
+
 struct S3DataLakeMetadataReadHelper
 {
     static std::shared_ptr<ReadBuffer> createReadBuffer(
diff --git a/src/Storages/FileLog/FileLogSource.h b/src/Storages/FileLog/FileLogSource.h
index 281b109ff4b..3ac2b407e10 100644
--- a/src/Storages/FileLog/FileLogSource.h
+++ b/src/Storages/FileLog/FileLogSource.h
@@ -30,7 +30,7 @@ public:
 
     void onFinish();
 
-    virtual ~FileLogSource() override;
+    ~FileLogSource() override;
 
 protected:
     Chunk generate() override;
diff --git a/src/Storages/Hive/HiveFile.h b/src/Storages/Hive/HiveFile.h
index 1f5e31f1d54..601c7f2a310 100644
--- a/src/Storages/Hive/HiveFile.h
+++ b/src/Storages/Hive/HiveFile.h
@@ -65,7 +65,7 @@ public:
         {ORC_INPUT_FORMAT, FileFormat::ORC},
     };
 
-    static inline bool isFormatClass(const String & format_class) { return VALID_HDFS_FORMATS.count(format_class) > 0; }
+    static inline bool isFormatClass(const String & format_class) { return VALID_HDFS_FORMATS.contains(format_class); }
     static inline FileFormat toFileFormat(const String & format_class)
     {
         if (isFormatClass(format_class))
diff --git a/src/Storages/IStorage.cpp b/src/Storages/IStorage.cpp
index 9852220241f..b532abc9074 100644
--- a/src/Storages/IStorage.cpp
+++ b/src/Storages/IStorage.cpp
@@ -305,7 +305,7 @@ void IStorage::backupData(BackupEntriesCollector &, const String &, const std::o
 void IStorage::restoreDataFromBackup(RestorerFromBackup & restorer, const String & data_path_in_backup, const std::optional<ASTs> &)
 {
     /// If an inherited class doesn't override restoreDataFromBackup() that means it doesn't backup any data.
-    auto filenames = restorer.getBackup()->listFiles(data_path_in_backup);
+    auto filenames = restorer.getBackup()->listFiles(data_path_in_backup, /*recursive*/ false);
     if (!filenames.empty())
         throw Exception(ErrorCodes::CANNOT_RESTORE_TABLE, "Cannot restore table {}: Folder {} in backup must be empty",
                         getStorageID().getFullTableName(), data_path_in_backup);
diff --git a/src/Storages/Kafka/KafkaConsumer.h b/src/Storages/Kafka/KafkaConsumer.h
index 9cc78d42856..f160d1c0855 100644
--- a/src/Storages/Kafka/KafkaConsumer.h
+++ b/src/Storages/Kafka/KafkaConsumer.h
@@ -184,7 +184,7 @@ private:
     std::atomic<UInt64> last_rebalance_timestamp_usec = 0;
     std::atomic<UInt64> num_rebalance_assignments = 0;
     std::atomic<UInt64> num_rebalance_revocations = 0;
-    std::atomic<bool> in_use = 0;
+    std::atomic<bool> in_use = false;
     /// Last used time (for TTL)
     std::atomic<UInt64> last_used_usec = 0;
 
diff --git a/src/Storages/Kafka/parseSyslogLevel.cpp b/src/Storages/Kafka/parseSyslogLevel.cpp
index f4358a13efe..43630a5001f 100644
--- a/src/Storages/Kafka/parseSyslogLevel.cpp
+++ b/src/Storages/Kafka/parseSyslogLevel.cpp
@@ -4,7 +4,7 @@
 /// Must be in a separate compilation unit due to macros overlaps:
 /// - syslog (LOG_DEBUG/...)
 /// - logger_useful.h (LOG_DEBUG(...)/...)
-std::pair<Poco::Message::Priority, DB::LogsLevel> parseSyslogLevel(const int level)
+std::pair<Poco::Message::Priority, DB::LogsLevel> parseSyslogLevel(int level)
 {
     using DB::LogsLevel;
     using Poco::Message;
diff --git a/src/Storages/Kafka/parseSyslogLevel.h b/src/Storages/Kafka/parseSyslogLevel.h
index f0fb4968631..a004fc3d29f 100644
--- a/src/Storages/Kafka/parseSyslogLevel.h
+++ b/src/Storages/Kafka/parseSyslogLevel.h
@@ -4,4 +4,4 @@
 #include <Poco/Message.h>
 #include <Common/CurrentThread.h>
 
-std::pair<Poco::Message::Priority, DB::LogsLevel> parseSyslogLevel(const int level);
+std::pair<Poco::Message::Priority, DB::LogsLevel> parseSyslogLevel(int level);
diff --git a/src/Storages/MergeTree/ColumnSizeEstimator.h b/src/Storages/MergeTree/ColumnSizeEstimator.h
index e512415ff05..1307a5f493e 100644
--- a/src/Storages/MergeTree/ColumnSizeEstimator.h
+++ b/src/Storages/MergeTree/ColumnSizeEstimator.h
@@ -23,9 +23,9 @@ public:
         : map(std::move(map_))
     {
         for (const auto & name : key_columns)
-            if (!map.count(name)) map[name] = 0;
+            if (!map.contains(name)) map[name] = 0;
         for (const auto & name : ordinary_columns)
-            if (!map.count(name)) map[name] = 0;
+            if (!map.contains(name)) map[name] = 0;
 
         for (const auto & name : key_columns)
             sum_index_columns += map.at(name);
diff --git a/src/Storages/MergeTree/IMergeTreeDataPart.h b/src/Storages/MergeTree/IMergeTreeDataPart.h
index 8bd32e777bc..57243335238 100644
--- a/src/Storages/MergeTree/IMergeTreeDataPart.h
+++ b/src/Storages/MergeTree/IMergeTreeDataPart.h
@@ -218,7 +218,6 @@ public:
 
     const MergeTreeData & storage;
 
-public:
     const String & name;    // const ref to private mutable_name
     MergeTreePartInfo info;
 
diff --git a/src/Storages/MergeTree/LeaderElection.h b/src/Storages/MergeTree/LeaderElection.h
index 3bd486fd54a..b7dc9c31655 100644
--- a/src/Storages/MergeTree/LeaderElection.h
+++ b/src/Storages/MergeTree/LeaderElection.h
@@ -19,7 +19,7 @@ namespace zkutil
   * For now, every replica can become leader if there is no leader among replicas with old version.
   */
 
-void checkNoOldLeaders(LoggerPtr log, ZooKeeper & zookeeper, const String path)
+inline void checkNoOldLeaders(LoggerPtr log, ZooKeeper & zookeeper, const String path)
 {
     /// Previous versions (before 21.12) used to create ephemeral sequential node path/leader_election-
     /// Replica with the lexicographically smallest node name becomes leader (before 20.6) or enables multi-leader mode (since 20.6)
diff --git a/src/Storages/MergeTree/MarkRange.h b/src/Storages/MergeTree/MarkRange.h
index f31d6a2a73b..1d9d0a1e27e 100644
--- a/src/Storages/MergeTree/MarkRange.h
+++ b/src/Storages/MergeTree/MarkRange.h
@@ -33,7 +33,7 @@ struct MarkRange
 
 struct MarkRanges : public std::deque<MarkRange>
 {
-    using std::deque<MarkRange>::deque;
+    using std::deque<MarkRange>::deque; /// NOLINT(modernize-type-traits)
 
     size_t getNumberOfMarks() const;
 
diff --git a/src/Storages/MergeTree/MergeTreeData.cpp b/src/Storages/MergeTree/MergeTreeData.cpp
index aacac5366a9..ba30c13049b 100644
--- a/src/Storages/MergeTree/MergeTreeData.cpp
+++ b/src/Storages/MergeTree/MergeTreeData.cpp
@@ -5435,7 +5435,7 @@ void MergeTreeData::restorePartsFromBackup(RestorerFromBackup & restorer, const
         partition_ids = getPartitionIDsFromQuery(*partitions, restorer.getContext());
 
     auto backup = restorer.getBackup();
-    Strings part_names = backup->listFiles(data_path_in_backup);
+    Strings part_names = backup->listFiles(data_path_in_backup, /*recursive*/ false);
     boost::remove_erase(part_names, "mutations");
 
     bool restore_broken_parts_as_detached = restorer.getRestoreSettings().restore_broken_parts_as_detached;
@@ -5516,7 +5516,7 @@ void MergeTreeData::restorePartFromBackup(std::shared_ptr<RestoredPartsHolder> r
         if (filename.ends_with(IMergeTreeDataPart::TXN_VERSION_METADATA_FILE_NAME))
             continue;
 
-        size_t file_size = backup->copyFileToDisk(part_path_in_backup_fs / filename, disk, temp_part_dir / filename);
+        size_t file_size = backup->copyFileToDisk(part_path_in_backup_fs / filename, disk, temp_part_dir / filename, WriteMode::Rewrite);
         reservation->update(reservation->getSize() - file_size);
     }
 
diff --git a/src/Storages/MergeTree/MergeTreeData.h b/src/Storages/MergeTree/MergeTreeData.h
index c864fce4659..8305c7c6ce9 100644
--- a/src/Storages/MergeTree/MergeTreeData.h
+++ b/src/Storages/MergeTree/MergeTreeData.h
@@ -1372,7 +1372,7 @@ protected:
                 latest_fail_time_us = static_cast<size_t>(Poco::Timestamp().epochMicroseconds());
             }
 
-            bool partCanBeMutated()
+            bool partCanBeMutated() const
             {
                 if (max_postpone_time_ms == 0)
                     return true;
diff --git a/src/Storages/MergeTree/MergeTreeMutationStatus.h b/src/Storages/MergeTree/MergeTreeMutationStatus.h
index b6f62e1218f..4bc9549872c 100644
--- a/src/Storages/MergeTree/MergeTreeMutationStatus.h
+++ b/src/Storages/MergeTree/MergeTreeMutationStatus.h
@@ -13,6 +13,7 @@ namespace DB
 
 struct MergeTreeMutationStatus
 {
+/// NOLINTBEGIN(readability-redundant-string-init)
     String id = "";
     String command = "";
     time_t create_time = 0;
@@ -30,6 +31,7 @@ struct MergeTreeMutationStatus
 
     /// FIXME: currently unused, but would be much better to report killed mutations with this flag.
     bool is_killed = false;
+/// NOLINTEND(readability-redundant-string-init)
 };
 
 /// Check mutation status and throw exception in case of error during mutation
diff --git a/src/Storages/MergeTree/RPNBuilder.h b/src/Storages/MergeTree/RPNBuilder.h
index d750c02d3e1..f8924ff6186 100644
--- a/src/Storages/MergeTree/RPNBuilder.h
+++ b/src/Storages/MergeTree/RPNBuilder.h
@@ -226,7 +226,7 @@ private:
                       * - in this case `n - 1` elements are added (where `n` is the number of arguments).
                       */
                     if (argument_index != 0 || element.function == RPNElement::FUNCTION_NOT)
-                        rpn_elements.emplace_back(std::move(element));
+                        rpn_elements.emplace_back(std::move(element)); /// NOLINT(bugprone-use-after-move,hicpp-invalid-access-moved)
                 }
 
                 if (arguments_size == 0 && function_node.getFunctionName() == "indexHint")
diff --git a/src/Storages/MergeTree/RangesInDataPart.h b/src/Storages/MergeTree/RangesInDataPart.h
index 0bc5829c82e..e275f2c27e7 100644
--- a/src/Storages/MergeTree/RangesInDataPart.h
+++ b/src/Storages/MergeTree/RangesInDataPart.h
@@ -67,7 +67,7 @@ struct RangesInDataPart
 
 struct RangesInDataParts: public std::vector<RangesInDataPart>
 {
-    using std::vector<RangesInDataPart>::vector;
+    using std::vector<RangesInDataPart>::vector; /// NOLINT(modernize-type-traits)
 
     RangesInDataPartsDescription getDescriptions() const;
 
diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeQuorumAddedParts.h b/src/Storages/MergeTree/ReplicatedMergeTreeQuorumAddedParts.h
index f0f737cb1e6..81045460ab7 100644
--- a/src/Storages/MergeTree/ReplicatedMergeTreeQuorumAddedParts.h
+++ b/src/Storages/MergeTree/ReplicatedMergeTreeQuorumAddedParts.h
@@ -66,7 +66,7 @@ struct ReplicatedMergeTreeQuorumAddedParts
     }
 
     /// Read added blocks when node in ZooKeeper supports only one partition.
-    PartitionIdToPartName readV1(ReadBuffer & in)
+    PartitionIdToPartName readV1(ReadBuffer & in) const
     {
         PartitionIdToPartName parts_in_quorum;
 
diff --git a/src/Storages/MergeTree/checkDataPart.cpp b/src/Storages/MergeTree/checkDataPart.cpp
index aea0020f89d..2d7727b57cb 100644
--- a/src/Storages/MergeTree/checkDataPart.cpp
+++ b/src/Storages/MergeTree/checkDataPart.cpp
@@ -56,7 +56,7 @@ bool isNotEnoughMemoryErrorCode(int code)
         || code == ErrorCodes::CANNOT_MREMAP;
 }
 
-bool isRetryableException(const std::exception_ptr exception_ptr)
+bool isRetryableException(std::exception_ptr exception_ptr)
 {
     try
     {
diff --git a/src/Storages/MergeTree/checkDataPart.h b/src/Storages/MergeTree/checkDataPart.h
index d0e48b6f80a..ada89a54dcc 100644
--- a/src/Storages/MergeTree/checkDataPart.h
+++ b/src/Storages/MergeTree/checkDataPart.h
@@ -13,6 +13,6 @@ IMergeTreeDataPart::Checksums checkDataPart(
     std::function<bool()> is_cancelled = []{ return false; });
 
 bool isNotEnoughMemoryErrorCode(int code);
-bool isRetryableException(const std::exception_ptr exception_ptr);
+bool isRetryableException(std::exception_ptr exception_ptr);
 
 }
diff --git a/src/Storages/RabbitMQ/RabbitMQProducer.cpp b/src/Storages/RabbitMQ/RabbitMQProducer.cpp
index 7ad83213b9b..f0fbc8c5c06 100644
--- a/src/Storages/RabbitMQ/RabbitMQProducer.cpp
+++ b/src/Storages/RabbitMQ/RabbitMQProducer.cpp
@@ -27,9 +27,9 @@ RabbitMQProducer::RabbitMQProducer(
     const RabbitMQConfiguration & configuration_,
     const Names & routing_keys_,
     const String & exchange_name_,
-    const AMQP::ExchangeType exchange_type_,
-    const size_t channel_id_base_,
-    const bool persistent_,
+    AMQP::ExchangeType exchange_type_,
+    size_t channel_id_base_,
+    bool persistent_,
     std::atomic<bool> & shutdown_called_,
     LoggerPtr log_)
     : AsynchronousMessageProducer(log_)
diff --git a/src/Storages/RabbitMQ/RabbitMQProducer.h b/src/Storages/RabbitMQ/RabbitMQProducer.h
index a790eda0d08..e43518aec06 100644
--- a/src/Storages/RabbitMQ/RabbitMQProducer.h
+++ b/src/Storages/RabbitMQ/RabbitMQProducer.h
@@ -20,9 +20,9 @@ public:
         const RabbitMQConfiguration & configuration_,
         const Names & routing_keys_,
         const String & exchange_name_,
-        const AMQP::ExchangeType exchange_type_,
-        const size_t channel_id_base_,
-        const bool persistent_,
+        AMQP::ExchangeType exchange_type_,
+        size_t channel_id_base_,
+        bool persistent_,
         std::atomic<bool> & shutdown_called_,
         LoggerPtr log_);
 
diff --git a/src/Storages/S3Queue/StorageS3Queue.h b/src/Storages/S3Queue/StorageS3Queue.h
index bdd3ab7b687..fe86759ca62 100644
--- a/src/Storages/S3Queue/StorageS3Queue.h
+++ b/src/Storages/S3Queue/StorageS3Queue.h
@@ -14,11 +14,6 @@
 #include <Storages/StorageFactory.h>
 
 
-namespace Aws::S3
-{
-class Client;
-}
-
 namespace DB
 {
 class S3QueueFilesMetadata;
diff --git a/src/Storages/StorageInMemoryMetadata.h b/src/Storages/StorageInMemoryMetadata.h
index 69cd3422a7d..afc3bc81ef6 100644
--- a/src/Storages/StorageInMemoryMetadata.h
+++ b/src/Storages/StorageInMemoryMetadata.h
@@ -72,8 +72,8 @@ struct StorageInMemoryMetadata
     StorageInMemoryMetadata(const StorageInMemoryMetadata & other);
     StorageInMemoryMetadata & operator=(const StorageInMemoryMetadata & other);
 
-    StorageInMemoryMetadata(StorageInMemoryMetadata && other) = default; /// NOLINT
-    StorageInMemoryMetadata & operator=(StorageInMemoryMetadata && other) = default; /// NOLINT
+    StorageInMemoryMetadata(StorageInMemoryMetadata && other) = default; /// NOLINT(hicpp-noexcept-move,performance-noexcept-move-constructor)
+    StorageInMemoryMetadata & operator=(StorageInMemoryMetadata && other) = default; /// NOLINT(hicpp-noexcept-move,performance-noexcept-move-constructor)
 
     /// NOTE: Thread unsafe part. You should not modify same StorageInMemoryMetadata
     /// structure from different threads. It should be used as MultiVersion
diff --git a/src/Storages/StorageMongoDBSocketFactory.h b/src/Storages/StorageMongoDBSocketFactory.h
index 5fc423c63cb..2dc9cbddf8a 100644
--- a/src/Storages/StorageMongoDBSocketFactory.h
+++ b/src/Storages/StorageMongoDBSocketFactory.h
@@ -9,7 +9,7 @@ namespace DB
 class StorageMongoDBSocketFactory : public Poco::MongoDB::Connection::SocketFactory
 {
 public:
-    virtual Poco::Net::StreamSocket createSocket(const std::string & host, int port, Poco::Timespan connectTimeout, bool secure) override;
+    Poco::Net::StreamSocket createSocket(const std::string & host, int port, Poco::Timespan connectTimeout, bool secure) override;
 
 private:
     static Poco::Net::StreamSocket createPlainSocket(const std::string & host, int port, Poco::Timespan connectTimeout);
diff --git a/src/Storages/StorageS3.h b/src/Storages/StorageS3.h
index 7c5db76aadf..f570a7e3a58 100644
--- a/src/Storages/StorageS3.h
+++ b/src/Storages/StorageS3.h
@@ -26,11 +26,6 @@
 #include <Storages/prepareReadingFromFormat.h>
 #include <IO/S3/BlobStorageLogWriter.h>
 
-namespace Aws::S3
-{
-    class Client;
-}
-
 namespace DB
 {
 
diff --git a/src/Storages/StorageView.cpp b/src/Storages/StorageView.cpp
index 673ca61cd50..db568a1d5ab 100644
--- a/src/Storages/StorageView.cpp
+++ b/src/Storages/StorageView.cpp
@@ -110,7 +110,7 @@ StorageView::StorageView(
     const ASTCreateQuery & query,
     const ColumnsDescription & columns_,
     const String & comment,
-    const bool is_parameterized_view_)
+    bool is_parameterized_view_)
     : IStorage(table_id_)
 {
     StorageInMemoryMetadata storage_metadata;
diff --git a/src/Storages/StorageView.h b/src/Storages/StorageView.h
index 4d265eed86b..053fca807d5 100644
--- a/src/Storages/StorageView.h
+++ b/src/Storages/StorageView.h
@@ -16,7 +16,7 @@ public:
         const ASTCreateQuery & query,
         const ColumnsDescription & columns_,
         const String & comment,
-        const bool is_parameterized_view_=false);
+        bool is_parameterized_view_ = false);
 
     std::string getName() const override { return "View"; }
     bool isView() const override { return true; }
@@ -45,7 +45,7 @@ public:
         replaceWithSubquery(select_query, metadata_snapshot->getSelectQuery().inner_query->clone(), view_name, parameterized_view);
     }
 
-    static void replaceWithSubquery(ASTSelectQuery & outer_query, ASTPtr view_query, ASTPtr & view_name, const bool parameterized_view);
+    static void replaceWithSubquery(ASTSelectQuery & outer_query, ASTPtr view_query, ASTPtr & view_name, bool parameterized_view);
     static ASTPtr restoreViewName(ASTSelectQuery & select_query, const ASTPtr & view_name);
 
 protected:
diff --git a/src/Storages/System/StorageSystemPartsBase.h b/src/Storages/System/StorageSystemPartsBase.h
index 0527a99b841..10d1a3a2e0e 100644
--- a/src/Storages/System/StorageSystemPartsBase.h
+++ b/src/Storages/System/StorageSystemPartsBase.h
@@ -94,7 +94,7 @@ protected:
         // nullptr means table was dropped while acquiring the lock
         return info.table_lock != nullptr;
     }
-protected:
+
     String query_id;
     Settings settings;
 
diff --git a/src/Storages/System/StorageSystemTimeZones.h b/src/Storages/System/StorageSystemTimeZones.h
index 160e8854e3e..3f9db2f53e2 100644
--- a/src/Storages/System/StorageSystemTimeZones.h
+++ b/src/Storages/System/StorageSystemTimeZones.h
@@ -17,7 +17,6 @@ public:
 
     using IStorageSystemOneBlock::IStorageSystemOneBlock;
 
-public:
     std::string getName() const override { return "SystemTimeZones"; }
 
     static ColumnsDescription getColumnsDescription();

From e5baeaba9c81d9ad1d49215f573b82c5ec4254d3 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Fri, 15 Mar 2024 13:28:42 +0000
Subject: [PATCH 429/443] Fix FastTest

---
 src/Interpreters/joinDispatch.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Interpreters/joinDispatch.h b/src/Interpreters/joinDispatch.h
index 324d6a4d434..dccbe68fdb6 100644
--- a/src/Interpreters/joinDispatch.h
+++ b/src/Interpreters/joinDispatch.h
@@ -113,7 +113,7 @@ inline bool joinDispatch(JoinKind kind, JoinStrictness strictness, std::vector<c
         {
             using MapType = typename MapGetter<KINDS[i], STRICTNESSES[j]>::Map;
             std::vector<const MapType *> v;
-            v.resize(mapsv.size());
+            v.reserve(mapsv.size());
             for (const auto & el : mapsv)
                 v.push_back(&std::get<MapType>(*el));
 

From 3cff8206b32419bfcceeee8c9fe0fa02e3631a85 Mon Sep 17 00:00:00 2001
From: Antonio Andelic <antonio@clickhouse.com>
Date: Fri, 15 Mar 2024 16:31:20 +0000
Subject: [PATCH 430/443] Split 02125_lz4_compression_bug

---
 .../0_stateless/02125_lz4_compression_bug.lib | 19 ++++++++
 .../02125_lz4_compression_bug.reference       | 45 -------------------
 .../0_stateless/02125_lz4_compression_bug.sh  | 21 ---------
 .../02125_lz4_compression_bug_CSV.reference   |  4 ++
 .../02125_lz4_compression_bug_CSV.sh          |  6 +++
 ...mpression_bug_JSONCompactEachRow.reference |  4 ++
 ..._lz4_compression_bug_JSONCompactEachRow.sh |  6 +++
 ..._lz4_compression_bug_JSONEachRow.reference |  4 ++
 .../02125_lz4_compression_bug_JSONEachRow.sh  |  6 +++
 ...mpression_bug_JSONStringsEachRow.reference |  4 ++
 ..._lz4_compression_bug_JSONStringsEachRow.sh |  6 +++
 ...02125_lz4_compression_bug_Native.reference |  4 ++
 .../02125_lz4_compression_bug_Native.sh       |  6 +++
 .../02125_lz4_compression_bug_TSKV.reference  |  4 ++
 .../02125_lz4_compression_bug_TSKV.sh         |  6 +++
 .../02125_lz4_compression_bug_TSV.reference   |  4 ++
 .../02125_lz4_compression_bug_TSV.sh          |  6 +++
 ...02125_lz4_compression_bug_Values.reference |  4 ++
 .../02125_lz4_compression_bug_Values.sh       |  6 +++
 19 files changed, 99 insertions(+), 66 deletions(-)
 create mode 100755 tests/queries/0_stateless/02125_lz4_compression_bug.lib
 delete mode 100644 tests/queries/0_stateless/02125_lz4_compression_bug.reference
 delete mode 100755 tests/queries/0_stateless/02125_lz4_compression_bug.sh
 create mode 100644 tests/queries/0_stateless/02125_lz4_compression_bug_CSV.reference
 create mode 100755 tests/queries/0_stateless/02125_lz4_compression_bug_CSV.sh
 create mode 100644 tests/queries/0_stateless/02125_lz4_compression_bug_JSONCompactEachRow.reference
 create mode 100755 tests/queries/0_stateless/02125_lz4_compression_bug_JSONCompactEachRow.sh
 create mode 100644 tests/queries/0_stateless/02125_lz4_compression_bug_JSONEachRow.reference
 create mode 100755 tests/queries/0_stateless/02125_lz4_compression_bug_JSONEachRow.sh
 create mode 100644 tests/queries/0_stateless/02125_lz4_compression_bug_JSONStringsEachRow.reference
 create mode 100755 tests/queries/0_stateless/02125_lz4_compression_bug_JSONStringsEachRow.sh
 create mode 100644 tests/queries/0_stateless/02125_lz4_compression_bug_Native.reference
 create mode 100755 tests/queries/0_stateless/02125_lz4_compression_bug_Native.sh
 create mode 100644 tests/queries/0_stateless/02125_lz4_compression_bug_TSKV.reference
 create mode 100755 tests/queries/0_stateless/02125_lz4_compression_bug_TSKV.sh
 create mode 100644 tests/queries/0_stateless/02125_lz4_compression_bug_TSV.reference
 create mode 100755 tests/queries/0_stateless/02125_lz4_compression_bug_TSV.sh
 create mode 100644 tests/queries/0_stateless/02125_lz4_compression_bug_Values.reference
 create mode 100755 tests/queries/0_stateless/02125_lz4_compression_bug_Values.sh

diff --git a/tests/queries/0_stateless/02125_lz4_compression_bug.lib b/tests/queries/0_stateless/02125_lz4_compression_bug.lib
new file mode 100755
index 00000000000..d13932aa166
--- /dev/null
+++ b/tests/queries/0_stateless/02125_lz4_compression_bug.lib
@@ -0,0 +1,19 @@
+#!/usr/bin/env bash
+# Tags: no-fasttest
+
+CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
+# shellcheck source=../shell_config.sh
+. "$CURDIR"/../shell_config.sh
+
+format=$1
+
+${CLICKHOUSE_CLIENT} --query "DROP TABLE IF EXISTS file"
+${CLICKHOUSE_CLIENT} --query "CREATE TABLE file (x UInt64) ENGINE = File($format, '${CLICKHOUSE_DATABASE}/data.$format.lz4')"
+for size in 10000 100000 1000000 2500000
+do
+    ${CLICKHOUSE_CLIENT} --query "TRUNCATE TABLE file"
+    ${CLICKHOUSE_CLIENT} --query "INSERT INTO file SELECT * FROM numbers($size)"
+    ${CLICKHOUSE_CLIENT} --query "SELECT max(x) FROM file"
+done
+
+${CLICKHOUSE_CLIENT} --query "DROP TABLE file"
diff --git a/tests/queries/0_stateless/02125_lz4_compression_bug.reference b/tests/queries/0_stateless/02125_lz4_compression_bug.reference
deleted file mode 100644
index 713767e1d44..00000000000
--- a/tests/queries/0_stateless/02125_lz4_compression_bug.reference
+++ /dev/null
@@ -1,45 +0,0 @@
-Native
-9999
-99999
-999999
-2499999
-Values
-9999
-99999
-999999
-2499999
-JSONCompactEachRow
-9999
-99999
-999999
-2499999
-TSKV
-9999
-99999
-999999
-2499999
-TSV
-9999
-99999
-999999
-2499999
-CSV
-9999
-99999
-999999
-2499999
-JSONEachRow
-9999
-99999
-999999
-2499999
-JSONCompactEachRow
-9999
-99999
-999999
-2499999
-JSONStringsEachRow
-9999
-99999
-999999
-2499999
diff --git a/tests/queries/0_stateless/02125_lz4_compression_bug.sh b/tests/queries/0_stateless/02125_lz4_compression_bug.sh
deleted file mode 100755
index 77646b01298..00000000000
--- a/tests/queries/0_stateless/02125_lz4_compression_bug.sh
+++ /dev/null
@@ -1,21 +0,0 @@
-#!/usr/bin/env bash
-# Tags: no-parallel, no-fasttest
-
-CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
-# shellcheck source=../shell_config.sh
-. "$CURDIR"/../shell_config.sh
-
-for format in Native Values JSONCompactEachRow TSKV TSV CSV JSONEachRow JSONCompactEachRow JSONStringsEachRow
-do
-    echo $format
-    ${CLICKHOUSE_CLIENT} --query "DROP TABLE IF EXISTS file"
-    ${CLICKHOUSE_CLIENT} --query "CREATE TABLE file (x UInt64) ENGINE = File($format, '${CLICKHOUSE_DATABASE}/data.$format.lz4')"
-    for size in 10000 100000 1000000 2500000
-    do
-        ${CLICKHOUSE_CLIENT} --query "TRUNCATE TABLE file"
-        ${CLICKHOUSE_CLIENT} --query "INSERT INTO file SELECT * FROM numbers($size)"
-        ${CLICKHOUSE_CLIENT} --query "SELECT max(x) FROM file"
-    done
-done
-
-${CLICKHOUSE_CLIENT} --query "DROP TABLE file"
diff --git a/tests/queries/0_stateless/02125_lz4_compression_bug_CSV.reference b/tests/queries/0_stateless/02125_lz4_compression_bug_CSV.reference
new file mode 100644
index 00000000000..b4ace07c0ac
--- /dev/null
+++ b/tests/queries/0_stateless/02125_lz4_compression_bug_CSV.reference
@@ -0,0 +1,4 @@
+9999
+99999
+999999
+2499999
diff --git a/tests/queries/0_stateless/02125_lz4_compression_bug_CSV.sh b/tests/queries/0_stateless/02125_lz4_compression_bug_CSV.sh
new file mode 100755
index 00000000000..ef38226a3fe
--- /dev/null
+++ b/tests/queries/0_stateless/02125_lz4_compression_bug_CSV.sh
@@ -0,0 +1,6 @@
+#!/usr/bin/env bash
+# Tags: no-fasttest
+
+CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
+# shellcheck source=../shell_config.sh
+${CUR_DIR}/02125_lz4_compression_bug.lib CSV
\ No newline at end of file
diff --git a/tests/queries/0_stateless/02125_lz4_compression_bug_JSONCompactEachRow.reference b/tests/queries/0_stateless/02125_lz4_compression_bug_JSONCompactEachRow.reference
new file mode 100644
index 00000000000..b4ace07c0ac
--- /dev/null
+++ b/tests/queries/0_stateless/02125_lz4_compression_bug_JSONCompactEachRow.reference
@@ -0,0 +1,4 @@
+9999
+99999
+999999
+2499999
diff --git a/tests/queries/0_stateless/02125_lz4_compression_bug_JSONCompactEachRow.sh b/tests/queries/0_stateless/02125_lz4_compression_bug_JSONCompactEachRow.sh
new file mode 100755
index 00000000000..41510edd7df
--- /dev/null
+++ b/tests/queries/0_stateless/02125_lz4_compression_bug_JSONCompactEachRow.sh
@@ -0,0 +1,6 @@
+#!/usr/bin/env bash
+# Tags: no-fasttest
+
+CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
+# shellcheck source=../shell_config.sh
+${CUR_DIR}/02125_lz4_compression_bug.lib JSONCompactEachRow
\ No newline at end of file
diff --git a/tests/queries/0_stateless/02125_lz4_compression_bug_JSONEachRow.reference b/tests/queries/0_stateless/02125_lz4_compression_bug_JSONEachRow.reference
new file mode 100644
index 00000000000..b4ace07c0ac
--- /dev/null
+++ b/tests/queries/0_stateless/02125_lz4_compression_bug_JSONEachRow.reference
@@ -0,0 +1,4 @@
+9999
+99999
+999999
+2499999
diff --git a/tests/queries/0_stateless/02125_lz4_compression_bug_JSONEachRow.sh b/tests/queries/0_stateless/02125_lz4_compression_bug_JSONEachRow.sh
new file mode 100755
index 00000000000..c4bfc4274de
--- /dev/null
+++ b/tests/queries/0_stateless/02125_lz4_compression_bug_JSONEachRow.sh
@@ -0,0 +1,6 @@
+#!/usr/bin/env bash
+# Tags: no-fasttest
+
+CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
+# shellcheck source=../shell_config.sh
+${CUR_DIR}/02125_lz4_compression_bug.lib JSONEachRow
\ No newline at end of file
diff --git a/tests/queries/0_stateless/02125_lz4_compression_bug_JSONStringsEachRow.reference b/tests/queries/0_stateless/02125_lz4_compression_bug_JSONStringsEachRow.reference
new file mode 100644
index 00000000000..b4ace07c0ac
--- /dev/null
+++ b/tests/queries/0_stateless/02125_lz4_compression_bug_JSONStringsEachRow.reference
@@ -0,0 +1,4 @@
+9999
+99999
+999999
+2499999
diff --git a/tests/queries/0_stateless/02125_lz4_compression_bug_JSONStringsEachRow.sh b/tests/queries/0_stateless/02125_lz4_compression_bug_JSONStringsEachRow.sh
new file mode 100755
index 00000000000..87f30a56da5
--- /dev/null
+++ b/tests/queries/0_stateless/02125_lz4_compression_bug_JSONStringsEachRow.sh
@@ -0,0 +1,6 @@
+#!/usr/bin/env bash
+# Tags: no-fasttest
+
+CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
+# shellcheck source=../shell_config.sh
+${CUR_DIR}/02125_lz4_compression_bug.lib JSONStringsEachRow
\ No newline at end of file
diff --git a/tests/queries/0_stateless/02125_lz4_compression_bug_Native.reference b/tests/queries/0_stateless/02125_lz4_compression_bug_Native.reference
new file mode 100644
index 00000000000..b4ace07c0ac
--- /dev/null
+++ b/tests/queries/0_stateless/02125_lz4_compression_bug_Native.reference
@@ -0,0 +1,4 @@
+9999
+99999
+999999
+2499999
diff --git a/tests/queries/0_stateless/02125_lz4_compression_bug_Native.sh b/tests/queries/0_stateless/02125_lz4_compression_bug_Native.sh
new file mode 100755
index 00000000000..523d40e0522
--- /dev/null
+++ b/tests/queries/0_stateless/02125_lz4_compression_bug_Native.sh
@@ -0,0 +1,6 @@
+#!/usr/bin/env bash
+# Tags: no-fasttest
+
+CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
+# shellcheck source=../shell_config.sh
+${CUR_DIR}/02125_lz4_compression_bug.lib Native
\ No newline at end of file
diff --git a/tests/queries/0_stateless/02125_lz4_compression_bug_TSKV.reference b/tests/queries/0_stateless/02125_lz4_compression_bug_TSKV.reference
new file mode 100644
index 00000000000..b4ace07c0ac
--- /dev/null
+++ b/tests/queries/0_stateless/02125_lz4_compression_bug_TSKV.reference
@@ -0,0 +1,4 @@
+9999
+99999
+999999
+2499999
diff --git a/tests/queries/0_stateless/02125_lz4_compression_bug_TSKV.sh b/tests/queries/0_stateless/02125_lz4_compression_bug_TSKV.sh
new file mode 100755
index 00000000000..b4f4045d6f6
--- /dev/null
+++ b/tests/queries/0_stateless/02125_lz4_compression_bug_TSKV.sh
@@ -0,0 +1,6 @@
+#!/usr/bin/env bash
+# Tags: no-fasttest
+
+CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
+# shellcheck source=../shell_config.sh
+${CUR_DIR}/02125_lz4_compression_bug.lib TSKV
\ No newline at end of file
diff --git a/tests/queries/0_stateless/02125_lz4_compression_bug_TSV.reference b/tests/queries/0_stateless/02125_lz4_compression_bug_TSV.reference
new file mode 100644
index 00000000000..b4ace07c0ac
--- /dev/null
+++ b/tests/queries/0_stateless/02125_lz4_compression_bug_TSV.reference
@@ -0,0 +1,4 @@
+9999
+99999
+999999
+2499999
diff --git a/tests/queries/0_stateless/02125_lz4_compression_bug_TSV.sh b/tests/queries/0_stateless/02125_lz4_compression_bug_TSV.sh
new file mode 100755
index 00000000000..ca50929c958
--- /dev/null
+++ b/tests/queries/0_stateless/02125_lz4_compression_bug_TSV.sh
@@ -0,0 +1,6 @@
+#!/usr/bin/env bash
+# Tags: no-fasttest
+
+CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
+# shellcheck source=../shell_config.sh
+${CUR_DIR}/02125_lz4_compression_bug.lib TSV
\ No newline at end of file
diff --git a/tests/queries/0_stateless/02125_lz4_compression_bug_Values.reference b/tests/queries/0_stateless/02125_lz4_compression_bug_Values.reference
new file mode 100644
index 00000000000..b4ace07c0ac
--- /dev/null
+++ b/tests/queries/0_stateless/02125_lz4_compression_bug_Values.reference
@@ -0,0 +1,4 @@
+9999
+99999
+999999
+2499999
diff --git a/tests/queries/0_stateless/02125_lz4_compression_bug_Values.sh b/tests/queries/0_stateless/02125_lz4_compression_bug_Values.sh
new file mode 100755
index 00000000000..68f3d90dbb4
--- /dev/null
+++ b/tests/queries/0_stateless/02125_lz4_compression_bug_Values.sh
@@ -0,0 +1,6 @@
+#!/usr/bin/env bash
+# Tags: no-fasttest
+
+CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
+# shellcheck source=../shell_config.sh
+${CUR_DIR}/02125_lz4_compression_bug.lib Values
\ No newline at end of file

From c98a165980cb57edf86e4b7152b66e6c97b6f991 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Mon, 18 Mar 2024 08:32:16 +0000
Subject: [PATCH 431/443] Fix build

---
 src/Dictionaries/RangeHashedDictionary.h           | 14 +++++++-------
 .../RangeHashedDictionaryGetItemsImpl.txx          |  2 +-
 ...ngeHashedDictionaryGetItemsShortCircuitImpl.txx |  2 +-
 3 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/src/Dictionaries/RangeHashedDictionary.h b/src/Dictionaries/RangeHashedDictionary.h
index 0469e82d7be..c208863905b 100644
--- a/src/Dictionaries/RangeHashedDictionary.h
+++ b/src/Dictionaries/RangeHashedDictionary.h
@@ -288,7 +288,7 @@ private:
 extern template class RangeHashedDictionary<DictionaryKeyType::Simple>;
 extern template class RangeHashedDictionary<DictionaryKeyType::Complex>;
 
-namespace
+namespace impl
 {
     template <typename F>
     void callOnRangeType(const DataTypePtr & range_type, F && func)
@@ -465,7 +465,7 @@ ColumnUInt8::Ptr RangeHashedDictionary<dictionary_key_type>::hasKeys(const Colum
     auto & out = result->getData();
     size_t keys_found = 0;
 
-    callOnRangeType(dict_struct.range_min->type, [&](const auto & types)
+    impl::callOnRangeType(dict_struct.range_min->type, [&](const auto & types)
     {
         using Types = std::decay_t<decltype(types)>;
         using RangeColumnType = typename Types::LeftType;
@@ -523,7 +523,7 @@ void RangeHashedDictionary<dictionary_key_type>::createAttributes()
                             getDictionaryID().getNameForLogs());
     }
 
-    callOnRangeType(dict_struct.range_min->type, [&](const auto & types)
+    impl::callOnRangeType(dict_struct.range_min->type, [&](const auto & types)
     {
         using Types = std::decay_t<decltype(types)>;
         using RangeColumnType = typename Types::LeftType;
@@ -553,7 +553,7 @@ void RangeHashedDictionary<dictionary_key_type>::loadData()
         updateData();
     }
 
-    callOnRangeType(dict_struct.range_min->type, [&](const auto & types)
+    impl::callOnRangeType(dict_struct.range_min->type, [&](const auto & types)
     {
         using Types = std::decay_t<decltype(types)>;
         using RangeColumnType = typename Types::LeftType;
@@ -573,7 +573,7 @@ void RangeHashedDictionary<dictionary_key_type>::loadData()
 template <DictionaryKeyType dictionary_key_type>
 void RangeHashedDictionary<dictionary_key_type>::calculateBytesAllocated()
 {
-    callOnRangeType(dict_struct.range_min->type, [&](const auto & types)
+    impl::callOnRangeType(dict_struct.range_min->type, [&](const auto & types)
     {
         using Types = std::decay_t<decltype(types)>;
         using RangeColumnType = typename Types::LeftType;
@@ -783,7 +783,7 @@ void RangeHashedDictionary<dictionary_key_type>::blockToAttributes(const Block &
         max_range_null_map = &max_range_column_nullable->getNullMapColumn().getData();
     }
 
-    callOnRangeType(dict_struct.range_min->type, [&](const auto & types)
+    impl::callOnRangeType(dict_struct.range_min->type, [&](const auto & types)
     {
         using Types = std::decay_t<decltype(types)>;
         using RangeColumnType = typename Types::LeftType;
@@ -930,7 +930,7 @@ Pipe RangeHashedDictionary<dictionary_key_type>::read(const Names & column_names
 
     PaddedPODArray<KeyType> keys;
 
-    callOnRangeType(dict_struct.range_min->type, [&](const auto & types)
+    impl::callOnRangeType(dict_struct.range_min->type, [&](const auto & types)
     {
         using Types = std::decay_t<decltype(types)>;
         using RangeColumnType = typename Types::LeftType;
diff --git a/src/Dictionaries/RangeHashedDictionaryGetItemsImpl.txx b/src/Dictionaries/RangeHashedDictionaryGetItemsImpl.txx
index 9da2b0faf4a..8ed4445f236 100644
--- a/src/Dictionaries/RangeHashedDictionaryGetItemsImpl.txx
+++ b/src/Dictionaries/RangeHashedDictionaryGetItemsImpl.txx
@@ -37,7 +37,7 @@ void RangeHashedDictionary<dictionary_key_type>::getItemsImpl(
     DictionaryKeysExtractor<dictionary_key_type> keys_extractor(key_columns_copy, arena_holder.getComplexKeyArena());
     const size_t keys_size = keys_extractor.getKeysSize();
 
-    callOnRangeType(
+    impl::callOnRangeType(
         dict_struct.range_min->type,
         [&](const auto & types)
         {
diff --git a/src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImpl.txx b/src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImpl.txx
index 5807af519f9..63c29f8cc34 100644
--- a/src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImpl.txx
+++ b/src/Dictionaries/RangeHashedDictionaryGetItemsShortCircuitImpl.txx
@@ -37,7 +37,7 @@ size_t RangeHashedDictionary<dictionary_key_type>::getItemsShortCircuitImpl(
     const size_t keys_size = keys_extractor.getKeysSize();
     default_mask.resize(keys_size);
 
-    callOnRangeType(
+    impl::callOnRangeType(
         dict_struct.range_min->type,
         [&](const auto & types)
         {

From 6549a96ced628dbbaa5b1c32304459ae6d40930a Mon Sep 17 00:00:00 2001
From: Antonio Andelic <antonio2368@users.noreply.github.com>
Date: Mon, 18 Mar 2024 10:38:28 +0100
Subject: [PATCH 432/443] Revert "Analyzer: Fix virtual columns in
 StorageMerge"

---
 .../InterpreterSelectQueryAnalyzer.cpp        |  1 -
 src/Storages/StorageMerge.cpp                 | 93 ++++++-------------
 tests/analyzer_tech_debt.txt                  |  1 +
 3 files changed, 27 insertions(+), 68 deletions(-)

diff --git a/src/Interpreters/InterpreterSelectQueryAnalyzer.cpp b/src/Interpreters/InterpreterSelectQueryAnalyzer.cpp
index 539d7a59f6f..922f4a99b4a 100644
--- a/src/Interpreters/InterpreterSelectQueryAnalyzer.cpp
+++ b/src/Interpreters/InterpreterSelectQueryAnalyzer.cpp
@@ -86,7 +86,6 @@ void replaceStorageInQueryTree(QueryTreeNodePtr & query_tree, const ContextPtr &
             continue;
 
         auto replacement_table_expression = std::make_shared<TableNode>(storage, context);
-        replacement_table_expression->setAlias(node->getAlias());
 
         if (auto table_expression_modifiers = table_node.getTableExpressionModifiers())
             replacement_table_expression->setTableExpressionModifiers(*table_expression_modifiers);
diff --git a/src/Storages/StorageMerge.cpp b/src/Storages/StorageMerge.cpp
index 3a4ec6f9b17..9161b2773da 100644
--- a/src/Storages/StorageMerge.cpp
+++ b/src/Storages/StorageMerge.cpp
@@ -885,7 +885,6 @@ SelectQueryInfo ReadFromMerge::getModifiedQueryInfo(const ContextMutablePtr & mo
     if (modified_query_info.table_expression)
     {
         auto replacement_table_expression = std::make_shared<TableNode>(storage, storage_lock, storage_snapshot_);
-        replacement_table_expression->setAlias(modified_query_info.table_expression->getAlias());
         if (query_info.table_expression_modifiers)
             replacement_table_expression->setTableExpressionModifiers(*query_info.table_expression_modifiers);
 
@@ -1026,7 +1025,7 @@ QueryPipelineBuilderPtr ReadFromMerge::createSources(
     const auto & [database_name, storage, _, table_name] = storage_with_lock;
     bool allow_experimental_analyzer = context->getSettingsRef().allow_experimental_analyzer;
     auto storage_stage
-        = storage->getQueryProcessingStage(context, processed_stage, storage_snapshot_, modified_query_info);
+        = storage->getQueryProcessingStage(context, QueryProcessingStage::Complete, storage_snapshot_, modified_query_info);
 
     builder = plan.buildQueryPipeline(
         QueryPlanOptimizationSettings::fromContext(context), BuildQueryPipelineSettings::fromContext(context));
@@ -1053,80 +1052,40 @@ QueryPipelineBuilderPtr ReadFromMerge::createSources(
 
         Block pipe_header = builder->getHeader();
 
-        if (allow_experimental_analyzer)
+        if (has_database_virtual_column && common_header.has("_database") && !pipe_header.has("_database"))
         {
-            String table_alias = modified_query_info.query_tree->as<QueryNode>()->getJoinTree()->as<TableNode>()->getAlias();
+            ColumnWithTypeAndName column;
+            column.name = "_database";
+            column.type = std::make_shared<DataTypeLowCardinality>(std::make_shared<DataTypeString>());
+            column.column = column.type->createColumnConst(0, Field(database_name));
 
-            String database_column = table_alias.empty() || processed_stage == QueryProcessingStage::FetchColumns ? "_database" : table_alias + "._database";
-            String table_column = table_alias.empty() || processed_stage == QueryProcessingStage::FetchColumns ? "_table" : table_alias + "._table";
+            auto adding_column_dag = ActionsDAG::makeAddingColumnActions(std::move(column));
+            auto adding_column_actions = std::make_shared<ExpressionActions>(
+                std::move(adding_column_dag), ExpressionActionsSettings::fromContext(context, CompileExpressions::yes));
 
-            if (has_database_virtual_column && common_header.has(database_column)
-                && (storage_stage == QueryProcessingStage::FetchColumns || !pipe_header.has("'" + database_name + "'_String")))
-            {
-                ColumnWithTypeAndName column;
-                column.name = database_column;
-                column.type = std::make_shared<DataTypeLowCardinality>(std::make_shared<DataTypeString>());
-                column.column = column.type->createColumnConst(0, Field(database_name));
-
-                auto adding_column_dag = ActionsDAG::makeAddingColumnActions(std::move(column));
-                auto adding_column_actions = std::make_shared<ExpressionActions>(
-                    std::move(adding_column_dag), ExpressionActionsSettings::fromContext(context, CompileExpressions::yes));
-
-                builder->addSimpleTransform([&](const Block & stream_header)
-                                            { return std::make_shared<ExpressionTransform>(stream_header, adding_column_actions); });
-            }
-
-            if (has_table_virtual_column && common_header.has(table_column)
-                && (storage_stage == QueryProcessingStage::FetchColumns || !pipe_header.has("'" + table_name + "'_String")))
-            {
-                ColumnWithTypeAndName column;
-                column.name = table_column;
-                column.type = std::make_shared<DataTypeLowCardinality>(std::make_shared<DataTypeString>());
-                column.column = column.type->createColumnConst(0, Field(table_name));
-
-                auto adding_column_dag = ActionsDAG::makeAddingColumnActions(std::move(column));
-                auto adding_column_actions = std::make_shared<ExpressionActions>(
-                    std::move(adding_column_dag), ExpressionActionsSettings::fromContext(context, CompileExpressions::yes));
-
-                builder->addSimpleTransform([&](const Block & stream_header)
-                                            { return std::make_shared<ExpressionTransform>(stream_header, adding_column_actions); });
-            }
+            builder->addSimpleTransform([&](const Block & stream_header)
+                                        { return std::make_shared<ExpressionTransform>(stream_header, adding_column_actions); });
         }
-        else
+
+        if (has_table_virtual_column && common_header.has("_table") && !pipe_header.has("_table"))
         {
-            if (has_database_virtual_column && common_header.has("_database") && !pipe_header.has("_database"))
-            {
-                ColumnWithTypeAndName column;
-                column.name = "_database";
-                column.type = std::make_shared<DataTypeLowCardinality>(std::make_shared<DataTypeString>());
-                column.column = column.type->createColumnConst(0, Field(database_name));
+            ColumnWithTypeAndName column;
+            column.name = "_table";
+            column.type = std::make_shared<DataTypeLowCardinality>(std::make_shared<DataTypeString>());
+            column.column = column.type->createColumnConst(0, Field(table_name));
 
-                auto adding_column_dag = ActionsDAG::makeAddingColumnActions(std::move(column));
-                auto adding_column_actions = std::make_shared<ExpressionActions>(
-                    std::move(adding_column_dag), ExpressionActionsSettings::fromContext(context, CompileExpressions::yes));
-                builder->addSimpleTransform([&](const Block & stream_header)
-                                            { return std::make_shared<ExpressionTransform>(stream_header, adding_column_actions); });
-            }
+            auto adding_column_dag = ActionsDAG::makeAddingColumnActions(std::move(column));
+            auto adding_column_actions = std::make_shared<ExpressionActions>(
+                std::move(adding_column_dag), ExpressionActionsSettings::fromContext(context, CompileExpressions::yes));
 
-            if (has_table_virtual_column && common_header.has("_table") && !pipe_header.has("_table"))
-            {
-                ColumnWithTypeAndName column;
-                column.name = "_table";
-                column.type = std::make_shared<DataTypeLowCardinality>(std::make_shared<DataTypeString>());
-                column.column = column.type->createColumnConst(0, Field(table_name));
-
-                auto adding_column_dag = ActionsDAG::makeAddingColumnActions(std::move(column));
-                auto adding_column_actions = std::make_shared<ExpressionActions>(
-                    std::move(adding_column_dag), ExpressionActionsSettings::fromContext(context, CompileExpressions::yes));
-                builder->addSimpleTransform([&](const Block & stream_header)
-                                            { return std::make_shared<ExpressionTransform>(stream_header, adding_column_actions); });
-            }
+            builder->addSimpleTransform([&](const Block & stream_header)
+                                        { return std::make_shared<ExpressionTransform>(stream_header, adding_column_actions); });
         }
 
         /// Subordinary tables could have different but convertible types, like numeric types of different width.
         /// We must return streams with structure equals to structure of Merge table.
         convertAndFilterSourceStream(
-            header, modified_query_info, storage_snapshot_, aliases, row_policy_data_opt, context, *builder, storage_stage);
+            header, modified_query_info, storage_snapshot_, aliases, row_policy_data_opt, context, *builder, processed_stage);
     }
 
     return builder;
@@ -1157,13 +1116,13 @@ QueryPlan ReadFromMerge::createPlanForTable(
     bool allow_experimental_analyzer = modified_context->getSettingsRef().allow_experimental_analyzer;
 
     auto storage_stage = storage->getQueryProcessingStage(modified_context,
-        processed_stage,
+        QueryProcessingStage::Complete,
         storage_snapshot_,
         modified_query_info);
 
     QueryPlan plan;
 
-    if (processed_stage <= storage_stage)
+    if (processed_stage <= storage_stage || (allow_experimental_analyzer && processed_stage == QueryProcessingStage::FetchColumns))
     {
         /// If there are only virtual columns in query, you must request at least one other column.
         if (real_column_names.empty())
@@ -1208,7 +1167,7 @@ QueryPlan ReadFromMerge::createPlanForTable(
                 row_policy_data_opt->addStorageFilter(source_step_with_filter);
         }
     }
-    else if (processed_stage > storage_stage || allow_experimental_analyzer)
+    else if (processed_stage > storage_stage || (allow_experimental_analyzer && processed_stage != QueryProcessingStage::FetchColumns))
     {
         /// Maximum permissible parallelism is streams_num
         modified_context->setSetting("max_threads", streams_num);
diff --git a/tests/analyzer_tech_debt.txt b/tests/analyzer_tech_debt.txt
index 7cd73705e77..bcf0eee0d0e 100644
--- a/tests/analyzer_tech_debt.txt
+++ b/tests/analyzer_tech_debt.txt
@@ -1,4 +1,5 @@
 00223_shard_distributed_aggregation_memory_efficient
+00717_merge_and_distributed
 00725_memory_tracking
 01062_pm_all_join_with_block_continuation
 01083_expressions_in_engine_arguments

From dc4a2064e4c7720d0b027746f28e1237c7ca58f3 Mon Sep 17 00:00:00 2001
From: Antonio Andelic <antonio@clickhouse.com>
Date: Mon, 18 Mar 2024 10:51:46 +0100
Subject: [PATCH 433/443] Fix clang-tidy build

---
 utils/check-marks/CMakeLists.txt        | 2 +-
 utils/check-mysql-binlog/CMakeLists.txt | 2 +-
 utils/keeper-data-dumper/CMakeLists.txt | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/utils/check-marks/CMakeLists.txt b/utils/check-marks/CMakeLists.txt
index 05546a2989b..857dc51d40a 100644
--- a/utils/check-marks/CMakeLists.txt
+++ b/utils/check-marks/CMakeLists.txt
@@ -1,2 +1,2 @@
 clickhouse_add_executable (check-marks main.cpp)
-target_link_libraries(check-marks PRIVATE dbms boost::program_options)
+target_link_libraries(check-marks PRIVATE dbms clickhouse_functions boost::program_options)
diff --git a/utils/check-mysql-binlog/CMakeLists.txt b/utils/check-mysql-binlog/CMakeLists.txt
index cbbecd456a0..0ece1684e6b 100644
--- a/utils/check-mysql-binlog/CMakeLists.txt
+++ b/utils/check-mysql-binlog/CMakeLists.txt
@@ -1,2 +1,2 @@
 clickhouse_add_executable(check-mysql-binlog main.cpp)
-target_link_libraries(check-mysql-binlog PRIVATE dbms boost::program_options)
+target_link_libraries(check-mysql-binlog PRIVATE dbms clickhouse_functions boost::program_options)
diff --git a/utils/keeper-data-dumper/CMakeLists.txt b/utils/keeper-data-dumper/CMakeLists.txt
index 136d6f2268c..71f29781b73 100644
--- a/utils/keeper-data-dumper/CMakeLists.txt
+++ b/utils/keeper-data-dumper/CMakeLists.txt
@@ -4,4 +4,4 @@ if (NOT TARGET ch_contrib::nuraft)
 endif ()
 
 clickhouse_add_executable(keeper-data-dumper main.cpp)
-target_link_libraries(keeper-data-dumper PRIVATE dbms)
+target_link_libraries(keeper-data-dumper PRIVATE dbms clickhouse_functions)

From fe867410591b7d0a8b16d8023be7f1938d91bc89 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Mon, 18 Mar 2024 10:44:56 +0000
Subject: [PATCH 434/443] Some fixups

---
 programs/keeper/Keeper.cpp                | 16 ++++++--------
 programs/server/Server.cpp                |  5 +----
 src/Common/CgroupsMemoryUsageObserver.cpp | 13 ++++--------
 src/Common/CgroupsMemoryUsageObserver.h   | 26 ++++++++++++-----------
 4 files changed, 25 insertions(+), 35 deletions(-)

diff --git a/programs/keeper/Keeper.cpp b/programs/keeper/Keeper.cpp
index 9263719da2a..a558ed64bf9 100644
--- a/programs/keeper/Keeper.cpp
+++ b/programs/keeper/Keeper.cpp
@@ -624,21 +624,17 @@ try
     buildLoggers(config(), logger());
     main_config_reloader->start();
 
-    std::optional<CgroupsMemoryUsageObserver> observer;
+    std::optional<CgroupsMemoryUsageObserver> cgroups_memory_usage_observer;
     try
     {
         auto wait_time = config().getUInt64("keeper_server.cgroups_memory_observer_wait_time", 15);
         if (wait_time != 0)
         {
-            observer.emplace(std::chrono::seconds(wait_time));
-            /// Keeper use setting max_memory_usage_soft_limit to control memory limit. When memory usage
-            /// hits the limit, Keeper will refuse all the request util memory usage is lower.
-            /// We don't call setLimits() here because of this.
-            observer->setOnMemoryAmountAvailableChanged([&]()
-            {
-                main_config_reloader->reload();
-            });
-            observer->startThread();
+            cgroups_memory_usage_observer.emplace(std::chrono::seconds(wait_time));
+            /// Not calling cgroups_memory_usage_observer->setLimits() here (as for the normal ClickHouse server) because Keeper controls
+            /// its memory usage by other means (via setting 'max_memory_usage_soft_limit').
+            cgroups_memory_usage_observer->setOnMemoryAmountAvailableChangedFn([&]() { main_config_reloader->reload(); });
+            cgroups_memory_usage_observer->startThread();
         }
     }
     catch (Exception &)
diff --git a/programs/server/Server.cpp b/programs/server/Server.cpp
index 78d95d8340f..d1957fa53a8 100644
--- a/programs/server/Server.cpp
+++ b/programs/server/Server.cpp
@@ -1702,10 +1702,7 @@ try
 
     if (cgroups_memory_usage_observer)
     {
-        cgroups_memory_usage_observer->setOnMemoryAmountAvailableChanged([&]()
-        {
-            main_config_reloader->reload();
-        });
+        cgroups_memory_usage_observer->setOnMemoryAmountAvailableChangedFn([&]() { main_config_reloader->reload(); });
         cgroups_memory_usage_observer->startThread();
     }
 
diff --git a/src/Common/CgroupsMemoryUsageObserver.cpp b/src/Common/CgroupsMemoryUsageObserver.cpp
index 661b80a6059..dd68bd0da64 100644
--- a/src/Common/CgroupsMemoryUsageObserver.cpp
+++ b/src/Common/CgroupsMemoryUsageObserver.cpp
@@ -84,7 +84,7 @@ void CgroupsMemoryUsageObserver::setMemoryUsageLimits(uint64_t hard_limit_, uint
             mallctl("arena." STRINGIFY(MALLCTL_ARENAS_ALL) ".purge", nullptr, nullptr, nullptr, 0);
 #endif
             /// Reset current usage in memory tracker. Expect zero for free_memory_in_allocator_arenas as we just purged them.
-            uint64_t memory_usage = readMemoryUsage();
+            uint64_t memory_usage = memory_usage_file.readMemoryUsage();
             MemoryTracker::setRSS(memory_usage, 0);
 
             LOG_INFO(log, "Purged jemalloc arenas. Current memory usage is {}", ReadableSize(memory_usage));
@@ -98,17 +98,12 @@ void CgroupsMemoryUsageObserver::setMemoryUsageLimits(uint64_t hard_limit_, uint
     LOG_INFO(log, "Set new limits, soft limit: {}, hard limit: {}", ReadableSize(soft_limit_), ReadableSize(hard_limit_));
 }
 
-void CgroupsMemoryUsageObserver::setOnMemoryAmountAvailableChanged(OnMemoryAmountAvailableChangedFn on_memory_amount_available_changed_)
+void CgroupsMemoryUsageObserver::setOnMemoryAmountAvailableChangedFn(OnMemoryAmountAvailableChangedFn on_memory_amount_available_changed_)
 {
-    std::lock_guard<std::mutex> memory_amount_change_lock(memory_amount_change_mutex);
+    std::lock_guard<std::mutex> memory_amount_available_changed_lock(memory_amount_available_changed_mutex);
     on_memory_amount_available_changed = on_memory_amount_available_changed_;
 }
 
-uint64_t CgroupsMemoryUsageObserver::readMemoryUsage() const
-{
-    return memory_usage_file.readMemoryUsage();
-}
-
 namespace
 {
 
@@ -299,7 +294,7 @@ void CgroupsMemoryUsageObserver::runThread()
             {
                 LOG_INFO(log, "Memory amount available to the process changed from {} to {}", ReadableSize(last_available_memory_amount), ReadableSize(available_memory_amount));
                 last_available_memory_amount = available_memory_amount;
-                std::lock_guard<std::mutex> memory_amount_change_lock(memory_amount_change_mutex);
+                std::lock_guard<std::mutex> memory_amount_available_changed_lock(memory_amount_available_changed_mutex);
                 on_memory_amount_available_changed();
             }
 
diff --git a/src/Common/CgroupsMemoryUsageObserver.h b/src/Common/CgroupsMemoryUsageObserver.h
index 1da49f8f777..ecb1fd8360f 100644
--- a/src/Common/CgroupsMemoryUsageObserver.h
+++ b/src/Common/CgroupsMemoryUsageObserver.h
@@ -2,7 +2,6 @@
 
 #include <Common/ThreadPool.h>
 
-#include <atomic>
 #include <chrono>
 #include <mutex>
 
@@ -14,11 +13,16 @@ namespace DB
 ///    You can specify soft or hard memory limits:
 ///    - When the soft memory limit is hit, drop jemalloc cache.
 ///    - When the hard memory limit is hit, update MemoryTracking metric to throw memory exceptions faster.
+///    The goal of this is to avoid that the process hits the maximum allowed memory limit at which there is a good
+///    chance that the Limux OOM killer terminates it. All of this is done is because internal memory tracking in
+///    ClickHouse can unfortunately under-estimate the actually used memory.
 /// 2. Periodically reads the the maximum memory available to the process (which can change due to cgroups settings).
-///    You can specify a callback to react on changes. The callback typically performs circular logic: It reloads the
-///    configuration (e.g. server configuration file), which will check the memory amount again and re-calculate
-///    soft/hard limits (see 1.) and update the value of setting `max_server_memory_usage` for clickhouse-server or
-///    `max_memory_usage_soft_limit` for clickhouse-keeper.
+///    You can specify a callback to react on changes. The callback typically reloads the configuration, i.e. Server
+///    or Keeper configuration file. This reloads settings 'max_server_memory_usage' (Server) and 'max_memory_usage_soft_limit'
+///    (Keeper) from which various other internal limits are calculated, including the soft and hard limits for (1.).
+///    The goal of this is to provide elasticity when the container is scaled-up/scaled-down. The mechanism (polling
+///    cgroups) is quite implicit, unfortuantely there is currently no better way to communicate memory threshold changes
+///    to the database.
 #if defined(OS_LINUX)
 class CgroupsMemoryUsageObserver
 {
@@ -36,12 +40,10 @@ public:
     ~CgroupsMemoryUsageObserver();
 
     void setMemoryUsageLimits(uint64_t hard_limit_, uint64_t soft_limit_);
-    void setOnMemoryAmountAvailableChanged(OnMemoryAmountAvailableChangedFn on_memory_amount_available_changed);
+    void setOnMemoryAmountAvailableChangedFn(OnMemoryAmountAvailableChangedFn on_memory_amount_available_changed_);
 
     void startThread();
 
-    uint64_t readMemoryUsage() const;
-
 private:
     LoggerPtr log;
 
@@ -52,8 +54,9 @@ private:
     size_t soft_limit TSA_GUARDED_BY(limit_mutex) = 0;
     OnMemoryLimitFn on_hard_limit TSA_GUARDED_BY(limit_mutex);
     OnMemoryLimitFn on_soft_limit TSA_GUARDED_BY(limit_mutex);
-    std::mutex memory_amount_change_mutex;
-    OnMemoryAmountAvailableChangedFn on_memory_amount_available_changed TSA_GUARDED_BY(memory_amount_change_mutex);
+
+    std::mutex memory_amount_available_changed_mutex;
+    OnMemoryAmountAvailableChangedFn on_memory_amount_available_changed TSA_GUARDED_BY(memory_amount_available_changed_mutex);
 
     uint64_t last_memory_usage = 0;        /// how much memory does the process use
     uint64_t last_available_memory_amount; /// how much memory can the process use
@@ -93,9 +96,8 @@ public:
     explicit CgroupsMemoryUsageObserver(std::chrono::seconds) {}
 
     void setMemoryUsageLimits(uint64_t, uint64_t) {}
-    void setOnMemoryAmountAvailableChanged(OnMemoryAmountAvailableChangedFn) {}
+    void setOnMemoryAmountAvailableChangedFn(OnMemoryAmountAvailableChangedFn) {}
     void startThread() {}
-    size_t readMemoryUsage() { return 0; }
 };
 #endif
 

From 9b9f409805acb2f3d8c4427537bac91168d7f33c Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Mon, 18 Mar 2024 10:45:57 +0000
Subject: [PATCH 435/443] Fix spelling

---
 src/Common/CgroupsMemoryUsageObserver.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Common/CgroupsMemoryUsageObserver.h b/src/Common/CgroupsMemoryUsageObserver.h
index ecb1fd8360f..51c5cd08124 100644
--- a/src/Common/CgroupsMemoryUsageObserver.h
+++ b/src/Common/CgroupsMemoryUsageObserver.h
@@ -21,7 +21,7 @@ namespace DB
 ///    or Keeper configuration file. This reloads settings 'max_server_memory_usage' (Server) and 'max_memory_usage_soft_limit'
 ///    (Keeper) from which various other internal limits are calculated, including the soft and hard limits for (1.).
 ///    The goal of this is to provide elasticity when the container is scaled-up/scaled-down. The mechanism (polling
-///    cgroups) is quite implicit, unfortuantely there is currently no better way to communicate memory threshold changes
+///    cgroups) is quite implicit, unfortunately there is currently no better way to communicate memory threshold changes
 ///    to the database.
 #if defined(OS_LINUX)
 class CgroupsMemoryUsageObserver

From 30b757b40dac46e07842c2fbe6a155b1ee99c065 Mon Sep 17 00:00:00 2001
From: Nikita Mikhaylov <mikhaylovnikitka@gmail.com>
Date: Mon, 18 Mar 2024 12:06:45 +0100
Subject: [PATCH 436/443] Check boundaries for some settings in parallel
 replicas (#61455)

---
 src/Common/NaNUtils.h                         | 12 +++++++++++
 src/Interpreters/InterpreterSelectQuery.cpp   |  8 ++++++--
 .../QueryPlan/ReadFromMergeTree.cpp           | 20 ++++++++++++++++---
 3 files changed, 35 insertions(+), 5 deletions(-)

diff --git a/src/Common/NaNUtils.h b/src/Common/NaNUtils.h
index 1c5a619e919..2ff3e2f5661 100644
--- a/src/Common/NaNUtils.h
+++ b/src/Common/NaNUtils.h
@@ -25,6 +25,18 @@ inline bool isFinite(T x)
         return true;
 }
 
+template <typename T>
+bool canConvertTo(Float64 x)
+{
+    if constexpr (std::is_floating_point_v<T>)
+        return true;
+    if (!isFinite(x))
+        return false;
+    if (x > Float64(std::numeric_limits<T>::max()) || x < Float64(std::numeric_limits<T>::lowest()))
+        return false;
+
+    return true;
+}
 
 template <typename T>
 T NaNOrZero()
diff --git a/src/Interpreters/InterpreterSelectQuery.cpp b/src/Interpreters/InterpreterSelectQuery.cpp
index 07f4e94680c..7c87dadfce6 100644
--- a/src/Interpreters/InterpreterSelectQuery.cpp
+++ b/src/Interpreters/InterpreterSelectQuery.cpp
@@ -96,6 +96,7 @@
 #include <Common/scope_guard_safe.h>
 #include <Common/typeid_cast.h>
 #include <Common/ProfileEvents.h>
+#include <Common/NaNUtils.h>
 
 
 namespace ProfileEvents
@@ -2553,10 +2554,13 @@ void InterpreterSelectQuery::executeFetchColumns(QueryProcessingStage::Enum proc
         /// If necessary, we request more sources than the number of threads - to distribute the work evenly over the threads.
         if (max_streams > 1 && !is_sync_remote)
         {
-            if (auto streams_with_ratio = max_streams * settings.max_streams_to_max_threads_ratio; streams_with_ratio < SIZE_MAX)
+            if (auto streams_with_ratio = max_streams * settings.max_streams_to_max_threads_ratio; canConvertTo<size_t>(streams_with_ratio))
                 max_streams = static_cast<size_t>(streams_with_ratio);
             else
-                throw Exception(ErrorCodes::PARAMETER_OUT_OF_BOUND, "Exceeded limit for `max_streams` with `max_streams_to_max_threads_ratio`. Make sure that `max_streams * max_streams_to_max_threads_ratio` not exceeds {}, current value: {}", SIZE_MAX, streams_with_ratio);
+                throw Exception(ErrorCodes::PARAMETER_OUT_OF_BOUND,
+                    "Exceeded limit for `max_streams` with `max_streams_to_max_threads_ratio`. "
+                    "Make sure that `max_streams * max_streams_to_max_threads_ratio` is in some reasonable boundaries, current value: {}",
+                    streams_with_ratio);
         }
 
         auto & prewhere_info = analysis_result.prewhere_info;
diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp
index fb92be6eed9..f1ce5a7802f 100644
--- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp
+++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp
@@ -131,6 +131,7 @@ namespace ErrorCodes
     extern const int LOGICAL_ERROR;
     extern const int TOO_MANY_ROWS;
     extern const int CANNOT_PARSE_TEXT;
+    extern const int PARAMETER_OUT_OF_BOUND;
 }
 
 static MergeTreeReaderSettings getMergeTreeReaderSettings(
@@ -348,7 +349,14 @@ Pipe ReadFromMergeTree::readFromPoolParallelReplicas(
 
     /// We have a special logic for local replica. It has to read less data, because in some cases it should
     /// merge states of aggregate functions or do some other important stuff other than reading from Disk.
-    pool_settings.min_marks_for_concurrent_read = static_cast<size_t>(pool_settings.min_marks_for_concurrent_read * context->getSettingsRef().parallel_replicas_single_task_marks_count_multiplier);
+    const auto multiplier = context->getSettingsRef().parallel_replicas_single_task_marks_count_multiplier;
+    if (auto result = pool_settings.min_marks_for_concurrent_read * multiplier; canConvertTo<size_t>(result))
+        pool_settings.min_marks_for_concurrent_read = static_cast<size_t>(result);
+    else
+        throw Exception(ErrorCodes::PARAMETER_OUT_OF_BOUND,
+            "Exceeded limit for the number of marks per a single task for parallel replicas. "
+            "Make sure that `parallel_replicas_single_task_marks_count_multiplier` is in some reasonable boundaries, current value is: {}",
+            multiplier);
 
     auto pool = std::make_shared<MergeTreeReadPoolParallelReplicas>(
         std::move(extension),
@@ -512,8 +520,14 @@ Pipe ReadFromMergeTree::readInOrder(
             .columns_to_read = required_columns,
         };
 
-        pool_settings.min_marks_for_concurrent_read = static_cast<size_t>(
-            pool_settings.min_marks_for_concurrent_read * context->getSettingsRef().parallel_replicas_single_task_marks_count_multiplier);
+        const auto multiplier = context->getSettingsRef().parallel_replicas_single_task_marks_count_multiplier;
+        if (auto result = pool_settings.min_marks_for_concurrent_read * multiplier; canConvertTo<size_t>(result))
+            pool_settings.min_marks_for_concurrent_read = static_cast<size_t>(result);
+        else
+            throw Exception(ErrorCodes::PARAMETER_OUT_OF_BOUND,
+                "Exceeded limit for the number of marks per a single task for parallel replicas. "
+                "Make sure that `parallel_replicas_single_task_marks_count_multiplier` is in some reasonable boundaries, current value is: {}",
+                multiplier);
 
         CoordinationMode mode = read_type == ReadType::InOrder
             ? CoordinationMode::WithOrder

From 47e3eccc59ed62149aff6fe68068b6ca4cbc1dad Mon Sep 17 00:00:00 2001
From: Antonio Andelic <antonio@clickhouse.com>
Date: Mon, 18 Mar 2024 13:41:25 +0100
Subject: [PATCH 437/443] Fix repeat with non native integers

---
 src/Functions/repeat.cpp                      | 32 ++++++++++++++-----
 ...3_repeat_with_nonnative_integers.reference |  4 +++
 .../03013_repeat_with_nonnative_integers.sql  |  4 +++
 3 files changed, 32 insertions(+), 8 deletions(-)
 create mode 100644 tests/queries/0_stateless/03013_repeat_with_nonnative_integers.reference
 create mode 100644 tests/queries/0_stateless/03013_repeat_with_nonnative_integers.sql

diff --git a/src/Functions/repeat.cpp b/src/Functions/repeat.cpp
index 11a2ca37a3b..6f2078b7e48 100644
--- a/src/Functions/repeat.cpp
+++ b/src/Functions/repeat.cpp
@@ -44,7 +44,7 @@ struct RepeatImpl
         ColumnString::Offsets & res_offsets,
         T repeat_time)
     {
-        repeat_time = repeat_time < 0 ? 0 : repeat_time;
+        repeat_time = repeat_time < 0 ? static_cast<T>(0) : repeat_time;
         checkRepeatTime(repeat_time);
 
         UInt64 data_size = 0;
@@ -76,7 +76,7 @@ struct RepeatImpl
         res_offsets.assign(offsets);
         for (UInt64 i = 0; i < col_num.size(); ++i)
         {
-            T repeat_time = col_num[i] < 0 ? 0 : col_num[i];
+            T repeat_time = col_num[i] < 0 ? static_cast<T>(0) : col_num[i];
             size_t repeated_size = (offsets[i] - offsets[i - 1] - 1) * repeat_time + 1;
             checkStringSize(repeated_size);
             data_size += repeated_size;
@@ -86,7 +86,7 @@ struct RepeatImpl
 
         for (UInt64 i = 0; i < col_num.size(); ++i)
         {
-            T repeat_time = col_num[i] < 0 ? 0 : col_num[i];
+            T repeat_time = col_num[i] < 0 ? static_cast<T>(0) : col_num[i];
             checkRepeatTime(repeat_time);
             process(data.data() + offsets[i - 1], res_data.data() + res_offsets[i - 1], offsets[i] - offsets[i - 1], repeat_time);
         }
@@ -105,7 +105,7 @@ struct RepeatImpl
         UInt64 col_size = col_num.size();
         for (UInt64 i = 0; i < col_size; ++i)
         {
-            T repeat_time = col_num[i] < 0 ? 0 : col_num[i];
+            T repeat_time = col_num[i] < 0 ? static_cast<T>(0) : col_num[i];
             size_t repeated_size = str_size * repeat_time + 1;
             checkStringSize(repeated_size);
             data_size += repeated_size;
@@ -114,7 +114,7 @@ struct RepeatImpl
         res_data.resize(data_size);
         for (UInt64 i = 0; i < col_size; ++i)
         {
-            T repeat_time = col_num[i] < 0 ? 0 : col_num[i];
+            T repeat_time = col_num[i] < 0 ? static_cast<T>(0) : col_num[i];
             checkRepeatTime(repeat_time);
             process(
                 reinterpret_cast<UInt8 *>(const_cast<char *>(copy_str.data())),
@@ -169,8 +169,19 @@ class FunctionRepeat : public IFunction
     template <typename F>
     static bool castType(const IDataType * type, F && f)
     {
-        return castTypeToEither<DataTypeInt8, DataTypeInt16, DataTypeInt32, DataTypeInt64,
-            DataTypeUInt8, DataTypeUInt16, DataTypeUInt32, DataTypeUInt64>(type, std::forward<F>(f));
+        return castTypeToEither<
+            DataTypeInt8,
+            DataTypeInt16,
+            DataTypeInt32,
+            DataTypeInt64,
+            DataTypeInt128,
+            DataTypeInt256,
+            DataTypeUInt8,
+            DataTypeUInt16,
+            DataTypeUInt32,
+            DataTypeUInt64,
+            DataTypeUInt128,
+            DataTypeUInt256>(type, std::forward<F>(f));
     }
 
 public:
@@ -208,7 +219,7 @@ public:
             if (const ColumnConst * col_num_const = checkAndGetColumn<ColumnConst>(col_num.get()))
             {
                 auto col_res = ColumnString::create();
-                castType(arguments[1].type.get(), [&](const auto & type)
+                auto success = castType(arguments[1].type.get(), [&](const auto & type)
                 {
                     using DataType = std::decay_t<decltype(type)>;
                     using T = typename DataType::FieldType;
@@ -216,6 +227,11 @@ public:
                     RepeatImpl::vectorStrConstRepeat(col->getChars(), col->getOffsets(), col_res->getChars(), col_res->getOffsets(), times);
                     return true;
                 });
+
+                if (!success)
+                    throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column type {} of 'n' of function {}",
+                        arguments[1].column->getName(), getName());
+
                 return col_res;
             }
             else if (castType(arguments[1].type.get(), [&](const auto & type)
diff --git a/tests/queries/0_stateless/03013_repeat_with_nonnative_integers.reference b/tests/queries/0_stateless/03013_repeat_with_nonnative_integers.reference
new file mode 100644
index 00000000000..50cb7002a32
--- /dev/null
+++ b/tests/queries/0_stateless/03013_repeat_with_nonnative_integers.reference
@@ -0,0 +1,4 @@
+000000000000
+000000000000
+000000000000
+000000000000
diff --git a/tests/queries/0_stateless/03013_repeat_with_nonnative_integers.sql b/tests/queries/0_stateless/03013_repeat_with_nonnative_integers.sql
new file mode 100644
index 00000000000..0dbe98994b9
--- /dev/null
+++ b/tests/queries/0_stateless/03013_repeat_with_nonnative_integers.sql
@@ -0,0 +1,4 @@
+SELECT repeat(toString(number), toUInt256(12)) FROM numbers(1);
+SELECT repeat(toString(number), toUInt128(12)) FROM numbers(1);
+SELECT repeat(toString(number), toInt256(12)) FROM numbers(1);
+SELECT repeat(toString(number), toInt128(12)) FROM numbers(1);

From a65f027cf72d42f2d9263afe8fe58801ec1b3313 Mon Sep 17 00:00:00 2001
From: Dmitry Novik <mrnovikd@gmail.com>
Date: Mon, 18 Mar 2024 14:00:20 +0100
Subject: [PATCH 438/443] Revert "Revert "Analyzer: Fix virtual columns in
 StorageMerge""

---
 .../InterpreterSelectQueryAnalyzer.cpp        |  1 +
 src/Storages/StorageMerge.cpp                 | 93 +++++++++++++------
 tests/analyzer_tech_debt.txt                  |  1 -
 3 files changed, 68 insertions(+), 27 deletions(-)

diff --git a/src/Interpreters/InterpreterSelectQueryAnalyzer.cpp b/src/Interpreters/InterpreterSelectQueryAnalyzer.cpp
index 922f4a99b4a..539d7a59f6f 100644
--- a/src/Interpreters/InterpreterSelectQueryAnalyzer.cpp
+++ b/src/Interpreters/InterpreterSelectQueryAnalyzer.cpp
@@ -86,6 +86,7 @@ void replaceStorageInQueryTree(QueryTreeNodePtr & query_tree, const ContextPtr &
             continue;
 
         auto replacement_table_expression = std::make_shared<TableNode>(storage, context);
+        replacement_table_expression->setAlias(node->getAlias());
 
         if (auto table_expression_modifiers = table_node.getTableExpressionModifiers())
             replacement_table_expression->setTableExpressionModifiers(*table_expression_modifiers);
diff --git a/src/Storages/StorageMerge.cpp b/src/Storages/StorageMerge.cpp
index 9161b2773da..3a4ec6f9b17 100644
--- a/src/Storages/StorageMerge.cpp
+++ b/src/Storages/StorageMerge.cpp
@@ -885,6 +885,7 @@ SelectQueryInfo ReadFromMerge::getModifiedQueryInfo(const ContextMutablePtr & mo
     if (modified_query_info.table_expression)
     {
         auto replacement_table_expression = std::make_shared<TableNode>(storage, storage_lock, storage_snapshot_);
+        replacement_table_expression->setAlias(modified_query_info.table_expression->getAlias());
         if (query_info.table_expression_modifiers)
             replacement_table_expression->setTableExpressionModifiers(*query_info.table_expression_modifiers);
 
@@ -1025,7 +1026,7 @@ QueryPipelineBuilderPtr ReadFromMerge::createSources(
     const auto & [database_name, storage, _, table_name] = storage_with_lock;
     bool allow_experimental_analyzer = context->getSettingsRef().allow_experimental_analyzer;
     auto storage_stage
-        = storage->getQueryProcessingStage(context, QueryProcessingStage::Complete, storage_snapshot_, modified_query_info);
+        = storage->getQueryProcessingStage(context, processed_stage, storage_snapshot_, modified_query_info);
 
     builder = plan.buildQueryPipeline(
         QueryPlanOptimizationSettings::fromContext(context), BuildQueryPipelineSettings::fromContext(context));
@@ -1052,40 +1053,80 @@ QueryPipelineBuilderPtr ReadFromMerge::createSources(
 
         Block pipe_header = builder->getHeader();
 
-        if (has_database_virtual_column && common_header.has("_database") && !pipe_header.has("_database"))
+        if (allow_experimental_analyzer)
         {
-            ColumnWithTypeAndName column;
-            column.name = "_database";
-            column.type = std::make_shared<DataTypeLowCardinality>(std::make_shared<DataTypeString>());
-            column.column = column.type->createColumnConst(0, Field(database_name));
+            String table_alias = modified_query_info.query_tree->as<QueryNode>()->getJoinTree()->as<TableNode>()->getAlias();
 
-            auto adding_column_dag = ActionsDAG::makeAddingColumnActions(std::move(column));
-            auto adding_column_actions = std::make_shared<ExpressionActions>(
-                std::move(adding_column_dag), ExpressionActionsSettings::fromContext(context, CompileExpressions::yes));
+            String database_column = table_alias.empty() || processed_stage == QueryProcessingStage::FetchColumns ? "_database" : table_alias + "._database";
+            String table_column = table_alias.empty() || processed_stage == QueryProcessingStage::FetchColumns ? "_table" : table_alias + "._table";
 
-            builder->addSimpleTransform([&](const Block & stream_header)
-                                        { return std::make_shared<ExpressionTransform>(stream_header, adding_column_actions); });
+            if (has_database_virtual_column && common_header.has(database_column)
+                && (storage_stage == QueryProcessingStage::FetchColumns || !pipe_header.has("'" + database_name + "'_String")))
+            {
+                ColumnWithTypeAndName column;
+                column.name = database_column;
+                column.type = std::make_shared<DataTypeLowCardinality>(std::make_shared<DataTypeString>());
+                column.column = column.type->createColumnConst(0, Field(database_name));
+
+                auto adding_column_dag = ActionsDAG::makeAddingColumnActions(std::move(column));
+                auto adding_column_actions = std::make_shared<ExpressionActions>(
+                    std::move(adding_column_dag), ExpressionActionsSettings::fromContext(context, CompileExpressions::yes));
+
+                builder->addSimpleTransform([&](const Block & stream_header)
+                                            { return std::make_shared<ExpressionTransform>(stream_header, adding_column_actions); });
+            }
+
+            if (has_table_virtual_column && common_header.has(table_column)
+                && (storage_stage == QueryProcessingStage::FetchColumns || !pipe_header.has("'" + table_name + "'_String")))
+            {
+                ColumnWithTypeAndName column;
+                column.name = table_column;
+                column.type = std::make_shared<DataTypeLowCardinality>(std::make_shared<DataTypeString>());
+                column.column = column.type->createColumnConst(0, Field(table_name));
+
+                auto adding_column_dag = ActionsDAG::makeAddingColumnActions(std::move(column));
+                auto adding_column_actions = std::make_shared<ExpressionActions>(
+                    std::move(adding_column_dag), ExpressionActionsSettings::fromContext(context, CompileExpressions::yes));
+
+                builder->addSimpleTransform([&](const Block & stream_header)
+                                            { return std::make_shared<ExpressionTransform>(stream_header, adding_column_actions); });
+            }
         }
-
-        if (has_table_virtual_column && common_header.has("_table") && !pipe_header.has("_table"))
+        else
         {
-            ColumnWithTypeAndName column;
-            column.name = "_table";
-            column.type = std::make_shared<DataTypeLowCardinality>(std::make_shared<DataTypeString>());
-            column.column = column.type->createColumnConst(0, Field(table_name));
+            if (has_database_virtual_column && common_header.has("_database") && !pipe_header.has("_database"))
+            {
+                ColumnWithTypeAndName column;
+                column.name = "_database";
+                column.type = std::make_shared<DataTypeLowCardinality>(std::make_shared<DataTypeString>());
+                column.column = column.type->createColumnConst(0, Field(database_name));
 
-            auto adding_column_dag = ActionsDAG::makeAddingColumnActions(std::move(column));
-            auto adding_column_actions = std::make_shared<ExpressionActions>(
-                std::move(adding_column_dag), ExpressionActionsSettings::fromContext(context, CompileExpressions::yes));
+                auto adding_column_dag = ActionsDAG::makeAddingColumnActions(std::move(column));
+                auto adding_column_actions = std::make_shared<ExpressionActions>(
+                    std::move(adding_column_dag), ExpressionActionsSettings::fromContext(context, CompileExpressions::yes));
+                builder->addSimpleTransform([&](const Block & stream_header)
+                                            { return std::make_shared<ExpressionTransform>(stream_header, adding_column_actions); });
+            }
 
-            builder->addSimpleTransform([&](const Block & stream_header)
-                                        { return std::make_shared<ExpressionTransform>(stream_header, adding_column_actions); });
+            if (has_table_virtual_column && common_header.has("_table") && !pipe_header.has("_table"))
+            {
+                ColumnWithTypeAndName column;
+                column.name = "_table";
+                column.type = std::make_shared<DataTypeLowCardinality>(std::make_shared<DataTypeString>());
+                column.column = column.type->createColumnConst(0, Field(table_name));
+
+                auto adding_column_dag = ActionsDAG::makeAddingColumnActions(std::move(column));
+                auto adding_column_actions = std::make_shared<ExpressionActions>(
+                    std::move(adding_column_dag), ExpressionActionsSettings::fromContext(context, CompileExpressions::yes));
+                builder->addSimpleTransform([&](const Block & stream_header)
+                                            { return std::make_shared<ExpressionTransform>(stream_header, adding_column_actions); });
+            }
         }
 
         /// Subordinary tables could have different but convertible types, like numeric types of different width.
         /// We must return streams with structure equals to structure of Merge table.
         convertAndFilterSourceStream(
-            header, modified_query_info, storage_snapshot_, aliases, row_policy_data_opt, context, *builder, processed_stage);
+            header, modified_query_info, storage_snapshot_, aliases, row_policy_data_opt, context, *builder, storage_stage);
     }
 
     return builder;
@@ -1116,13 +1157,13 @@ QueryPlan ReadFromMerge::createPlanForTable(
     bool allow_experimental_analyzer = modified_context->getSettingsRef().allow_experimental_analyzer;
 
     auto storage_stage = storage->getQueryProcessingStage(modified_context,
-        QueryProcessingStage::Complete,
+        processed_stage,
         storage_snapshot_,
         modified_query_info);
 
     QueryPlan plan;
 
-    if (processed_stage <= storage_stage || (allow_experimental_analyzer && processed_stage == QueryProcessingStage::FetchColumns))
+    if (processed_stage <= storage_stage)
     {
         /// If there are only virtual columns in query, you must request at least one other column.
         if (real_column_names.empty())
@@ -1167,7 +1208,7 @@ QueryPlan ReadFromMerge::createPlanForTable(
                 row_policy_data_opt->addStorageFilter(source_step_with_filter);
         }
     }
-    else if (processed_stage > storage_stage || (allow_experimental_analyzer && processed_stage != QueryProcessingStage::FetchColumns))
+    else if (processed_stage > storage_stage || allow_experimental_analyzer)
     {
         /// Maximum permissible parallelism is streams_num
         modified_context->setSetting("max_threads", streams_num);
diff --git a/tests/analyzer_tech_debt.txt b/tests/analyzer_tech_debt.txt
index bcf0eee0d0e..7cd73705e77 100644
--- a/tests/analyzer_tech_debt.txt
+++ b/tests/analyzer_tech_debt.txt
@@ -1,5 +1,4 @@
 00223_shard_distributed_aggregation_memory_efficient
-00717_merge_and_distributed
 00725_memory_tracking
 01062_pm_all_join_with_block_continuation
 01083_expressions_in_engine_arguments

From 836fe46c3371e99b8e6657d7fbeccff2f1cc9af2 Mon Sep 17 00:00:00 2001
From: Dmitry Novik <n0vik@clickhouse.com>
Date: Mon, 18 Mar 2024 15:50:58 +0100
Subject: [PATCH 439/443] Disable new analyzer in MutationsInterpreter
 #ci_set_analyzer

---
 src/Interpreters/MutationsInterpreter.cpp | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/src/Interpreters/MutationsInterpreter.cpp b/src/Interpreters/MutationsInterpreter.cpp
index 3b1a499255b..45ca9afce7c 100644
--- a/src/Interpreters/MutationsInterpreter.cpp
+++ b/src/Interpreters/MutationsInterpreter.cpp
@@ -402,10 +402,17 @@ MutationsInterpreter::MutationsInterpreter(
     , metadata_snapshot(metadata_snapshot_)
     , commands(std::move(commands_))
     , available_columns(std::move(available_columns_))
-    , context(Context::createCopy(context_))
     , settings(std::move(settings_))
     , select_limits(SelectQueryOptions().analyze(!settings.can_execute).ignoreLimits())
 {
+    auto new_context = Context::createCopy(context_);
+    if (new_context->getSettingsRef().allow_experimental_analyzer)
+    {
+        new_context->setSetting("allow_experimental_analyzer", false);
+        LOG_DEBUG(&Poco::Logger::get("MutationsInterpreter"), "Will use old analyzer to prepare mutation");
+    }
+    context = std::move(new_context);
+
     prepare(!settings.can_execute);
 }
 

From 15896ecaf67843d5a56105f438efd1c868c29b15 Mon Sep 17 00:00:00 2001
From: Nikita Mikhaylov <mikhaylovnikitka@gmail.com>
Date: Mon, 18 Mar 2024 16:58:57 +0100
Subject: [PATCH 440/443] Add table `system.keywords` [Part 1] (#51808)

---
 programs/keeper-client/Commands.cpp           |  19 +-
 src/Access/Common/AuthenticationType.cpp      |  26 +-
 src/Access/Common/AuthenticationType.h        |   5 +-
 src/Common/HashTable/StringHashMap.h          |   6 +-
 src/Common/HashTable/StringHashTable.h        |   6 +-
 src/Common/IntervalKind.cpp                   |   2 +-
 .../MySQL/tryParseTableIDFromDDL.cpp          |  14 +-
 src/Functions/DateTimeTransforms.h            |   1 -
 .../JSONPath/Parsers/ParserJSONPathRange.cpp  |   2 +-
 src/Interpreters/SessionLog.cpp               |   2 +-
 src/Interpreters/SystemLog.cpp                |  33 +-
 src/Parsers/ASTQueryWithOnCluster.cpp         |   2 +-
 src/Parsers/Access/ParserCreateQuotaQuery.cpp |  40 +-
 src/Parsers/Access/ParserCreateRoleQuery.cpp  |  20 +-
 .../Access/ParserCreateRowPolicyQuery.cpp     |  38 +-
 .../ParserCreateSettingsProfileQuery.cpp      |  22 +-
 src/Parsers/Access/ParserCreateUserQuery.cpp  |  80 +--
 .../Access/ParserDropAccessEntityQuery.cpp    |  12 +-
 src/Parsers/Access/ParserGrantQuery.cpp       |  28 +-
 .../Access/ParserMoveAccessEntityQuery.cpp    |  10 +-
 src/Parsers/Access/ParserPublicSSHKey.cpp     |   4 +-
 src/Parsers/Access/ParserRolesOrUsersSet.cpp  |  10 +-
 src/Parsers/Access/ParserRowPolicyName.cpp    |  12 +-
 src/Parsers/Access/ParserSetRoleQuery.cpp     |   8 +-
 .../Access/ParserSettingsProfileElement.cpp   |  22 +-
 .../Access/ParserShowAccessEntitiesQuery.cpp  |  14 +-
 src/Parsers/Access/ParserShowAccessQuery.h    |   2 +-
 .../ParserShowCreateAccessEntityQuery.cpp     |  12 +-
 src/Parsers/Access/ParserShowGrantsQuery.cpp  |   4 +-
 .../Access/ParserShowPrivilegesQuery.cpp      |   2 +-
 src/Parsers/Access/parseUserName.cpp          |   2 +-
 src/Parsers/CommonParsers.cpp                 | 109 ++++
 src/Parsers/CommonParsers.h                   | 580 +++++++++++++++++-
 src/Parsers/ExpressionElementParsers.cpp      |  99 +--
 src/Parsers/ExpressionListParsers.cpp         | 134 ++--
 .../KustoFunctions/KQLStringFunctions.cpp     |   4 +-
 src/Parsers/Kusto/ParserKQLMVExpand.cpp       |  12 +-
 src/Parsers/Kusto/ParserKQLMakeSeries.cpp     |   6 +-
 src/Parsers/Kusto/ParserKQLQuery.cpp          |   2 +-
 src/Parsers/MySQL/ASTAlterCommand.cpp         |  64 +-
 src/Parsers/MySQL/ASTAlterQuery.cpp           |   2 +-
 src/Parsers/MySQL/ASTCreateQuery.cpp          |  12 +-
 src/Parsers/MySQL/ASTDeclareConstraint.cpp    |  12 +-
 src/Parsers/MySQL/ASTDeclareIndex.cpp         |  34 +-
 src/Parsers/MySQL/ASTDeclareOption.cpp        |   2 +-
 src/Parsers/MySQL/ASTDeclarePartition.cpp     |   8 +-
 .../MySQL/ASTDeclarePartitionOptions.cpp      |  24 +-
 src/Parsers/MySQL/ASTDeclareReference.cpp     |  24 +-
 src/Parsers/MySQL/ASTDeclareSubPartition.cpp  |   2 +-
 src/Parsers/MySQL/ASTDeclareTableOptions.cpp  |   8 +-
 src/Parsers/MySQL/ASTDropQuery.cpp            |  24 +-
 .../ParserAlterNamedCollectionQuery.cpp       |  17 +-
 src/Parsers/ParserAlterQuery.cpp              | 180 +++---
 src/Parsers/ParserBackupQuery.cpp             |  40 +-
 src/Parsers/ParserCase.cpp                    |  10 +-
 src/Parsers/ParserCheckQuery.cpp              |   8 +-
 src/Parsers/ParserCreateFunctionQuery.cpp     |  12 +-
 src/Parsers/ParserCreateIndexQuery.cpp        |  16 +-
 src/Parsers/ParserCreateQuery.cpp             | 230 +++----
 src/Parsers/ParserCreateQuery.h               |  36 +-
 src/Parsers/ParserDataType.cpp                |  48 +-
 src/Parsers/ParserDatabaseOrNone.cpp          |   2 +-
 src/Parsers/ParserDeleteQuery.cpp             |  10 +-
 src/Parsers/ParserDescribeCacheQuery.cpp      |   6 +-
 src/Parsers/ParserDescribeTableQuery.cpp      |   8 +-
 src/Parsers/ParserDictionary.cpp              |  12 +-
 .../ParserDictionaryAttributeDeclaration.cpp  |  12 +-
 src/Parsers/ParserDropFunctionQuery.cpp       |   8 +-
 src/Parsers/ParserDropIndexQuery.cpp          |   8 +-
 .../ParserDropNamedCollectionQuery.cpp        |   8 +-
 src/Parsers/ParserDropQuery.cpp               |  28 +-
 src/Parsers/ParserExplainQuery.cpp            |  18 +-
 src/Parsers/ParserExternalDDLQuery.cpp        |  10 +-
 src/Parsers/ParserInsertQuery.cpp             |  24 +-
 src/Parsers/ParserKillQueryQuery.cpp          |  20 +-
 src/Parsers/ParserOptimizeQuery.cpp           |  14 +-
 src/Parsers/ParserPartition.cpp               |   4 +-
 src/Parsers/ParserProjectionSelectQuery.cpp   |   8 +-
 src/Parsers/ParserQueryWithOutput.cpp         |  16 +-
 src/Parsers/ParserRefreshStrategy.cpp         |  19 +-
 src/Parsers/ParserRenameQuery.cpp             |  22 +-
 src/Parsers/ParserSelectQuery.cpp             |  58 +-
 src/Parsers/ParserSetQuery.cpp                |  14 +-
 src/Parsers/ParserShowColumnsQuery.cpp        |  20 +-
 src/Parsers/ParserShowEngineQuery.h           |   2 +-
 src/Parsers/ParserShowFunctionsQuery.cpp      |   4 +-
 src/Parsers/ParserShowIndexesQuery.cpp        |  12 +-
 src/Parsers/ParserShowProcesslistQuery.h      |   2 +-
 src/Parsers/ParserShowSettingQuery.cpp        |   2 +-
 src/Parsers/ParserShowTablesQuery.cpp         |  40 +-
 src/Parsers/ParserSystemQuery.cpp             |  83 +--
 src/Parsers/ParserTablePropertiesQuery.cpp    |  16 +-
 src/Parsers/ParserTablesInSelectQuery.cpp     |  46 +-
 src/Parsers/ParserTransactionControl.cpp      |   8 +-
 src/Parsers/ParserUndropQuery.cpp             |   8 +-
 src/Parsers/ParserUseQuery.cpp                |   2 +-
 src/Parsers/ParserWatchQuery.cpp              |   6 +-
 src/Parsers/ParserWithElement.cpp             |   2 +-
 src/Parsers/parseIntervalKind.cpp             |  65 +-
 src/Storages/System/StorageSystemKeywords.cpp |  29 +
 src/Storages/System/StorageSystemKeywords.h   |  28 +
 src/Storages/System/attachSystemTables.cpp    |   2 +
 102 files changed, 1848 insertions(+), 1077 deletions(-)
 create mode 100644 src/Storages/System/StorageSystemKeywords.cpp
 create mode 100644 src/Storages/System/StorageSystemKeywords.h

diff --git a/programs/keeper-client/Commands.cpp b/programs/keeper-client/Commands.cpp
index 62b082ce15a..2ec43ae15d0 100644
--- a/programs/keeper-client/Commands.cpp
+++ b/programs/keeper-client/Commands.cpp
@@ -2,6 +2,7 @@
 #include "Commands.h"
 #include <queue>
 #include "KeeperClient.h"
+#include "Parsers/CommonParsers.h"
 
 
 namespace DB
@@ -106,13 +107,13 @@ bool CreateCommand::parse(IParser::Pos & pos, std::shared_ptr<ASTKeeperQuery> &
 
     int mode = zkutil::CreateMode::Persistent;
 
-    if (ParserKeyword{"PERSISTENT"}.ignore(pos, expected))
+    if (ParserKeyword(Keyword::PERSISTENT).ignore(pos, expected))
         mode = zkutil::CreateMode::Persistent;
-    else if (ParserKeyword{"EPHEMERAL"}.ignore(pos, expected))
+    else if (ParserKeyword(Keyword::EPHEMERAL).ignore(pos, expected))
         mode = zkutil::CreateMode::Ephemeral;
-    else if (ParserKeyword{"EPHEMERAL SEQUENTIAL"}.ignore(pos, expected))
+    else if (ParserKeyword(Keyword::EPHEMERAL_SEQUENTIAL).ignore(pos, expected))
         mode = zkutil::CreateMode::EphemeralSequential;
-    else if (ParserKeyword{"PERSISTENT SEQUENTIAL"}.ignore(pos, expected))
+    else if (ParserKeyword(Keyword::PERSISTENT_SEQUENTIAL).ignore(pos, expected))
         mode = zkutil::CreateMode::PersistentSequential;
 
     node->args.push_back(std::move(mode));
@@ -382,12 +383,16 @@ void RMRCommand::execute(const ASTKeeperQuery * query, KeeperClient * client) co
 
 bool ReconfigCommand::parse(IParser::Pos & pos, std::shared_ptr<ASTKeeperQuery> & node, DB::Expected & expected) const
 {
+    ParserKeyword s_add(Keyword::ADD);
+    ParserKeyword s_remove(Keyword::REMOVE);
+    ParserKeyword s_set(Keyword::SET);
+
     ReconfigCommand::Operation operation;
-    if (ParserKeyword{"ADD"}.ignore(pos, expected))
+    if (s_add.ignore(pos, expected))
         operation = ReconfigCommand::Operation::ADD;
-    else if (ParserKeyword{"REMOVE"}.ignore(pos, expected))
+    else if (s_remove.ignore(pos, expected))
         operation = ReconfigCommand::Operation::REMOVE;
-    else if (ParserKeyword{"SET"}.ignore(pos, expected))
+    else if (s_set.ignore(pos, expected))
         operation = ReconfigCommand::Operation::SET;
     else
         return false;
diff --git a/src/Access/Common/AuthenticationType.cpp b/src/Access/Common/AuthenticationType.cpp
index a61dc33fc0e..2cc126ad9b7 100644
--- a/src/Access/Common/AuthenticationType.cpp
+++ b/src/Access/Common/AuthenticationType.cpp
@@ -13,63 +13,63 @@ namespace ErrorCodes
 
 const AuthenticationTypeInfo & AuthenticationTypeInfo::get(AuthenticationType type_)
 {
-    static constexpr auto make_info = [](const char * raw_name_, bool is_password_ = false)
+    static constexpr auto make_info = [](Keyword keyword_, bool is_password_ = false)
     {
-        String init_name = raw_name_;
+        String init_name = String(toStringView(keyword_));
         boost::to_lower(init_name);
-        return AuthenticationTypeInfo{raw_name_, std::move(init_name), is_password_};
+        return AuthenticationTypeInfo{keyword_, std::move(init_name), is_password_};
     };
 
     switch (type_)
     {
         case AuthenticationType::NO_PASSWORD:
         {
-            static const auto info = make_info("NO_PASSWORD");
+            static const auto info = make_info(Keyword::NO_PASSWORD);
             return info;
         }
         case AuthenticationType::PLAINTEXT_PASSWORD:
         {
-            static const auto info = make_info("PLAINTEXT_PASSWORD", true);
+            static const auto info = make_info(Keyword::PLAINTEXT_PASSWORD, true);
             return info;
         }
         case AuthenticationType::SHA256_PASSWORD:
         {
-            static const auto info = make_info("SHA256_PASSWORD", true);
+            static const auto info = make_info(Keyword::SHA256_PASSWORD, true);
             return info;
         }
         case AuthenticationType::DOUBLE_SHA1_PASSWORD:
         {
-            static const auto info = make_info("DOUBLE_SHA1_PASSWORD", true);
+            static const auto info = make_info(Keyword::DOUBLE_SHA1_PASSWORD, true);
             return info;
         }
         case AuthenticationType::LDAP:
         {
-            static const auto info = make_info("LDAP");
+            static const auto info = make_info(Keyword::LDAP);
             return info;
         }
         case AuthenticationType::KERBEROS:
         {
-            static const auto info = make_info("KERBEROS");
+            static const auto info = make_info(Keyword::KERBEROS);
             return info;
         }
         case AuthenticationType::SSL_CERTIFICATE:
         {
-            static const auto info = make_info("SSL_CERTIFICATE");
+            static const auto info = make_info(Keyword::SSL_CERTIFICATE);
             return info;
         }
         case AuthenticationType::BCRYPT_PASSWORD:
         {
-            static const auto info = make_info("BCRYPT_PASSWORD", true);
+            static const auto info = make_info(Keyword::BCRYPT_PASSWORD, true);
             return info;
         }
         case AuthenticationType::SSH_KEY:
         {
-            static const auto info = make_info("SSH_KEY");
+            static const auto info = make_info(Keyword::SSH_KEY);
             return info;
         }
         case AuthenticationType::HTTP:
         {
-            static const auto info = make_info("HTTP");
+            static const auto info = make_info(Keyword::HTTP);
             return info;
         }
         case AuthenticationType::MAX:
diff --git a/src/Access/Common/AuthenticationType.h b/src/Access/Common/AuthenticationType.h
index e81b9e6e439..48ace3ca00a 100644
--- a/src/Access/Common/AuthenticationType.h
+++ b/src/Access/Common/AuthenticationType.h
@@ -1,6 +1,7 @@
 #pragma once
 
 #include <base/types.h>
+#include <Parsers/CommonParsers.h>
 
 namespace DB
 {
@@ -45,7 +46,7 @@ enum class AuthenticationType
 
 struct AuthenticationTypeInfo
 {
-    const char * const raw_name;
+    Keyword keyword; // Keyword used in parser
     const String name; /// Lowercased with underscores, e.g. "sha256_password".
     bool is_password;
     static const AuthenticationTypeInfo & get(AuthenticationType type_);
@@ -53,7 +54,7 @@ struct AuthenticationTypeInfo
 
 inline String toString(AuthenticationType type_)
 {
-    return AuthenticationTypeInfo::get(type_).raw_name;
+    return String(toStringView(AuthenticationTypeInfo::get(type_).keyword));
 }
 
 }
diff --git a/src/Common/HashTable/StringHashMap.h b/src/Common/HashTable/StringHashMap.h
index 828a54d357a..2599422ff3c 100644
--- a/src/Common/HashTable/StringHashMap.h
+++ b/src/Common/HashTable/StringHashMap.h
@@ -12,7 +12,7 @@ struct StringHashMapCell : public HashMapCell<Key, TMapped, StringHashTableHash,
     using Base::Base;
     static constexpr bool need_zero_value_storage = false;
     // external
-    StringRef getKey() const { return toStringRef(this->value.first); } /// NOLINT
+    StringRef getKey() const { return toStringView(this->value.first); } /// NOLINT
     // internal
     static const Key & getKey(const value_type & value_) { return value_.first; }
 };
@@ -32,7 +32,7 @@ struct StringHashMapCell<StringKey16, TMapped> : public HashMapCell<StringKey16,
     void setZero() { this->value.first.items[1] = 0; }
 
     // external
-    StringRef getKey() const { return toStringRef(this->value.first); } /// NOLINT
+    StringRef getKey() const { return toStringView(this->value.first); } /// NOLINT
     // internal
     static const StringKey16 & getKey(const value_type & value_) { return value_.first; }
 };
@@ -53,7 +53,7 @@ struct StringHashMapCell<StringKey24, TMapped> : public HashMapCell<StringKey24,
     void setZero() { this->value.first.c = 0; }
 
     // external
-    StringRef getKey() const { return toStringRef(this->value.first); } /// NOLINT
+    StringRef getKey() const { return toStringView(this->value.first); } /// NOLINT
     // internal
     static const StringKey24 & getKey(const value_type & value_) { return value_.first; }
 };
diff --git a/src/Common/HashTable/StringHashTable.h b/src/Common/HashTable/StringHashTable.h
index f3ba54e7d6e..30137d4ca6e 100644
--- a/src/Common/HashTable/StringHashTable.h
+++ b/src/Common/HashTable/StringHashTable.h
@@ -19,7 +19,7 @@ struct StringKey24
     bool operator==(const StringKey24 rhs) const { return a == rhs.a && b == rhs.b && c == rhs.c; }
 };
 
-inline StringRef ALWAYS_INLINE toStringRef(const StringKey8 & n)
+inline StringRef ALWAYS_INLINE toStringView(const StringKey8 & n)
 {
     assert(n != 0);
 #if __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__
@@ -28,7 +28,7 @@ inline StringRef ALWAYS_INLINE toStringRef(const StringKey8 & n)
     return {reinterpret_cast<const char *>(&n), 8ul - (std::countl_zero(n) >> 3)};
 #endif
 }
-inline StringRef ALWAYS_INLINE toStringRef(const StringKey16 & n)
+inline StringRef ALWAYS_INLINE toStringView(const StringKey16 & n)
 {
     assert(n.items[1] != 0);
 #if __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__
@@ -37,7 +37,7 @@ inline StringRef ALWAYS_INLINE toStringRef(const StringKey16 & n)
     return {reinterpret_cast<const char *>(&n), 16ul - (std::countl_zero(n.items[1]) >> 3)};
 #endif
 }
-inline StringRef ALWAYS_INLINE toStringRef(const StringKey24 & n)
+inline StringRef ALWAYS_INLINE toStringView(const StringKey24 & n)
 {
     assert(n.c != 0);
 #if __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__
diff --git a/src/Common/IntervalKind.cpp b/src/Common/IntervalKind.cpp
index 659f8fa5c6c..7ba32a689f5 100644
--- a/src/Common/IntervalKind.cpp
+++ b/src/Common/IntervalKind.cpp
@@ -240,7 +240,7 @@ const char * IntervalKind::toNameOfFunctionExtractTimePart() const
             return "toDayOfMonth";
         case IntervalKind::Kind::Week:
             // TODO: SELECT toRelativeWeekNum(toDate('2017-06-15')) - toRelativeWeekNum(toStartOfYear(toDate('2017-06-15')))
-            // else if (ParserKeyword("WEEK").ignore(pos, expected))
+            // else if (ParserKeyword(Keyword::WEEK).ignore(pos, expected))
             //    function_name = "toRelativeWeekNum";
             throw Exception(ErrorCodes::SYNTAX_ERROR, "The syntax 'EXTRACT(WEEK FROM date)' is not supported, cannot extract the number of a week");
         case IntervalKind::Kind::Month:
diff --git a/src/Databases/MySQL/tryParseTableIDFromDDL.cpp b/src/Databases/MySQL/tryParseTableIDFromDDL.cpp
index 4fe0f44c767..bc7f2507e57 100644
--- a/src/Databases/MySQL/tryParseTableIDFromDDL.cpp
+++ b/src/Databases/MySQL/tryParseTableIDFromDDL.cpp
@@ -12,23 +12,23 @@ StorageID tryParseTableIDFromDDL(const String & query, const String & default_da
     Tokens tokens(query.data(), query.data() + query.size());
     IParser::Pos pos(tokens, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS);
     Expected expected;
-    if (ParserKeyword("CREATE TEMPORARY TABLE").ignore(pos, expected) || ParserKeyword("CREATE TABLE").ignore(pos, expected))
+    if (ParserKeyword(Keyword::CREATE_TEMPORARY_TABLE).ignore(pos, expected) || ParserKeyword(Keyword::CREATE_TABLE).ignore(pos, expected))
     {
-        ParserKeyword("IF NOT EXISTS").ignore(pos, expected);
+        ParserKeyword(Keyword::IF_NOT_EXISTS).ignore(pos, expected);
         is_ddl = true;
     }
-    else if (ParserKeyword("ALTER TABLE").ignore(pos, expected) || ParserKeyword("RENAME TABLE").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::ALTER_TABLE).ignore(pos, expected) || ParserKeyword(Keyword::RENAME_TABLE).ignore(pos, expected))
     {
         is_ddl = true;
     }
-    else if (ParserKeyword("DROP TABLE").ignore(pos, expected) || ParserKeyword("DROP TEMPORARY TABLE").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::DROP_TABLE).ignore(pos, expected) || ParserKeyword(Keyword::DROP_TEMPORARY_TABLE).ignore(pos, expected))
     {
-        ParserKeyword("IF EXISTS").ignore(pos, expected);
+        ParserKeyword(Keyword::IF_EXISTS).ignore(pos, expected);
         is_ddl = true;
     }
-    else if (ParserKeyword("TRUNCATE").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::TRUNCATE).ignore(pos, expected))
     {
-        ParserKeyword("TABLE").ignore(pos, expected);
+        ParserKeyword(Keyword::TABLE).ignore(pos, expected);
         is_ddl = true;
     }
 
diff --git a/src/Functions/DateTimeTransforms.h b/src/Functions/DateTimeTransforms.h
index 70b2a7a83b4..f5072de24f3 100644
--- a/src/Functions/DateTimeTransforms.h
+++ b/src/Functions/DateTimeTransforms.h
@@ -2094,7 +2094,6 @@ struct Transformer
                     || std::is_same_v<Additions, DateTimeAccurateOrNullConvertStrategyAdditions>)
                 {
                     bool is_valid_input = vec_from[i] >= 0 && vec_from[i] <= 0xFFFFFFFFL;
-
                     if (!is_valid_input)
                     {
                         if constexpr (std::is_same_v<Additions, DateTimeAccurateOrNullConvertStrategyAdditions>)
diff --git a/src/Functions/JSONPath/Parsers/ParserJSONPathRange.cpp b/src/Functions/JSONPath/Parsers/ParserJSONPathRange.cpp
index 03c006774c0..fb74018b330 100644
--- a/src/Functions/JSONPath/Parsers/ParserJSONPathRange.cpp
+++ b/src/Functions/JSONPath/Parsers/ParserJSONPathRange.cpp
@@ -55,7 +55,7 @@ bool ParserJSONPathRange::parseImpl(Pos & pos, ASTPtr & node, Expected & expecte
         }
         else if (pos->type == TokenType::BareWord)
         {
-            if (!ParserKeyword("TO").ignore(pos, expected))
+            if (!ParserKeyword(Keyword::TO).ignore(pos, expected))
             {
                 return false;
             }
diff --git a/src/Interpreters/SessionLog.cpp b/src/Interpreters/SessionLog.cpp
index dc0ac963d0b..adb94cae0c2 100644
--- a/src/Interpreters/SessionLog.cpp
+++ b/src/Interpreters/SessionLog.cpp
@@ -77,7 +77,7 @@ ColumnsDescription SessionLogElement::getColumnsDescription()
             {"Logout",                 static_cast<Int8>(SESSION_LOGOUT)}
         });
 
-#define AUTH_TYPE_NAME_AND_VALUE(v) std::make_pair(AuthenticationTypeInfo::get(v).raw_name, static_cast<Int8>(v))
+#define AUTH_TYPE_NAME_AND_VALUE(v) std::make_pair(toString(v), static_cast<Int8>(v))
     auto identified_with_column = std::make_shared<DataTypeEnum8>(
         DataTypeEnum8::Values
         {
diff --git a/src/Interpreters/SystemLog.cpp b/src/Interpreters/SystemLog.cpp
index e4cbbb8f5f7..efb2559ce42 100644
--- a/src/Interpreters/SystemLog.cpp
+++ b/src/Interpreters/SystemLog.cpp
@@ -1,8 +1,18 @@
-#include <IO/WriteHelpers.h>
+#include <Interpreters/SystemLog.h>
+
+#include <base/scope_guard.h>
+#include <Common/logger_useful.h>
+#include <Common/MemoryTrackerBlockerInThread.h>
+#include <Common/quoteString.h>
+#include <Common/setThreadName.h>
 #include <Interpreters/AsynchronousInsertLog.h>
 #include <Interpreters/AsynchronousMetricLog.h>
+#include <Interpreters/BackupLog.h>
+#include <Interpreters/BlobStorageLog.h>
 #include <Interpreters/Context.h>
 #include <Interpreters/CrashLog.h>
+#include <Interpreters/FilesystemCacheLog.h>
+#include <Interpreters/FilesystemReadPrefetchesLog.h>
 #include <Interpreters/InterpreterCreateQuery.h>
 #include <Interpreters/InterpreterInsertQuery.h>
 #include <Interpreters/InterpreterRenameQuery.h>
@@ -10,38 +20,31 @@
 #include <Interpreters/OpenTelemetrySpanLog.h>
 #include <Interpreters/PartLog.h>
 #include <Interpreters/ProcessorsProfileLog.h>
-#include <Interpreters/BlobStorageLog.h>
 #include <Interpreters/QueryLog.h>
 #include <Interpreters/QueryThreadLog.h>
 #include <Interpreters/QueryViewsLog.h>
+#include <Interpreters/S3QueueLog.h>
 #include <Interpreters/SessionLog.h>
 #include <Interpreters/TextLog.h>
 #include <Interpreters/TraceLog.h>
 #include <Interpreters/TransactionsInfoLog.h>
-#include <Interpreters/FilesystemCacheLog.h>
-#include <Interpreters/FilesystemReadPrefetchesLog.h>
-#include <Interpreters/S3QueueLog.h>
 #include <Interpreters/ZooKeeperLog.h>
-#include <Interpreters/BackupLog.h>
+#include <IO/WriteHelpers.h>
 #include <Parsers/ASTCreateQuery.h>
 #include <Parsers/ASTFunction.h>
 #include <Parsers/ASTIndexDeclaration.h>
 #include <Parsers/ASTInsertQuery.h>
 #include <Parsers/ASTRenameQuery.h>
-#include <Parsers/ParserCreateQuery.h>
+#include <Parsers/CommonParsers.h>
 #include <Parsers/formatAST.h>
 #include <Parsers/parseQuery.h>
+#include <Parsers/ParserCreateQuery.h>
+#include <Poco/Util/AbstractConfiguration.h>
 #include <Processors/Executors/PushingPipelineExecutor.h>
 #include <Storages/IStorage.h>
 #include <Storages/MergeTree/MergeTreeSettings.h>
-#include <base/scope_guard.h>
-#include <fmt/core.h>
-#include <Poco/Util/AbstractConfiguration.h>
-#include "Common/quoteString.h"
-#include <Common/MemoryTrackerBlockerInThread.h>
-#include <Common/logger_useful.h>
-#include <Common/setThreadName.h>
 
+#include <fmt/core.h>
 
 namespace DB
 {
@@ -91,7 +94,7 @@ namespace
             if (!storage_p.parse(pos, storage, expected))
                 return false;
 
-            ParserKeyword s_comment("COMMENT");
+            ParserKeyword s_comment(Keyword::COMMENT);
             ParserStringLiteral string_literal_parser;
             ASTPtr comment;
 
diff --git a/src/Parsers/ASTQueryWithOnCluster.cpp b/src/Parsers/ASTQueryWithOnCluster.cpp
index d0b6d82ab13..1281a97fa00 100644
--- a/src/Parsers/ASTQueryWithOnCluster.cpp
+++ b/src/Parsers/ASTQueryWithOnCluster.cpp
@@ -19,7 +19,7 @@ std::string ASTQueryWithOnCluster::getRewrittenQueryWithoutOnCluster(const Witho
 
 bool ASTQueryWithOnCluster::parse(Pos & pos, std::string & cluster_str, Expected & expected)
 {
-    if (!ParserKeyword{"CLUSTER"}.ignore(pos, expected))
+    if (!ParserKeyword(Keyword::CLUSTER).ignore(pos, expected))
         return false;
 
     return parseIdentifierOrStringLiteral(pos, expected, cluster_str);
diff --git a/src/Parsers/Access/ParserCreateQuotaQuery.cpp b/src/Parsers/Access/ParserCreateQuotaQuery.cpp
index bc3512e3fc0..ddfdbe38903 100644
--- a/src/Parsers/Access/ParserCreateQuotaQuery.cpp
+++ b/src/Parsers/Access/ParserCreateQuotaQuery.cpp
@@ -33,7 +33,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (!ParserKeyword{"RENAME TO"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::RENAME_TO}.ignore(pos, expected))
                 return false;
 
             return parseIdentifierOrStringLiteral(pos, expected, new_name);
@@ -44,13 +44,13 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (ParserKeyword{"NOT KEYED"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::NOT_KEYED}.ignore(pos, expected))
             {
                 key_type = QuotaKeyType::NONE;
                 return true;
             }
 
-            if (!ParserKeyword{"KEY BY"}.ignore(pos, expected) && !ParserKeyword{"KEYED BY"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::KEY_BY}.ignore(pos, expected) && !ParserKeyword{Keyword::KEYED_BY}.ignore(pos, expected))
                 return false;
 
             Strings names;
@@ -84,7 +84,7 @@ namespace
         {
             for (auto qt : collections::range(QuotaType::MAX))
             {
-                if (ParserKeyword{QuotaTypeInfo::get(qt).keyword}.ignore(pos, expected))
+                if (ParserKeyword::createDeprecated(QuotaTypeInfo::get(qt).keyword).ignore(pos, expected))
                 {
                     quota_type = qt;
                     return true;
@@ -141,7 +141,7 @@ namespace
 
         auto parse_limit = [&]
         {
-            max_prefix_encountered |= ParserKeyword{"MAX"}.ignore(pos, expected);
+            max_prefix_encountered |= ParserKeyword{Keyword::MAX}.ignore(pos, expected);
 
             QuotaType quota_type;
             if (!parseQuotaType(pos, expected, quota_type))
@@ -153,7 +153,7 @@ namespace
             }
             else
             {
-                if (!ParserKeyword{"MAX"}.ignore(pos, expected))
+                if (!ParserKeyword{Keyword::MAX}.ignore(pos, expected))
                     return false;
             }
 
@@ -178,13 +178,13 @@ namespace
 
         auto parse_interval_with_limits = [&]
         {
-            if (!ParserKeyword{"FOR"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::FOR}.ignore(pos, expected))
                 return false;
 
             ASTCreateQuotaQuery::Limits limits;
-            limits.randomize_interval = ParserKeyword{"RANDOMIZED"}.ignore(pos, expected);
+            limits.randomize_interval = ParserKeyword{Keyword::RANDOMIZED}.ignore(pos, expected);
 
-            ParserKeyword{"INTERVAL"}.ignore(pos, expected);
+            ParserKeyword{Keyword::INTERVAL}.ignore(pos, expected);
 
             ASTPtr num_intervals_ast;
             if (!ParserNumber{}.parse(pos, num_intervals_ast, expected))
@@ -199,11 +199,11 @@ namespace
             limits.duration = std::chrono::seconds(static_cast<UInt64>(num_intervals * interval_kind.toAvgSeconds()));
             std::vector<std::pair<QuotaType, QuotaValue>> new_limits;
 
-            if (ParserKeyword{"NO LIMITS"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::NO_LIMITS}.ignore(pos, expected))
             {
                 limits.drop = true;
             }
-            else if (ParserKeyword{"TRACKING ONLY"}.ignore(pos, expected))
+            else if (ParserKeyword{Keyword::TRACKING_ONLY}.ignore(pos, expected))
             {
             }
             else if (parseLimits(pos, expected, new_limits))
@@ -232,7 +232,7 @@ namespace
             ASTPtr node;
             ParserRolesOrUsersSet roles_p;
             roles_p.allowAll().allowRoles().allowUsers().allowCurrentUser().useIDMode(id_mode);
-            if (!ParserKeyword{"TO"}.ignore(pos, expected) || !roles_p.parse(pos, node, expected))
+            if (!ParserKeyword{Keyword::TO}.ignore(pos, expected) || !roles_p.parse(pos, node, expected))
                 return false;
 
             roles = std::static_pointer_cast<ASTRolesOrUsersSet>(node);
@@ -244,7 +244,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            return ParserKeyword{"ON"}.ignore(pos, expected) && ASTQueryWithOnCluster::parse(pos, cluster, expected);
+            return ParserKeyword{Keyword::ON}.ignore(pos, expected) && ASTQueryWithOnCluster::parse(pos, cluster, expected);
         });
     }
 }
@@ -255,14 +255,14 @@ bool ParserCreateQuotaQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe
     bool alter = false;
     if (attach_mode)
     {
-        if (!ParserKeyword{"ATTACH QUOTA"}.ignore(pos, expected))
+        if (!ParserKeyword{Keyword::ATTACH_QUOTA}.ignore(pos, expected))
             return false;
     }
     else
     {
-        if (ParserKeyword{"ALTER QUOTA"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::ALTER_QUOTA}.ignore(pos, expected))
             alter = true;
-        else if (!ParserKeyword{"CREATE QUOTA"}.ignore(pos, expected))
+        else if (!ParserKeyword{Keyword::CREATE_QUOTA}.ignore(pos, expected))
             return false;
     }
 
@@ -271,14 +271,14 @@ bool ParserCreateQuotaQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe
     bool or_replace = false;
     if (alter)
     {
-        if (ParserKeyword{"IF EXISTS"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::IF_EXISTS}.ignore(pos, expected))
             if_exists = true;
     }
     else
     {
-        if (ParserKeyword{"IF NOT EXISTS"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::IF_NOT_EXISTS}.ignore(pos, expected))
             if_not_exists = true;
-        else if (ParserKeyword{"OR REPLACE"}.ignore(pos, expected))
+        else if (ParserKeyword{Keyword::OR_REPLACE}.ignore(pos, expected))
             or_replace = true;
     }
 
@@ -313,7 +313,7 @@ bool ParserCreateQuotaQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe
         if (cluster.empty() && parseOnCluster(pos, expected, cluster))
             continue;
 
-        if (storage_name.empty() && ParserKeyword{"IN"}.ignore(pos, expected) && parseAccessStorageName(pos, expected, storage_name))
+        if (storage_name.empty() && ParserKeyword{Keyword::IN}.ignore(pos, expected) && parseAccessStorageName(pos, expected, storage_name))
             continue;
 
         break;
diff --git a/src/Parsers/Access/ParserCreateRoleQuery.cpp b/src/Parsers/Access/ParserCreateRoleQuery.cpp
index 99a97f6901d..cdcb399e408 100644
--- a/src/Parsers/Access/ParserCreateRoleQuery.cpp
+++ b/src/Parsers/Access/ParserCreateRoleQuery.cpp
@@ -18,7 +18,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (!ParserKeyword{"RENAME TO"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::RENAME_TO}.ignore(pos, expected))
                 return false;
 
             return parseRoleName(pos, expected, new_name);
@@ -29,7 +29,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (!ParserKeyword{"SETTINGS"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::SETTINGS}.ignore(pos, expected))
                 return false;
 
             ASTPtr new_settings_ast;
@@ -47,7 +47,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            return ParserKeyword{"ON"}.ignore(pos, expected) && ASTQueryWithOnCluster::parse(pos, cluster, expected);
+            return ParserKeyword{Keyword::ON}.ignore(pos, expected) && ASTQueryWithOnCluster::parse(pos, cluster, expected);
         });
     }
 }
@@ -58,14 +58,14 @@ bool ParserCreateRoleQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
     bool alter = false;
     if (attach_mode)
     {
-        if (!ParserKeyword{"ATTACH ROLE"}.ignore(pos, expected))
+        if (!ParserKeyword{Keyword::ATTACH_ROLE}.ignore(pos, expected))
             return false;
     }
     else
     {
-        if (ParserKeyword{"ALTER ROLE"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::ALTER_ROLE}.ignore(pos, expected))
             alter = true;
-        else if (!ParserKeyword{"CREATE ROLE"}.ignore(pos, expected))
+        else if (!ParserKeyword{Keyword::CREATE_ROLE}.ignore(pos, expected))
             return false;
     }
 
@@ -74,14 +74,14 @@ bool ParserCreateRoleQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
     bool or_replace = false;
     if (alter)
     {
-        if (ParserKeyword{"IF EXISTS"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::IF_EXISTS}.ignore(pos, expected))
             if_exists = true;
     }
     else
     {
-        if (ParserKeyword{"IF NOT EXISTS"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::IF_NOT_EXISTS}.ignore(pos, expected))
             if_not_exists = true;
-        else if (ParserKeyword{"OR REPLACE"}.ignore(pos, expected))
+        else if (ParserKeyword{Keyword::OR_REPLACE}.ignore(pos, expected))
             or_replace = true;
     }
 
@@ -112,7 +112,7 @@ bool ParserCreateRoleQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
         if (cluster.empty() && parseOnCluster(pos, expected, cluster))
             continue;
 
-        if (storage_name.empty() && ParserKeyword{"IN"}.ignore(pos, expected) && parseAccessStorageName(pos, expected, storage_name))
+        if (storage_name.empty() && ParserKeyword{Keyword::IN}.ignore(pos, expected) && parseAccessStorageName(pos, expected, storage_name))
             continue;
 
         break;
diff --git a/src/Parsers/Access/ParserCreateRowPolicyQuery.cpp b/src/Parsers/Access/ParserCreateRowPolicyQuery.cpp
index f9e9466e35d..b49e153c5b0 100644
--- a/src/Parsers/Access/ParserCreateRowPolicyQuery.cpp
+++ b/src/Parsers/Access/ParserCreateRowPolicyQuery.cpp
@@ -23,7 +23,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (!ParserKeyword{"RENAME TO"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::RENAME_TO}.ignore(pos, expected))
                 return false;
 
             return parseIdentifierOrStringLiteral(pos, expected, new_short_name);
@@ -34,16 +34,16 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (!ParserKeyword{"AS"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::AS}.ignore(pos, expected))
                 return false;
 
-            if (ParserKeyword{"RESTRICTIVE"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::RESTRICTIVE}.ignore(pos, expected))
             {
                 is_restrictive = true;
                 return true;
             }
 
-            if (!ParserKeyword{"PERMISSIVE"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::PERMISSIVE}.ignore(pos, expected))
                 return false;
 
             is_restrictive = false;
@@ -55,7 +55,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (ParserKeyword("NONE").ignore(pos, expected))
+            if (ParserKeyword(Keyword::NONE).ignore(pos, expected))
             {
                 expr = nullptr;
                 return true;
@@ -89,7 +89,7 @@ namespace
 
         auto parse_command = [&]
         {
-            if (ParserKeyword{"ALL"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::ALL}.ignore(pos, expected))
             {
                 addAllCommands(res_commands);
                 return true;
@@ -98,7 +98,7 @@ namespace
             for (auto filter_type : collections::range(RowPolicyFilterType::MAX))
             {
                 std::string_view command = RowPolicyFilterTypeInfo::get(filter_type).command;
-                if (ParserKeyword{command.data()}.ignore(pos, expected))
+                if (ParserKeyword::createDeprecated(command.data()).ignore(pos, expected))
                 {
                     res_commands.emplace(command);
                     return true;
@@ -125,7 +125,7 @@ namespace
         {
             boost::container::flat_set<std::string_view> commands;
 
-            if (ParserKeyword{"FOR"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::FOR}.ignore(pos, expected))
             {
                 if (!parseCommands(pos, expected, commands))
                     return false;
@@ -135,12 +135,12 @@ namespace
 
             std::optional<ASTPtr> filter;
             std::optional<ASTPtr> check;
-            if (ParserKeyword{"USING"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::USING}.ignore(pos, expected))
             {
                 if (!parseFilterExpression(pos, expected, filter.emplace()))
                     return false;
             }
-            if (ParserKeyword{"WITH CHECK"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::WITH_CHECK}.ignore(pos, expected))
             {
                 if (!parseFilterExpression(pos, expected, check.emplace()))
                     return false;
@@ -179,7 +179,7 @@ namespace
         return IParserBase::wrapParseImpl(pos, [&]
         {
             ASTPtr ast;
-            if (!ParserKeyword{"TO"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::TO}.ignore(pos, expected))
                 return false;
 
             ParserRolesOrUsersSet roles_p;
@@ -196,7 +196,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            return ParserKeyword{"ON"}.ignore(pos, expected) && ASTQueryWithOnCluster::parse(pos, cluster, expected);
+            return ParserKeyword{Keyword::ON}.ignore(pos, expected) && ASTQueryWithOnCluster::parse(pos, cluster, expected);
         });
     }
 }
@@ -207,14 +207,14 @@ bool ParserCreateRowPolicyQuery::parseImpl(Pos & pos, ASTPtr & node, Expected &
     bool alter = false;
     if (attach_mode)
     {
-        if (!ParserKeyword{"ATTACH POLICY"}.ignore(pos, expected) && !ParserKeyword{"ATTACH ROW POLICY"}.ignore(pos, expected))
+        if (!ParserKeyword{Keyword::ATTACH_POLICY}.ignore(pos, expected) && !ParserKeyword{Keyword::ATTACH_ROW_POLICY}.ignore(pos, expected))
             return false;
     }
     else
     {
-        if (ParserKeyword{"ALTER POLICY"}.ignore(pos, expected) || ParserKeyword{"ALTER ROW POLICY"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::ALTER_POLICY}.ignore(pos, expected) || ParserKeyword{Keyword::ALTER_ROW_POLICY}.ignore(pos, expected))
             alter = true;
-        else if (!ParserKeyword{"CREATE POLICY"}.ignore(pos, expected) && !ParserKeyword{"CREATE ROW POLICY"}.ignore(pos, expected))
+        else if (!ParserKeyword{Keyword::CREATE_POLICY}.ignore(pos, expected) && !ParserKeyword{Keyword::CREATE_ROW_POLICY}.ignore(pos, expected))
             return false;
     }
 
@@ -223,14 +223,14 @@ bool ParserCreateRowPolicyQuery::parseImpl(Pos & pos, ASTPtr & node, Expected &
     bool or_replace = false;
     if (alter)
     {
-        if (ParserKeyword{"IF EXISTS"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::IF_EXISTS}.ignore(pos, expected))
             if_exists = true;
     }
     else
     {
-        if (ParserKeyword{"IF NOT EXISTS"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::IF_NOT_EXISTS}.ignore(pos, expected))
             if_not_exists = true;
-        else if (ParserKeyword{"OR REPLACE"}.ignore(pos, expected))
+        else if (ParserKeyword{Keyword::OR_REPLACE}.ignore(pos, expected))
             or_replace = true;
     }
 
@@ -273,7 +273,7 @@ bool ParserCreateRowPolicyQuery::parseImpl(Pos & pos, ASTPtr & node, Expected &
         if (cluster.empty() && parseOnCluster(pos, expected, cluster))
             continue;
 
-        if (storage_name.empty() && ParserKeyword{"IN"}.ignore(pos, expected) && parseAccessStorageName(pos, expected, storage_name))
+        if (storage_name.empty() && ParserKeyword{Keyword::IN}.ignore(pos, expected) && parseAccessStorageName(pos, expected, storage_name))
             continue;
 
         break;
diff --git a/src/Parsers/Access/ParserCreateSettingsProfileQuery.cpp b/src/Parsers/Access/ParserCreateSettingsProfileQuery.cpp
index fcb6ebd8806..6526aaf4fc5 100644
--- a/src/Parsers/Access/ParserCreateSettingsProfileQuery.cpp
+++ b/src/Parsers/Access/ParserCreateSettingsProfileQuery.cpp
@@ -20,7 +20,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (!ParserKeyword{"RENAME TO"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::RENAME_TO}.ignore(pos, expected))
                 return false;
 
             return parseIdentifierOrStringLiteral(pos, expected, new_name);
@@ -31,7 +31,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (!ParserKeyword{"SETTINGS"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::SETTINGS}.ignore(pos, expected))
                 return false;
 
             ASTPtr new_settings_ast;
@@ -50,7 +50,7 @@ namespace
         return IParserBase::wrapParseImpl(pos, [&]
         {
             ASTPtr ast;
-            if (!ParserKeyword{"TO"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::TO}.ignore(pos, expected))
                 return false;
 
             ParserRolesOrUsersSet roles_p;
@@ -67,7 +67,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            return ParserKeyword{"ON"}.ignore(pos, expected) && ASTQueryWithOnCluster::parse(pos, cluster, expected);
+            return ParserKeyword{Keyword::ON}.ignore(pos, expected) && ASTQueryWithOnCluster::parse(pos, cluster, expected);
         });
     }
 }
@@ -78,14 +78,14 @@ bool ParserCreateSettingsProfileQuery::parseImpl(Pos & pos, ASTPtr & node, Expec
     bool alter = false;
     if (attach_mode)
     {
-        if (!ParserKeyword{"ATTACH SETTINGS PROFILE"}.ignore(pos, expected) && !ParserKeyword{"ATTACH PROFILE"}.ignore(pos, expected))
+        if (!ParserKeyword{Keyword::ATTACH_SETTINGS_PROFILE}.ignore(pos, expected) && !ParserKeyword{Keyword::ATTACH_PROFILE}.ignore(pos, expected))
             return false;
     }
     else
     {
-        if (ParserKeyword{"ALTER SETTINGS PROFILE"}.ignore(pos, expected) || ParserKeyword{"ALTER PROFILE"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::ALTER_SETTINGS_PROFILE}.ignore(pos, expected) || ParserKeyword{Keyword::ALTER_PROFILE}.ignore(pos, expected))
             alter = true;
-        else if (!ParserKeyword{"CREATE SETTINGS PROFILE"}.ignore(pos, expected) && !ParserKeyword{"CREATE PROFILE"}.ignore(pos, expected))
+        else if (!ParserKeyword{Keyword::CREATE_SETTINGS_PROFILE}.ignore(pos, expected) && !ParserKeyword{Keyword::CREATE_PROFILE}.ignore(pos, expected))
             return false;
     }
 
@@ -94,14 +94,14 @@ bool ParserCreateSettingsProfileQuery::parseImpl(Pos & pos, ASTPtr & node, Expec
     bool or_replace = false;
     if (alter)
     {
-        if (ParserKeyword{"IF EXISTS"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::IF_EXISTS}.ignore(pos, expected))
             if_exists = true;
     }
     else
     {
-        if (ParserKeyword{"IF NOT EXISTS"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::IF_NOT_EXISTS}.ignore(pos, expected))
             if_not_exists = true;
-        else if (ParserKeyword{"OR REPLACE"}.ignore(pos, expected))
+        else if (ParserKeyword{Keyword::OR_REPLACE}.ignore(pos, expected))
             or_replace = true;
     }
 
@@ -132,7 +132,7 @@ bool ParserCreateSettingsProfileQuery::parseImpl(Pos & pos, ASTPtr & node, Expec
         if (cluster.empty() && parseOnCluster(pos, expected, cluster))
             continue;
 
-        if (storage_name.empty() && ParserKeyword{"IN"}.ignore(pos, expected) && parseAccessStorageName(pos, expected, storage_name))
+        if (storage_name.empty() && ParserKeyword{Keyword::IN}.ignore(pos, expected) && parseAccessStorageName(pos, expected, storage_name))
             continue;
 
         break;
diff --git a/src/Parsers/Access/ParserCreateUserQuery.cpp b/src/Parsers/Access/ParserCreateUserQuery.cpp
index 8e5a4d789c6..d4729ab796a 100644
--- a/src/Parsers/Access/ParserCreateUserQuery.cpp
+++ b/src/Parsers/Access/ParserCreateUserQuery.cpp
@@ -32,7 +32,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (!ParserKeyword{"RENAME TO"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::RENAME_TO}.ignore(pos, expected))
                 return false;
 
             String maybe_new_name;
@@ -48,7 +48,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (ParserKeyword{"NOT IDENTIFIED"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::NOT_IDENTIFIED}.ignore(pos, expected))
             {
                 auth_data = std::make_shared<ASTAuthenticationData>();
                 auth_data->type = AuthenticationType::NO_PASSWORD;
@@ -56,7 +56,7 @@ namespace
                 return true;
             }
 
-            if (!ParserKeyword{"IDENTIFIED"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::IDENTIFIED}.ignore(pos, expected))
                 return false;
 
             std::optional<AuthenticationType> type;
@@ -69,11 +69,11 @@ namespace
             bool expect_public_ssh_key = false;
             bool expect_http_auth_server = false;
 
-            if (ParserKeyword{"WITH"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::WITH}.ignore(pos, expected))
             {
                 for (auto check_type : collections::range(AuthenticationType::MAX))
                 {
-                    if (ParserKeyword{AuthenticationTypeInfo::get(check_type).raw_name}.ignore(pos, expected))
+                    if (ParserKeyword{AuthenticationTypeInfo::get(check_type).keyword}.ignore(pos, expected))
                     {
                         type = check_type;
 
@@ -96,17 +96,17 @@ namespace
 
                 if (!type)
                 {
-                    if (ParserKeyword{"SHA256_HASH"}.ignore(pos, expected))
+                    if (ParserKeyword{Keyword::SHA256_HASH}.ignore(pos, expected))
                     {
                         type = AuthenticationType::SHA256_PASSWORD;
                         expect_hash = true;
                     }
-                    else if (ParserKeyword{"DOUBLE_SHA1_HASH"}.ignore(pos, expected))
+                    else if (ParserKeyword{Keyword::DOUBLE_SHA1_HASH}.ignore(pos, expected))
                     {
                         type = AuthenticationType::DOUBLE_SHA1_PASSWORD;
                         expect_hash = true;
                     }
-                    else if (ParserKeyword{"BCRYPT_HASH"}.ignore(pos, expected))
+                    else if (ParserKeyword{Keyword::BCRYPT_HASH}.ignore(pos, expected))
                     {
                         type = AuthenticationType::BCRYPT_PASSWORD;
                         expect_hash = true;
@@ -128,12 +128,12 @@ namespace
 
             if (expect_password || expect_hash)
             {
-                if (!ParserKeyword{"BY"}.ignore(pos, expected) || !ParserStringAndSubstitution{}.parse(pos, value, expected))
+                if (!ParserKeyword{Keyword::BY}.ignore(pos, expected) || !ParserStringAndSubstitution{}.parse(pos, value, expected))
                     return false;
 
                 if (expect_hash && type == AuthenticationType::SHA256_PASSWORD)
                 {
-                    if (ParserKeyword{"SALT"}.ignore(pos, expected))
+                    if (ParserKeyword{Keyword::SALT}.ignore(pos, expected))
                     {
                         if (!ParserStringAndSubstitution{}.parse(pos, parsed_salt, expected))
                             return false;
@@ -142,12 +142,12 @@ namespace
             }
             else if (expect_ldap_server_name)
             {
-                if (!ParserKeyword{"SERVER"}.ignore(pos, expected) || !ParserStringAndSubstitution{}.parse(pos, value, expected))
+                if (!ParserKeyword{Keyword::SERVER}.ignore(pos, expected) || !ParserStringAndSubstitution{}.parse(pos, value, expected))
                     return false;
             }
             else if (expect_kerberos_realm)
             {
-                if (ParserKeyword{"REALM"}.ignore(pos, expected))
+                if (ParserKeyword{Keyword::REALM}.ignore(pos, expected))
                 {
                     if (!ParserStringAndSubstitution{}.parse(pos, value, expected))
                         return false;
@@ -155,7 +155,7 @@ namespace
             }
             else if (expect_common_names)
             {
-                if (!ParserKeyword{"CN"}.ignore(pos, expected))
+                if (!ParserKeyword{Keyword::CN}.ignore(pos, expected))
                     return false;
 
                 if (!ParserList{std::make_unique<ParserStringAndSubstitution>(), std::make_unique<ParserToken>(TokenType::Comma), false}.parse(pos, common_names, expected))
@@ -163,7 +163,7 @@ namespace
             }
             else if (expect_public_ssh_key)
             {
-                if (!ParserKeyword{"BY"}.ignore(pos, expected))
+                if (!ParserKeyword{Keyword::BY}.ignore(pos, expected))
                     return false;
 
                 if (!ParserList{std::make_unique<ParserPublicSSHKey>(), std::make_unique<ParserToken>(TokenType::Comma), false}.parse(pos, common_names, expected))
@@ -171,12 +171,12 @@ namespace
             }
             else if (expect_http_auth_server)
             {
-                if (!ParserKeyword{"SERVER"}.ignore(pos, expected))
+                if (!ParserKeyword{Keyword::SERVER}.ignore(pos, expected))
                     return false;
                 if (!ParserStringAndSubstitution{}.parse(pos, value, expected))
                     return false;
 
-                if (ParserKeyword{"SCHEME"}.ignore(pos, expected))
+                if (ParserKeyword{Keyword::SCHEME}.ignore(pos, expected))
                 {
                     if (!ParserStringAndSubstitution{}.parse(pos, http_auth_scheme, expected))
                         return false;
@@ -215,22 +215,22 @@ namespace
 
         auto parse_host = [&]
         {
-            if (ParserKeyword{"NONE"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::NONE}.ignore(pos, expected))
                 return true;
 
-            if (ParserKeyword{"ANY"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::ANY}.ignore(pos, expected))
             {
                 res_hosts.addAnyHost();
                 return true;
             }
 
-            if (ParserKeyword{"LOCAL"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::LOCAL}.ignore(pos, expected))
             {
                 res_hosts.addLocalHost();
                 return true;
             }
 
-            if (ParserKeyword{"REGEXP"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::REGEXP}.ignore(pos, expected))
             {
                 ASTPtr ast;
                 if (!ParserList{std::make_unique<ParserStringLiteral>(), std::make_unique<ParserToken>(TokenType::Comma), false}.parse(pos, ast, expected))
@@ -241,7 +241,7 @@ namespace
                 return true;
             }
 
-            if (ParserKeyword{"NAME"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::NAME}.ignore(pos, expected))
             {
                 ASTPtr ast;
                 if (!ParserList{std::make_unique<ParserStringLiteral>(), std::make_unique<ParserToken>(TokenType::Comma), false}.parse(pos, ast, expected))
@@ -253,7 +253,7 @@ namespace
                 return true;
             }
 
-            if (ParserKeyword{"IP"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::IP}.ignore(pos, expected))
             {
                 ASTPtr ast;
                 if (!ParserList{std::make_unique<ParserStringLiteral>(), std::make_unique<ParserToken>(TokenType::Comma), false}.parse(pos, ast, expected))
@@ -265,7 +265,7 @@ namespace
                 return true;
             }
 
-            if (ParserKeyword{"LIKE"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::LIKE}.ignore(pos, expected))
             {
                 ASTPtr ast;
                 if (!ParserList{std::make_unique<ParserStringLiteral>(), std::make_unique<ParserToken>(TokenType::Comma), false}.parse(pos, ast, expected))
@@ -292,10 +292,10 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (!prefix.empty() && !ParserKeyword{prefix}.ignore(pos, expected))
+            if (!prefix.empty() && !ParserKeyword::createDeprecated(prefix).ignore(pos, expected))
                 return false;
 
-            if (!ParserKeyword{"HOST"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::HOST}.ignore(pos, expected))
                 return false;
 
             AllowedClientHosts res_hosts;
@@ -312,7 +312,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (!ParserKeyword{"DEFAULT ROLE"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::DEFAULT_ROLE}.ignore(pos, expected))
                 return false;
 
             ASTPtr ast;
@@ -332,7 +332,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (!ParserKeyword{"SETTINGS"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::SETTINGS}.ignore(pos, expected))
                 return false;
 
             ASTPtr new_settings_ast;
@@ -350,7 +350,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (!ParserKeyword{"GRANTEES"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::GRANTEES}.ignore(pos, expected))
                 return false;
 
             ASTPtr ast;
@@ -368,7 +368,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            return ParserKeyword{"ON"}.ignore(pos, expected) && ASTQueryWithOnCluster::parse(pos, cluster, expected);
+            return ParserKeyword{Keyword::ON}.ignore(pos, expected) && ASTQueryWithOnCluster::parse(pos, cluster, expected);
         });
     }
 
@@ -376,7 +376,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (!ParserKeyword{"DEFAULT DATABASE"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::DEFAULT_DATABASE}.ignore(pos, expected))
                 return false;
 
             ASTPtr ast;
@@ -393,7 +393,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (!ParserKeyword{"VALID UNTIL"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::VALID_UNTIL}.ignore(pos, expected))
                 return false;
 
             ParserStringAndSubstitution until_p;
@@ -409,14 +409,14 @@ bool ParserCreateUserQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
     bool alter = false;
     if (attach_mode)
     {
-        if (!ParserKeyword{"ATTACH USER"}.ignore(pos, expected))
+        if (!ParserKeyword{Keyword::ATTACH_USER}.ignore(pos, expected))
             return false;
     }
     else
     {
-        if (ParserKeyword{"ALTER USER"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::ALTER_USER}.ignore(pos, expected))
             alter = true;
-        else if (!ParserKeyword{"CREATE USER"}.ignore(pos, expected))
+        else if (!ParserKeyword{Keyword::CREATE_USER}.ignore(pos, expected))
             return false;
     }
 
@@ -425,14 +425,14 @@ bool ParserCreateUserQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
     bool or_replace = false;
     if (alter)
     {
-        if (ParserKeyword{"IF EXISTS"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::IF_EXISTS}.ignore(pos, expected))
             if_exists = true;
     }
     else
     {
-        if (ParserKeyword{"IF NOT EXISTS"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::IF_NOT_EXISTS}.ignore(pos, expected))
             if_not_exists = true;
-        else if (ParserKeyword{"OR REPLACE"}.ignore(pos, expected))
+        else if (ParserKeyword{Keyword::OR_REPLACE}.ignore(pos, expected))
             or_replace = true;
     }
 
@@ -508,7 +508,7 @@ bool ParserCreateUserQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
             if (!new_name && (names->size() == 1) && parseRenameTo(pos, expected, new_name))
                 continue;
 
-            if (parseHosts(pos, expected, "ADD", new_hosts))
+            if (parseHosts(pos, expected, toStringView(Keyword::ADD), new_hosts))
             {
                 if (!add_hosts)
                     add_hosts.emplace();
@@ -516,7 +516,7 @@ bool ParserCreateUserQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
                 continue;
             }
 
-            if (parseHosts(pos, expected, "DROP", new_hosts))
+            if (parseHosts(pos, expected, toStringView(Keyword::DROP), new_hosts))
             {
                 if (!remove_hosts)
                     remove_hosts.emplace();
@@ -525,7 +525,7 @@ bool ParserCreateUserQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
             }
         }
 
-        if (storage_name.empty() && ParserKeyword{"IN"}.ignore(pos, expected) && parseAccessStorageName(pos, expected, storage_name))
+        if (storage_name.empty() && ParserKeyword{Keyword::IN}.ignore(pos, expected) && parseAccessStorageName(pos, expected, storage_name))
             continue;
 
         break;
diff --git a/src/Parsers/Access/ParserDropAccessEntityQuery.cpp b/src/Parsers/Access/ParserDropAccessEntityQuery.cpp
index 7fd34a6f8e4..973bf07e346 100644
--- a/src/Parsers/Access/ParserDropAccessEntityQuery.cpp
+++ b/src/Parsers/Access/ParserDropAccessEntityQuery.cpp
@@ -18,8 +18,8 @@ namespace
         for (auto i : collections::range(AccessEntityType::MAX))
         {
             const auto & type_info = AccessEntityTypeInfo::get(i);
-            if (ParserKeyword{type_info.name}.ignore(pos, expected)
-                || (!type_info.alias.empty() && ParserKeyword{type_info.alias}.ignore(pos, expected)))
+            if (ParserKeyword::createDeprecated(type_info.name).ignore(pos, expected)
+                || (!type_info.alias.empty() && ParserKeyword::createDeprecated(type_info.alias).ignore(pos, expected)))
             {
                 type = i;
                 return true;
@@ -33,7 +33,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            return ParserKeyword{"ON"}.ignore(pos, expected) && ASTQueryWithOnCluster::parse(pos, cluster, expected);
+            return ParserKeyword{Keyword::ON}.ignore(pos, expected) && ASTQueryWithOnCluster::parse(pos, cluster, expected);
         });
     }
 }
@@ -41,7 +41,7 @@ namespace
 
 bool ParserDropAccessEntityQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    if (!ParserKeyword{"DROP"}.ignore(pos, expected))
+    if (!ParserKeyword{Keyword::DROP}.ignore(pos, expected))
         return false;
 
     AccessEntityType type;
@@ -49,7 +49,7 @@ bool ParserDropAccessEntityQuery::parseImpl(Pos & pos, ASTPtr & node, Expected &
         return false;
 
     bool if_exists = false;
-    if (ParserKeyword{"IF EXISTS"}.ignore(pos, expected))
+    if (ParserKeyword{Keyword::IF_EXISTS}.ignore(pos, expected))
         if_exists = true;
 
     Strings names;
@@ -78,7 +78,7 @@ bool ParserDropAccessEntityQuery::parseImpl(Pos & pos, ASTPtr & node, Expected &
             return false;
     }
 
-    if (ParserKeyword{"FROM"}.ignore(pos, expected))
+    if (ParserKeyword{Keyword::FROM}.ignore(pos, expected))
         parseAccessStorageName(pos, expected, storage_name);
 
     if (cluster.empty())
diff --git a/src/Parsers/Access/ParserGrantQuery.cpp b/src/Parsers/Access/ParserGrantQuery.cpp
index bdb338e9338..799cd65dd5c 100644
--- a/src/Parsers/Access/ParserGrantQuery.cpp
+++ b/src/Parsers/Access/ParserGrantQuery.cpp
@@ -28,7 +28,7 @@ namespace
             if (pos_->type != TokenType::BareWord)
                 return false;
             std::string_view word{pos_->begin, pos_->size()};
-            return !(boost::iequals(word, "ON") || boost::iequals(word, "TO") || boost::iequals(word, "FROM"));
+            return !(boost::iequals(word, toStringView(Keyword::ON)) || boost::iequals(word, toStringView(Keyword::TO)) || boost::iequals(word, toStringView(Keyword::FROM)));
         };
 
         expected.add(pos, "access type");
@@ -132,7 +132,7 @@ namespace
                         ++is_global_with_parameter;
                 }
 
-                if (!ParserKeyword{"ON"}.ignore(pos, expected))
+                if (!ParserKeyword{Keyword::ON}.ignore(pos, expected))
                     return false;
 
                 if (is_global_with_parameter && is_global_with_parameter == access_and_columns.size())
@@ -197,7 +197,7 @@ namespace
         {
             AccessRightsElement default_element(AccessType::ALL);
 
-            if (!ParserKeyword{"ON"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::ON}.ignore(pos, expected))
                 return false;
 
             String database_name;
@@ -265,7 +265,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (!ParserKeyword{is_revoke ? "FROM" : "TO"}.ignore(pos, expected))
+            if (!ParserKeyword{is_revoke ? Keyword::FROM : Keyword::TO}.ignore(pos, expected))
                 return false;
 
             ASTPtr ast;
@@ -283,7 +283,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            return ParserKeyword{"ON"}.ignore(pos, expected) && ASTQueryWithOnCluster::parse(pos, cluster, expected);
+            return ParserKeyword{Keyword::ON}.ignore(pos, expected) && ASTQueryWithOnCluster::parse(pos, cluster, expected);
         });
     }
 }
@@ -291,14 +291,14 @@ namespace
 
 bool ParserGrantQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    if (attach_mode && !ParserKeyword{"ATTACH"}.ignore(pos, expected))
+    if (attach_mode && !ParserKeyword{Keyword::ATTACH}.ignore(pos, expected))
         return false;
 
     bool is_replace = false;
     bool is_revoke = false;
-    if (ParserKeyword{"REVOKE"}.ignore(pos, expected))
+    if (ParserKeyword{Keyword::REVOKE}.ignore(pos, expected))
         is_revoke = true;
-    else if (!ParserKeyword{"GRANT"}.ignore(pos, expected))
+    else if (!ParserKeyword{Keyword::GRANT}.ignore(pos, expected))
         return false;
 
     String cluster;
@@ -308,9 +308,9 @@ bool ParserGrantQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
     bool admin_option = false;
     if (is_revoke)
     {
-        if (ParserKeyword{"GRANT OPTION FOR"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::GRANT_OPTION_FOR}.ignore(pos, expected))
             grant_option = true;
-        else if (ParserKeyword{"ADMIN OPTION FOR"}.ignore(pos, expected))
+        else if (ParserKeyword{Keyword::ADMIN_OPTION_FOR}.ignore(pos, expected))
             admin_option = true;
     }
 
@@ -318,7 +318,7 @@ bool ParserGrantQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
     std::shared_ptr<ASTRolesOrUsersSet> roles;
 
     bool current_grants = false;
-    if (!is_revoke && ParserKeyword{"CURRENT GRANTS"}.ignore(pos, expected))
+    if (!is_revoke && ParserKeyword{Keyword::CURRENT_GRANTS}.ignore(pos, expected))
     {
         current_grants = true;
         if (!parseCurrentGrants(pos, expected, elements))
@@ -342,12 +342,12 @@ bool ParserGrantQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 
     if (!is_revoke)
     {
-        if (ParserKeyword{"WITH GRANT OPTION"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::WITH_GRANT_OPTION}.ignore(pos, expected))
             grant_option = true;
-        else if (ParserKeyword{"WITH ADMIN OPTION"}.ignore(pos, expected))
+        else if (ParserKeyword{Keyword::WITH_ADMIN_OPTION}.ignore(pos, expected))
             admin_option = true;
 
-        if (ParserKeyword{"WITH REPLACE OPTION"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::WITH_REPLACE_OPTION}.ignore(pos, expected))
             is_replace = true;
     }
 
diff --git a/src/Parsers/Access/ParserMoveAccessEntityQuery.cpp b/src/Parsers/Access/ParserMoveAccessEntityQuery.cpp
index 798e200bfb3..571b03f6f3c 100644
--- a/src/Parsers/Access/ParserMoveAccessEntityQuery.cpp
+++ b/src/Parsers/Access/ParserMoveAccessEntityQuery.cpp
@@ -18,8 +18,8 @@ namespace
         for (auto i : collections::range(AccessEntityType::MAX))
         {
             const auto & type_info = AccessEntityTypeInfo::get(i);
-            if (ParserKeyword{type_info.name}.ignore(pos, expected)
-                || (!type_info.alias.empty() && ParserKeyword{type_info.alias}.ignore(pos, expected)))
+            if (ParserKeyword::createDeprecated(type_info.name).ignore(pos, expected)
+                || (!type_info.alias.empty() && ParserKeyword::createDeprecated(type_info.alias).ignore(pos, expected)))
             {
                 type = i;
                 return true;
@@ -33,7 +33,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            return ParserKeyword{"ON"}.ignore(pos, expected) && ASTQueryWithOnCluster::parse(pos, cluster, expected);
+            return ParserKeyword{Keyword::ON}.ignore(pos, expected) && ASTQueryWithOnCluster::parse(pos, cluster, expected);
         });
     }
 }
@@ -41,7 +41,7 @@ namespace
 
 bool ParserMoveAccessEntityQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    if (!ParserKeyword{"MOVE"}.ignore(pos, expected))
+    if (!ParserKeyword{Keyword::MOVE}.ignore(pos, expected))
         return false;
 
     AccessEntityType type;
@@ -74,7 +74,7 @@ bool ParserMoveAccessEntityQuery::parseImpl(Pos & pos, ASTPtr & node, Expected &
             return false;
     }
 
-    if (!ParserKeyword{"TO"}.ignore(pos, expected) || !parseAccessStorageName(pos, expected, storage_name))
+    if (!ParserKeyword{Keyword::TO}.ignore(pos, expected) || !parseAccessStorageName(pos, expected, storage_name))
         return false;
 
     if (cluster.empty())
diff --git a/src/Parsers/Access/ParserPublicSSHKey.cpp b/src/Parsers/Access/ParserPublicSSHKey.cpp
index 725fe67a75b..bc033e25bbb 100644
--- a/src/Parsers/Access/ParserPublicSSHKey.cpp
+++ b/src/Parsers/Access/ParserPublicSSHKey.cpp
@@ -15,11 +15,11 @@ namespace
         return IParserBase::wrapParseImpl(pos, [&]
         {
             String key_base64;
-            if (!ParserKeyword{"KEY"}.ignore(pos, expected) || !parseIdentifierOrStringLiteral(pos, expected, key_base64))
+            if (!ParserKeyword{Keyword::KEY}.ignore(pos, expected) || !parseIdentifierOrStringLiteral(pos, expected, key_base64))
                 return false;
 
             String type;
-            if (!ParserKeyword{"TYPE"}.ignore(pos, expected) || !parseIdentifierOrStringLiteral(pos, expected, type))
+            if (!ParserKeyword{Keyword::TYPE}.ignore(pos, expected) || !parseIdentifierOrStringLiteral(pos, expected, type))
                 return false;
 
             ast = std::make_shared<ASTPublicSSHKey>();
diff --git a/src/Parsers/Access/ParserRolesOrUsersSet.cpp b/src/Parsers/Access/ParserRolesOrUsersSet.cpp
index 6f426d89bb3..6c46bff459b 100644
--- a/src/Parsers/Access/ParserRolesOrUsersSet.cpp
+++ b/src/Parsers/Access/ParserRolesOrUsersSet.cpp
@@ -19,7 +19,7 @@ namespace
             if (!id_mode)
                 return parseRoleName(pos, expected, res);
 
-            if (!ParserKeyword{"ID"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::ID}.ignore(pos, expected))
                 return false;
             if (!ParserToken(TokenType::OpeningRoundBracket).ignore(pos, expected))
                 return false;
@@ -53,16 +53,16 @@ namespace
 
         auto parse_element = [&]
         {
-            if (ParserKeyword{"NONE"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::NONE}.ignore(pos, expected))
                 return true;
 
-            if (allow_all && ParserKeyword{"ALL"}.ignore(pos, expected))
+            if (allow_all && ParserKeyword{Keyword::ALL}.ignore(pos, expected))
             {
                 res_all = true;
                 return true;
             }
 
-            if (allow_any && ParserKeyword{"ANY"}.ignore(pos, expected))
+            if (allow_any && ParserKeyword{Keyword::ANY}.ignore(pos, expected))
             {
                 res_all = true;
                 return true;
@@ -102,7 +102,7 @@ namespace
         bool & except_current_user)
     {
         return IParserBase::wrapParseImpl(pos, [&] {
-            if (!ParserKeyword{"EXCEPT"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::EXCEPT}.ignore(pos, expected))
                 return false;
 
             bool unused;
diff --git a/src/Parsers/Access/ParserRowPolicyName.cpp b/src/Parsers/Access/ParserRowPolicyName.cpp
index efdff3c24bf..e15b8b471e6 100644
--- a/src/Parsers/Access/ParserRowPolicyName.cpp
+++ b/src/Parsers/Access/ParserRowPolicyName.cpp
@@ -16,7 +16,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            return ParserKeyword{"ON"}.ignore(pos, expected) && ASTQueryWithOnCluster::parse(pos, cluster, expected);
+            return ParserKeyword{Keyword::ON}.ignore(pos, expected) && ASTQueryWithOnCluster::parse(pos, cluster, expected);
         });
     }
 
@@ -39,13 +39,13 @@ namespace
                 res_table_name = RowPolicyName::ANY_TABLE_MARK;
             }
 
-            /// If table is specified without DB it cannot be followed by "ON"
-            /// (but can be followed by "ON CLUSTER").
+            /// If table is specified without DB it cannot be followed by Keyword::ON
+            /// (but can be followed by Keyword::ON CLUSTER).
             /// The following code is necessary to figure out while parsing something like
             /// policy1 ON table1, policy2 ON table2
             /// that policy2 is another policy, not another table.
             auto end_pos = pos;
-            if (res_database.empty() && ParserKeyword{"ON"}.ignore(pos, expected))
+            if (res_database.empty() && ParserKeyword{Keyword::ON}.ignore(pos, expected))
             {
                 String unused;
                 if (ASTQueryWithOnCluster::parse(pos, unused, expected))
@@ -65,7 +65,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            return ParserKeyword{"ON"}.ignore(pos, expected) && parseDBAndTableName(pos, expected, database, table_name);
+            return ParserKeyword{Keyword::ON}.ignore(pos, expected) && parseDBAndTableName(pos, expected, database, table_name);
         });
     }
 
@@ -74,7 +74,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (!ParserKeyword{"ON"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::ON}.ignore(pos, expected))
                 return false;
 
             std::vector<std::pair<String, String>> res;
diff --git a/src/Parsers/Access/ParserSetRoleQuery.cpp b/src/Parsers/Access/ParserSetRoleQuery.cpp
index 50ccc67a372..251b33e9c11 100644
--- a/src/Parsers/Access/ParserSetRoleQuery.cpp
+++ b/src/Parsers/Access/ParserSetRoleQuery.cpp
@@ -29,7 +29,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (!ParserKeyword{"TO"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::TO}.ignore(pos, expected))
                 return false;
 
             ASTPtr ast;
@@ -50,11 +50,11 @@ bool ParserSetRoleQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected
 {
     using Kind = ASTSetRoleQuery::Kind;
     Kind kind;
-    if (ParserKeyword{"SET ROLE DEFAULT"}.ignore(pos, expected))
+    if (ParserKeyword{Keyword::SET_ROLE_DEFAULT}.ignore(pos, expected))
         kind = Kind::SET_ROLE_DEFAULT;
-    else if (ParserKeyword{"SET ROLE"}.ignore(pos, expected))
+    else if (ParserKeyword{Keyword::SET_ROLE}.ignore(pos, expected))
         kind = Kind::SET_ROLE;
-    else if (ParserKeyword{"SET DEFAULT ROLE"}.ignore(pos, expected))
+    else if (ParserKeyword{Keyword::SET_DEFAULT_ROLE}.ignore(pos, expected))
         kind = Kind::SET_DEFAULT_ROLE;
     else
         return false;
diff --git a/src/Parsers/Access/ParserSettingsProfileElement.cpp b/src/Parsers/Access/ParserSettingsProfileElement.cpp
index 36330b96622..78002d1b425 100644
--- a/src/Parsers/Access/ParserSettingsProfileElement.cpp
+++ b/src/Parsers/Access/ParserSettingsProfileElement.cpp
@@ -15,12 +15,12 @@ namespace
 {
     bool parseProfileKeyword(IParserBase::Pos & pos, Expected & expected, bool use_inherit_keyword)
     {
-        if (ParserKeyword{"PROFILE"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::PROFILE}.ignore(pos, expected))
             return true;
 
-        if (use_inherit_keyword && ParserKeyword{"INHERIT"}.ignore(pos, expected))
+        if (use_inherit_keyword && ParserKeyword{Keyword::INHERIT}.ignore(pos, expected))
         {
-            ParserKeyword{"PROFILE"}.ignore(pos, expected);
+            ParserKeyword{Keyword::PROFILE}.ignore(pos, expected);
             return true;
         }
 
@@ -36,7 +36,7 @@ namespace
             if (!id_mode)
                 return parseIdentifierOrStringLiteral(pos, expected, res);
 
-            if (!ParserKeyword{"ID"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::ID}.ignore(pos, expected))
                 return false;
             if (!ParserToken(TokenType::OpeningRoundBracket).ignore(pos, expected))
                 return false;
@@ -73,8 +73,8 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            bool is_min_value = ParserKeyword{"MIN"}.ignore(pos, expected);
-            bool is_max_value = !is_min_value && ParserKeyword{"MAX"}.ignore(pos, expected);
+            bool is_min_value = ParserKeyword{Keyword::MIN}.ignore(pos, expected);
+            bool is_max_value = !is_min_value && ParserKeyword{Keyword::MAX}.ignore(pos, expected);
             if (!is_min_value && !is_max_value)
                 return false;
 
@@ -99,17 +99,17 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (ParserKeyword{"READONLY"}.ignore(pos, expected) || ParserKeyword{"CONST"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::READONLY}.ignore(pos, expected) || ParserKeyword{Keyword::CONST}.ignore(pos, expected))
             {
                 writability = SettingConstraintWritability::CONST;
                 return true;
             }
-            else if (ParserKeyword{"WRITABLE"}.ignore(pos, expected))
+            else if (ParserKeyword{Keyword::WRITABLE}.ignore(pos, expected))
             {
                 writability = SettingConstraintWritability::WRITABLE;
                 return true;
             }
-            else if (ParserKeyword{"CHANGEABLE_IN_READONLY"}.ignore(pos, expected))
+            else if (ParserKeyword{Keyword::CHANGEABLE_IN_READONLY}.ignore(pos, expected))
             {
                 writability = SettingConstraintWritability::CHANGEABLE_IN_READONLY;
                 return true;
@@ -151,7 +151,7 @@ namespace
             if (!has_value_or_constraint)
                 return false;
 
-            if (boost::iequals(res_setting_name, "PROFILE") && !res_value && !res_min_value && !res_max_value
+            if (boost::iequals(res_setting_name, toStringView(Keyword::PROFILE)) && !res_value && !res_min_value && !res_max_value
                 && res_writability == SettingConstraintWritability::CONST)
             {
                 /// Ambiguity: "profile readonly" can be treated either as a profile named "readonly" or
@@ -224,7 +224,7 @@ bool ParserSettingsProfileElements::parseImpl(Pos & pos, ASTPtr & node, Expected
 {
     std::vector<std::shared_ptr<ASTSettingsProfileElement>> elements;
 
-    if (ParserKeyword{"NONE"}.ignore(pos, expected))
+    if (ParserKeyword{Keyword::NONE}.ignore(pos, expected))
     {
     }
     else
diff --git a/src/Parsers/Access/ParserShowAccessEntitiesQuery.cpp b/src/Parsers/Access/ParserShowAccessEntitiesQuery.cpp
index 15cb815f457..898971a02b3 100644
--- a/src/Parsers/Access/ParserShowAccessEntitiesQuery.cpp
+++ b/src/Parsers/Access/ParserShowAccessEntitiesQuery.cpp
@@ -15,8 +15,8 @@ namespace
         for (auto i : collections::range(AccessEntityType::MAX))
         {
             const auto & type_info = AccessEntityTypeInfo::get(i);
-            if (ParserKeyword{type_info.plural_name}.ignore(pos, expected)
-                || (!type_info.plural_alias.empty() && ParserKeyword{type_info.plural_alias}.ignore(pos, expected)))
+            if (ParserKeyword::createDeprecated(type_info.plural_name).ignore(pos, expected)
+                || (!type_info.plural_alias.empty() && ParserKeyword::createDeprecated(type_info.plural_alias).ignore(pos, expected)))
             {
                 type = i;
                 return true;
@@ -29,7 +29,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            return ParserKeyword{"ON"}.ignore(pos, expected)
+            return ParserKeyword{Keyword::ON}.ignore(pos, expected)
                 && parseDatabaseAndTableNameOrAsterisks(pos, expected, database, any_database, table, any_table);
         });
     }
@@ -38,7 +38,7 @@ namespace
 
 bool ParserShowAccessEntitiesQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    if (!ParserKeyword{"SHOW"}.ignore(pos, expected))
+    if (!ParserKeyword{Keyword::SHOW}.ignore(pos, expected))
         return false;
 
     AccessEntityType type;
@@ -51,17 +51,17 @@ bool ParserShowAccessEntitiesQuery::parseImpl(Pos & pos, ASTPtr & node, Expected
     {
         all = true;
     }
-    else if (ParserKeyword{"CURRENT ROLES"}.ignore(pos, expected))
+    else if (ParserKeyword{Keyword::CURRENT_ROLES}.ignore(pos, expected))
     {
         type = AccessEntityType::ROLE;
         current_roles = true;
     }
-    else if (ParserKeyword{"ENABLED ROLES"}.ignore(pos, expected))
+    else if (ParserKeyword{Keyword::ENABLED_ROLES}.ignore(pos, expected))
     {
         type = AccessEntityType::ROLE;
         enabled_roles = true;
     }
-    else if (ParserKeyword{"CURRENT QUOTA"}.ignore(pos, expected) || ParserKeyword{"QUOTA"}.ignore(pos, expected))
+    else if (ParserKeyword{Keyword::CURRENT_QUOTA}.ignore(pos, expected) || ParserKeyword{Keyword::QUOTA}.ignore(pos, expected))
     {
         type = AccessEntityType::QUOTA;
         current_quota = true;
diff --git a/src/Parsers/Access/ParserShowAccessQuery.h b/src/Parsers/Access/ParserShowAccessQuery.h
index da0d6ff449f..51ccf84b349 100644
--- a/src/Parsers/Access/ParserShowAccessQuery.h
+++ b/src/Parsers/Access/ParserShowAccessQuery.h
@@ -20,7 +20,7 @@ protected:
     {
         auto query = std::make_shared<ASTShowAccessQuery>();
 
-        if (!ParserKeyword("SHOW ACCESS").ignore(pos, expected))
+        if (!ParserKeyword(Keyword::SHOW_ACCESS).ignore(pos, expected))
             return false;
 
         node = query;
diff --git a/src/Parsers/Access/ParserShowCreateAccessEntityQuery.cpp b/src/Parsers/Access/ParserShowCreateAccessEntityQuery.cpp
index 17caa6366f1..8e8321dad28 100644
--- a/src/Parsers/Access/ParserShowCreateAccessEntityQuery.cpp
+++ b/src/Parsers/Access/ParserShowCreateAccessEntityQuery.cpp
@@ -25,8 +25,8 @@ namespace
         for (auto i : collections::range(AccessEntityType::MAX))
         {
             const auto & type_info = AccessEntityTypeInfo::get(i);
-            if (ParserKeyword{type_info.name}.ignore(pos, expected)
-                || (!type_info.alias.empty() && ParserKeyword{type_info.alias}.ignore(pos, expected)))
+            if (ParserKeyword::createDeprecated(type_info.name).ignore(pos, expected)
+                || (!type_info.alias.empty() && ParserKeyword::createDeprecated(type_info.alias).ignore(pos, expected)))
             {
                 type = i;
                 plural = false;
@@ -37,8 +37,8 @@ namespace
         for (auto i : collections::range(AccessEntityType::MAX))
         {
             const auto & type_info = AccessEntityTypeInfo::get(i);
-            if (ParserKeyword{type_info.plural_name}.ignore(pos, expected)
-                || (!type_info.plural_alias.empty() && ParserKeyword{type_info.plural_alias}.ignore(pos, expected)))
+            if (ParserKeyword::createDeprecated(type_info.plural_name).ignore(pos, expected)
+                || (!type_info.plural_alias.empty() && ParserKeyword::createDeprecated(type_info.plural_alias).ignore(pos, expected)))
             {
                 type = i;
                 plural = true;
@@ -53,7 +53,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            return ParserKeyword{"ON"}.ignore(pos, expected)
+            return ParserKeyword{Keyword::ON}.ignore(pos, expected)
                 && parseDatabaseAndTableNameOrAsterisks(pos, expected, database, any_database, table, any_table);
         });
     }
@@ -62,7 +62,7 @@ namespace
 
 bool ParserShowCreateAccessEntityQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    if (!ParserKeyword{"SHOW CREATE"}.ignore(pos, expected))
+    if (!ParserKeyword{Keyword::SHOW_CREATE}.ignore(pos, expected))
         return false;
 
     AccessEntityType type;
diff --git a/src/Parsers/Access/ParserShowGrantsQuery.cpp b/src/Parsers/Access/ParserShowGrantsQuery.cpp
index 02d85d2f90b..092334e8a6f 100644
--- a/src/Parsers/Access/ParserShowGrantsQuery.cpp
+++ b/src/Parsers/Access/ParserShowGrantsQuery.cpp
@@ -10,12 +10,12 @@ namespace DB
 {
 bool ParserShowGrantsQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    if (!ParserKeyword{"SHOW GRANTS"}.ignore(pos, expected))
+    if (!ParserKeyword{Keyword::SHOW_GRANTS}.ignore(pos, expected))
         return false;
 
     std::shared_ptr<ASTRolesOrUsersSet> for_roles;
 
-    if (ParserKeyword{"FOR"}.ignore(pos, expected))
+    if (ParserKeyword{Keyword::FOR}.ignore(pos, expected))
     {
         ASTPtr for_roles_ast;
         ParserRolesOrUsersSet for_roles_p;
diff --git a/src/Parsers/Access/ParserShowPrivilegesQuery.cpp b/src/Parsers/Access/ParserShowPrivilegesQuery.cpp
index a120d4ed7c2..33bf118bdad 100644
--- a/src/Parsers/Access/ParserShowPrivilegesQuery.cpp
+++ b/src/Parsers/Access/ParserShowPrivilegesQuery.cpp
@@ -10,7 +10,7 @@ bool ParserShowPrivilegesQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & e
 {
     auto query = std::make_shared<ASTShowPrivilegesQuery>();
 
-    if (!ParserKeyword("SHOW PRIVILEGES").ignore(pos, expected))
+    if (!ParserKeyword(Keyword::SHOW_PRIVILEGES).ignore(pos, expected))
         return false;
 
     node = query;
diff --git a/src/Parsers/Access/parseUserName.cpp b/src/Parsers/Access/parseUserName.cpp
index a4c5c7e8946..2c638094fce 100644
--- a/src/Parsers/Access/parseUserName.cpp
+++ b/src/Parsers/Access/parseUserName.cpp
@@ -32,7 +32,7 @@ bool parseCurrentUserTag(IParser::Pos & pos, Expected & expected)
 {
     return IParserBase::wrapParseImpl(pos, [&]
     {
-        if (!ParserKeyword{"CURRENT_USER"}.ignore(pos, expected) && !ParserKeyword{"currentUser"}.ignore(pos, expected))
+        if (!ParserKeyword{Keyword::CURRENTUSER}.ignore(pos, expected) && !ParserKeyword{Keyword::CURRENT_USER}.ignore(pos, expected))
             return false;
 
         if (ParserToken{TokenType::OpeningRoundBracket}.ignore(pos, expected))
diff --git a/src/Parsers/CommonParsers.cpp b/src/Parsers/CommonParsers.cpp
index 275679d61f0..ed8d58d5a68 100644
--- a/src/Parsers/CommonParsers.cpp
+++ b/src/Parsers/CommonParsers.cpp
@@ -1,9 +1,117 @@
 #include <Parsers/CommonParsers.h>
+#include <Common/ErrorCodes.h>
 #include <base/find_symbols.h>
+#include <algorithm>
+#include <cctype>
 
 namespace DB
 {
 
+namespace ErrorCodes
+{
+    extern const int LOGICAL_ERROR;
+}
+
+namespace
+{
+using Strings = std::vector<String>;
+
+class KeyWordToStringConverter
+{
+public:
+    static const KeyWordToStringConverter & instance()
+    {
+        static const KeyWordToStringConverter res;
+        return res;
+    }
+
+    std::string_view convert(Keyword type) const
+    {
+        return mapping[static_cast<size_t>(type)];
+    }
+
+    const std::vector<String> & getMapping() const
+    {
+        return mapping;
+    }
+
+private:
+    KeyWordToStringConverter()
+    {
+#define KEYWORD_TYPE_TO_STRING_CONVERTER_ADD_TO_MAPPING(identifier, value) \
+        checkUnderscore(value); \
+        addToMapping(Keyword::identifier, value);
+        APPLY_FOR_PARSER_KEYWORDS(KEYWORD_TYPE_TO_STRING_CONVERTER_ADD_TO_MAPPING)
+#undef KEYWORD_TYPE_TO_STRING_CONVERTER_ADD_TO_MAPPING
+
+#define KEYWORD_TYPE_TO_STRING_CONVERTER_ADD_TO_MAPPING(identifier, value) \
+        addToMapping(Keyword::identifier, value);
+        APPLY_FOR_PARSER_KEYWORDS_WITH_UNDERSCORES(KEYWORD_TYPE_TO_STRING_CONVERTER_ADD_TO_MAPPING)
+#undef KEYWORD_TYPE_TO_STRING_CONVERTER_ADD_TO_MAPPING
+
+#ifndef NDEBUG
+#define KEYWORD_TYPE_TO_STRING_CONVERTER_ADD_TO_MAPPING(identifier, value) \
+        check(#identifier, value);
+        APPLY_FOR_PARSER_KEYWORDS(KEYWORD_TYPE_TO_STRING_CONVERTER_ADD_TO_MAPPING)
+        APPLY_FOR_PARSER_KEYWORDS_WITH_UNDERSCORES(KEYWORD_TYPE_TO_STRING_CONVERTER_ADD_TO_MAPPING)
+#undef KEYWORD_TYPE_TO_STRING_CONVERTER_ADD_TO_MAPPING
+#endif
+    }
+
+    void addToMapping(Keyword identifier, std::string_view value)
+    {
+        size_t index = static_cast<size_t>(identifier);
+        mapping.resize(std::max(index + 1, mapping.size()));
+        mapping[index] = value;
+    }
+
+    void checkUnderscore(std::string_view value)
+    {
+        if (value.contains('_'))
+            throw Exception(ErrorCodes::LOGICAL_ERROR,
+                "The keyword {} has underscore. If this is intentional, please declare it in another list.", value);
+    }
+
+    [[ maybe_unused ]] void check(std::string_view identifier, std::string_view value)
+    {
+        if (value == "TRUE" || value == "FALSE" || value == "NULL")
+            return;
+
+        if (identifier.size() != value.size())
+            throw Exception(ErrorCodes::LOGICAL_ERROR, "The length of the keyword identifier and the length of its value are different.");
+
+        for (size_t i = 0; i < identifier.size(); ++i)
+        {
+            if (std::tolower(identifier[i]) == '_' && std::tolower(value[i]) == ' ')
+                continue;
+            if (std::tolower(identifier[i]) == std::tolower(value[i]))
+                continue;
+            throw Exception(ErrorCodes::LOGICAL_ERROR,
+                "Keyword identifier {} differs from its value {} in {} position: {} and {}",
+                identifier, value, i, identifier[i], value[i]);
+        }
+
+    }
+
+    Strings mapping;
+};
+}
+
+
+std::string_view toStringView(Keyword type)
+{
+    return KeyWordToStringConverter::instance().convert(type);
+}
+
+const std::vector<String> & getAllKeyWords()
+{
+    return KeyWordToStringConverter::instance().getMapping();
+}
+
+ParserKeyword::ParserKeyword(Keyword keyword)
+    : s(toStringView(keyword))
+{}
+
 bool ParserKeyword::parseImpl(Pos & pos, [[maybe_unused]] ASTPtr & node, Expected & expected)
 {
     if (pos->type != TokenType::BareWord)
@@ -38,4 +146,5 @@ bool ParserKeyword::parseImpl(Pos & pos, [[maybe_unused]] ASTPtr & node, Expecte
     return true;
 }
 
+
 }
diff --git a/src/Parsers/CommonParsers.h b/src/Parsers/CommonParsers.h
index f6c5c9c0520..b2a07bc06bf 100644
--- a/src/Parsers/CommonParsers.h
+++ b/src/Parsers/CommonParsers.h
@@ -3,10 +3,575 @@
 #include <Parsers/IParserBase.h>
 
 #include <cassert>
+#include <string_view>
 
 namespace DB
 {
 
+#define APPLY_FOR_PARSER_KEYWORDS(MR_MACROS) \
+    MR_MACROS(ADD_COLUMN, "ADD COLUMN") \
+    MR_MACROS(ADD_CONSTRAINT, "ADD CONSTRAINT") \
+    MR_MACROS(ADD_INDEX, "ADD INDEX") \
+    MR_MACROS(ADD_PROJECTION, "ADD PROJECTION") \
+    MR_MACROS(ADD_STATISTIC, "ADD STATISTIC") \
+    MR_MACROS(ADD, "ADD") \
+    MR_MACROS(ADMIN_OPTION_FOR, "ADMIN OPTION FOR") \
+    MR_MACROS(AFTER, "AFTER") \
+    MR_MACROS(ALGORITHM, "ALGORITHM") \
+    MR_MACROS(ALIAS, "ALIAS") \
+    MR_MACROS(ALL, "ALL") \
+    MR_MACROS(ALTER_COLUMN, "ALTER COLUMN") \
+    MR_MACROS(ALTER_DATABASE, "ALTER DATABASE") \
+    MR_MACROS(ALTER_LIVE_VIEW, "ALTER LIVE VIEW") \
+    MR_MACROS(ALTER_POLICY, "ALTER POLICY") \
+    MR_MACROS(ALTER_PROFILE, "ALTER PROFILE") \
+    MR_MACROS(ALTER_QUOTA, "ALTER QUOTA") \
+    MR_MACROS(ALTER_ROLE, "ALTER ROLE") \
+    MR_MACROS(ALTER_ROW_POLICY, "ALTER ROW POLICY") \
+    MR_MACROS(ALTER_SETTINGS_PROFILE, "ALTER SETTINGS PROFILE") \
+    MR_MACROS(ALTER_TABLE, "ALTER TABLE") \
+    MR_MACROS(ALTER_TEMPORARY_TABLE, "ALTER TEMPORARY TABLE") \
+    MR_MACROS(ALTER_USER, "ALTER USER") \
+    MR_MACROS(ALTER, "ALTER") \
+    MR_MACROS(AND_STDOUT, "AND STDOUT") \
+    MR_MACROS(AND, "AND") \
+    MR_MACROS(ANTI, "ANTI") \
+    MR_MACROS(ANY, "ANY") \
+    MR_MACROS(APPEND, "APPEND") \
+    MR_MACROS(APPLY_DELETED_MASK, "APPLY DELETED MASK") \
+    MR_MACROS(APPLY, "APPLY") \
+    MR_MACROS(ARRAY_JOIN, "ARRAY JOIN") \
+    MR_MACROS(AS, "AS") \
+    MR_MACROS(ASC, "ASC") \
+    MR_MACROS(ASCENDING, "ASCENDING") \
+    MR_MACROS(ASOF, "ASOF") \
+    MR_MACROS(ASSUME, "ASSUME") \
+    MR_MACROS(AST, "AST") \
+    MR_MACROS(ASYNC, "ASYNC") \
+    MR_MACROS(ATTACH_PART, "ATTACH PART") \
+    MR_MACROS(ATTACH_PARTITION, "ATTACH PARTITION") \
+    MR_MACROS(ATTACH_POLICY, "ATTACH POLICY") \
+    MR_MACROS(ATTACH_PROFILE, "ATTACH PROFILE") \
+    MR_MACROS(ATTACH_QUOTA, "ATTACH QUOTA") \
+    MR_MACROS(ATTACH_ROLE, "ATTACH ROLE") \
+    MR_MACROS(ATTACH_ROW_POLICY, "ATTACH ROW POLICY") \
+    MR_MACROS(ATTACH_SETTINGS_PROFILE, "ATTACH SETTINGS PROFILE") \
+    MR_MACROS(ATTACH_USER, "ATTACH USER") \
+    MR_MACROS(ATTACH, "ATTACH") \
+    MR_MACROS(AZURE, "AZURE") \
+    MR_MACROS(BACKUP, "BACKUP") \
+    MR_MACROS(BAGEXPANSION, "bagexpansion") \
+    MR_MACROS(BEGIN_TRANSACTION, "BEGIN TRANSACTION") \
+    MR_MACROS(BETWEEN, "BETWEEN") \
+    MR_MACROS(BIDIRECTIONAL, "BIDIRECTIONAL") \
+    MR_MACROS(BOTH, "BOTH") \
+    MR_MACROS(BY, "BY") \
+    MR_MACROS(CASCADE, "CASCADE") \
+    MR_MACROS(CASE, "CASE") \
+    MR_MACROS(CAST, "CAST") \
+    MR_MACROS(CHANGE, "CHANGE") \
+    MR_MACROS(CHANGED, "CHANGED") \
+    MR_MACROS(CHAR_VARYING, "CHAR VARYING") \
+    MR_MACROS(CHAR, "CHAR") \
+    MR_MACROS(CHARACTER_LARGE_OBJECT, "CHARACTER LARGE OBJECT") \
+    MR_MACROS(CHARACTER_VARYING, "CHARACTER VARYING") \
+    MR_MACROS(CHARACTER, "CHARACTER") \
+    MR_MACROS(CHECK_ALL_TABLES, "CHECK ALL TABLES") \
+    MR_MACROS(CHECK_TABLE, "CHECK TABLE") \
+    MR_MACROS(CHECK, "CHECK") \
+    MR_MACROS(CLEANUP, "CLEANUP") \
+    MR_MACROS(CLEAR_COLUMN, "CLEAR COLUMN") \
+    MR_MACROS(CLEAR_INDEX, "CLEAR INDEX") \
+    MR_MACROS(CLEAR_PROJECTION, "CLEAR PROJECTION") \
+    MR_MACROS(CLEAR_STATISTIC, "CLEAR STATISTIC") \
+    MR_MACROS(CLUSTER, "CLUSTER") \
+    MR_MACROS(CLUSTERS, "CLUSTERS") \
+    MR_MACROS(CN, "CN") \
+    MR_MACROS(CODEC, "CODEC") \
+    MR_MACROS(COLLATE, "COLLATE") \
+    MR_MACROS(COLUMN, "COLUMN") \
+    MR_MACROS(COLUMNS, "COLUMNS") \
+    MR_MACROS(COMMENT_COLUMN, "COMMENT COLUMN") \
+    MR_MACROS(COMMENT, "COMMENT") \
+    MR_MACROS(COMMIT, "COMMIT") \
+    MR_MACROS(COMPRESSION, "COMPRESSION") \
+    MR_MACROS(CONST, "CONST") \
+    MR_MACROS(CONSTRAINT, "CONSTRAINT") \
+    MR_MACROS(CREATE_POLICY, "CREATE POLICY") \
+    MR_MACROS(CREATE_PROFILE, "CREATE PROFILE") \
+    MR_MACROS(CREATE_QUOTA, "CREATE QUOTA") \
+    MR_MACROS(CREATE_ROLE, "CREATE ROLE") \
+    MR_MACROS(CREATE_ROW_POLICY, "CREATE ROW POLICY") \
+    MR_MACROS(CREATE_SETTINGS_PROFILE, "CREATE SETTINGS PROFILE") \
+    MR_MACROS(CREATE_TABLE, "CREATE TABLE") \
+    MR_MACROS(CREATE_TEMPORARY_TABLE, "CREATE TEMPORARY TABLE") \
+    MR_MACROS(CREATE_USER, "CREATE USER") \
+    MR_MACROS(CREATE, "CREATE") \
+    MR_MACROS(CROSS, "CROSS") \
+    MR_MACROS(CUBE, "CUBE") \
+    MR_MACROS(CURRENT_GRANTS, "CURRENT GRANTS") \
+    MR_MACROS(CURRENT_QUOTA, "CURRENT QUOTA") \
+    MR_MACROS(CURRENT_ROLES, "CURRENT ROLES") \
+    MR_MACROS(CURRENT_ROW, "CURRENT ROW") \
+    MR_MACROS(CURRENT_TRANSACTION, "CURRENT TRANSACTION") \
+    MR_MACROS(CURRENTUSER, "CURRENTUSER") \
+    MR_MACROS(D, "D") \
+    MR_MACROS(DATABASE, "DATABASE") \
+    MR_MACROS(DATABASES, "DATABASES") \
+    MR_MACROS(DATE, "DATE") \
+    MR_MACROS(DAY, "DAY") \
+    MR_MACROS(DAYS, "DAYS") \
+    MR_MACROS(DD, "DD") \
+    MR_MACROS(DEDUPLICATE, "DEDUPLICATE") \
+    MR_MACROS(DEFAULT_DATABASE, "DEFAULT DATABASE") \
+    MR_MACROS(DEFAULT_ROLE, "DEFAULT ROLE") \
+    MR_MACROS(DEFAULT, "DEFAULT") \
+    MR_MACROS(DEFINER, "DEFINER") \
+    MR_MACROS(DELETE, "DELETE") \
+    MR_MACROS(DEPENDS_ON, "DEPENDS ON") \
+    MR_MACROS(DESC, "DESC") \
+    MR_MACROS(DESCENDING, "DESCENDING") \
+    MR_MACROS(DESCRIBE, "DESCRIBE") \
+    MR_MACROS(DETACH_PART, "DETACH PART") \
+    MR_MACROS(DETACH_PARTITION, "DETACH PARTITION") \
+    MR_MACROS(DETACH, "DETACH") \
+    MR_MACROS(DICTIONARIES, "DICTIONARIES") \
+    MR_MACROS(DICTIONARY, "DICTIONARY") \
+    MR_MACROS(DISK, "DISK") \
+    MR_MACROS(DISTINCT_ON, "DISTINCT ON") \
+    MR_MACROS(DISTINCT, "DISTINCT") \
+    MR_MACROS(DIV, "DIV") \
+    MR_MACROS(DROP_COLUMN, "DROP COLUMN") \
+    MR_MACROS(DROP_CONSTRAINT, "DROP CONSTRAINT") \
+    MR_MACROS(DROP_DEFAULT, "DROP DEFAULT") \
+    MR_MACROS(DROP_DETACHED_PART, "DROP DETACHED PART") \
+    MR_MACROS(DROP_DETACHED_PARTITION, "DROP DETACHED PARTITION") \
+    MR_MACROS(DROP_INDEX, "DROP INDEX") \
+    MR_MACROS(DROP_PART, "DROP PART") \
+    MR_MACROS(DROP_PARTITION, "DROP PARTITION") \
+    MR_MACROS(DROP_PROJECTION, "DROP PROJECTION") \
+    MR_MACROS(DROP_STATISTIC, "DROP STATISTIC") \
+    MR_MACROS(DROP_TABLE, "DROP TABLE") \
+    MR_MACROS(DROP_TEMPORARY_TABLE, "DROP TEMPORARY TABLE") \
+    MR_MACROS(DROP, "DROP") \
+    MR_MACROS(ELSE, "ELSE") \
+    MR_MACROS(EMPTY_AS, "EMPTY AS") \
+    MR_MACROS(EMPTY, "EMPTY") \
+    MR_MACROS(ENABLED_ROLES, "ENABLED ROLES") \
+    MR_MACROS(END, "END") \
+    MR_MACROS(ENFORCED, "ENFORCED") \
+    MR_MACROS(ENGINE, "ENGINE") \
+    MR_MACROS(EPHEMERAL_SEQUENTIAL, "EPHEMERAL SEQUENTIAL") \
+    MR_MACROS(EPHEMERAL, "EPHEMERAL") \
+    MR_MACROS(ESTIMATE, "ESTIMATE") \
+    MR_MACROS(EVENT, "EVENT") \
+    MR_MACROS(EVENTS, "EVENTS") \
+    MR_MACROS(EVERY, "EVERY") \
+    MR_MACROS(EXCEPT_DATABASE, "EXCEPT DATABASE") \
+    MR_MACROS(EXCEPT_DATABASES, "EXCEPT DATABASES") \
+    MR_MACROS(EXCEPT_TABLE, "EXCEPT TABLE") \
+    MR_MACROS(EXCEPT_TABLES, "EXCEPT TABLES") \
+    MR_MACROS(EXCEPT, "EXCEPT") \
+    MR_MACROS(EXCHANGE_DICTIONARIES, "EXCHANGE DICTIONARIES") \
+    MR_MACROS(EXCHANGE_TABLES, "EXCHANGE TABLES") \
+    MR_MACROS(EXISTS, "EXISTS") \
+    MR_MACROS(EXPLAIN, "EXPLAIN") \
+    MR_MACROS(EXPRESSION, "EXPRESSION") \
+    MR_MACROS(EXTENDED, "EXTENDED") \
+    MR_MACROS(EXTERNAL_DDL_FROM, "EXTERNAL DDL FROM") \
+    MR_MACROS(FALSE_KEYWORD, "FALSE") /*The name differs from the value*/ \
+    MR_MACROS(FETCH_PART, "FETCH PART") \
+    MR_MACROS(FETCH_PARTITION, "FETCH PARTITION") \
+    MR_MACROS(FETCH, "FETCH") \
+    MR_MACROS(FIELDS, "FIELDS") \
+    MR_MACROS(FILE, "FILE") \
+    MR_MACROS(FILESYSTEM_CACHE, "FILESYSTEM CACHE") \
+    MR_MACROS(FILESYSTEM_CACHES, "FILESYSTEM CACHES") \
+    MR_MACROS(FILTER, "FILTER") \
+    MR_MACROS(FINAL, "FINAL") \
+    MR_MACROS(FIRST, "FIRST") \
+    MR_MACROS(FOLLOWING, "FOLLOWING") \
+    MR_MACROS(FOR, "FOR") \
+    MR_MACROS(FOREIGN_KEY, "FOREIGN KEY") \
+    MR_MACROS(FOREIGN, "FOREIGN") \
+    MR_MACROS(FORGET_PARTITION, "FORGET PARTITION") \
+    MR_MACROS(FORMAT, "FORMAT") \
+    MR_MACROS(FREEZE, "FREEZE") \
+    MR_MACROS(FROM_INFILE, "FROM INFILE") \
+    MR_MACROS(FROM_SHARD, "FROM SHARD") \
+    MR_MACROS(FROM, "FROM") \
+    MR_MACROS(FULL, "FULL") \
+    MR_MACROS(FULLTEXT, "FULLTEXT") \
+    MR_MACROS(FUNCTION, "FUNCTION") \
+    MR_MACROS(GLOBAL_IN, "GLOBAL IN") \
+    MR_MACROS(GLOBAL_NOT_IN, "GLOBAL NOT IN") \
+    MR_MACROS(GLOBAL, "GLOBAL") \
+    MR_MACROS(GRANT_OPTION_FOR, "GRANT OPTION FOR") \
+    MR_MACROS(GRANT, "GRANT") \
+    MR_MACROS(GRANTEES, "GRANTEES") \
+    MR_MACROS(GRANULARITY, "GRANULARITY") \
+    MR_MACROS(GROUP_BY, "GROUP BY") \
+    MR_MACROS(GROUPING_SETS, "GROUPING SETS") \
+    MR_MACROS(GROUPS, "GROUPS") \
+    MR_MACROS(H, "H") \
+    MR_MACROS(HASH, "HASH") \
+    MR_MACROS(HAVING, "HAVING") \
+    MR_MACROS(HDFS, "HDFS") \
+    MR_MACROS(HH, "HH") \
+    MR_MACROS(HIERARCHICAL, "HIERARCHICAL") \
+    MR_MACROS(HOST, "HOST") \
+    MR_MACROS(HOUR, "HOUR") \
+    MR_MACROS(HOURS, "HOURS") \
+    MR_MACROS(HTTP, "HTTP") \
+    MR_MACROS(ID, "ID") \
+    MR_MACROS(IDENTIFIED, "IDENTIFIED") \
+    MR_MACROS(IF_EMPTY, "IF EMPTY") \
+    MR_MACROS(IF_EXISTS, "IF EXISTS") \
+    MR_MACROS(IF_NOT_EXISTS, "IF NOT EXISTS") \
+    MR_MACROS(IGNORE_NULLS, "IGNORE NULLS") \
+    MR_MACROS(ILIKE, "ILIKE") \
+    MR_MACROS(IN_PARTITION, "IN PARTITION") \
+    MR_MACROS(IN, "IN") \
+    MR_MACROS(INDEX, "INDEX") \
+    MR_MACROS(INDEXES, "INDEXES") \
+    MR_MACROS(INDICES, "INDICES") \
+    MR_MACROS(INHERIT, "INHERIT") \
+    MR_MACROS(INJECTIVE, "INJECTIVE") \
+    MR_MACROS(INNER, "INNER") \
+    MR_MACROS(INSERT_INTO, "INSERT INTO") \
+    MR_MACROS(INTERPOLATE, "INTERPOLATE") \
+    MR_MACROS(INTERSECT, "INTERSECT") \
+    MR_MACROS(INTERVAL, "INTERVAL") \
+    MR_MACROS(INTO_OUTFILE, "INTO OUTFILE") \
+    MR_MACROS(INVISIBLE, "INVISIBLE") \
+    MR_MACROS(INVOKER, "INVOKER") \
+    MR_MACROS(IP, "IP") \
+    MR_MACROS(IS_NOT_DISTINCT_FROM, "IS NOT DISTINCT FROM") \
+    MR_MACROS(IS_NOT_NULL, "IS NOT NULL") \
+    MR_MACROS(IS_NULL, "IS NULL") \
+    MR_MACROS(JOIN, "JOIN") \
+    MR_MACROS(KERBEROS, "KERBEROS") \
+    MR_MACROS(KEY_BY, "KEY BY") \
+    MR_MACROS(KEY, "KEY") \
+    MR_MACROS(KEYED_BY, "KEYED BY") \
+    MR_MACROS(KEYS, "KEYS") \
+    MR_MACROS(KILL, "KILL") \
+    MR_MACROS(KIND, "KIND") \
+    MR_MACROS(LARGE_OBJECT, "LARGE OBJECT") \
+    MR_MACROS(LAST, "LAST") \
+    MR_MACROS(LAYOUT, "LAYOUT") \
+    MR_MACROS(LDAP, "LDAP") \
+    MR_MACROS(LEADING, "LEADING") \
+    MR_MACROS(LEFT_ARRAY_JOIN, "LEFT ARRAY JOIN") \
+    MR_MACROS(LEFT, "LEFT") \
+    MR_MACROS(LESS_THAN, "LESS THAN") \
+    MR_MACROS(LEVEL, "LEVEL") \
+    MR_MACROS(LIFETIME, "LIFETIME") \
+    MR_MACROS(LIGHTWEIGHT, "LIGHTWEIGHT") \
+    MR_MACROS(LIKE, "LIKE") \
+    MR_MACROS(LIMIT, "LIMIT") \
+    MR_MACROS(LINEAR, "LINEAR") \
+    MR_MACROS(LIST, "LIST") \
+    MR_MACROS(LIVE, "LIVE") \
+    MR_MACROS(LOCAL, "LOCAL") \
+    MR_MACROS(M, "M") \
+    MR_MACROS(MATCH, "MATCH") \
+    MR_MACROS(MATERIALIZE_COLUMN, "MATERIALIZE COLUMN") \
+    MR_MACROS(MATERIALIZE_INDEX, "MATERIALIZE INDEX") \
+    MR_MACROS(MATERIALIZE_PROJECTION, "MATERIALIZE PROJECTION") \
+    MR_MACROS(MATERIALIZE_STATISTIC, "MATERIALIZE STATISTIC") \
+    MR_MACROS(MATERIALIZE_TTL, "MATERIALIZE TTL") \
+    MR_MACROS(MATERIALIZE, "MATERIALIZE") \
+    MR_MACROS(MATERIALIZED, "MATERIALIZED") \
+    MR_MACROS(MAX, "MAX") \
+    MR_MACROS(MCS, "MCS") \
+    MR_MACROS(MEMORY, "MEMORY") \
+    MR_MACROS(MERGES, "MERGES") \
+    MR_MACROS(MI, "MI") \
+    MR_MACROS(MICROSECOND, "MICROSECOND") \
+    MR_MACROS(MICROSECONDS, "MICROSECONDS") \
+    MR_MACROS(MILLISECOND, "MILLISECOND") \
+    MR_MACROS(MILLISECONDS, "MILLISECONDS") \
+    MR_MACROS(MIN, "MIN") \
+    MR_MACROS(MINUTE, "MINUTE") \
+    MR_MACROS(MINUTES, "MINUTES") \
+    MR_MACROS(MM, "MM") \
+    MR_MACROS(MOD, "MOD") \
+    MR_MACROS(MODIFY_COLUMN, "MODIFY COLUMN") \
+    MR_MACROS(MODIFY_COMMENT, "MODIFY COMMENT") \
+    MR_MACROS(MODIFY_ORDER_BY, "MODIFY ORDER BY") \
+    MR_MACROS(MODIFY_QUERY, "MODIFY QUERY") \
+    MR_MACROS(MODIFY_REFRESH, "MODIFY REFRESH") \
+    MR_MACROS(MODIFY_SAMPLE_BY, "MODIFY SAMPLE BY") \
+    MR_MACROS(MODIFY_SETTING, "MODIFY SETTING") \
+    MR_MACROS(MODIFY_SQL_SECURITY, "MODIFY SQL SECURITY") \
+    MR_MACROS(MODIFY_TTL, "MODIFY TTL") \
+    MR_MACROS(MODIFY, "MODIFY") \
+    MR_MACROS(MONTH, "MONTH") \
+    MR_MACROS(MONTHS, "MONTHS") \
+    MR_MACROS(MOVE_PART, "MOVE PART") \
+    MR_MACROS(MOVE_PARTITION, "MOVE PARTITION") \
+    MR_MACROS(MOVE, "MOVE") \
+    MR_MACROS(MS, "MS") \
+    MR_MACROS(MUTATION, "MUTATION") \
+    MR_MACROS(N, "N") \
+    MR_MACROS(NAME, "NAME") \
+    MR_MACROS(NAMED_COLLECTION, "NAMED COLLECTION") \
+    MR_MACROS(NANOSECOND, "NANOSECOND") \
+    MR_MACROS(NANOSECONDS, "NANOSECONDS") \
+    MR_MACROS(NEXT, "NEXT") \
+    MR_MACROS(NO_ACTION, "NO ACTION") \
+    MR_MACROS(NO_DELAY, "NO DELAY") \
+    MR_MACROS(NO_LIMITS, "NO LIMITS") \
+    MR_MACROS(NONE, "NONE") \
+    MR_MACROS(NOT_BETWEEN, "NOT BETWEEN") \
+    MR_MACROS(NOT_IDENTIFIED, "NOT IDENTIFIED") \
+    MR_MACROS(NOT_ILIKE, "NOT ILIKE") \
+    MR_MACROS(NOT_IN, "NOT IN") \
+    MR_MACROS(NOT_KEYED, "NOT KEYED") \
+    MR_MACROS(NOT_LIKE, "NOT LIKE") \
+    MR_MACROS(NOT_OVERRIDABLE, "NOT OVERRIDABLE") \
+    MR_MACROS(NOT, "NOT") \
+    MR_MACROS(NS, "NS") \
+    MR_MACROS(NULL_KEYWORD, "NULL") \
+    MR_MACROS(NULLS, "NULLS") \
+    MR_MACROS(OFFSET, "OFFSET") \
+    MR_MACROS(ON_DELETE, "ON DELETE") \
+    MR_MACROS(ON_UPDATE, "ON UPDATE") \
+    MR_MACROS(ON_VOLUME, "ON VOLUME") \
+    MR_MACROS(ON, "ON") \
+    MR_MACROS(ONLY, "ONLY") \
+    MR_MACROS(OPTIMIZE_TABLE, "OPTIMIZE TABLE") \
+    MR_MACROS(OR_REPLACE, "OR REPLACE") \
+    MR_MACROS(OR, "OR") \
+    MR_MACROS(ORDER_BY, "ORDER BY") \
+    MR_MACROS(OUTER, "OUTER") \
+    MR_MACROS(OVER, "OVER") \
+    MR_MACROS(OVERRIDABLE, "OVERRIDABLE") \
+    MR_MACROS(PART, "PART") \
+    MR_MACROS(PARTIAL, "PARTIAL") \
+    MR_MACROS(PARTITION_BY, "PARTITION BY") \
+    MR_MACROS(PARTITION, "PARTITION") \
+    MR_MACROS(PARTITIONS, "PARTITIONS") \
+    MR_MACROS(PASTE, "PASTE") \
+    MR_MACROS(PERIODIC_REFRESH, "PERIODIC REFRESH") \
+    MR_MACROS(PERMANENTLY, "PERMANENTLY") \
+    MR_MACROS(PERMISSIVE, "PERMISSIVE") \
+    MR_MACROS(PERSISTENT_SEQUENTIAL, "PERSISTENT SEQUENTIAL") \
+    MR_MACROS(PERSISTENT, "PERSISTENT") \
+    MR_MACROS(PIPELINE, "PIPELINE") \
+    MR_MACROS(PLAN, "PLAN") \
+    MR_MACROS(POPULATE, "POPULATE") \
+    MR_MACROS(PRECEDING, "PRECEDING") \
+    MR_MACROS(PRECISION, "PRECISION") \
+    MR_MACROS(PREWHERE, "PREWHERE") \
+    MR_MACROS(PRIMARY_KEY, "PRIMARY KEY") \
+    MR_MACROS(PRIMARY, "PRIMARY") \
+    MR_MACROS(PROFILE, "PROFILE") \
+    MR_MACROS(PROJECTION, "PROJECTION") \
+    MR_MACROS(PROTOBUF, "Protobuf") \
+    MR_MACROS(PULL, "PULL") \
+    MR_MACROS(Q, "Q") \
+    MR_MACROS(QQ, "QQ") \
+    MR_MACROS(QUARTER, "QUARTER") \
+    MR_MACROS(QUARTERS, "QUARTERS") \
+    MR_MACROS(QUERY_TREE, "QUERY TREE") \
+    MR_MACROS(QUERY, "QUERY") \
+    MR_MACROS(QUOTA, "QUOTA") \
+    MR_MACROS(RANDOMIZE_FOR, "RANDOMIZE FOR") \
+    MR_MACROS(RANDOMIZED, "RANDOMIZED") \
+    MR_MACROS(RANGE, "RANGE") \
+    MR_MACROS(READONLY, "READONLY") \
+    MR_MACROS(REALM, "REALM") \
+    MR_MACROS(RECOMPRESS, "RECOMPRESS") \
+    MR_MACROS(REFERENCES, "REFERENCES") \
+    MR_MACROS(REFRESH, "REFRESH") \
+    MR_MACROS(REGEXP, "REGEXP") \
+    MR_MACROS(REMOVE_SAMPLE_BY, "REMOVE SAMPLE BY") \
+    MR_MACROS(REMOVE_TTL, "REMOVE TTL") \
+    MR_MACROS(REMOVE, "REMOVE") \
+    MR_MACROS(RENAME_COLUMN, "RENAME COLUMN") \
+    MR_MACROS(RENAME_DATABASE, "RENAME DATABASE") \
+    MR_MACROS(RENAME_DICTIONARY, "RENAME DICTIONARY") \
+    MR_MACROS(RENAME_TABLE, "RENAME TABLE") \
+    MR_MACROS(RENAME_TO, "RENAME TO") \
+    MR_MACROS(RENAME, "RENAME") \
+    MR_MACROS(REPLACE_PARTITION, "REPLACE PARTITION") \
+    MR_MACROS(REPLACE, "REPLACE") \
+    MR_MACROS(RESET_SETTING, "RESET SETTING") \
+    MR_MACROS(RESPECT_NULLS, "RESPECT NULLS") \
+    MR_MACROS(RESTORE, "RESTORE") \
+    MR_MACROS(RESTRICT, "RESTRICT") \
+    MR_MACROS(RESTRICTIVE, "RESTRICTIVE") \
+    MR_MACROS(RESUME, "RESUME") \
+    MR_MACROS(REVOKE, "REVOKE") \
+    MR_MACROS(RIGHT, "RIGHT") \
+    MR_MACROS(ROLLBACK, "ROLLBACK") \
+    MR_MACROS(ROLLUP, "ROLLUP") \
+    MR_MACROS(ROW, "ROW") \
+    MR_MACROS(ROWS, "ROWS") \
+    MR_MACROS(S, "S") \
+    MR_MACROS(S3, "S3") \
+    MR_MACROS(SALT, "SALT") \
+    MR_MACROS(SAMPLE_BY, "SAMPLE BY") \
+    MR_MACROS(SAMPLE, "SAMPLE") \
+    MR_MACROS(SCHEME, "SCHEME") \
+    MR_MACROS(SECOND, "SECOND") \
+    MR_MACROS(SECONDS, "SECONDS") \
+    MR_MACROS(SELECT, "SELECT") \
+    MR_MACROS(SEMI, "SEMI") \
+    MR_MACROS(SERVER, "SERVER") \
+    MR_MACROS(SET_DEFAULT_ROLE, "SET DEFAULT ROLE") \
+    MR_MACROS(SET_DEFAULT, "SET DEFAULT") \
+    MR_MACROS(SET_FAKE_TIME, "SET FAKE TIME") \
+    MR_MACROS(SET_NULL, "SET NULL") \
+    MR_MACROS(SET_ROLE_DEFAULT, "SET ROLE DEFAULT") \
+    MR_MACROS(SET_ROLE, "SET ROLE") \
+    MR_MACROS(SET_TRANSACTION_SNAPSHOT, "SET TRANSACTION SNAPSHOT") \
+    MR_MACROS(SET, "SET") \
+    MR_MACROS(SETTINGS, "SETTINGS") \
+    MR_MACROS(SHOW_ACCESS, "SHOW ACCESS") \
+    MR_MACROS(SHOW_CREATE, "SHOW CREATE") \
+    MR_MACROS(SHOW_ENGINES, "SHOW ENGINES") \
+    MR_MACROS(SHOW_FUNCTIONS, "SHOW FUNCTIONS") \
+    MR_MACROS(SHOW_GRANTS, "SHOW GRANTS") \
+    MR_MACROS(SHOW_PRIVILEGES, "SHOW PRIVILEGES") \
+    MR_MACROS(SHOW_PROCESSLIST, "SHOW PROCESSLIST") \
+    MR_MACROS(SHOW_SETTING, "SHOW SETTING") \
+    MR_MACROS(SHOW, "SHOW") \
+    MR_MACROS(SIGNED, "SIGNED") \
+    MR_MACROS(SIMPLE, "SIMPLE") \
+    MR_MACROS(SOURCE, "SOURCE") \
+    MR_MACROS(SPATIAL, "SPATIAL") \
+    MR_MACROS(SQL_SECURITY, "SQL SECURITY") \
+    MR_MACROS(SS, "SS") \
+    MR_MACROS(STATISTIC, "STATISTIC") \
+    MR_MACROS(STEP, "STEP") \
+    MR_MACROS(STORAGE, "STORAGE") \
+    MR_MACROS(STRICT, "STRICT") \
+    MR_MACROS(SUBPARTITION_BY, "SUBPARTITION BY") \
+    MR_MACROS(SUBPARTITION, "SUBPARTITION") \
+    MR_MACROS(SUBPARTITIONS, "SUBPARTITIONS") \
+    MR_MACROS(SUSPEND, "SUSPEND") \
+    MR_MACROS(SYNC, "SYNC") \
+    MR_MACROS(SYNTAX, "SYNTAX") \
+    MR_MACROS(SYSTEM, "SYSTEM") \
+    MR_MACROS(TABLE_OVERRIDE, "TABLE OVERRIDE") \
+    MR_MACROS(TABLE, "TABLE") \
+    MR_MACROS(TABLES, "TABLES") \
+    MR_MACROS(TEMPORARY_TABLE, "TEMPORARY TABLE") \
+    MR_MACROS(TEMPORARY, "TEMPORARY") \
+    MR_MACROS(TEST, "TEST") \
+    MR_MACROS(THEN, "THEN") \
+    MR_MACROS(TIMESTAMP, "TIMESTAMP") \
+    MR_MACROS(TO_DISK, "TO DISK") \
+    MR_MACROS(TO_INNER_UUID, "TO INNER UUID") \
+    MR_MACROS(TO_SHARD, "TO SHARD") \
+    MR_MACROS(TO_TABLE, "TO TABLE") \
+    MR_MACROS(TO_VOLUME, "TO VOLUME") \
+    MR_MACROS(TO, "TO") \
+    MR_MACROS(TOP, "TOP") \
+    MR_MACROS(TOTALS, "TOTALS") \
+    MR_MACROS(TRACKING_ONLY, "TRACKING ONLY") \
+    MR_MACROS(TRAILING, "TRAILING") \
+    MR_MACROS(TRANSACTION, "TRANSACTION") \
+    MR_MACROS(TRIGGER, "TRIGGER") \
+    MR_MACROS(TRUE_KEYWORD, "TRUE") /*The name differs from the value*/ \
+    MR_MACROS(TRUNCATE, "TRUNCATE") \
+    MR_MACROS(TTL, "TTL") \
+    MR_MACROS(TYPE, "TYPE") \
+    MR_MACROS(TYPEOF, "TYPEOF") \
+    MR_MACROS(UNBOUNDED, "UNBOUNDED") \
+    MR_MACROS(UNDROP, "UNDROP") \
+    MR_MACROS(UNFREEZE, "UNFREEZE") \
+    MR_MACROS(UNION, "UNION") \
+    MR_MACROS(UNIQUE, "UNIQUE") \
+    MR_MACROS(UNSET_FAKE_TIME, "UNSET FAKE TIME") \
+    MR_MACROS(UNSIGNED, "UNSIGNED") \
+    MR_MACROS(UPDATE, "UPDATE") \
+    MR_MACROS(URL, "URL") \
+    MR_MACROS(USE, "USE") \
+    MR_MACROS(USING, "USING") \
+    MR_MACROS(UUID, "UUID") \
+    MR_MACROS(VALID_UNTIL, "VALID UNTIL") \
+    MR_MACROS(VALUES, "VALUES") \
+    MR_MACROS(VARYING, "VARYING") \
+    MR_MACROS(VIEW, "VIEW") \
+    MR_MACROS(VISIBLE, "VISIBLE") \
+    MR_MACROS(WATCH, "WATCH") \
+    MR_MACROS(WATERMARK, "WATERMARK") \
+    MR_MACROS(WEEK, "WEEK") \
+    MR_MACROS(WEEKS, "WEEKS") \
+    MR_MACROS(WHEN, "WHEN") \
+    MR_MACROS(WHERE, "WHERE") \
+    MR_MACROS(WINDOW, "WINDOW") \
+    MR_MACROS(WITH_ADMIN_OPTION, "WITH ADMIN OPTION") \
+    MR_MACROS(WITH_CHECK, "WITH CHECK") \
+    MR_MACROS(WITH_FILL, "WITH FILL") \
+    MR_MACROS(WITH_GRANT_OPTION, "WITH GRANT OPTION") \
+    MR_MACROS(WITH_NAME, "WITH NAME") \
+    MR_MACROS(WITH_REPLACE_OPTION, "WITH REPLACE OPTION") \
+    MR_MACROS(WITH_TIES, "WITH TIES") \
+    MR_MACROS(WITH, "WITH") \
+    MR_MACROS(WK, "WK") \
+    MR_MACROS(WRITABLE, "WRITABLE") \
+    MR_MACROS(WW, "WW") \
+    MR_MACROS(YEAR, "YEAR") \
+    MR_MACROS(YEARS, "YEARS") \
+    MR_MACROS(YY, "YY") \
+    MR_MACROS(YYYY, "YYYY") \
+    MR_MACROS(ZKPATH, "ZKPATH") \
+
+/// The list of keywords where underscore is intentional
+#define APPLY_FOR_PARSER_KEYWORDS_WITH_UNDERSCORES(MR_MACROS) \
+    MR_MACROS(ALLOWED_LATENESS, "ALLOWED_LATENESS") \
+    MR_MACROS(AUTO_INCREMENT, "AUTO_INCREMENT") \
+    MR_MACROS(BASE_BACKUP, "base_backup") \
+    MR_MACROS(BCRYPT_HASH, "BCRYPT_HASH") \
+    MR_MACROS(BCRYPT_PASSWORD, "BCRYPT_PASSWORD") \
+    MR_MACROS(CHANGEABLE_IN_READONLY, "CHANGEABLE_IN_READONLY") \
+    MR_MACROS(CLUSTER_HOST_IDS, "cluster_host_ids") \
+    MR_MACROS(CURRENT_USER, "CURRENT_USER") \
+    MR_MACROS(DOUBLE_SHA1_HASH, "DOUBLE_SHA1_HASH") \
+    MR_MACROS(DOUBLE_SHA1_PASSWORD, "DOUBLE_SHA1_PASSWORD") \
+    MR_MACROS(IS_OBJECT_ID, "IS_OBJECT_ID") \
+    MR_MACROS(NO_PASSWORD, "NO_PASSWORD") \
+    MR_MACROS(PART_MOVE_TO_SHARD, "PART_MOVE_TO_SHARD") \
+    MR_MACROS(PLAINTEXT_PASSWORD, "PLAINTEXT_PASSWORD") \
+    MR_MACROS(SHA256_HASH, "SHA256_HASH") \
+    MR_MACROS(SHA256_PASSWORD, "SHA256_PASSWORD") \
+    MR_MACROS(SQL_TSI_DAY, "SQL_TSI_DAY") \
+    MR_MACROS(SQL_TSI_HOUR, "SQL_TSI_HOUR") \
+    MR_MACROS(SQL_TSI_MICROSECOND, "SQL_TSI_MICROSECOND") \
+    MR_MACROS(SQL_TSI_MILLISECOND, "SQL_TSI_MILLISECOND") \
+    MR_MACROS(SQL_TSI_MINUTE, "SQL_TSI_MINUTE") \
+    MR_MACROS(SQL_TSI_MONTH, "SQL_TSI_MONTH") \
+    MR_MACROS(SQL_TSI_NANOSECOND, "SQL_TSI_NANOSECOND") \
+    MR_MACROS(SQL_TSI_QUARTER, "SQL_TSI_QUARTER") \
+    MR_MACROS(SQL_TSI_SECOND, "SQL_TSI_SECOND") \
+    MR_MACROS(SQL_TSI_WEEK, "SQL_TSI_WEEK") \
+    MR_MACROS(SQL_TSI_YEAR, "SQL_TSI_YEAR") \
+    MR_MACROS(SSH_KEY, "SSH_KEY") \
+    MR_MACROS(SSL_CERTIFICATE, "SSL_CERTIFICATE") \
+    MR_MACROS(STRICTLY_ASCENDING, "STRICTLY_ASCENDING") \
+    MR_MACROS(WITH_ITEMINDEX, "with_itemindex") \
+
+enum class Keyword : size_t
+{
+#define DECLARE_PARSER_KEYWORD_ENUM(identifier, name) \
+    identifier,
+
+    APPLY_FOR_PARSER_KEYWORDS(DECLARE_PARSER_KEYWORD_ENUM)
+    APPLY_FOR_PARSER_KEYWORDS_WITH_UNDERSCORES(DECLARE_PARSER_KEYWORD_ENUM)
+#undef DECLARE_PARSER_KEYWORD_ENUM
+};
+
+
+std::string_view toStringView(Keyword type);
+
+const std::vector<String> & getAllKeyWords();
+
+
 /** Parse specified keyword such as SELECT or compound keyword such as ORDER BY.
   * All case insensitive. Requires word boundary.
   * For compound keywords, any whitespace characters and comments could be in the middle.
@@ -17,9 +582,20 @@ class ParserKeyword : public IParserBase
 private:
     std::string_view s;
 
+    explicit ParserKeyword(std::string_view s_): s(s_) { assert(!s.empty()); }
+
 public:
-    //NOLINTNEXTLINE Want to be able to init ParserKeyword("literal")
-    constexpr ParserKeyword(std::string_view s_): s(s_) { assert(!s.empty()); }
+    static ParserKeyword createDeprecated(std::string_view s_)
+    {
+        return ParserKeyword(s_);
+    }
+
+    static std::shared_ptr<ParserKeyword> createDeprecatedPtr(std::string_view s_)
+    {
+        return std::shared_ptr<ParserKeyword>(new ParserKeyword(s_));
+    }
+
+    explicit ParserKeyword(Keyword keyword);
 
     constexpr const char * getName() const override { return s.data(); }
 
diff --git a/src/Parsers/ExpressionElementParsers.cpp b/src/Parsers/ExpressionElementParsers.cpp
index 486555ae86d..1135fe88c16 100644
--- a/src/Parsers/ExpressionElementParsers.cpp
+++ b/src/Parsers/ExpressionElementParsers.cpp
@@ -8,6 +8,7 @@
 #include <Common/typeid_cast.h>
 #include <Common/StringUtils/StringUtils.h>
 #include <Common/BinStringDecodeHelper.h>
+#include "Parsers/CommonParsers.h"
 
 #include <Parsers/DumpASTNode.h>
 #include <Parsers/ASTAsterisk.h>
@@ -290,7 +291,7 @@ bool ParserCompoundIdentifier::parseImpl(Pos & pos, ASTPtr & node, Expected & ex
             params.push_back(child->as<ASTIdentifier>()->getParam());
     }
 
-    ParserKeyword s_uuid("UUID");
+    ParserKeyword s_uuid(Keyword::UUID);
     UUID uuid = UUIDHelpers::Nil;
 
     if (table_name_with_optional_uuid)
@@ -337,7 +338,7 @@ bool ParserFilterClause::parseImpl(Pos & pos, ASTPtr & node, Expected & expected
         return false;
     }
 
-    ParserKeyword parser_where("WHERE");
+    ParserKeyword parser_where(Keyword::WHERE);
     if (!parser_where.ignore(pos, expected))
     {
         return false;
@@ -391,9 +392,9 @@ bool ParserWindowReference::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
 static bool tryParseFrameDefinition(ASTWindowDefinition * node, IParser::Pos & pos,
     Expected & expected)
 {
-    ParserKeyword keyword_rows("ROWS");
-    ParserKeyword keyword_groups("GROUPS");
-    ParserKeyword keyword_range("RANGE");
+    ParserKeyword keyword_rows(Keyword::ROWS);
+    ParserKeyword keyword_groups(Keyword::GROUPS);
+    ParserKeyword keyword_range(Keyword::RANGE);
 
     node->frame_is_default = false;
     if (keyword_rows.ignore(pos, expected))
@@ -415,12 +416,12 @@ static bool tryParseFrameDefinition(ASTWindowDefinition * node, IParser::Pos & p
         return true;
     }
 
-    ParserKeyword keyword_between("BETWEEN");
-    ParserKeyword keyword_unbounded("UNBOUNDED");
-    ParserKeyword keyword_preceding("PRECEDING");
-    ParserKeyword keyword_following("FOLLOWING");
-    ParserKeyword keyword_and("AND");
-    ParserKeyword keyword_current_row("CURRENT ROW");
+    ParserKeyword keyword_between(Keyword::BETWEEN);
+    ParserKeyword keyword_unbounded(Keyword::UNBOUNDED);
+    ParserKeyword keyword_preceding(Keyword::PRECEDING);
+    ParserKeyword keyword_following(Keyword::FOLLOWING);
+    ParserKeyword keyword_and(Keyword::AND);
+    ParserKeyword keyword_current_row(Keyword::CURRENT_ROW);
 
     // There are two variants of grammar for the frame:
     // 1) ROWS BETWEEN UNBOUNDED PRECEDING AND CURRENT ROW
@@ -524,10 +525,10 @@ static bool tryParseFrameDefinition(ASTWindowDefinition * node, IParser::Pos & p
 static bool parseWindowDefinitionParts(IParser::Pos & pos,
     ASTWindowDefinition & node, Expected & expected)
 {
-    ParserKeyword keyword_partition_by("PARTITION BY");
+    ParserKeyword keyword_partition_by(Keyword::PARTITION_BY);
     ParserNotEmptyExpressionList columns_partition_by(
         false /* we don't allow declaring aliases here*/);
-    ParserKeyword keyword_order_by("ORDER BY");
+    ParserKeyword keyword_order_by(Keyword::ORDER_BY);
     ParserOrderByExpressionList columns_order_by;
 
     if (keyword_partition_by.ignore(pos, expected))
@@ -626,7 +627,7 @@ bool ParserWindowList::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
         }
         elem->name = getIdentifierName(window_name_identifier);
 
-        ParserKeyword keyword_as("AS");
+        ParserKeyword keyword_as(Keyword::AS);
         if (!keyword_as.ignore(pos, expected))
         {
             return false;
@@ -854,7 +855,7 @@ bool ParserCastOperator::parseImpl(Pos & pos, ASTPtr & node, Expected & expected
 
 bool ParserNull::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword nested_parser("NULL");
+    ParserKeyword nested_parser(Keyword::NULL_KEYWORD);
     if (nested_parser.parse(pos, node, expected))
     {
         node = std::make_shared<ASTLiteral>(Null());
@@ -867,12 +868,12 @@ bool ParserNull::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 
 bool ParserBool::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    if (ParserKeyword("true").parse(pos, node, expected))
+    if (ParserKeyword(Keyword::TRUE_KEYWORD).parse(pos, node, expected))
     {
         node = std::make_shared<ASTLiteral>(true);
         return true;
     }
-    else if (ParserKeyword("false").parse(pos, node, expected))
+    else if (ParserKeyword(Keyword::FALSE_KEYWORD).parse(pos, node, expected))
     {
         node = std::make_shared<ASTLiteral>(false);
         return true;
@@ -1488,7 +1489,7 @@ const char * ParserAlias::restricted_keywords[] =
 
 bool ParserAlias::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_as("AS");
+    ParserKeyword s_as(Keyword::AS);
     ParserIdentifier id_p;
 
     bool has_as_word = s_as.ignore(pos, expected);
@@ -1519,11 +1520,11 @@ bool ParserAlias::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 
 bool ParserColumnsTransformers::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword apply("APPLY");
-    ParserKeyword except("EXCEPT");
-    ParserKeyword replace("REPLACE");
-    ParserKeyword as("AS");
-    ParserKeyword strict("STRICT");
+    ParserKeyword apply(Keyword::APPLY);
+    ParserKeyword except(Keyword::EXCEPT);
+    ParserKeyword replace(Keyword::REPLACE);
+    ParserKeyword as(Keyword::AS);
+    ParserKeyword strict(Keyword::STRICT);
 
     if (allowed_transformers.isSet(ColumnTransformer::APPLY) && apply.ignore(pos, expected))
     {
@@ -1847,7 +1848,7 @@ static bool parseColumnsMatcherBody(IParser::Pos & pos, ASTPtr & node, Expected
 
 bool ParserColumnsMatcher::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword columns("COLUMNS");
+    ParserKeyword columns(Keyword::COLUMNS);
 
     if (!columns.ignore(pos, expected))
         return false;
@@ -2052,18 +2053,18 @@ bool ParserWithOptionalAlias::parseImpl(Pos & pos, ASTPtr & node, Expected & exp
 bool ParserOrderByElement::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
     ParserExpressionWithOptionalAlias elem_p(false);
-    ParserKeyword ascending("ASCENDING");
-    ParserKeyword descending("DESCENDING");
-    ParserKeyword asc("ASC");
-    ParserKeyword desc("DESC");
-    ParserKeyword nulls("NULLS");
-    ParserKeyword first("FIRST");
-    ParserKeyword last("LAST");
-    ParserKeyword collate("COLLATE");
-    ParserKeyword with_fill("WITH FILL");
-    ParserKeyword from("FROM");
-    ParserKeyword to("TO");
-    ParserKeyword step("STEP");
+    ParserKeyword ascending(Keyword::ASCENDING);
+    ParserKeyword descending(Keyword::DESCENDING);
+    ParserKeyword asc(Keyword::ASC);
+    ParserKeyword desc(Keyword::DESC);
+    ParserKeyword nulls(Keyword::NULLS);
+    ParserKeyword first(Keyword::FIRST);
+    ParserKeyword last(Keyword::LAST);
+    ParserKeyword collate(Keyword::COLLATE);
+    ParserKeyword with_fill(Keyword::WITH_FILL);
+    ParserKeyword from(Keyword::FROM);
+    ParserKeyword to(Keyword::TO);
+    ParserKeyword step(Keyword::STEP);
     ParserStringLiteral collate_locale_parser;
     ParserExpressionWithOptionalAlias exp_parser(false);
 
@@ -2139,7 +2140,7 @@ bool ParserOrderByElement::parseImpl(Pos & pos, ASTPtr & node, Expected & expect
 
 bool ParserInterpolateElement::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword as("AS");
+    ParserKeyword as(Keyword::AS);
     ParserExpression element_p;
     ParserIdentifier ident_p;
 
@@ -2210,18 +2211,18 @@ bool ParserFunctionWithKeyValueArguments::parseImpl(Pos & pos, ASTPtr & node, Ex
 
 bool ParserTTLElement::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_to_disk("TO DISK");
-    ParserKeyword s_to_volume("TO VOLUME");
-    ParserKeyword s_if_exists("IF EXISTS");
-    ParserKeyword s_delete("DELETE");
-    ParserKeyword s_where("WHERE");
-    ParserKeyword s_group_by("GROUP BY");
-    ParserKeyword s_set("SET");
-    ParserKeyword s_recompress("RECOMPRESS");
-    ParserKeyword s_codec("CODEC");
-    ParserKeyword s_materialize("MATERIALIZE");
-    ParserKeyword s_remove("REMOVE");
-    ParserKeyword s_modify("MODIFY");
+    ParserKeyword s_to_disk(Keyword::TO_DISK);
+    ParserKeyword s_to_volume(Keyword::TO_VOLUME);
+    ParserKeyword s_if_exists(Keyword::IF_EXISTS);
+    ParserKeyword s_delete(Keyword::DELETE);
+    ParserKeyword s_where(Keyword::WHERE);
+    ParserKeyword s_group_by(Keyword::GROUP_BY);
+    ParserKeyword s_set(Keyword::SET);
+    ParserKeyword s_recompress(Keyword::RECOMPRESS);
+    ParserKeyword s_codec(Keyword::CODEC);
+    ParserKeyword s_materialize(Keyword::MATERIALIZE);
+    ParserKeyword s_remove(Keyword::REMOVE);
+    ParserKeyword s_modify(Keyword::MODIFY);
 
     ParserIdentifier parser_identifier;
     ParserStringLiteral parser_string_literal;
diff --git a/src/Parsers/ExpressionListParsers.cpp b/src/Parsers/ExpressionListParsers.cpp
index 1e7d0158878..05691529f43 100644
--- a/src/Parsers/ExpressionListParsers.cpp
+++ b/src/Parsers/ExpressionListParsers.cpp
@@ -69,11 +69,11 @@ bool ParserList::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 bool ParserUnionList::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
     ParserUnionQueryElement elem_parser;
-    ParserKeyword s_union_parser("UNION");
-    ParserKeyword s_all_parser("ALL");
-    ParserKeyword s_distinct_parser("DISTINCT");
-    ParserKeyword s_except_parser("EXCEPT");
-    ParserKeyword s_intersect_parser("INTERSECT");
+    ParserKeyword s_union_parser(Keyword::UNION);
+    ParserKeyword s_all_parser(Keyword::ALL);
+    ParserKeyword s_distinct_parser(Keyword::DISTINCT);
+    ParserKeyword s_except_parser(Keyword::EXCEPT);
+    ParserKeyword s_intersect_parser(Keyword::INTERSECT);
     ASTs elements;
 
     auto parse_element = [&]
@@ -135,7 +135,7 @@ static bool parseOperator(IParser::Pos & pos, std::string_view op, Expected & ex
 {
     if (!op.empty() && isWordCharASCII(op.front()))
     {
-        return ParserKeyword(op).ignore(pos, expected);
+        return ParserKeyword::createDeprecated(op).ignore(pos, expected);
     }
     else if (op.length() == pos->size() && 0 == memcmp(op.data(), pos->begin, pos->size()))
     {
@@ -806,7 +806,6 @@ struct ParserExpressionImpl
 {
     static const std::vector<std::pair<std::string_view, Operator>> operators_table;
     static const std::vector<std::pair<std::string_view, Operator>> unary_operators_table;
-    static const std::array<std::string_view, 1> overlapping_operators_to_skip;
 
     static const Operator finish_between_operator;
 
@@ -819,8 +818,8 @@ struct ParserExpressionImpl
     ParserSubstitution substitution_parser;
     ParserMySQLGlobalVariable mysql_global_variable_parser;
 
-    ParserKeyword any_parser{"ANY"};
-    ParserKeyword all_parser{"ALL"};
+    ParserKeyword any_parser{Keyword::ANY};
+    ParserKeyword all_parser{Keyword::ALL};
 
     // Recursion
     ParserQualifiedAsterisk qualified_asterisk_parser;
@@ -885,7 +884,7 @@ public:
             if (test_pos.isValid() && test_pos->type != TokenType::Semicolon)
             {
                 /// If we can't parse FROM then return
-                if (!ParserKeyword("FROM").ignore(test_pos, test_expected))
+                if (!ParserKeyword(Keyword::FROM).ignore(test_pos, test_expected))
                     return true;
 
                 // If there is a comma after 'from' then the first one was a name of a column
@@ -893,7 +892,7 @@ public:
                     return true;
 
                 /// If we parse a second FROM then the first one was a name of a column
-                if (ParserKeyword("FROM").ignore(test_pos, test_expected))
+                if (ParserKeyword(Keyword::FROM).ignore(test_pos, test_expected))
                     return true;
 
                 /// If we parse an explicit alias to FROM, then it was a name of a column
@@ -981,8 +980,8 @@ public:
             auto pos_after_bracket = pos;
             auto old_expected = expected;
 
-            ParserKeyword all("ALL");
-            ParserKeyword distinct("DISTINCT");
+            ParserKeyword all(Keyword::ALL);
+            ParserKeyword distinct(Keyword::DISTINCT);
 
             if (all.ignore(pos, expected))
                 has_all = true;
@@ -1065,8 +1064,8 @@ public:
                     auto pos_after_bracket = pos;
                     auto old_expected = expected;
 
-                    ParserKeyword all("ALL");
-                    ParserKeyword distinct("DISTINCT");
+                    ParserKeyword all(Keyword::ALL);
+                    ParserKeyword distinct(Keyword::DISTINCT);
 
                     if (all.ignore(pos, expected))
                         has_all = true;
@@ -1112,10 +1111,10 @@ public:
                 function_node->children.push_back(function_node->parameters);
             }
 
-            ParserKeyword filter("FILTER");
-            ParserKeyword over("OVER");
-            ParserKeyword respect_nulls("RESPECT NULLS");
-            ParserKeyword ignore_nulls("IGNORE NULLS");
+            ParserKeyword filter(Keyword::FILTER);
+            ParserKeyword over(Keyword::OVER);
+            ParserKeyword respect_nulls(Keyword::RESPECT_NULLS);
+            ParserKeyword ignore_nulls(Keyword::IGNORE_NULLS);
 
             if (filter.ignore(pos, expected))
             {
@@ -1267,7 +1266,7 @@ public:
         /// 0. Parse all the cases (-> 1)
         /// 1. Parse closing token (finished)
 
-        ParserKeyword as_keyword_parser("AS");
+        ParserKeyword as_keyword_parser(Keyword::AS);
         ASTPtr alias;
 
         /// expr AS type
@@ -1348,7 +1347,7 @@ public:
                 if (elements.size() != 2)
                     return false;
 
-                elements = {makeASTFunction("CAST", elements[0], elements[1])};
+                elements = {makeASTFunction(toString(toStringView(Keyword::CAST)), elements[0], elements[1])};
                 finished = true;
                 return true;
             }
@@ -1374,7 +1373,7 @@ public:
         if (state == 0)
         {
             IParser::Pos begin = pos;
-            ParserKeyword s_from("FROM");
+            ParserKeyword s_from(Keyword::FROM);
 
             if (parseIntervalKind(pos, expected, interval_kind) && s_from.ignore(pos, expected))
             {
@@ -1447,7 +1446,7 @@ public:
         if (state == 0)
         {
             if (ParserToken(TokenType::Comma).ignore(pos, expected) ||
-                ParserKeyword("FROM").ignore(pos, expected))
+                ParserKeyword(Keyword::FROM).ignore(pos, expected))
             {
                 action = Action::OPERAND;
 
@@ -1461,7 +1460,7 @@ public:
         if (state == 1)
         {
             if (ParserToken(TokenType::Comma).ignore(pos, expected) ||
-                ParserKeyword("FOR").ignore(pos, expected))
+                ParserKeyword(Keyword::FOR).ignore(pos, expected))
             {
                 action = Action::OPERAND;
 
@@ -1518,7 +1517,7 @@ public:
 
                 state = 1;
             }
-            if (ParserKeyword("IN").ignore(pos, expected))
+            if (ParserKeyword(Keyword::IN).ignore(pos, expected))
             {
                 action = Action::OPERAND;
 
@@ -1610,18 +1609,18 @@ public:
         {
             if (!trim_left && !trim_right)
             {
-                if (ParserKeyword("BOTH").ignore(pos, expected))
+                if (ParserKeyword(Keyword::BOTH).ignore(pos, expected))
                 {
                     trim_left = true;
                     trim_right = true;
                     char_override = true;
                 }
-                else if (ParserKeyword("LEADING").ignore(pos, expected))
+                else if (ParserKeyword(Keyword::LEADING).ignore(pos, expected))
                 {
                     trim_left = true;
                     char_override = true;
                 }
-                else if (ParserKeyword("TRAILING").ignore(pos, expected))
+                else if (ParserKeyword(Keyword::TRAILING).ignore(pos, expected))
                 {
                     trim_right = true;
                     char_override = true;
@@ -1645,7 +1644,7 @@ public:
 
         if (state == 1)
         {
-            if (ParserKeyword("FROM").ignore(pos, expected))
+            if (ParserKeyword(Keyword::FROM).ignore(pos, expected))
             {
                 action = Action::OPERAND;
 
@@ -2002,7 +2001,7 @@ public:
         if (state == 0)
         {
             auto old_pos = pos;
-            has_case_expr = !ParserKeyword("WHEN").ignore(pos, expected);
+            has_case_expr = !ParserKeyword(Keyword::WHEN).ignore(pos, expected);
             pos = old_pos;
 
             state = 1;
@@ -2010,7 +2009,7 @@ public:
 
         if (state == 1)
         {
-            if (ParserKeyword("WHEN").ignore(pos, expected))
+            if (ParserKeyword(Keyword::WHEN).ignore(pos, expected))
             {
                 if ((has_case_expr || !elements.empty()) && !mergeElement())
                     return false;
@@ -2018,7 +2017,7 @@ public:
                 action = Action::OPERAND;
                 state = 2;
             }
-            else if (ParserKeyword("ELSE").ignore(pos, expected))
+            else if (ParserKeyword(Keyword::ELSE).ignore(pos, expected))
             {
                 if (!mergeElement())
                     return false;
@@ -2026,7 +2025,7 @@ public:
                 action = Action::OPERAND;
                 state = 3;
             }
-            else if (ParserKeyword("END").ignore(pos, expected))
+            else if (ParserKeyword(Keyword::END).ignore(pos, expected))
             {
                 if (!mergeElement())
                     return false;
@@ -2045,7 +2044,7 @@ public:
 
         if (state == 2)
         {
-            if (ParserKeyword("THEN").ignore(pos, expected))
+            if (ParserKeyword(Keyword::THEN).ignore(pos, expected))
             {
                 if (!mergeElement())
                     return false;
@@ -2057,7 +2056,7 @@ public:
 
         if (state == 3)
         {
-            if (ParserKeyword("END").ignore(pos, expected))
+            if (ParserKeyword(Keyword::END).ignore(pos, expected))
             {
                 if (!mergeElement())
                     return false;
@@ -2119,7 +2118,7 @@ public:
                 return true;
             }
 
-            if (!ParserKeyword{"ELSE"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::ELSE}.ignore(pos, expected))
                 return false;
 
             state = 1;
@@ -2303,7 +2302,7 @@ bool ParseDateOperatorExpression(IParser::Pos & pos, ASTPtr & node, Expected & e
     auto begin = pos;
 
     /// If no DATE keyword, go to the nested parser.
-    if (!ParserKeyword("DATE").ignore(pos, expected))
+    if (!ParserKeyword(Keyword::DATE).ignore(pos, expected))
         return false;
 
     ASTPtr expr;
@@ -2322,7 +2321,7 @@ bool ParseTimestampOperatorExpression(IParser::Pos & pos, ASTPtr & node, Expecte
     auto begin = pos;
 
     /// If no TIMESTAMP keyword, go to the nested parser.
-    if (!ParserKeyword("TIMESTAMP").ignore(pos, expected))
+    if (!ParserKeyword(Keyword::TIMESTAMP).ignore(pos, expected))
         return false;
 
     ASTPtr expr;
@@ -2378,13 +2377,13 @@ const std::vector<std::pair<std::string_view, Operator>> ParserExpressionImpl::o
     {"->",            Operator("lambda",          1,  2, OperatorType::Lambda)},
     {"?",             Operator("",                2,  0, OperatorType::StartIf)},
     {":",             Operator("if",              3,  3, OperatorType::FinishIf)},
-    {"OR",            Operator("or",              3,  2, OperatorType::Mergeable)},
-    {"AND",           Operator("and",             4,  2, OperatorType::Mergeable)},
-    {"IS NOT DISTINCT FROM", Operator("isNotDistinctFrom", 6, 2)},
-    {"IS NULL",       Operator("isNull",          6,  1, OperatorType::IsNull)},
-    {"IS NOT NULL",   Operator("isNotNull",       6,  1, OperatorType::IsNull)},
-    {"BETWEEN",       Operator("",                7,  0, OperatorType::StartBetween)},
-    {"NOT BETWEEN",   Operator("",                7,  0, OperatorType::StartNotBetween)},
+    {toStringView(Keyword::OR),            Operator("or",              3,  2, OperatorType::Mergeable)},
+    {toStringView(Keyword::AND),           Operator("and",             4,  2, OperatorType::Mergeable)},
+    {toStringView(Keyword::IS_NOT_DISTINCT_FROM), Operator("isNotDistinctFrom", 6, 2)},
+    {toStringView(Keyword::IS_NULL),       Operator("isNull",          6,  1, OperatorType::IsNull)},
+    {toStringView(Keyword::IS_NOT_NULL),   Operator("isNotNull",       6,  1, OperatorType::IsNull)},
+    {toStringView(Keyword::BETWEEN),       Operator("",                7,  0, OperatorType::StartBetween)},
+    {toStringView(Keyword::NOT_BETWEEN),   Operator("",                7,  0, OperatorType::StartNotBetween)},
     {"==",            Operator("equals",          9,  2, OperatorType::Comparison)},
     {"!=",            Operator("notEquals",       9,  2, OperatorType::Comparison)},
     {"<=>",           Operator("isNotDistinctFrom", 9, 2, OperatorType::Comparison)},
@@ -2394,15 +2393,15 @@ const std::vector<std::pair<std::string_view, Operator>> ParserExpressionImpl::o
     {"<",             Operator("less",            9,  2, OperatorType::Comparison)},
     {">",             Operator("greater",         9,  2, OperatorType::Comparison)},
     {"=",             Operator("equals",          9,  2, OperatorType::Comparison)},
-    {"LIKE",          Operator("like",            9,  2)},
-    {"ILIKE",         Operator("ilike",           9,  2)},
-    {"NOT LIKE",      Operator("notLike",         9,  2)},
-    {"NOT ILIKE",     Operator("notILike",        9,  2)},
-    {"REGEXP",        Operator("match",           9,  2)},
-    {"IN",            Operator("in",              9,  2)},
-    {"NOT IN",        Operator("notIn",           9,  2)},
-    {"GLOBAL IN",     Operator("globalIn",        9,  2)},
-    {"GLOBAL NOT IN", Operator("globalNotIn",     9,  2)},
+    {toStringView(Keyword::LIKE),          Operator("like",            9,  2)},
+    {toStringView(Keyword::ILIKE),         Operator("ilike",           9,  2)},
+    {toStringView(Keyword::NOT_LIKE),      Operator("notLike",         9,  2)},
+    {toStringView(Keyword::NOT_ILIKE),     Operator("notILike",        9,  2)},
+    {toStringView(Keyword::REGEXP),        Operator("match",           9,  2)},
+    {toStringView(Keyword::IN),            Operator("in",              9,  2)},
+    {toStringView(Keyword::NOT_IN),        Operator("notIn",           9,  2)},
+    {toStringView(Keyword::GLOBAL_IN),     Operator("globalIn",        9,  2)},
+    {toStringView(Keyword::GLOBAL_NOT_IN), Operator("globalNotIn",     9,  2)},
     {"||",            Operator("concat",          10, 2, OperatorType::Mergeable)},
     {"+",             Operator("plus",            11, 2)},
     {"-",             Operator("minus",           11, 2)},
@@ -2410,26 +2409,22 @@ const std::vector<std::pair<std::string_view, Operator>> ParserExpressionImpl::o
     {"*",             Operator("multiply",        12, 2)},
     {"/",             Operator("divide",          12, 2)},
     {"%",             Operator("modulo",          12, 2)},
-    {"MOD",           Operator("modulo",          12, 2)},
-    {"DIV",           Operator("intDiv",          12, 2)},
+    {toStringView(Keyword::MOD),           Operator("modulo",          12, 2)},
+    {toStringView(Keyword::DIV),           Operator("intDiv",          12, 2)},
     {".",             Operator("tupleElement",    14, 2, OperatorType::TupleElement)},
     {"[",             Operator("arrayElement",    14, 2, OperatorType::ArrayElement)},
-    {"::",            Operator("CAST",            14, 2, OperatorType::Cast)},
+    {"::",            Operator(toString(toStringView(Keyword::CAST)),            14, 2, OperatorType::Cast)},
 };
 
 const std::vector<std::pair<std::string_view, Operator>> ParserExpressionImpl::unary_operators_table
 {
-    {"NOT",           Operator("not",             5,  1, OperatorType::Not)},
-    {"-",             Operator("negate",          13, 1)},
-    {"−",             Operator("negate",          13, 1)}
+    {toStringView(Keyword::NOT), Operator("not", 5, 1, OperatorType::Not)},
+    {"-", Operator("negate", 13, 1)},
+    {"−", Operator("negate", 13, 1)},
 };
 
 const Operator ParserExpressionImpl::finish_between_operator("", 8, 0, OperatorType::FinishBetween);
 
-const std::array<std::string_view, 1> ParserExpressionImpl::overlapping_operators_to_skip
-{
-    "IN PARTITION"
-};
 
 bool ParserExpressionImpl::parse(std::unique_ptr<Layer> start, IParser::Pos & pos, ASTPtr & node, Expected & expected)
 {
@@ -2516,7 +2511,7 @@ Action ParserExpressionImpl::tryParseOperand(Layers & layers, IParser::Pos & pos
         if (layers.back()->isCurrentElementEmpty())
         {
             auto old_pos = pos;
-            ParserKeyword s_settings("SETTINGS");
+            ParserKeyword s_settings(Keyword::SETTINGS);
             if (s_settings.ignore(pos, expected))
             {
                 ParserSetQuery parser_settings(true);
@@ -2608,13 +2603,13 @@ Action ParserExpressionImpl::tryParseOperand(Layers & layers, IParser::Pos & pos
     auto current_checkpoint = layers.back()->current_checkpoint;
     layers.back()->current_checkpoint = Checkpoint::None;
 
-    if (current_checkpoint != Checkpoint::Interval && parseOperator(pos, "INTERVAL", expected))
+    if (current_checkpoint != Checkpoint::Interval && parseOperator(pos, toStringView(Keyword::INTERVAL), expected))
     {
         layers.back()->saved_checkpoint = {old_pos, Checkpoint::Interval};
         layers.push_back(std::make_unique<IntervalLayer>());
         return Action::OPERAND;
     }
-    else if (current_checkpoint != Checkpoint::Case && parseOperator(pos, "CASE", expected))
+    else if (current_checkpoint != Checkpoint::Case && parseOperator(pos, toStringView(Keyword::CASE), expected))
     {
         layers.back()->saved_checkpoint = {old_pos, Checkpoint::Case};
         layers.push_back(std::make_unique<CaseLayer>());
@@ -2689,9 +2684,8 @@ Action ParserExpressionImpl::tryParseOperator(Layers & layers, IParser::Pos & po
     ///
     /// 'IN PARTITION' here is not an 'IN' operator, so we should stop parsing immediately
     Expected stub;
-    for (const auto & it : overlapping_operators_to_skip)
-        if (ParserKeyword{it}.checkWithoutMoving(pos, stub))
-            return Action::NONE;
+    if (ParserKeyword(Keyword::IN_PARTITION).checkWithoutMoving(pos, stub))
+        return Action::NONE;
 
     /// Try to find operators from 'operators_table'
     auto cur_op = operators_table.begin();
diff --git a/src/Parsers/Kusto/KustoFunctions/KQLStringFunctions.cpp b/src/Parsers/Kusto/KustoFunctions/KQLStringFunctions.cpp
index 18c986c2191..991a3139fcb 100644
--- a/src/Parsers/Kusto/KustoFunctions/KQLStringFunctions.cpp
+++ b/src/Parsers/Kusto/KustoFunctions/KQLStringFunctions.cpp
@@ -102,7 +102,7 @@ bool CountOf::convertImpl(String & out, IParser::Pos & pos)
 
 bool Extract::convertImpl(String & out, IParser::Pos & pos)
 {
-    ParserKeyword s_kql("typeof");
+    ParserKeyword s_kql(Keyword::TYPEOF);
     ParserToken open_bracket(TokenType::OpeningRoundBracket);
     ParserToken close_bracket(TokenType::ClosingRoundBracket);
     Expected expected;
@@ -243,7 +243,7 @@ bool ExtractAll::convertImpl(String & out, IParser::Pos & pos)
 bool ExtractJSON::convertImpl(String & out, IParser::Pos & pos)
 {
     String datatype = "String";
-    ParserKeyword s_kql("typeof");
+    ParserKeyword s_kql(Keyword::TYPEOF);
     ParserToken open_bracket(TokenType::OpeningRoundBracket);
     ParserToken close_bracket(TokenType::ClosingRoundBracket);
     Expected expected;
diff --git a/src/Parsers/Kusto/ParserKQLMVExpand.cpp b/src/Parsers/Kusto/ParserKQLMVExpand.cpp
index d174e9b5911..9beb1c39e34 100644
--- a/src/Parsers/Kusto/ParserKQLMVExpand.cpp
+++ b/src/Parsers/Kusto/ParserKQLMVExpand.cpp
@@ -39,8 +39,8 @@ bool ParserKQLMVExpand::parseColumnArrayExprs(ColumnArrayExprs & column_array_ex
     ParserToken close_bracket(TokenType::ClosingRoundBracket);
     ParserToken comma(TokenType::Comma);
 
-    ParserKeyword s_to("to");
-    ParserKeyword s_type("typeof");
+    ParserKeyword s_to(Keyword::TO);
+    ParserKeyword s_type(Keyword::TYPEOF);
     uint16_t bracket_count = 0;
     Pos expr_begin_pos = pos;
     Pos expr_end_pos = pos;
@@ -144,10 +144,10 @@ bool ParserKQLMVExpand::parseColumnArrayExprs(ColumnArrayExprs & column_array_ex
 
 bool ParserKQLMVExpand::parserMVExpand(KQLMVExpand & kql_mv_expand, Pos & pos, Expected & expected)
 {
-    ParserKeyword s_bagexpansion("bagexpansion");
-    ParserKeyword s_kind("kind");
-    ParserKeyword s_with_itemindex("with_itemindex");
-    ParserKeyword s_limit("limit");
+    ParserKeyword s_bagexpansion(Keyword::BAGEXPANSION);
+    ParserKeyword s_kind(Keyword::KIND);
+    ParserKeyword s_with_itemindex(Keyword::WITH_ITEMINDEX);
+    ParserKeyword s_limit(Keyword::LIMIT);
 
     ParserToken equals(TokenType::Equals);
     ParserToken comma(TokenType::Comma);
diff --git a/src/Parsers/Kusto/ParserKQLMakeSeries.cpp b/src/Parsers/Kusto/ParserKQLMakeSeries.cpp
index 4759efc0025..f1ad9d9738b 100644
--- a/src/Parsers/Kusto/ParserKQLMakeSeries.cpp
+++ b/src/Parsers/Kusto/ParserKQLMakeSeries.cpp
@@ -34,7 +34,7 @@ bool ParserKQLMakeSeries ::parseAggregationColumns(AggregationColumns & aggregat
          "variance"});
 
     Expected expected;
-    ParserKeyword s_default("default");
+    ParserKeyword s_default(Keyword::DEFAULT);
     ParserToken equals(TokenType::Equals);
     ParserToken open_bracket(TokenType::OpeningRoundBracket);
     ParserToken close_bracket(TokenType::ClosingRoundBracket);
@@ -364,8 +364,8 @@ bool ParserKQLMakeSeries ::makeSeries(KQLMakeSeries & kql_make_series, ASTPtr &
 bool ParserKQLMakeSeries ::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
     auto begin = pos;
-    ParserKeyword s_on("on");
-    ParserKeyword s_by("by");
+    ParserKeyword s_on(Keyword::ON);
+    ParserKeyword s_by(Keyword::BY);
 
     ParserToken equals(TokenType::Equals);
     ParserToken comma(TokenType::Comma);
diff --git a/src/Parsers/Kusto/ParserKQLQuery.cpp b/src/Parsers/Kusto/ParserKQLQuery.cpp
index 6fd9c95ec6f..99b2d1da890 100644
--- a/src/Parsers/Kusto/ParserKQLQuery.cpp
+++ b/src/Parsers/Kusto/ParserKQLQuery.cpp
@@ -399,7 +399,7 @@ bool ParserKQLQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
                     if (!isValidKQLPos(pos))
                         return false;
 
-                    ParserKeyword s_by("by");
+                    ParserKeyword s_by(Keyword::BY);
                     if (s_by.ignore(pos, expected))
                     {
                         kql_operator = "order by";
diff --git a/src/Parsers/MySQL/ASTAlterCommand.cpp b/src/Parsers/MySQL/ASTAlterCommand.cpp
index 10446d40172..733cb657efc 100644
--- a/src/Parsers/MySQL/ASTAlterCommand.cpp
+++ b/src/Parsers/MySQL/ASTAlterCommand.cpp
@@ -55,7 +55,7 @@ static inline bool parseAddCommand(IParser::Pos & pos, ASTPtr & node, Expected &
     else
     {
         alter_command->type = ASTAlterCommand::ADD_COLUMN;
-        ParserKeyword("COLUMN").ignore(pos, expected);
+        ParserKeyword(Keyword::COLUMN).ignore(pos, expected);
 
         if (ParserToken(TokenType::OpeningRoundBracket).ignore(pos, expected))
         {
@@ -76,9 +76,9 @@ static inline bool parseAddCommand(IParser::Pos & pos, ASTPtr & node, Expected &
             additional_columns = std::make_shared<ASTExpressionList>();
             additional_columns->children.emplace_back(declare_column);
 
-            if (ParserKeyword("FIRST").ignore(pos, expected))
+            if (ParserKeyword(Keyword::FIRST).ignore(pos, expected))
                 alter_command->first = true;
-            else if (ParserKeyword("AFTER").ignore(pos, expected))
+            else if (ParserKeyword(Keyword::AFTER).ignore(pos, expected))
             {
                 ASTPtr after_column;
                 ParserIdentifier identifier_p;
@@ -103,12 +103,12 @@ static inline bool parseDropCommand(IParser::Pos & pos, ASTPtr & node, Expected
 
     auto alter_command = std::make_shared<ASTAlterCommand>();
 
-    if (ParserKeyword("PRIMARY KEY").ignore(pos, expected))
+    if (ParserKeyword(Keyword::PRIMARY_KEY).ignore(pos, expected))
     {
         alter_command->index_type = "PRIMARY_KEY";
         alter_command->type = ASTAlterCommand::DROP_INDEX;
     }
-    else if (ParserKeyword("FOREIGN KEY").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::FOREIGN_KEY).ignore(pos, expected))
     {
         if (!identifier_p.parse(pos, name, expected))
             return false;
@@ -117,7 +117,7 @@ static inline bool parseDropCommand(IParser::Pos & pos, ASTPtr & node, Expected
         alter_command->type = ASTAlterCommand::DROP_INDEX;
         alter_command->index_name = getIdentifierName(name);
     }
-    else if (ParserKeyword("INDEX").ignore(pos, expected) || ParserKeyword("KEY").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::INDEX).ignore(pos, expected) || ParserKeyword(Keyword::KEY).ignore(pos, expected))
     {
         if (!identifier_p.parse(pos, name, expected))
             return false;
@@ -126,7 +126,7 @@ static inline bool parseDropCommand(IParser::Pos & pos, ASTPtr & node, Expected
         alter_command->type = ASTAlterCommand::DROP_INDEX;
         alter_command->index_name = getIdentifierName(name);
     }
-    else if (ParserKeyword("CONSTRAINT").ignore(pos, expected) || ParserKeyword("CHECK").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::CONSTRAINT).ignore(pos, expected) || ParserKeyword(Keyword::CHECK).ignore(pos, expected))
     {
         if (!identifier_p.parse(pos, name, expected))
             return false;
@@ -136,7 +136,7 @@ static inline bool parseDropCommand(IParser::Pos & pos, ASTPtr & node, Expected
     }
     else
     {
-        ParserKeyword("COLUMN").ignore(pos, expected);
+        ParserKeyword(Keyword::COLUMN).ignore(pos, expected);
 
         if (!identifier_p.parse(pos, name, expected))
             return false;
@@ -156,30 +156,30 @@ static inline bool parseAlterCommand(IParser::Pos & pos, ASTPtr & node, Expected
     ParserIdentifier identifier_p;
     auto alter_command = std::make_shared<ASTAlterCommand>();
 
-    if (ParserKeyword("INDEX").ignore(pos, expected))
+    if (ParserKeyword(Keyword::INDEX).ignore(pos, expected))
     {
         /// ALTER INDEX index_name {VISIBLE | INVISIBLE}
 
         if (!identifier_p.parse(pos, name, expected))
             return false;
 
-        alter_command->index_visible = ParserKeyword("VISIBLE").ignore(pos, expected);
+        alter_command->index_visible = ParserKeyword(Keyword::VISIBLE).ignore(pos, expected);
 
-        if (!alter_command->index_visible && !ParserKeyword("INVISIBLE").ignore(pos, expected))
+        if (!alter_command->index_visible && !ParserKeyword(Keyword::INVISIBLE).ignore(pos, expected))
             return false;
 
         alter_command->type = ASTAlterCommand::MODIFY_INDEX_VISIBLE;
         alter_command->index_name = getIdentifierName(name);
     }
-    else if (ParserKeyword("CHECK").ignore(pos, expected) || ParserKeyword("CONSTRAINT").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::CHECK).ignore(pos, expected) || ParserKeyword(Keyword::CONSTRAINT).ignore(pos, expected))
     {
         /// ALTER {CHECK | CONSTRAINT} symbol [NOT] ENFORCED
         if (!identifier_p.parse(pos, name, expected))
             return false;
 
-        alter_command->not_check_enforced = ParserKeyword("NOT").ignore(pos, expected);
+        alter_command->not_check_enforced = ParserKeyword(Keyword::NOT).ignore(pos, expected);
 
-        if (!ParserKeyword("ENFORCED").ignore(pos, expected))
+        if (!ParserKeyword(Keyword::ENFORCED).ignore(pos, expected))
             return false;
 
         alter_command->type = ASTAlterCommand::MODIFY_CHECK;
@@ -189,14 +189,14 @@ static inline bool parseAlterCommand(IParser::Pos & pos, ASTPtr & node, Expected
     {
         /// ALTER [COLUMN] col_name {SET DEFAULT {literal | (expr)} | DROP DEFAULT}
 
-        ParserKeyword("COLUMN").ignore(pos, expected);
+        ParserKeyword(Keyword::COLUMN).ignore(pos, expected);
 
         if (!identifier_p.parse(pos, name, expected))
             return false;
 
-        if (ParserKeyword("DROP DEFAULT").ignore(pos, expected))
+        if (ParserKeyword(Keyword::DROP_DEFAULT).ignore(pos, expected))
             alter_command->type = ASTAlterCommand::DROP_COLUMN_DEFAULT;
-        else if (ParserKeyword("SET DEFAULT").ignore(pos, expected))
+        else if (ParserKeyword(Keyword::SET_DEFAULT).ignore(pos, expected))
         {
             ASTPtr default_expression;
             ParserExpression expression_p;
@@ -225,12 +225,12 @@ static inline bool parseRenameCommand(IParser::Pos & pos, ASTPtr & node, Expecte
     ParserIdentifier identifier_p;
     auto alter_command = std::make_shared<ASTAlterCommand>();
 
-    if (ParserKeyword("COLUMN").ignore(pos, expected))
+    if (ParserKeyword(Keyword::COLUMN).ignore(pos, expected))
     {
         if (!identifier_p.parse(pos, old_name, expected))
             return false;
 
-        if (!ParserKeyword("TO").ignore(pos, expected))
+        if (!ParserKeyword(Keyword::TO).ignore(pos, expected))
             return false;
 
         if (!identifier_p.parse(pos, new_name, expected))
@@ -240,7 +240,7 @@ static inline bool parseRenameCommand(IParser::Pos & pos, ASTPtr & node, Expecte
         alter_command->old_name = getIdentifierName(old_name);
         alter_command->column_name = getIdentifierName(new_name);
     }
-    else if (ParserKeyword("TO").ignore(pos, expected) || ParserKeyword("AS").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::TO).ignore(pos, expected) || ParserKeyword(Keyword::AS).ignore(pos, expected))
     {
         if (!ParserCompoundIdentifier(true).parse(pos, new_name, expected))
             return false;
@@ -250,12 +250,12 @@ static inline bool parseRenameCommand(IParser::Pos & pos, ASTPtr & node, Expecte
         alter_command->new_table_name = new_table_id.table_name;
         alter_command->new_database_name = new_table_id.database_name;
     }
-    else if (ParserKeyword("INDEX").ignore(pos, expected) || ParserKeyword("KEY").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::INDEX).ignore(pos, expected) || ParserKeyword(Keyword::KEY).ignore(pos, expected))
     {
         if (!identifier_p.parse(pos, old_name, expected))
             return false;
 
-        if (!ParserKeyword("TO").ignore(pos, expected))
+        if (!ParserKeyword(Keyword::TO).ignore(pos, expected))
             return false;
 
         if (!identifier_p.parse(pos, new_name, expected))
@@ -283,7 +283,7 @@ static inline bool parseOtherCommand(IParser::Pos & pos, ASTPtr & node, Expected
 {
     auto alter_command = std::make_shared<ASTAlterCommand>();
 
-    if (ParserKeyword("ORDER BY").ignore(pos, expected))
+    if (ParserKeyword(Keyword::ORDER_BY).ignore(pos, expected))
     {
         /// ORDER BY col_name [, col_name] ...
         ASTPtr columns;
@@ -335,7 +335,7 @@ static inline bool parseModifyCommand(IParser::Pos & pos, ASTPtr & node, Expecte
     ASTPtr old_column_name;
     auto alter_command = std::make_shared<ASTAlterCommand>();
 
-    ParserKeyword("COLUMN").ignore(pos, expected);
+    ParserKeyword(Keyword::COLUMN).ignore(pos, expected);
     if (exists_old_column_name && !ParserIdentifier().parse(pos, old_column_name, expected))
         return false;
 
@@ -343,9 +343,9 @@ static inline bool parseModifyCommand(IParser::Pos & pos, ASTPtr & node, Expecte
     if (!ParserDeclareColumn().parse(pos, additional_column, expected))
         return false;
 
-    if (ParserKeyword("FIRST").ignore(pos, expected))
+    if (ParserKeyword(Keyword::FIRST).ignore(pos, expected))
         alter_command->first = true;
-    else if (ParserKeyword("AFTER").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::AFTER).ignore(pos, expected))
     {
         ASTPtr after_column;
         ParserIdentifier identifier_p;
@@ -368,12 +368,12 @@ static inline bool parseModifyCommand(IParser::Pos & pos, ASTPtr & node, Expecte
 
 bool ParserAlterCommand::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword k_add("ADD");
-    ParserKeyword k_drop("DROP");
-    ParserKeyword k_alter("ALTER");
-    ParserKeyword k_rename("RENAME");
-    ParserKeyword k_modify("MODIFY");
-    ParserKeyword k_change("CHANGE");
+    ParserKeyword k_add(Keyword::ADD);
+    ParserKeyword k_drop(Keyword::DROP);
+    ParserKeyword k_alter(Keyword::ALTER);
+    ParserKeyword k_rename(Keyword::RENAME);
+    ParserKeyword k_modify(Keyword::MODIFY);
+    ParserKeyword k_change(Keyword::CHANGE);
 
     if (k_add.ignore(pos, expected))
         return parseAddCommand(pos, node, expected);
diff --git a/src/Parsers/MySQL/ASTAlterQuery.cpp b/src/Parsers/MySQL/ASTAlterQuery.cpp
index 59f0ada4e32..4e8c2cb5a6b 100644
--- a/src/Parsers/MySQL/ASTAlterQuery.cpp
+++ b/src/Parsers/MySQL/ASTAlterQuery.cpp
@@ -33,7 +33,7 @@ bool ParserAlterQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected & e
     ASTPtr table;
     ASTPtr command_list;
 
-    if (!ParserKeyword("ALTER TABLE").ignore(pos, expected))
+    if (!ParserKeyword(Keyword::ALTER_TABLE).ignore(pos, expected))
         return false;
 
     if (!ParserCompoundIdentifier(true).parse(pos, table, expected))
diff --git a/src/Parsers/MySQL/ASTCreateQuery.cpp b/src/Parsers/MySQL/ASTCreateQuery.cpp
index 227ac62f86d..d8b053591d6 100644
--- a/src/Parsers/MySQL/ASTCreateQuery.cpp
+++ b/src/Parsers/MySQL/ASTCreateQuery.cpp
@@ -51,29 +51,29 @@ bool ParserCreateQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
     bool is_temporary = false;
     bool if_not_exists = false;
 
-    if (!ParserKeyword("CREATE").ignore(pos, expected))
+    if (!ParserKeyword(Keyword::CREATE).ignore(pos, expected))
         return false;
 
-    if (ParserKeyword("TEMPORARY").ignore(pos, expected))
+    if (ParserKeyword(Keyword::TEMPORARY).ignore(pos, expected))
         is_temporary = true;
 
-    if (!ParserKeyword("TABLE").ignore(pos, expected))
+    if (!ParserKeyword(Keyword::TABLE).ignore(pos, expected))
         return false;
 
-    if (ParserKeyword("IF NOT EXISTS").ignore(pos, expected))
+    if (ParserKeyword(Keyword::IF_NOT_EXISTS).ignore(pos, expected))
         if_not_exists = true;
 
     if (!ParserCompoundIdentifier(true).parse(pos, table, expected))
         return false;
 
-    if (ParserKeyword("LIKE").ignore(pos, expected))
+    if (ParserKeyword(Keyword::LIKE).ignore(pos, expected))
     {
         if (!ParserCompoundIdentifier(true).parse(pos, like_table, expected))
             return false;
     }
     else if (ParserToken(TokenType::OpeningRoundBracket).ignore(pos, expected))
     {
-        if (ParserKeyword("LIKE").ignore(pos, expected))
+        if (ParserKeyword(Keyword::LIKE).ignore(pos, expected))
         {
             if (!ParserCompoundIdentifier(true).parse(pos, like_table, expected))
                 return false;
diff --git a/src/Parsers/MySQL/ASTDeclareConstraint.cpp b/src/Parsers/MySQL/ASTDeclareConstraint.cpp
index 96184dfc89d..414e259f08d 100644
--- a/src/Parsers/MySQL/ASTDeclareConstraint.cpp
+++ b/src/Parsers/MySQL/ASTDeclareConstraint.cpp
@@ -32,22 +32,22 @@ bool ParserDeclareConstraint::parseImpl(IParser::Pos & pos, ASTPtr & node, Expec
     ASTPtr index_check_expression;
     ParserExpression p_expression;
 
-    if (ParserKeyword("CONSTRAINT").ignore(pos, expected))
+    if (ParserKeyword(Keyword::CONSTRAINT).ignore(pos, expected))
     {
-        if (!ParserKeyword("CHECK").checkWithoutMoving(pos, expected))
+        if (!ParserKeyword(Keyword::CHECK).checkWithoutMoving(pos, expected))
             ParserIdentifier().parse(pos, constraint_symbol, expected);
     }
 
 
-    if (!ParserKeyword("CHECK").ignore(pos, expected))
+    if (!ParserKeyword(Keyword::CHECK).ignore(pos, expected))
         return false;
 
     if (!p_expression.parse(pos, index_check_expression, expected))
         return false;
 
-    if (ParserKeyword("NOT").ignore(pos, expected))
+    if (ParserKeyword(Keyword::NOT).ignore(pos, expected))
     {
-        if (!ParserKeyword("ENFORCED").ignore(pos, expected))
+        if (!ParserKeyword(Keyword::ENFORCED).ignore(pos, expected))
             return false;
 
         enforced = false;
@@ -55,7 +55,7 @@ bool ParserDeclareConstraint::parseImpl(IParser::Pos & pos, ASTPtr & node, Expec
     else
     {
         enforced = true;
-        ParserKeyword("ENFORCED").ignore(pos, expected);
+        ParserKeyword(Keyword::ENFORCED).ignore(pos, expected);
     }
 
     auto declare_constraint = std::make_shared<ASTDeclareConstraint>();
diff --git a/src/Parsers/MySQL/ASTDeclareIndex.cpp b/src/Parsers/MySQL/ASTDeclareIndex.cpp
index c5b4686e976..146f803e650 100644
--- a/src/Parsers/MySQL/ASTDeclareIndex.cpp
+++ b/src/Parsers/MySQL/ASTDeclareIndex.cpp
@@ -24,8 +24,8 @@ protected:
         if (!p_expression.parse(pos, node, expected))
             return false;
 
-        ParserKeyword("ASC").ignore(pos, expected);
-        ParserKeyword("DESC").ignore(pos, expected);
+        ParserKeyword(Keyword::ASC).ignore(pos, expected);
+        ParserKeyword(Keyword::DESC).ignore(pos, expected);
         return true;
     }
 };
@@ -61,12 +61,12 @@ ASTPtr ASTDeclareIndex::clone() const
 static inline bool parseDeclareOrdinaryIndex(IParser::Pos & pos, String & index_name, String & index_type, Expected & expected)
 {
     ASTPtr temp_node;
-    ParserKeyword k_key("KEY");
-    ParserKeyword k_index("INDEX");
+    ParserKeyword k_key(Keyword::KEY);
+    ParserKeyword k_index(Keyword::INDEX);
 
     ParserIdentifier p_identifier;
 
-    if (ParserKeyword("SPATIAL").ignore(pos, expected))
+    if (ParserKeyword(Keyword::SPATIAL).ignore(pos, expected))
     {
         if (!k_key.ignore(pos, expected))
             k_index.ignore(pos, expected);
@@ -75,7 +75,7 @@ static inline bool parseDeclareOrdinaryIndex(IParser::Pos & pos, String & index_
         if (p_identifier.parse(pos, temp_node, expected))
             index_name = temp_node->as<ASTIdentifier>()->name();
     }
-    else if (ParserKeyword("FULLTEXT").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::FULLTEXT).ignore(pos, expected))
     {
         if (!k_key.ignore(pos, expected))
             k_index.ignore(pos, expected);
@@ -96,7 +96,7 @@ static inline bool parseDeclareOrdinaryIndex(IParser::Pos & pos, String & index_
         if (p_identifier.parse(pos, temp_node, expected))
             index_name = temp_node->as<ASTIdentifier>()->name();
 
-        if (ParserKeyword("USING").ignore(pos, expected))
+        if (ParserKeyword(Keyword::USING).ignore(pos, expected))
         {
             if (!p_identifier.parse(pos, temp_node, expected))
                 return false;
@@ -113,11 +113,11 @@ static inline bool parseDeclareConstraintIndex(IParser::Pos & pos, String & inde
     ASTPtr temp_node;
     ParserIdentifier p_identifier;
 
-    if (ParserKeyword("CONSTRAINT").ignore(pos, expected))
+    if (ParserKeyword(Keyword::CONSTRAINT).ignore(pos, expected))
     {
 
-        if (!ParserKeyword("PRIMARY").checkWithoutMoving(pos, expected) && !ParserKeyword("UNIQUE").checkWithoutMoving(pos, expected)
-            && !ParserKeyword("FOREIGN").checkWithoutMoving(pos, expected))
+        if (!ParserKeyword(Keyword::PRIMARY).checkWithoutMoving(pos, expected) && !ParserKeyword(Keyword::UNIQUE).checkWithoutMoving(pos, expected)
+            && !ParserKeyword(Keyword::FOREIGN).checkWithoutMoving(pos, expected))
         {
             if (!p_identifier.parse(pos, temp_node, expected))
                 return false;
@@ -126,16 +126,16 @@ static inline bool parseDeclareConstraintIndex(IParser::Pos & pos, String & inde
         }
     }
 
-    if (ParserKeyword("UNIQUE").ignore(pos, expected))
+    if (ParserKeyword(Keyword::UNIQUE).ignore(pos, expected))
     {
-        if (!ParserKeyword("KEY").ignore(pos, expected))
-            ParserKeyword("INDEX").ignore(pos, expected);
+        if (!ParserKeyword(Keyword::KEY).ignore(pos, expected))
+            ParserKeyword(Keyword::INDEX).ignore(pos, expected);
 
         if (p_identifier.parse(pos, temp_node, expected))
             index_name = temp_node->as<ASTIdentifier>()->name();  /// reset index_name
 
         index_type = "UNIQUE_BTREE"; /// default btree index_type
-        if (ParserKeyword("USING").ignore(pos, expected))
+        if (ParserKeyword(Keyword::USING).ignore(pos, expected))
         {
             if (!p_identifier.parse(pos, temp_node, expected))
                 return false;
@@ -143,10 +143,10 @@ static inline bool parseDeclareConstraintIndex(IParser::Pos & pos, String & inde
             index_type = "UNIQUE_" + temp_node->as<ASTIdentifier>()->name();
         }
     }
-    else if (ParserKeyword("PRIMARY KEY").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::PRIMARY_KEY).ignore(pos, expected))
     {
         index_type = "PRIMARY_KEY_BTREE"; /// default btree index_type
-        if (ParserKeyword("USING").ignore(pos, expected))
+        if (ParserKeyword(Keyword::USING).ignore(pos, expected))
         {
             if (!p_identifier.parse(pos, temp_node, expected))
                 return false;
@@ -154,7 +154,7 @@ static inline bool parseDeclareConstraintIndex(IParser::Pos & pos, String & inde
             index_type = "PRIMARY_KEY_" + temp_node->as<ASTIdentifier>()->name();
         }
     }
-    else if (ParserKeyword("FOREIGN KEY").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::FOREIGN_KEY).ignore(pos, expected))
     {
         index_type = "FOREIGN";
         if (p_identifier.parse(pos, temp_node, expected))
diff --git a/src/Parsers/MySQL/ASTDeclareOption.cpp b/src/Parsers/MySQL/ASTDeclareOption.cpp
index 740d049bafd..f4be38f4f2d 100644
--- a/src/Parsers/MySQL/ASTDeclareOption.cpp
+++ b/src/Parsers/MySQL/ASTDeclareOption.cpp
@@ -23,7 +23,7 @@ bool ParserDeclareOptionImpl<recursive>::parseImpl(Pos & pos, ASTPtr & node, Exp
     {
         auto iterator = usage_parsers_cached.find(usage_name);
         if (iterator == usage_parsers_cached.end())
-            iterator = usage_parsers_cached.insert(std::make_pair(usage_name, std::make_shared<ParserKeyword>(usage_name))).first;
+            iterator = usage_parsers_cached.insert(std::make_pair(usage_name, ParserKeyword::createDeprecatedPtr(usage_name))).first;
 
         return iterator->second;
     };
diff --git a/src/Parsers/MySQL/ASTDeclarePartition.cpp b/src/Parsers/MySQL/ASTDeclarePartition.cpp
index 76f864fcc44..2647131fde5 100644
--- a/src/Parsers/MySQL/ASTDeclarePartition.cpp
+++ b/src/Parsers/MySQL/ASTDeclarePartition.cpp
@@ -46,7 +46,7 @@ ASTPtr ASTDeclarePartition::clone() const
 
 bool ParserDeclarePartition::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected & expected)
 {
-    if (!ParserKeyword{"PARTITION"}.ignore(pos, expected))
+    if (!ParserKeyword{Keyword::PARTITION}.ignore(pos, expected))
         return false;
 
     ASTPtr options;
@@ -60,15 +60,15 @@ bool ParserDeclarePartition::parseImpl(IParser::Pos & pos, ASTPtr & node, Expect
     if (!p_identifier.parse(pos, partition_name, expected))
         return false;
 
-    ParserKeyword p_values("VALUES");
+    ParserKeyword p_values(Keyword::VALUES);
     if (p_values.ignore(pos, expected))
     {
-        if (ParserKeyword{"IN"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::IN}.ignore(pos, expected))
         {
             if (!p_expression.parse(pos, in_expression, expected))
                 return false;
         }
-        else if (ParserKeyword{"LESS THAN"}.ignore(pos, expected))
+        else if (ParserKeyword{Keyword::LESS_THAN}.ignore(pos, expected))
         {
             if (!p_expression.parse(pos, less_than, expected))
                 return false;
diff --git a/src/Parsers/MySQL/ASTDeclarePartitionOptions.cpp b/src/Parsers/MySQL/ASTDeclarePartitionOptions.cpp
index 199a5e2dd26..744beffe8e1 100644
--- a/src/Parsers/MySQL/ASTDeclarePartitionOptions.cpp
+++ b/src/Parsers/MySQL/ASTDeclarePartitionOptions.cpp
@@ -49,30 +49,30 @@ static inline bool parsePartitionExpression(IParser::Pos & pos, std::string & ty
 {
     ASTPtr expression;
     ParserExpression p_expression;
-    if (!subpartition && ParserKeyword("LIST").ignore(pos, expected))
+    if (!subpartition && ParserKeyword(Keyword::LIST).ignore(pos, expected))
     {
         type = "list";
-        ParserKeyword("COLUMNS").ignore(pos, expected);
+        ParserKeyword(Keyword::COLUMNS).ignore(pos, expected);
         if (!p_expression.parse(pos, expression, expected))
             return false;
     }
-    else if (!subpartition && ParserKeyword("RANGE").ignore(pos, expected))
+    else if (!subpartition && ParserKeyword(Keyword::RANGE).ignore(pos, expected))
     {
         type = "range";
-        ParserKeyword("COLUMNS").ignore(pos, expected);
+        ParserKeyword(Keyword::COLUMNS).ignore(pos, expected);
         if (!p_expression.parse(pos, expression, expected))
             return false;
     }
     else
     {
-        if (ParserKeyword("LINEAR").ignore(pos, expected))
+        if (ParserKeyword(Keyword::LINEAR).ignore(pos, expected))
             type = "linear_";
 
-        if (ParserKeyword("KEY").ignore(pos, expected))
+        if (ParserKeyword(Keyword::KEY).ignore(pos, expected))
         {
             type += "key";
 
-            if (ParserKeyword("ALGORITHM").ignore(pos, expected))
+            if (ParserKeyword(Keyword::ALGORITHM).ignore(pos, expected))
             {
                 if (!ParserToken(TokenType::Equals).ignore(pos, expected))
                     return false;
@@ -93,7 +93,7 @@ static inline bool parsePartitionExpression(IParser::Pos & pos, std::string & ty
             if (!p_expression.parse(pos, expression, expected))
                 return false;
         }
-        else if (ParserKeyword("HASH").ignore(pos, expected))
+        else if (ParserKeyword(Keyword::HASH).ignore(pos, expected))
         {
             type += "hash";
             if (!p_expression.parse(pos, expression, expected))
@@ -117,25 +117,25 @@ bool ParserDeclarePartitionOptions::parseImpl(Pos & pos, ASTPtr & node, Expected
     ASTPtr subpartition_expression;
     ASTPtr declare_partitions;
 
-    if (!ParserKeyword("PARTITION BY").ignore(pos, expected))
+    if (!ParserKeyword(Keyword::PARTITION_BY).ignore(pos, expected))
         return false;
 
     if (!parsePartitionExpression(pos, partition_type, partition_expression, expected))
         return false;
 
-    if (ParserKeyword("PARTITIONS").ignore(pos, expected))
+    if (ParserKeyword(Keyword::PARTITIONS).ignore(pos, expected))
     {
         ParserLiteral p_literal;
         if (!p_literal.parse(pos, partition_numbers, expected))
             return false;
     }
 
-    if (ParserKeyword("SUBPARTITION BY").ignore(pos, expected))
+    if (ParserKeyword(Keyword::SUBPARTITION_BY).ignore(pos, expected))
     {
         if (!parsePartitionExpression(pos, subpartition_type, subpartition_expression, expected, true))
             return false;
 
-        if (ParserKeyword("SUBPARTITIONS").ignore(pos, expected))
+        if (ParserKeyword(Keyword::SUBPARTITIONS).ignore(pos, expected))
         {
             ParserLiteral p_literal;
             if (!p_literal.parse(pos, subpartition_numbers, expected))
diff --git a/src/Parsers/MySQL/ASTDeclareReference.cpp b/src/Parsers/MySQL/ASTDeclareReference.cpp
index 862d35e2b76..c8d6c4cd2ee 100644
--- a/src/Parsers/MySQL/ASTDeclareReference.cpp
+++ b/src/Parsers/MySQL/ASTDeclareReference.cpp
@@ -13,15 +13,15 @@ namespace MySQLParser
 
 bool parseReferenceOption(IParser::Pos & pos, ASTDeclareReference::ReferenceOption & option, Expected & expected)
 {
-    if (ParserKeyword("RESTRICT").ignore(pos, expected))
+    if (ParserKeyword(Keyword::RESTRICT).ignore(pos, expected))
         option = ASTDeclareReference::RESTRICT;
-    else if (ParserKeyword("CASCADE").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::CASCADE).ignore(pos, expected))
         option = ASTDeclareReference::CASCADE;
-    else if (ParserKeyword("SET NULL").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::SET_NULL).ignore(pos, expected))
         option = ASTDeclareReference::SET_NULL;
-    else if (ParserKeyword("NO ACTION").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::NO_ACTION).ignore(pos, expected))
         option = ASTDeclareReference::NO_ACTION;
-    else if (ParserKeyword("SET DEFAULT").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::SET_DEFAULT).ignore(pos, expected))
         option = ASTDeclareReference::SET_DEFAULT;
     else
         return false;
@@ -53,7 +53,7 @@ bool ParserDeclareReference::parseImpl(IParser::Pos & pos, ASTPtr & node, Expect
     ASTDeclareReference::ReferenceOption delete_option = ASTDeclareReference::RESTRICT;
     ASTDeclareReference::ReferenceOption update_option = ASTDeclareReference::RESTRICT;
 
-    if (!ParserKeyword("REFERENCES").ignore(pos, expected))
+    if (!ParserKeyword(Keyword::REFERENCES).ignore(pos, expected))
         return false;
 
     if (!p_identifier.parse(pos, table_name, expected))
@@ -62,13 +62,13 @@ bool ParserDeclareReference::parseImpl(IParser::Pos & pos, ASTPtr & node, Expect
     if (!p_expression.parse(pos, expression, expected))
         return false;
 
-    if (ParserKeyword("MATCH").ignore(pos, expected))
+    if (ParserKeyword(Keyword::MATCH).ignore(pos, expected))
     {
-        if (ParserKeyword("FULL").ignore(pos, expected))
+        if (ParserKeyword(Keyword::FULL).ignore(pos, expected))
             match_kind = ASTDeclareReference::MATCH_FULL;
-        else if (ParserKeyword("SIMPLE").ignore(pos, expected))
+        else if (ParserKeyword(Keyword::SIMPLE).ignore(pos, expected))
             match_kind = ASTDeclareReference::MATCH_SIMPLE;
-        else if (ParserKeyword("PARTIAL").ignore(pos, expected))
+        else if (ParserKeyword(Keyword::PARTIAL).ignore(pos, expected))
             match_kind = ASTDeclareReference::MATCH_PARTIAL;
         else
             return false;
@@ -76,12 +76,12 @@ bool ParserDeclareReference::parseImpl(IParser::Pos & pos, ASTPtr & node, Expect
 
     while (true)
     {
-        if (ParserKeyword("ON DELETE").ignore(pos, expected))
+        if (ParserKeyword(Keyword::ON_DELETE).ignore(pos, expected))
         {
             if (!parseReferenceOption(pos, delete_option, expected))
                 return false;
         }
-        else if (ParserKeyword("ON UPDATE").ignore(pos, expected))
+        else if (ParserKeyword(Keyword::ON_UPDATE).ignore(pos, expected))
         {
             if (!parseReferenceOption(pos, update_option, expected))
                 return false;
diff --git a/src/Parsers/MySQL/ASTDeclareSubPartition.cpp b/src/Parsers/MySQL/ASTDeclareSubPartition.cpp
index d77fba271c4..e0f8098501e 100644
--- a/src/Parsers/MySQL/ASTDeclareSubPartition.cpp
+++ b/src/Parsers/MySQL/ASTDeclareSubPartition.cpp
@@ -13,7 +13,7 @@ namespace MySQLParser
 
 bool ParserDeclareSubPartition::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    if (!ParserKeyword{"SUBPARTITION"}.ignore(pos, expected))
+    if (!ParserKeyword{Keyword::SUBPARTITION}.ignore(pos, expected))
         return false;
 
     ASTPtr options;
diff --git a/src/Parsers/MySQL/ASTDeclareTableOptions.cpp b/src/Parsers/MySQL/ASTDeclareTableOptions.cpp
index c903c7d2fa7..05008e79f19 100644
--- a/src/Parsers/MySQL/ASTDeclareTableOptions.cpp
+++ b/src/Parsers/MySQL/ASTDeclareTableOptions.cpp
@@ -23,7 +23,7 @@ protected:
     {
         if constexpr (allow_default)
         {
-            if (ParserKeyword("DEFAULT").ignore(pos, expected))
+            if (ParserKeyword(Keyword::DEFAULT).ignore(pos, expected))
             {
                 node = std::make_shared<ASTIdentifier>("DEFAULT");
                 return true;
@@ -49,11 +49,11 @@ protected:
         if (!p_identifier.parse(pos, node, expected))
             return false;
 
-        if (ParserKeyword("STORAGE").ignore(pos, expected))
+        if (ParserKeyword(Keyword::STORAGE).ignore(pos, expected))
         {
-            if (!ParserKeyword("DISK").ignore(pos, expected))
+            if (!ParserKeyword(Keyword::DISK).ignore(pos, expected))
             {
-                if (!ParserKeyword("MEMORY").ignore(pos, expected))
+                if (!ParserKeyword(Keyword::MEMORY).ignore(pos, expected))
                     return false;
             }
         }
diff --git a/src/Parsers/MySQL/ASTDropQuery.cpp b/src/Parsers/MySQL/ASTDropQuery.cpp
index 890451e3e55..0879186a775 100644
--- a/src/Parsers/MySQL/ASTDropQuery.cpp
+++ b/src/Parsers/MySQL/ASTDropQuery.cpp
@@ -23,20 +23,20 @@ ASTPtr ASTDropQuery::clone() const
 
 bool ParserDropQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_drop("DROP");
-    ParserKeyword s_truncate("TRUNCATE");
-    ParserKeyword s_table("TABLE");
-    ParserKeyword s_database("DATABASE");
-    ParserKeyword s_if_exists("IF EXISTS");
-    ParserKeyword s_view("VIEW");
-    ParserKeyword on("ON");
+    ParserKeyword s_drop(Keyword::DROP);
+    ParserKeyword s_truncate(Keyword::TRUNCATE);
+    ParserKeyword s_table(Keyword::TABLE);
+    ParserKeyword s_database(Keyword::DATABASE);
+    ParserKeyword s_if_exists(Keyword::IF_EXISTS);
+    ParserKeyword s_view(Keyword::VIEW);
+    ParserKeyword on(Keyword::ON);
     ParserIdentifier name_p(false);
 
-    ParserKeyword s_event("EVENT");
-    ParserKeyword s_function("FUNCTION");
-    ParserKeyword s_index("INDEX");
-    ParserKeyword s_server("SERVER");
-    ParserKeyword s_trigger("TRIGGER");
+    ParserKeyword s_event(Keyword::EVENT);
+    ParserKeyword s_function(Keyword::FUNCTION);
+    ParserKeyword s_index(Keyword::INDEX);
+    ParserKeyword s_server(Keyword::SERVER);
+    ParserKeyword s_trigger(Keyword::TRIGGER);
 
     auto query = std::make_shared<ASTDropQuery>();
     node = query;
diff --git a/src/Parsers/ParserAlterNamedCollectionQuery.cpp b/src/Parsers/ParserAlterNamedCollectionQuery.cpp
index 9b33aa253ba..3c6f764a190 100644
--- a/src/Parsers/ParserAlterNamedCollectionQuery.cpp
+++ b/src/Parsers/ParserAlterNamedCollectionQuery.cpp
@@ -11,15 +11,16 @@ namespace DB
 
 bool ParserAlterNamedCollectionQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_alter("ALTER");
-    ParserKeyword s_collection("NAMED COLLECTION");
-    ParserKeyword s_if_exists("IF EXISTS");
-    ParserKeyword s_on("ON");
-    ParserKeyword s_delete("DELETE");
+    ParserKeyword s_alter(Keyword::ALTER);
+    ParserKeyword s_collection(Keyword::NAMED_COLLECTION);
+    ParserKeyword s_if_exists(Keyword::IF_EXISTS);
+    ParserKeyword s_on(Keyword::ON);
+    ParserKeyword s_delete(Keyword::DELETE);
+
     ParserIdentifier name_p;
-    ParserKeyword s_set("SET");
-    ParserKeyword s_overridable("OVERRIDABLE");
-    ParserKeyword s_not_overridable("NOT OVERRIDABLE");
+    ParserKeyword s_set(Keyword::SET);
+    ParserKeyword s_overridable(Keyword::OVERRIDABLE);
+    ParserKeyword s_not_overridable(Keyword::NOT_OVERRIDABLE);
     ParserToken s_comma(TokenType::Comma);
 
     String cluster_str;
diff --git a/src/Parsers/ParserAlterQuery.cpp b/src/Parsers/ParserAlterQuery.cpp
index 4bc95e67afb..c0c43381585 100644
--- a/src/Parsers/ParserAlterQuery.cpp
+++ b/src/Parsers/ParserAlterQuery.cpp
@@ -24,97 +24,97 @@ bool ParserAlterCommand::parseImpl(Pos & pos, ASTPtr & node, Expected & expected
     auto command = std::make_shared<ASTAlterCommand>();
     node = command;
 
-    ParserKeyword s_add_column("ADD COLUMN");
-    ParserKeyword s_drop_column("DROP COLUMN");
-    ParserKeyword s_clear_column("CLEAR COLUMN");
-    ParserKeyword s_modify_column("MODIFY COLUMN");
-    ParserKeyword s_alter_column("ALTER COLUMN");
-    ParserKeyword s_rename_column("RENAME COLUMN");
-    ParserKeyword s_comment_column("COMMENT COLUMN");
-    ParserKeyword s_materialize_column("MATERIALIZE COLUMN");
+    ParserKeyword s_add_column(Keyword::ADD_COLUMN);
+    ParserKeyword s_drop_column(Keyword::DROP_COLUMN);
+    ParserKeyword s_clear_column(Keyword::CLEAR_COLUMN);
+    ParserKeyword s_modify_column(Keyword::MODIFY_COLUMN);
+    ParserKeyword s_alter_column(Keyword::ALTER_COLUMN);
+    ParserKeyword s_rename_column(Keyword::RENAME_COLUMN);
+    ParserKeyword s_comment_column(Keyword::COMMENT_COLUMN);
+    ParserKeyword s_materialize_column(Keyword::MATERIALIZE_COLUMN);
 
-    ParserKeyword s_modify_order_by("MODIFY ORDER BY");
-    ParserKeyword s_modify_sample_by("MODIFY SAMPLE BY");
-    ParserKeyword s_modify_ttl("MODIFY TTL");
-    ParserKeyword s_materialize_ttl("MATERIALIZE TTL");
-    ParserKeyword s_modify_setting("MODIFY SETTING");
-    ParserKeyword s_reset_setting("RESET SETTING");
-    ParserKeyword s_modify_query("MODIFY QUERY");
-    ParserKeyword s_modify_sql_security("MODIFY SQL SECURITY");
-    ParserKeyword s_modify_refresh("MODIFY REFRESH");
+    ParserKeyword s_modify_order_by(Keyword::MODIFY_ORDER_BY);
+    ParserKeyword s_modify_sample_by(Keyword::MODIFY_SAMPLE_BY);
+    ParserKeyword s_modify_ttl(Keyword::MODIFY_TTL);
+    ParserKeyword s_materialize_ttl(Keyword::MATERIALIZE_TTL);
+    ParserKeyword s_modify_setting(Keyword::MODIFY_SETTING);
+    ParserKeyword s_reset_setting(Keyword::RESET_SETTING);
+    ParserKeyword s_modify_query(Keyword::MODIFY_QUERY);
+    ParserKeyword s_modify_sql_security(Keyword::MODIFY_SQL_SECURITY);
+    ParserKeyword s_modify_refresh(Keyword::MODIFY_REFRESH);
 
-    ParserKeyword s_add_index("ADD INDEX");
-    ParserKeyword s_drop_index("DROP INDEX");
-    ParserKeyword s_clear_index("CLEAR INDEX");
-    ParserKeyword s_materialize_index("MATERIALIZE INDEX");
+    ParserKeyword s_add_index(Keyword::ADD_INDEX);
+    ParserKeyword s_drop_index(Keyword::DROP_INDEX);
+    ParserKeyword s_clear_index(Keyword::CLEAR_INDEX);
+    ParserKeyword s_materialize_index(Keyword::MATERIALIZE_INDEX);
 
-    ParserKeyword s_add_statistic("ADD STATISTIC");
-    ParserKeyword s_drop_statistic("DROP STATISTIC");
-    ParserKeyword s_clear_statistic("CLEAR STATISTIC");
-    ParserKeyword s_materialize_statistic("MATERIALIZE STATISTIC");
+    ParserKeyword s_add_statistic(Keyword::ADD_STATISTIC);
+    ParserKeyword s_drop_statistic(Keyword::DROP_STATISTIC);
+    ParserKeyword s_clear_statistic(Keyword::CLEAR_STATISTIC);
+    ParserKeyword s_materialize_statistic(Keyword::MATERIALIZE_STATISTIC);
 
-    ParserKeyword s_add_constraint("ADD CONSTRAINT");
-    ParserKeyword s_drop_constraint("DROP CONSTRAINT");
+    ParserKeyword s_add_constraint(Keyword::ADD_CONSTRAINT);
+    ParserKeyword s_drop_constraint(Keyword::DROP_CONSTRAINT);
 
-    ParserKeyword s_add_projection("ADD PROJECTION");
-    ParserKeyword s_drop_projection("DROP PROJECTION");
-    ParserKeyword s_clear_projection("CLEAR PROJECTION");
-    ParserKeyword s_materialize_projection("MATERIALIZE PROJECTION");
-    ParserKeyword s_modify_comment("MODIFY COMMENT");
+    ParserKeyword s_add_projection(Keyword::ADD_PROJECTION);
+    ParserKeyword s_drop_projection(Keyword::DROP_PROJECTION);
+    ParserKeyword s_clear_projection(Keyword::CLEAR_PROJECTION);
+    ParserKeyword s_materialize_projection(Keyword::MATERIALIZE_PROJECTION);
+    ParserKeyword s_modify_comment(Keyword::MODIFY_COMMENT);
 
-    ParserKeyword s_add("ADD");
-    ParserKeyword s_drop("DROP");
-    ParserKeyword s_modify("MODIFY");
+    ParserKeyword s_add(Keyword::ADD);
+    ParserKeyword s_drop(Keyword::DROP);
+    ParserKeyword s_modify(Keyword::MODIFY);
 
-    ParserKeyword s_attach_partition("ATTACH PARTITION");
-    ParserKeyword s_attach_part("ATTACH PART");
-    ParserKeyword s_detach_partition("DETACH PARTITION");
-    ParserKeyword s_detach_part("DETACH PART");
-    ParserKeyword s_drop_partition("DROP PARTITION");
-    ParserKeyword s_drop_part("DROP PART");
-    ParserKeyword s_forget_partition("FORGET PARTITION");
-    ParserKeyword s_move_partition("MOVE PARTITION");
-    ParserKeyword s_move_part("MOVE PART");
-    ParserKeyword s_drop_detached_partition("DROP DETACHED PARTITION");
-    ParserKeyword s_drop_detached_part("DROP DETACHED PART");
-    ParserKeyword s_fetch_partition("FETCH PARTITION");
-    ParserKeyword s_fetch_part("FETCH PART");
-    ParserKeyword s_replace_partition("REPLACE PARTITION");
-    ParserKeyword s_freeze("FREEZE");
-    ParserKeyword s_unfreeze("UNFREEZE");
-    ParserKeyword s_partition("PARTITION");
+    ParserKeyword s_attach_partition(Keyword::ATTACH_PARTITION);
+    ParserKeyword s_attach_part(Keyword::ATTACH_PART);
+    ParserKeyword s_detach_partition(Keyword::DETACH_PARTITION);
+    ParserKeyword s_detach_part(Keyword::DETACH_PART);
+    ParserKeyword s_drop_partition(Keyword::DROP_PARTITION);
+    ParserKeyword s_drop_part(Keyword::DROP_PART);
+    ParserKeyword s_forget_partition(Keyword::FORGET_PARTITION);
+    ParserKeyword s_move_partition(Keyword::MOVE_PARTITION);
+    ParserKeyword s_move_part(Keyword::MOVE_PART);
+    ParserKeyword s_drop_detached_partition(Keyword::DROP_DETACHED_PARTITION);
+    ParserKeyword s_drop_detached_part(Keyword::DROP_DETACHED_PART);
+    ParserKeyword s_fetch_partition(Keyword::FETCH_PARTITION);
+    ParserKeyword s_fetch_part(Keyword::FETCH_PART);
+    ParserKeyword s_replace_partition(Keyword::REPLACE_PARTITION);
+    ParserKeyword s_freeze(Keyword::FREEZE);
+    ParserKeyword s_unfreeze(Keyword::UNFREEZE);
+    ParserKeyword s_partition(Keyword::PARTITION);
 
-    ParserKeyword s_first("FIRST");
-    ParserKeyword s_after("AFTER");
-    ParserKeyword s_if_not_exists("IF NOT EXISTS");
-    ParserKeyword s_if_exists("IF EXISTS");
-    ParserKeyword s_from("FROM");
-    ParserKeyword s_in_partition("IN PARTITION");
-    ParserKeyword s_with("WITH");
-    ParserKeyword s_name("NAME");
+    ParserKeyword s_first(Keyword::FIRST);
+    ParserKeyword s_after(Keyword::AFTER);
+    ParserKeyword s_if_not_exists(Keyword::IF_NOT_EXISTS);
+    ParserKeyword s_if_exists(Keyword::IF_EXISTS);
+    ParserKeyword s_from(Keyword::FROM);
+    ParserKeyword s_in_partition(Keyword::IN_PARTITION);
+    ParserKeyword s_with(Keyword::WITH);
+    ParserKeyword s_name(Keyword::NAME);
 
-    ParserKeyword s_to_disk("TO DISK");
-    ParserKeyword s_to_volume("TO VOLUME");
-    ParserKeyword s_to_table("TO TABLE");
-    ParserKeyword s_to_shard("TO SHARD");
+    ParserKeyword s_to_disk(Keyword::TO_DISK);
+    ParserKeyword s_to_volume(Keyword::TO_VOLUME);
+    ParserKeyword s_to_table(Keyword::TO_TABLE);
+    ParserKeyword s_to_shard(Keyword::TO_SHARD);
 
-    ParserKeyword s_delete("DELETE");
-    ParserKeyword s_update("UPDATE");
-    ParserKeyword s_where("WHERE");
-    ParserKeyword s_to("TO");
+    ParserKeyword s_delete(Keyword::DELETE);
+    ParserKeyword s_update(Keyword::UPDATE);
+    ParserKeyword s_where(Keyword::WHERE);
+    ParserKeyword s_to(Keyword::TO);
 
-    ParserKeyword s_remove("REMOVE");
-    ParserKeyword s_default("DEFAULT");
-    ParserKeyword s_materialized("MATERIALIZED");
-    ParserKeyword s_alias("ALIAS");
-    ParserKeyword s_comment("COMMENT");
-    ParserKeyword s_codec("CODEC");
-    ParserKeyword s_ttl("TTL");
-    ParserKeyword s_settings("SETTINGS");
+    ParserKeyword s_remove(Keyword::REMOVE);
+    ParserKeyword s_default(Keyword::DEFAULT);
+    ParserKeyword s_materialized(Keyword::MATERIALIZED);
+    ParserKeyword s_alias(Keyword::ALIAS);
+    ParserKeyword s_comment(Keyword::COMMENT);
+    ParserKeyword s_codec(Keyword::CODEC);
+    ParserKeyword s_ttl(Keyword::TTL);
+    ParserKeyword s_settings(Keyword::SETTINGS);
 
-    ParserKeyword s_remove_ttl("REMOVE TTL");
-    ParserKeyword s_remove_sample_by("REMOVE SAMPLE BY");
-    ParserKeyword s_apply_deleted_mask("APPLY DELETED MASK");
+    ParserKeyword s_remove_ttl(Keyword::REMOVE_TTL);
+    ParserKeyword s_remove_sample_by(Keyword::REMOVE_SAMPLE_BY);
+    ParserKeyword s_apply_deleted_mask(Keyword::APPLY_DELETED_MASK);
 
     ParserToken parser_opening_round_bracket(TokenType::OpeningRoundBracket);
     ParserToken parser_closing_round_bracket(TokenType::ClosingRoundBracket);
@@ -720,19 +720,19 @@ bool ParserAlterCommand::parseImpl(Pos & pos, ASTPtr & node, Expected & expected
                 if (s_remove.ignore(pos, expected))
                 {
                     if (s_default.ignore(pos, expected))
-                        command->remove_property = "DEFAULT";
+                        command->remove_property = toStringView(Keyword::DEFAULT);
                     else if (s_materialized.ignore(pos, expected))
-                        command->remove_property = "MATERIALIZED";
+                        command->remove_property = toStringView(Keyword::MATERIALIZED);
                     else if (s_alias.ignore(pos, expected))
-                        command->remove_property = "ALIAS";
+                        command->remove_property = toStringView(Keyword::ALIAS);
                     else if (s_comment.ignore(pos, expected))
-                        command->remove_property = "COMMENT";
+                        command->remove_property = toStringView(Keyword::COMMENT);
                     else if (s_codec.ignore(pos, expected))
-                        command->remove_property = "CODEC";
+                        command->remove_property = toStringView(Keyword::CODEC);
                     else if (s_ttl.ignore(pos, expected))
-                        command->remove_property = "TTL";
+                        command->remove_property = toStringView(Keyword::TTL);
                     else if (s_settings.ignore(pos, expected))
-                        command->remove_property = "SETTINGS";
+                        command->remove_property = toStringView(Keyword::SETTINGS);
                     else
                         return false;
                 }
@@ -981,9 +981,9 @@ bool ParserAlterQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
     auto query = std::make_shared<ASTAlterQuery>();
     node = query;
 
-    ParserKeyword s_alter_table("ALTER TABLE");
-    ParserKeyword s_alter_temporary_table("ALTER TEMPORARY TABLE");
-    ParserKeyword s_alter_database("ALTER DATABASE");
+    ParserKeyword s_alter_table(Keyword::ALTER_TABLE);
+    ParserKeyword s_alter_temporary_table(Keyword::ALTER_TEMPORARY_TABLE);
+    ParserKeyword s_alter_database(Keyword::ALTER_DATABASE);
 
     ASTAlterQuery::AlterObjectType alter_object_type;
 
@@ -1009,7 +1009,7 @@ bool ParserAlterQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
             return false;
 
         String cluster_str;
-        if (ParserKeyword{"ON"}.ignore(pos, expected))
+        if (ParserKeyword(Keyword::ON).ignore(pos, expected))
         {
             if (!ASTQueryWithOnCluster::parse(pos, cluster_str, expected))
                 return false;
diff --git a/src/Parsers/ParserBackupQuery.cpp b/src/Parsers/ParserBackupQuery.cpp
index cbe4567ee90..5eab0cb288d 100644
--- a/src/Parsers/ParserBackupQuery.cpp
+++ b/src/Parsers/ParserBackupQuery.cpp
@@ -24,7 +24,7 @@ namespace
 
     bool parsePartitions(IParser::Pos & pos, Expected & expected, std::optional<ASTs> & partitions)
     {
-        if (!ParserKeyword{"PARTITION"}.ignore(pos, expected) && !ParserKeyword{"PARTITIONS"}.ignore(pos, expected))
+        if (!ParserKeyword(Keyword::PARTITION).ignore(pos, expected) && !ParserKeyword(Keyword::PARTITIONS).ignore(pos, expected))
             return false;
 
         ASTs result;
@@ -47,7 +47,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (!ParserKeyword{"EXCEPT DATABASE"}.ignore(pos, expected) && !ParserKeyword{"EXCEPT DATABASES"}.ignore(pos, expected))
+            if (!ParserKeyword(Keyword::EXCEPT_DATABASE).ignore(pos, expected) && !ParserKeyword(Keyword::EXCEPT_DATABASES).ignore(pos, expected))
                 return false;
 
             std::set<String> result;
@@ -71,7 +71,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (!ParserKeyword{"EXCEPT TABLE"}.ignore(pos, expected) && !ParserKeyword{"EXCEPT TABLES"}.ignore(pos, expected))
+            if (!ParserKeyword(Keyword::EXCEPT_TABLE).ignore(pos, expected) && !ParserKeyword(Keyword::EXCEPT_TABLES).ignore(pos, expected))
                 return false;
 
             std::set<DatabaseAndTableName> result;
@@ -107,8 +107,8 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (ParserKeyword{"TABLE"}.ignore(pos, expected) || ParserKeyword{"DICTIONARY"}.ignore(pos, expected) ||
-                ParserKeyword{"VIEW"}.ignore(pos, expected))
+            if (ParserKeyword(Keyword::TABLE).ignore(pos, expected) || ParserKeyword(Keyword::DICTIONARY).ignore(pos, expected) ||
+                ParserKeyword(Keyword::VIEW).ignore(pos, expected))
             {
                 element.type = ElementType::TABLE;
                 if (!parseDatabaseAndTableName(pos, expected, element.database_name, element.table_name))
@@ -116,7 +116,7 @@ namespace
 
                 element.new_database_name = element.database_name;
                 element.new_table_name = element.table_name;
-                if (ParserKeyword("AS").ignore(pos, expected))
+                if (ParserKeyword(Keyword::AS).ignore(pos, expected))
                 {
                     if (!parseDatabaseAndTableName(pos, expected, element.new_database_name, element.new_table_name))
                         return false;
@@ -126,7 +126,7 @@ namespace
                 return true;
             }
 
-            if (ParserKeyword{"TEMPORARY TABLE"}.ignore(pos, expected))
+            if (ParserKeyword(Keyword::TEMPORARY_TABLE).ignore(pos, expected))
             {
                 element.type = ElementType::TEMPORARY_TABLE;
 
@@ -136,7 +136,7 @@ namespace
                 element.table_name = getIdentifierName(ast);
                 element.new_table_name = element.table_name;
 
-                if (ParserKeyword("AS").ignore(pos, expected))
+                if (ParserKeyword(Keyword::AS).ignore(pos, expected))
                 {
                     ast = nullptr;
                     if (!ParserIdentifier{}.parse(pos, ast, expected))
@@ -147,7 +147,7 @@ namespace
                 return true;
             }
 
-            if (ParserKeyword{"DATABASE"}.ignore(pos, expected))
+            if (ParserKeyword(Keyword::DATABASE).ignore(pos, expected))
             {
                 element.type = ElementType::DATABASE;
 
@@ -157,7 +157,7 @@ namespace
                 element.database_name = getIdentifierName(ast);
                 element.new_database_name = element.database_name;
 
-                if (ParserKeyword("AS").ignore(pos, expected))
+                if (ParserKeyword(Keyword::AS).ignore(pos, expected))
                 {
                     ast = nullptr;
                     if (!ParserIdentifier{}.parse(pos, ast, expected))
@@ -169,7 +169,7 @@ namespace
                 return true;
             }
 
-            if (ParserKeyword{"ALL"}.ignore(pos, expected))
+            if (ParserKeyword(Keyword::ALL).ignore(pos, expected))
             {
                 element.type = ElementType::ALL;
                 parseExceptDatabases(pos, expected, element.except_databases);
@@ -219,7 +219,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            return ParserKeyword{"base_backup"}.ignore(pos, expected)
+            return ParserKeyword{Keyword::BASE_BACKUP}.ignore(pos, expected)
                 && ParserToken(TokenType::Equals).ignore(pos, expected)
                 && parseBackupName(pos, expected, base_backup_name);
         });
@@ -234,7 +234,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            return ParserKeyword{"cluster_host_ids"}.ignore(pos, expected)
+            return ParserKeyword{Keyword::CLUSTER_HOST_IDS}.ignore(pos, expected)
                 && ParserToken(TokenType::Equals).ignore(pos, expected)
                 && parseClusterHostIDs(pos, expected, cluster_host_ids);
         });
@@ -244,7 +244,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            if (!ParserKeyword{"SETTINGS"}.ignore(pos, expected))
+            if (!ParserKeyword(Keyword::SETTINGS).ignore(pos, expected))
                 return false;
 
             SettingsChanges settings_changes;
@@ -291,9 +291,9 @@ namespace
     bool parseSyncOrAsync(IParser::Pos & pos, Expected & expected, ASTPtr & settings)
     {
         bool async;
-        if (ParserKeyword{"ASYNC"}.ignore(pos, expected))
+        if (ParserKeyword(Keyword::ASYNC).ignore(pos, expected))
             async = true;
-        else if (ParserKeyword{"SYNC"}.ignore(pos, expected))
+        else if (ParserKeyword(Keyword::SYNC).ignore(pos, expected))
             async = false;
         else
             return false;
@@ -318,7 +318,7 @@ namespace
     {
         return IParserBase::wrapParseImpl(pos, [&]
         {
-            return ParserKeyword{"ON"}.ignore(pos, expected) && ASTQueryWithOnCluster::parse(pos, cluster, expected);
+            return ParserKeyword(Keyword::ON).ignore(pos, expected) && ASTQueryWithOnCluster::parse(pos, cluster, expected);
         });
     }
 }
@@ -327,9 +327,9 @@ namespace
 bool ParserBackupQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
     Kind kind;
-    if (ParserKeyword{"BACKUP"}.ignore(pos, expected))
+    if (ParserKeyword(Keyword::BACKUP).ignore(pos, expected))
         kind = Kind::BACKUP;
-    else if (ParserKeyword{"RESTORE"}.ignore(pos, expected))
+    else if (ParserKeyword(Keyword::RESTORE).ignore(pos, expected))
         kind = Kind::RESTORE;
     else
         return false;
@@ -341,7 +341,7 @@ bool ParserBackupQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
     String cluster;
     parseOnCluster(pos, expected, cluster);
 
-    if (!ParserKeyword{(kind == Kind::BACKUP) ? "TO" : "FROM"}.ignore(pos, expected))
+    if (!ParserKeyword((kind == Kind::BACKUP) ? Keyword::TO : Keyword::FROM).ignore(pos, expected))
         return false;
 
     ASTPtr backup_name;
diff --git a/src/Parsers/ParserCase.cpp b/src/Parsers/ParserCase.cpp
index 521eec212da..e73ff0e0844 100644
--- a/src/Parsers/ParserCase.cpp
+++ b/src/Parsers/ParserCase.cpp
@@ -10,11 +10,11 @@ namespace DB
 
 bool ParserCase::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_case{"CASE"};
-    ParserKeyword s_when{"WHEN"};
-    ParserKeyword s_then{"THEN"};
-    ParserKeyword s_else{"ELSE"};
-    ParserKeyword s_end{ "END"};
+    ParserKeyword s_case{Keyword::CASE};
+    ParserKeyword s_when{Keyword::WHEN};
+    ParserKeyword s_then{Keyword::THEN};
+    ParserKeyword s_else{Keyword::ELSE};
+    ParserKeyword s_end{ Keyword::END};
     ParserExpressionWithOptionalAlias p_expr{false};
 
     if (!s_case.ignore(pos, expected))
diff --git a/src/Parsers/ParserCheckQuery.cpp b/src/Parsers/ParserCheckQuery.cpp
index a840cfbae5e..42716ba7f2c 100644
--- a/src/Parsers/ParserCheckQuery.cpp
+++ b/src/Parsers/ParserCheckQuery.cpp
@@ -12,7 +12,7 @@ namespace DB
 
 bool ParserCheckQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_check_table("CHECK ALL TABLES");
+    ParserKeyword s_check_table(Keyword::CHECK_ALL_TABLES);
     if (s_check_table.ignore(pos, expected))
     {
         auto query = std::make_shared<ASTCheckAllTablesQuery>();
@@ -25,9 +25,9 @@ bool ParserCheckQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 
 bool ParserCheckQuery::parseCheckTable(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_check_table("CHECK TABLE");
-    ParserKeyword s_partition("PARTITION");
-    ParserKeyword s_part("PART");
+    ParserKeyword s_check_table(Keyword::CHECK_TABLE);
+    ParserKeyword s_partition(Keyword::PARTITION);
+    ParserKeyword s_part(Keyword::PART);
     ParserToken s_dot(TokenType::Dot);
 
     ParserPartition partition_parser;
diff --git a/src/Parsers/ParserCreateFunctionQuery.cpp b/src/Parsers/ParserCreateFunctionQuery.cpp
index 2b3cf98a8a7..b912016a439 100644
--- a/src/Parsers/ParserCreateFunctionQuery.cpp
+++ b/src/Parsers/ParserCreateFunctionQuery.cpp
@@ -13,13 +13,13 @@ namespace DB
 
 bool ParserCreateFunctionQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_create("CREATE");
-    ParserKeyword s_function("FUNCTION");
-    ParserKeyword s_or_replace("OR REPLACE");
-    ParserKeyword s_if_not_exists("IF NOT EXISTS");
-    ParserKeyword s_on("ON");
+    ParserKeyword s_create(Keyword::CREATE);
+    ParserKeyword s_function(Keyword::FUNCTION);
+    ParserKeyword s_or_replace(Keyword::OR_REPLACE);
+    ParserKeyword s_if_not_exists(Keyword::IF_NOT_EXISTS);
+    ParserKeyword s_on(Keyword::ON);
     ParserIdentifier function_name_p;
-    ParserKeyword s_as("AS");
+    ParserKeyword s_as(Keyword::AS);
     ParserExpression lambda_p;
 
     ASTPtr function_name;
diff --git a/src/Parsers/ParserCreateIndexQuery.cpp b/src/Parsers/ParserCreateIndexQuery.cpp
index e654656b68a..3b1b9d8ec84 100644
--- a/src/Parsers/ParserCreateIndexQuery.cpp
+++ b/src/Parsers/ParserCreateIndexQuery.cpp
@@ -15,11 +15,12 @@ namespace DB
 
 bool ParserCreateIndexDeclaration::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_type("TYPE");
-    ParserKeyword s_granularity("GRANULARITY");
+    ParserKeyword s_type(Keyword::TYPE);
+    ParserKeyword s_granularity(Keyword::GRANULARITY);
     ParserToken open(TokenType::OpeningRoundBracket);
     ParserToken close(TokenType::ClosingRoundBracket);
     ParserOrderByExpressionList order_list;
+
     ParserDataType data_type_p;
     ParserExpression expression_p;
     ParserUnsignedInteger granularity_p;
@@ -79,11 +80,12 @@ bool ParserCreateIndexQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expect
     auto query = std::make_shared<ASTCreateIndexQuery>();
     node = query;
 
-    ParserKeyword s_create("CREATE");
-    ParserKeyword s_unique("UNIQUE");
-    ParserKeyword s_index("INDEX");
-    ParserKeyword s_if_not_exists("IF NOT EXISTS");
-    ParserKeyword s_on("ON");
+    ParserKeyword s_create(Keyword::CREATE);
+    ParserKeyword s_unique(Keyword::UNIQUE);
+    ParserKeyword s_index(Keyword::INDEX);
+    ParserKeyword s_if_not_exists(Keyword::IF_NOT_EXISTS);
+    ParserKeyword s_on(Keyword::ON);
+
     ParserIdentifier index_name_p;
     ParserCreateIndexDeclaration parser_create_idx_decl;
 
diff --git a/src/Parsers/ParserCreateQuery.cpp b/src/Parsers/ParserCreateQuery.cpp
index 8ebadf4606f..f2c09e9b050 100644
--- a/src/Parsers/ParserCreateQuery.cpp
+++ b/src/Parsers/ParserCreateQuery.cpp
@@ -40,7 +40,7 @@ namespace
 
 ASTPtr parseComment(IParser::Pos & pos, Expected & expected)
 {
-    ParserKeyword s_comment("COMMENT");
+    ParserKeyword s_comment(Keyword::COMMENT);
     ParserStringLiteral string_literal_parser;
     ASTPtr comment;
 
@@ -88,7 +88,11 @@ bool ParserNestedTable::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 bool ParserSQLSecurity::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
     ParserToken s_eq(TokenType::Equals);
-    ParserKeyword s_definer("DEFINER");
+    ParserKeyword s_definer(Keyword::DEFINER);
+    ParserKeyword s_current_user{Keyword::CURRENT_USER};
+    ParserKeyword s_sql_security{Keyword::SQL_SECURITY};
+    ParserKeyword s_invoker{Keyword::INVOKER};
+    ParserKeyword s_none{Keyword::NONE};
 
     bool is_definer_current_user = false;
     ASTPtr definer;
@@ -99,7 +103,7 @@ bool ParserSQLSecurity::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
         if (!definer && s_definer.ignore(pos, expected))
         {
             s_eq.ignore(pos, expected);
-            if (ParserKeyword{"CURRENT_USER"}.ignore(pos, expected))
+            if (s_current_user.ignore(pos, expected))
                 is_definer_current_user = true;
             else if (!ParserUserNameWithHost{}.parse(pos, definer, expected))
                 return false;
@@ -107,13 +111,13 @@ bool ParserSQLSecurity::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
             continue;
         }
 
-        if (!type && ParserKeyword{"SQL SECURITY"}.ignore(pos, expected))
+        if (!type && s_sql_security.ignore(pos, expected))
         {
             if (s_definer.ignore(pos, expected))
                 type = SQLSecurityType::DEFINER;
-            else if (ParserKeyword{"INVOKER"}.ignore(pos, expected))
+            else if (s_invoker.ignore(pos, expected))
                 type = SQLSecurityType::INVOKER;
-            else if (ParserKeyword{"NONE"}.ignore(pos, expected))
+            else if (s_none.ignore(pos, expected))
                 type = SQLSecurityType::NONE;
             else
                 return false;
@@ -170,8 +174,8 @@ bool ParserNameList::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 
 bool ParserIndexDeclaration::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_type("TYPE");
-    ParserKeyword s_granularity("GRANULARITY");
+    ParserKeyword s_type(Keyword::TYPE);
+    ParserKeyword s_granularity(Keyword::GRANULARITY);
 
     ParserIdentifier name_p;
     ParserDataType data_type_p;
@@ -225,7 +229,7 @@ bool ParserIndexDeclaration::parseImpl(Pos & pos, ASTPtr & node, Expected & expe
 
 bool ParserStatisticDeclaration::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_type("TYPE");
+    ParserKeyword s_type(Keyword::TYPE);
 
     ParserList columns_p(std::make_unique<ParserIdentifier>(), std::make_unique<ParserToken>(TokenType::Comma), false);
     ParserIdentifier type_p;
@@ -252,8 +256,8 @@ bool ParserStatisticDeclaration::parseImpl(Pos & pos, ASTPtr & node, Expected &
 
 bool ParserConstraintDeclaration::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_check("CHECK");
-    ParserKeyword s_assume("ASSUME");
+    ParserKeyword s_check(Keyword::CHECK);
+    ParserKeyword s_assume(Keyword::ASSUME);
 
     ParserIdentifier name_p;
     ParserExpression expression_p;
@@ -317,7 +321,7 @@ bool ParserProjectionDeclaration::parseImpl(Pos & pos, ASTPtr & node, Expected &
 
 bool ParserForeignKeyDeclaration::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_references("REFERENCES");
+    ParserKeyword s_references(Keyword::REFERENCES);
     ParserCompoundIdentifier table_name_p(true, true);
     ParserExpression expression_p;
 
@@ -336,20 +340,20 @@ bool ParserForeignKeyDeclaration::parseImpl(Pos & pos, ASTPtr & node, Expected &
     if (!expression_p.parse(pos, expr, expected))
         return false;
 
-    ParserKeyword s_on("ON");
+    ParserKeyword s_on(Keyword::ON);
     while (s_on.ignore(pos, expected))
     {
-        ParserKeyword s_delete("DELETE");
-        ParserKeyword s_update("UPDATE");
+        ParserKeyword s_delete(Keyword::DELETE);
+        ParserKeyword s_update(Keyword::UPDATE);
 
         if (!s_delete.ignore(pos, expected) && !s_update.ignore(pos, expected))
             return false;
 
-        ParserKeyword s_restrict("RESTRICT");
-        ParserKeyword s_cascade("CASCADE");
-        ParserKeyword s_set_null("SET NULL");
-        ParserKeyword s_no_action("NO ACTION");
-        ParserKeyword s_set_default("SET DEFAULT");
+        ParserKeyword s_restrict(Keyword::RESTRICT);
+        ParserKeyword s_cascade(Keyword::CASCADE);
+        ParserKeyword s_set_null(Keyword::SET_NULL);
+        ParserKeyword s_no_action(Keyword::NO_ACTION);
+        ParserKeyword s_set_default(Keyword::SET_DEFAULT);
 
         if (!s_restrict.ignore(pos, expected) && !s_cascade.ignore(pos, expected) &&
             !s_set_null.ignore(pos, expected) && !s_no_action.ignore(pos, expected) &&
@@ -368,11 +372,11 @@ bool ParserForeignKeyDeclaration::parseImpl(Pos & pos, ASTPtr & node, Expected &
 
 bool ParserTablePropertyDeclaration::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_index("INDEX");
-    ParserKeyword s_constraint("CONSTRAINT");
-    ParserKeyword s_projection("PROJECTION");
-    ParserKeyword s_foreign_key("FOREIGN KEY");
-    ParserKeyword s_primary_key("PRIMARY KEY");
+    ParserKeyword s_index(Keyword::INDEX);
+    ParserKeyword s_constraint(Keyword::CONSTRAINT);
+    ParserKeyword s_projection(Keyword::PROJECTION);
+    ParserKeyword s_foreign_key(Keyword::FOREIGN_KEY);
+    ParserKeyword s_primary_key(Keyword::PRIMARY_KEY);
 
     ParserIndexDeclaration index_p;
     ParserConstraintDeclaration constraint_p;
@@ -512,14 +516,14 @@ bool ParserTablePropertiesDeclarationList::parseImpl(Pos & pos, ASTPtr & node, E
 
 bool ParserStorage::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_engine("ENGINE");
+    ParserKeyword s_engine(Keyword::ENGINE);
     ParserToken s_eq(TokenType::Equals);
-    ParserKeyword s_partition_by("PARTITION BY");
-    ParserKeyword s_primary_key("PRIMARY KEY");
-    ParserKeyword s_order_by("ORDER BY");
-    ParserKeyword s_sample_by("SAMPLE BY");
-    ParserKeyword s_ttl("TTL");
-    ParserKeyword s_settings("SETTINGS");
+    ParserKeyword s_partition_by(Keyword::PARTITION_BY);
+    ParserKeyword s_primary_key(Keyword::PRIMARY_KEY);
+    ParserKeyword s_order_by(Keyword::ORDER_BY);
+    ParserKeyword s_sample_by(Keyword::SAMPLE_BY);
+    ParserKeyword s_ttl(Keyword::TTL);
+    ParserKeyword s_settings(Keyword::SETTINGS);
 
     ParserIdentifierWithOptionalParameters ident_with_optional_params_p;
     ParserExpression expression_p;
@@ -651,16 +655,16 @@ bool ParserStorage::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 
 bool ParserCreateTableQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_create("CREATE");
-    ParserKeyword s_attach("ATTACH");
-    ParserKeyword s_replace("REPLACE");
-    ParserKeyword s_or_replace("OR REPLACE");
-    ParserKeyword s_temporary("TEMPORARY");
-    ParserKeyword s_table("TABLE");
-    ParserKeyword s_if_not_exists("IF NOT EXISTS");
+    ParserKeyword s_create(Keyword::CREATE);
+    ParserKeyword s_attach(Keyword::ATTACH);
+    ParserKeyword s_replace(Keyword::REPLACE);
+    ParserKeyword s_or_replace(Keyword::OR_REPLACE);
+    ParserKeyword s_temporary(Keyword::TEMPORARY);
+    ParserKeyword s_table(Keyword::TABLE);
+    ParserKeyword s_if_not_exists(Keyword::IF_NOT_EXISTS);
     ParserCompoundIdentifier table_name_p(/*table_name_with_optional_uuid*/ true, /*allow_query_parameter*/ true);
-    ParserKeyword s_from("FROM");
-    ParserKeyword s_on("ON");
+    ParserKeyword s_from(Keyword::FROM);
+    ParserKeyword s_on(Keyword::ON);
     ParserToken s_dot(TokenType::Dot);
     ParserToken s_comma(TokenType::Comma);
     ParserToken s_lparen(TokenType::OpeningRoundBracket);
@@ -756,13 +760,13 @@ bool ParserCreateTableQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe
 
     auto need_parse_as_select = [&is_create_empty, &pos, &expected]()
     {
-        if (ParserKeyword{"EMPTY AS"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::EMPTY_AS}.ignore(pos, expected))
         {
             is_create_empty = true;
             return true;
         }
 
-        return ParserKeyword{"AS"}.ignore(pos, expected);
+        return ParserKeyword{Keyword::AS}.ignore(pos, expected);
     };
 
     /// List of columns.
@@ -895,13 +899,13 @@ bool ParserCreateTableQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe
 
 bool ParserCreateLiveViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_create("CREATE");
-    ParserKeyword s_attach("ATTACH");
-    ParserKeyword s_if_not_exists("IF NOT EXISTS");
+    ParserKeyword s_create(Keyword::CREATE);
+    ParserKeyword s_attach(Keyword::ATTACH);
+    ParserKeyword s_if_not_exists(Keyword::IF_NOT_EXISTS);
     ParserCompoundIdentifier table_name_p(/*table_name_with_optional_uuid*/ true, /*allow_query_parameter*/ true);
-    ParserKeyword s_as("AS");
-    ParserKeyword s_view("VIEW");
-    ParserKeyword s_live("LIVE");
+    ParserKeyword s_as(Keyword::AS);
+    ParserKeyword s_view(Keyword::VIEW);
+    ParserKeyword s_live(Keyword::LIVE);
     ParserToken s_dot(TokenType::Dot);
     ParserToken s_lparen(TokenType::OpeningRoundBracket);
     ParserToken s_rparen(TokenType::ClosingRoundBracket);
@@ -945,14 +949,14 @@ bool ParserCreateLiveViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & e
     if (!table_name_p.parse(pos, table, expected))
         return false;
 
-    if (ParserKeyword{"ON"}.ignore(pos, expected))
+    if (ParserKeyword{Keyword::ON}.ignore(pos, expected))
     {
         if (!ASTQueryWithOnCluster::parse(pos, cluster_str, expected))
             return false;
     }
 
     // TO [db.]table
-    if (ParserKeyword{"TO"}.ignore(pos, expected))
+    if (ParserKeyword{Keyword::TO}.ignore(pos, expected))
     {
         if (!table_name_p.parse(pos, to_table, expected))
             return false;
@@ -1018,15 +1022,21 @@ bool ParserCreateLiveViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & e
 
 bool ParserCreateWindowViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_create("CREATE");
-    ParserKeyword s_temporary("TEMPORARY");
-    ParserKeyword s_attach("ATTACH");
-    ParserKeyword s_if_not_exists("IF NOT EXISTS");
+    ParserKeyword s_create(Keyword::CREATE);
+    ParserKeyword s_temporary(Keyword::TEMPORARY);
+    ParserKeyword s_attach(Keyword::ATTACH);
+    ParserKeyword s_if_not_exists(Keyword::IF_NOT_EXISTS);
     ParserCompoundIdentifier table_name_p(/*table_name_with_optional_uuid*/ true, /*allow_query_parameter*/ true);
-    ParserKeyword s_as("AS");
-    ParserKeyword s_view("VIEW");
-    ParserKeyword s_window("WINDOW");
-    ParserKeyword s_populate("POPULATE");
+    ParserKeyword s_as(Keyword::AS);
+    ParserKeyword s_view(Keyword::VIEW);
+    ParserKeyword s_window(Keyword::WINDOW);
+    ParserKeyword s_populate(Keyword::POPULATE);
+    ParserKeyword s_on(Keyword::ON);
+    ParserKeyword s_to(Keyword::TO);
+    ParserKeyword s_inner(Keyword::INNER);
+    ParserKeyword s_watermark(Keyword::WATERMARK);
+    ParserKeyword s_allowed_lateness(Keyword::ALLOWED_LATENESS);
+    ParserKeyword s_empty(Keyword::EMPTY);
     ParserToken s_dot(TokenType::Dot);
     ParserToken s_eq(TokenType::Equals);
     ParserToken s_lparen(TokenType::OpeningRoundBracket);
@@ -1079,14 +1089,14 @@ bool ParserCreateWindowViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected &
     if (!table_name_p.parse(pos, table, expected))
         return false;
 
-    if (ParserKeyword{"ON"}.ignore(pos, expected))
+    if (s_on.ignore(pos, expected))
     {
         if (!ASTQueryWithOnCluster::parse(pos, cluster_str, expected))
             return false;
     }
 
     // TO [db.]table
-    if (ParserKeyword{"TO"}.ignore(pos, expected))
+    if (s_to.ignore(pos, expected))
     {
         if (!table_name_p.parse(pos, to_table, expected))
             return false;
@@ -1102,7 +1112,7 @@ bool ParserCreateWindowViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected &
             return false;
     }
 
-    if (ParserKeyword{"INNER"}.ignore(pos, expected))
+    if (s_inner.ignore(pos, expected))
     {
         /// Inner table ENGINE for WINDOW VIEW
         storage_inner.parse(pos, inner_storage, expected);
@@ -1115,13 +1125,13 @@ bool ParserCreateWindowViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected &
     }
 
     // WATERMARK
-    if (ParserKeyword{"WATERMARK"}.ignore(pos, expected))
+    if (s_watermark.ignore(pos, expected))
     {
         s_eq.ignore(pos, expected);
 
-        if (ParserKeyword("STRICTLY_ASCENDING").ignore(pos,expected))
+        if (ParserKeyword(Keyword::STRICTLY_ASCENDING).ignore(pos,expected))
             is_watermark_strictly_ascending = true;
-        else if (ParserKeyword("ASCENDING").ignore(pos,expected))
+        else if (ParserKeyword(Keyword::ASCENDING).ignore(pos,expected))
             is_watermark_ascending = true;
         else if (watermark_p.parse(pos, watermark, expected))
             is_watermark_bounded = true;
@@ -1130,7 +1140,7 @@ bool ParserCreateWindowViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected &
     }
 
     // ALLOWED LATENESS
-    if (ParserKeyword{"ALLOWED_LATENESS"}.ignore(pos, expected))
+    if (s_allowed_lateness.ignore(pos, expected))
     {
         s_eq.ignore(pos, expected);
         allowed_lateness = true;
@@ -1141,7 +1151,7 @@ bool ParserCreateWindowViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected &
 
     if (s_populate.ignore(pos, expected))
         is_populate = true;
-    else if (ParserKeyword{"EMPTY"}.ignore(pos, expected))
+    else if (s_empty.ignore(pos, expected))
         is_create_empty = true;
 
     /// AS SELECT ...
@@ -1194,19 +1204,19 @@ bool ParserCreateWindowViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected &
 
 bool ParserTableOverrideDeclaration::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_table_override("TABLE OVERRIDE");
+    ParserKeyword s_table_override(Keyword::TABLE_OVERRIDE);
     ParserIdentifier table_name_p;
     ParserToken lparen_p(TokenType::OpeningRoundBracket);
     ParserToken rparen_p(TokenType::ClosingRoundBracket);
     ParserTablePropertiesDeclarationList table_properties_p;
     ParserExpression expression_p;
     ParserTTLExpressionList parser_ttl_list;
-    ParserKeyword s_columns("COLUMNS");
-    ParserKeyword s_partition_by("PARTITION BY");
-    ParserKeyword s_primary_key("PRIMARY KEY");
-    ParserKeyword s_order_by("ORDER BY");
-    ParserKeyword s_sample_by("SAMPLE BY");
-    ParserKeyword s_ttl("TTL");
+    ParserKeyword s_columns(Keyword::COLUMNS);
+    ParserKeyword s_partition_by(Keyword::PARTITION_BY);
+    ParserKeyword s_primary_key(Keyword::PRIMARY_KEY);
+    ParserKeyword s_order_by(Keyword::ORDER_BY);
+    ParserKeyword s_sample_by(Keyword::SAMPLE_BY);
+    ParserKeyword s_ttl(Keyword::TTL);
     ASTPtr table_name;
     ASTPtr columns;
     ASTPtr partition_by;
@@ -1332,10 +1342,12 @@ bool ParserTableOverridesDeclarationList::parseImpl(Pos & pos, ASTPtr & node, Ex
 
 bool ParserCreateDatabaseQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_create("CREATE");
-    ParserKeyword s_attach("ATTACH");
-    ParserKeyword s_database("DATABASE");
-    ParserKeyword s_if_not_exists("IF NOT EXISTS");
+    ParserKeyword s_create(Keyword::CREATE);
+    ParserKeyword s_attach(Keyword::ATTACH);
+    ParserKeyword s_database(Keyword::DATABASE);
+    ParserKeyword s_if_not_exists(Keyword::IF_NOT_EXISTS);
+    ParserKeyword s_on(Keyword::ON);
+    ParserKeyword s_uuid(Keyword::UUID);
     ParserStorage storage_p{ParserStorage::DATABASE_ENGINE};
     ParserIdentifier name_p(true);
     ParserTableOverridesDeclarationList table_overrides_p;
@@ -1366,7 +1378,7 @@ bool ParserCreateDatabaseQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & e
     if (!name_p.parse(pos, database, expected))
         return false;
 
-    if (ParserKeyword("UUID").ignore(pos, expected))
+    if (s_uuid.ignore(pos, expected))
     {
         ParserStringLiteral uuid_p;
         ASTPtr ast_uuid;
@@ -1375,7 +1387,7 @@ bool ParserCreateDatabaseQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & e
         uuid = parseFromString<UUID>(ast_uuid->as<ASTLiteral>()->value.get<String>());
     }
 
-    if (ParserKeyword{"ON"}.ignore(pos, expected))
+    if (s_on.ignore(pos, expected))
     {
         if (!ASTQueryWithOnCluster::parse(pos, cluster_str, expected))
             return false;
@@ -1411,17 +1423,18 @@ bool ParserCreateDatabaseQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & e
 
 bool ParserCreateViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_create("CREATE");
-    ParserKeyword s_attach("ATTACH");
-    ParserKeyword s_if_not_exists("IF NOT EXISTS");
+    ParserKeyword s_create(Keyword::CREATE);
+    ParserKeyword s_attach(Keyword::ATTACH);
+    ParserKeyword s_if_not_exists(Keyword::IF_NOT_EXISTS);
     ParserCompoundIdentifier table_name_p(/*table_name_with_optional_uuid*/ true, /*allow_query_parameter*/ true);
     ParserCompoundIdentifier to_table_name_p(/*table_name_with_optional_uuid*/ true, /*allow_query_parameter*/ false);
-    ParserKeyword s_as("AS");
-    ParserKeyword s_view("VIEW");
-    ParserKeyword s_materialized("MATERIALIZED");
-    ParserKeyword s_populate("POPULATE");
-    ParserKeyword s_empty("EMPTY");
-    ParserKeyword s_or_replace("OR REPLACE");
+    ParserKeyword s_as(Keyword::AS);
+    ParserKeyword s_view(Keyword::VIEW);
+    ParserKeyword s_materialized(Keyword::MATERIALIZED);
+    ParserKeyword s_populate(Keyword::POPULATE);
+    ParserKeyword s_empty(Keyword::EMPTY);
+    ParserKeyword s_or_replace(Keyword::OR_REPLACE);
+    ParserKeyword s_to{Keyword::TO};
     ParserToken s_dot(TokenType::Dot);
     ParserToken s_lparen(TokenType::OpeningRoundBracket);
     ParserToken s_rparen(TokenType::ClosingRoundBracket);
@@ -1484,13 +1497,13 @@ bool ParserCreateViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
     if (!table_name_p.parse(pos, table, expected))
         return false;
 
-    if (ParserKeyword{"ON"}.ignore(pos, expected))
+    if (ParserKeyword{Keyword::ON}.ignore(pos, expected))
     {
         if (!ASTQueryWithOnCluster::parse(pos, cluster_str, expected))
             return false;
     }
 
-    if (ParserKeyword{"REFRESH"}.ignore(pos, expected))
+    if (ParserKeyword{Keyword::REFRESH}.ignore(pos, expected))
     {
         // REFRESH only with materialized views
         if (!is_materialized_view)
@@ -1499,13 +1512,13 @@ bool ParserCreateViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
             return false;
     }
 
-    if (is_materialized_view && ParserKeyword{"TO INNER UUID"}.ignore(pos, expected))
+    if (is_materialized_view && ParserKeyword{Keyword::TO_INNER_UUID}.ignore(pos, expected))
     {
         ParserStringLiteral literal_p;
         if (!literal_p.parse(pos, to_inner_uuid, expected))
             return false;
     }
-    else if (is_materialized_view && ParserKeyword{"TO"}.ignore(pos, expected))
+    else if (is_materialized_view && ParserKeyword{Keyword::TO}.ignore(pos, expected))
     {
         // TO [db.]table
         if (!table_name_p.parse(pos, to_table, expected))
@@ -1535,7 +1548,7 @@ bool ParserCreateViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
             else if (s_empty.ignore(pos, expected))
                 is_create_empty = true;
 
-            if (ParserKeyword{"TO"}.ignore(pos, expected))
+            if (s_to.ignore(pos, expected))
                 throw Exception(
                     ErrorCodes::SYNTAX_ERROR, "When creating a materialized view you can't declare both 'ENGINE' and 'TO [db].[table]'");
         }
@@ -1617,13 +1630,13 @@ bool ParserCreateViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
 
 bool ParserCreateNamedCollectionQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_create("CREATE");
-    ParserKeyword s_named_collection("NAMED COLLECTION");
-    ParserKeyword s_if_not_exists("IF NOT EXISTS");
-    ParserKeyword s_on("ON");
-    ParserKeyword s_as("AS");
-    ParserKeyword s_not_overridable("NOT OVERRIDABLE");
-    ParserKeyword s_overridable("OVERRIDABLE");
+    ParserKeyword s_create(Keyword::CREATE);
+    ParserKeyword s_named_collection(Keyword::NAMED_COLLECTION);
+    ParserKeyword s_if_not_exists(Keyword::IF_NOT_EXISTS);
+    ParserKeyword s_on(Keyword::ON);
+    ParserKeyword s_as(Keyword::AS);
+    ParserKeyword s_not_overridable(Keyword::NOT_OVERRIDABLE);
+    ParserKeyword s_overridable(Keyword::OVERRIDABLE);
     ParserIdentifier name_p;
     ParserToken s_comma(TokenType::Comma);
 
@@ -1644,6 +1657,7 @@ bool ParserCreateNamedCollectionQuery::parseImpl(Pos & pos, ASTPtr & node, Expec
     if (!name_p.parse(pos, collection_name, expected))
         return false;
 
+
     if (s_on.ignore(pos, expected))
     {
         if (!ASTQueryWithOnCluster::parse(pos, cluster_str, expected))
@@ -1685,13 +1699,13 @@ bool ParserCreateNamedCollectionQuery::parseImpl(Pos & pos, ASTPtr & node, Expec
 
 bool ParserCreateDictionaryQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_create("CREATE");
-    ParserKeyword s_attach("ATTACH");
-    ParserKeyword s_replace("REPLACE");
-    ParserKeyword s_or_replace("OR REPLACE");
-    ParserKeyword s_dictionary("DICTIONARY");
-    ParserKeyword s_if_not_exists("IF NOT EXISTS");
-    ParserKeyword s_on("ON");
+    ParserKeyword s_create(Keyword::CREATE);
+    ParserKeyword s_attach(Keyword::ATTACH);
+    ParserKeyword s_replace(Keyword::REPLACE);
+    ParserKeyword s_or_replace(Keyword::OR_REPLACE);
+    ParserKeyword s_dictionary(Keyword::DICTIONARY);
+    ParserKeyword s_if_not_exists(Keyword::IF_NOT_EXISTS);
+    ParserKeyword s_on(Keyword::ON);
     ParserCompoundIdentifier dict_name_p(/*table_name_with_optional_uuid*/ true, /*allow_query_parameter*/ true);
     ParserToken s_left_paren(TokenType::OpeningRoundBracket);
     ParserToken s_right_paren(TokenType::ClosingRoundBracket);
diff --git a/src/Parsers/ParserCreateQuery.h b/src/Parsers/ParserCreateQuery.h
index 30bce57f9d9..d001c097114 100644
--- a/src/Parsers/ParserCreateQuery.h
+++ b/src/Parsers/ParserCreateQuery.h
@@ -129,24 +129,24 @@ using ParserCompoundColumnDeclaration = IParserColumnDeclaration<ParserCompoundI
 template <typename NameParser>
 bool IParserColumnDeclaration<NameParser>::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_default{"DEFAULT"};
-    ParserKeyword s_null{"NULL"};
-    ParserKeyword s_not{"NOT"};
-    ParserKeyword s_materialized{"MATERIALIZED"};
-    ParserKeyword s_ephemeral{"EPHEMERAL"};
-    ParserKeyword s_alias{"ALIAS"};
-    ParserKeyword s_auto_increment{"AUTO_INCREMENT"};
-    ParserKeyword s_comment{"COMMENT"};
-    ParserKeyword s_codec{"CODEC"};
-    ParserKeyword s_stat{"STATISTIC"};
-    ParserKeyword s_ttl{"TTL"};
-    ParserKeyword s_remove{"REMOVE"};
-    ParserKeyword s_modify_setting("MODIFY SETTING");
-    ParserKeyword s_reset_setting("RESET SETTING");
-    ParserKeyword s_settings("SETTINGS");
-    ParserKeyword s_type{"TYPE"};
-    ParserKeyword s_collate{"COLLATE"};
-    ParserKeyword s_primary_key{"PRIMARY KEY"};
+    ParserKeyword s_default{Keyword::DEFAULT};
+    ParserKeyword s_null{Keyword::NULL_KEYWORD};
+    ParserKeyword s_not{Keyword::NOT};
+    ParserKeyword s_materialized{Keyword::MATERIALIZED};
+    ParserKeyword s_ephemeral{Keyword::EPHEMERAL};
+    ParserKeyword s_alias{Keyword::ALIAS};
+    ParserKeyword s_auto_increment{Keyword::AUTO_INCREMENT};
+    ParserKeyword s_comment{Keyword::COMMENT};
+    ParserKeyword s_codec{Keyword::CODEC};
+    ParserKeyword s_stat{Keyword::STATISTIC};
+    ParserKeyword s_ttl{Keyword::TTL};
+    ParserKeyword s_remove{Keyword::REMOVE};
+    ParserKeyword s_modify_setting(Keyword::MODIFY_SETTING);
+    ParserKeyword s_reset_setting(Keyword::RESET_SETTING);
+    ParserKeyword s_settings(Keyword::SETTINGS);
+    ParserKeyword s_type{Keyword::TYPE};
+    ParserKeyword s_collate{Keyword::COLLATE};
+    ParserKeyword s_primary_key{Keyword::PRIMARY_KEY};
 
     NameParser name_parser;
     ParserDataType type_parser;
diff --git a/src/Parsers/ParserDataType.cpp b/src/Parsers/ParserDataType.cpp
index b75f17dca72..fcf189e51f4 100644
--- a/src/Parsers/ParserDataType.cpp
+++ b/src/Parsers/ParserDataType.cpp
@@ -64,39 +64,39 @@ bool ParserDataType::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
     /// only for certain first words, otherwise we don't know how many words to parse
     if (type_name_upper == "NATIONAL")
     {
-        if (ParserKeyword("CHARACTER LARGE OBJECT").ignore(pos))
-            type_name_suffix = "CHARACTER LARGE OBJECT";
-        else if (ParserKeyword("CHARACTER VARYING").ignore(pos))
-            type_name_suffix = "CHARACTER VARYING";
-        else if (ParserKeyword("CHAR VARYING").ignore(pos))
-            type_name_suffix = "CHAR VARYING";
-        else if (ParserKeyword("CHARACTER").ignore(pos))
-            type_name_suffix = "CHARACTER";
-        else if (ParserKeyword("CHAR").ignore(pos))
-            type_name_suffix = "CHAR";
+        if (ParserKeyword(Keyword::CHARACTER_LARGE_OBJECT).ignore(pos))
+            type_name_suffix = toStringView(Keyword::CHARACTER_LARGE_OBJECT);
+        else if (ParserKeyword(Keyword::CHARACTER_VARYING).ignore(pos))
+            type_name_suffix = toStringView(Keyword::CHARACTER_VARYING);
+        else if (ParserKeyword(Keyword::CHAR_VARYING).ignore(pos))
+            type_name_suffix = toStringView(Keyword::CHAR_VARYING);
+        else if (ParserKeyword(Keyword::CHARACTER).ignore(pos))
+            type_name_suffix = toStringView(Keyword::CHARACTER);
+        else if (ParserKeyword(Keyword::CHAR).ignore(pos))
+            type_name_suffix = toStringView(Keyword::CHAR);
     }
     else if (type_name_upper == "BINARY" ||
              type_name_upper == "CHARACTER" ||
              type_name_upper == "CHAR" ||
              type_name_upper == "NCHAR")
     {
-        if (ParserKeyword("LARGE OBJECT").ignore(pos))
-            type_name_suffix = "LARGE OBJECT";
-        else if (ParserKeyword("VARYING").ignore(pos))
-            type_name_suffix = "VARYING";
+        if (ParserKeyword(Keyword::LARGE_OBJECT).ignore(pos))
+            type_name_suffix = toStringView(Keyword::LARGE_OBJECT);
+        else if (ParserKeyword(Keyword::VARYING).ignore(pos))
+            type_name_suffix = toStringView(Keyword::VARYING);
     }
     else if (type_name_upper == "DOUBLE")
     {
-        if (ParserKeyword("PRECISION").ignore(pos))
-            type_name_suffix = "PRECISION";
+        if (ParserKeyword(Keyword::PRECISION).ignore(pos))
+            type_name_suffix = toStringView(Keyword::PRECISION);
     }
     else if (type_name_upper.find("INT") != std::string::npos)
     {
         /// Support SIGNED and UNSIGNED integer type modifiers for compatibility with MySQL
-        if (ParserKeyword("SIGNED").ignore(pos))
-            type_name_suffix = "SIGNED";
-        else if (ParserKeyword("UNSIGNED").ignore(pos))
-            type_name_suffix = "UNSIGNED";
+        if (ParserKeyword(Keyword::SIGNED).ignore(pos))
+            type_name_suffix = toStringView(Keyword::SIGNED);
+        else if (ParserKeyword(Keyword::UNSIGNED).ignore(pos))
+            type_name_suffix = toStringView(Keyword::UNSIGNED);
         else if (pos->type == TokenType::OpeningRoundBracket)
         {
             ++pos;
@@ -105,10 +105,10 @@ bool ParserDataType::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
             if (pos->type != TokenType::ClosingRoundBracket)
                return false;
             ++pos;
-            if (ParserKeyword("SIGNED").ignore(pos))
-                type_name_suffix = "SIGNED";
-            else if (ParserKeyword("UNSIGNED").ignore(pos))
-                type_name_suffix = "UNSIGNED";
+            if (ParserKeyword(Keyword::SIGNED).ignore(pos))
+                type_name_suffix = toStringView(Keyword::SIGNED);
+            else if (ParserKeyword(Keyword::UNSIGNED).ignore(pos))
+                type_name_suffix = toStringView(Keyword::UNSIGNED);
         }
 
     }
diff --git a/src/Parsers/ParserDatabaseOrNone.cpp b/src/Parsers/ParserDatabaseOrNone.cpp
index c53c547aedd..6b319c80f11 100644
--- a/src/Parsers/ParserDatabaseOrNone.cpp
+++ b/src/Parsers/ParserDatabaseOrNone.cpp
@@ -10,7 +10,7 @@ bool ParserDatabaseOrNone::parseImpl(Pos & pos, ASTPtr & node, Expected & expect
     auto result = std::make_shared<ASTDatabaseOrNone>();
     node = result;
 
-    if (ParserKeyword{"NONE"}.ignore(pos, expected))
+    if (ParserKeyword(Keyword::NONE).ignore(pos, expected))
     {
         result->none = true;
         return true;
diff --git a/src/Parsers/ParserDeleteQuery.cpp b/src/Parsers/ParserDeleteQuery.cpp
index 7b27651d82d..00fac45c8ed 100644
--- a/src/Parsers/ParserDeleteQuery.cpp
+++ b/src/Parsers/ParserDeleteQuery.cpp
@@ -13,12 +13,12 @@ bool ParserDeleteQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
     auto query = std::make_shared<ASTDeleteQuery>();
     node = query;
 
-    ParserKeyword s_delete("DELETE");
-    ParserKeyword s_from("FROM");
-    ParserKeyword s_where("WHERE");
+    ParserKeyword s_delete(Keyword::DELETE);
+    ParserKeyword s_from(Keyword::FROM);
+    ParserKeyword s_where(Keyword::WHERE);
     ParserExpression parser_exp_elem;
-    ParserKeyword s_settings("SETTINGS");
-    ParserKeyword s_on{"ON"};
+    ParserKeyword s_settings(Keyword::SETTINGS);
+    ParserKeyword s_on{Keyword::ON};
 
     if (s_delete.ignore(pos, expected))
     {
diff --git a/src/Parsers/ParserDescribeCacheQuery.cpp b/src/Parsers/ParserDescribeCacheQuery.cpp
index c6c9cbe8bc0..d0219600892 100644
--- a/src/Parsers/ParserDescribeCacheQuery.cpp
+++ b/src/Parsers/ParserDescribeCacheQuery.cpp
@@ -9,9 +9,9 @@ namespace DB
 
 bool ParserDescribeCacheQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword p_describe("DESCRIBE");
-    ParserKeyword p_desc("DESC");
-    ParserKeyword p_cache("FILESYSTEM CACHE");
+    ParserKeyword p_describe(Keyword::DESCRIBE);
+    ParserKeyword p_desc(Keyword::DESC);
+    ParserKeyword p_cache(Keyword::FILESYSTEM_CACHE);
     ParserLiteral p_cache_name;
 
     if ((!p_describe.ignore(pos, expected) && !p_desc.ignore(pos, expected))
diff --git a/src/Parsers/ParserDescribeTableQuery.cpp b/src/Parsers/ParserDescribeTableQuery.cpp
index fcfc4799dbe..92c0cfacd9b 100644
--- a/src/Parsers/ParserDescribeTableQuery.cpp
+++ b/src/Parsers/ParserDescribeTableQuery.cpp
@@ -14,10 +14,10 @@ namespace DB
 
 bool ParserDescribeTableQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_describe("DESCRIBE");
-    ParserKeyword s_desc("DESC");
-    ParserKeyword s_table("TABLE");
-    ParserKeyword s_settings("SETTINGS");
+    ParserKeyword s_describe(Keyword::DESCRIBE);
+    ParserKeyword s_desc(Keyword::DESC);
+    ParserKeyword s_table(Keyword::TABLE);
+    ParserKeyword s_settings(Keyword::SETTINGS);
     ParserToken s_dot(TokenType::Dot);
     ParserIdentifier name_p;
     ParserSetQuery parser_settings(true);
diff --git a/src/Parsers/ParserDictionary.cpp b/src/Parsers/ParserDictionary.cpp
index ef914e2264a..539ce30e010 100644
--- a/src/Parsers/ParserDictionary.cpp
+++ b/src/Parsers/ParserDictionary.cpp
@@ -165,12 +165,12 @@ bool ParserDictionarySettings::parseImpl(Pos & pos, ASTPtr & node, Expected & ex
 
 bool ParserDictionary::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword primary_key_keyword("PRIMARY KEY");
-    ParserKeyword source_keyword("SOURCE");
-    ParserKeyword lifetime_keyword("LIFETIME");
-    ParserKeyword range_keyword("RANGE");
-    ParserKeyword layout_keyword("LAYOUT");
-    ParserKeyword settings_keyword("SETTINGS");
+    ParserKeyword primary_key_keyword(Keyword::PRIMARY_KEY);
+    ParserKeyword source_keyword(Keyword::SOURCE);
+    ParserKeyword lifetime_keyword(Keyword::LIFETIME);
+    ParserKeyword range_keyword(Keyword::RANGE);
+    ParserKeyword layout_keyword(Keyword::LAYOUT);
+    ParserKeyword settings_keyword(Keyword::SETTINGS);
     ParserToken open(TokenType::OpeningRoundBracket);
     ParserToken close(TokenType::ClosingRoundBracket);
     ParserFunctionWithKeyValueArguments key_value_pairs_p;
diff --git a/src/Parsers/ParserDictionaryAttributeDeclaration.cpp b/src/Parsers/ParserDictionaryAttributeDeclaration.cpp
index 9910c73e541..376cad136ae 100644
--- a/src/Parsers/ParserDictionaryAttributeDeclaration.cpp
+++ b/src/Parsers/ParserDictionaryAttributeDeclaration.cpp
@@ -12,12 +12,12 @@ bool ParserDictionaryAttributeDeclaration::parseImpl(Pos & pos, ASTPtr & node, E
 {
     ParserIdentifier name_parser;
     ParserDataType type_parser;
-    ParserKeyword s_default{"DEFAULT"};
-    ParserKeyword s_expression{"EXPRESSION"};
-    ParserKeyword s_hierarchical{"HIERARCHICAL"};
-    ParserKeyword s_bidirectional{"BIDIRECTIONAL"};
-    ParserKeyword s_injective{"INJECTIVE"};
-    ParserKeyword s_is_object_id{"IS_OBJECT_ID"};
+    ParserKeyword s_default{Keyword::DEFAULT};
+    ParserKeyword s_expression{Keyword::EXPRESSION};
+    ParserKeyword s_hierarchical{Keyword::HIERARCHICAL};
+    ParserKeyword s_bidirectional{Keyword::BIDIRECTIONAL};
+    ParserKeyword s_injective{Keyword::INJECTIVE};
+    ParserKeyword s_is_object_id{Keyword::IS_OBJECT_ID};
     ParserLiteral default_parser;
     ParserArrayOfLiterals array_literals_parser;
     ParserExpression expression_parser;
diff --git a/src/Parsers/ParserDropFunctionQuery.cpp b/src/Parsers/ParserDropFunctionQuery.cpp
index 3f35124f9ab..d98acef3252 100644
--- a/src/Parsers/ParserDropFunctionQuery.cpp
+++ b/src/Parsers/ParserDropFunctionQuery.cpp
@@ -9,10 +9,10 @@ namespace DB
 
 bool ParserDropFunctionQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_drop("DROP");
-    ParserKeyword s_function("FUNCTION");
-    ParserKeyword s_if_exists("IF EXISTS");
-    ParserKeyword s_on("ON");
+    ParserKeyword s_drop(Keyword::DROP);
+    ParserKeyword s_function(Keyword::FUNCTION);
+    ParserKeyword s_if_exists(Keyword::IF_EXISTS);
+    ParserKeyword s_on(Keyword::ON);
     ParserIdentifier function_name_p;
 
     String cluster_str;
diff --git a/src/Parsers/ParserDropIndexQuery.cpp b/src/Parsers/ParserDropIndexQuery.cpp
index 5843d8b94bf..1363c00f252 100644
--- a/src/Parsers/ParserDropIndexQuery.cpp
+++ b/src/Parsers/ParserDropIndexQuery.cpp
@@ -13,10 +13,10 @@ bool ParserDropIndexQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected
     auto query = std::make_shared<ASTDropIndexQuery>();
     node = query;
 
-    ParserKeyword s_drop("DROP");
-    ParserKeyword s_index("INDEX");
-    ParserKeyword s_on("ON");
-    ParserKeyword s_if_exists("IF EXISTS");
+    ParserKeyword s_drop(Keyword::DROP);
+    ParserKeyword s_index(Keyword::INDEX);
+    ParserKeyword s_on(Keyword::ON);
+    ParserKeyword s_if_exists(Keyword::IF_EXISTS);
     ParserIdentifier index_name_p;
 
     String cluster_str;
diff --git a/src/Parsers/ParserDropNamedCollectionQuery.cpp b/src/Parsers/ParserDropNamedCollectionQuery.cpp
index b0b010b5ef6..b438f3b5f4f 100644
--- a/src/Parsers/ParserDropNamedCollectionQuery.cpp
+++ b/src/Parsers/ParserDropNamedCollectionQuery.cpp
@@ -9,10 +9,10 @@ namespace DB
 
 bool ParserDropNamedCollectionQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_drop("DROP");
-    ParserKeyword s_collection("NAMED COLLECTION");
-    ParserKeyword s_if_exists("IF EXISTS");
-    ParserKeyword s_on("ON");
+    ParserKeyword s_drop(Keyword::DROP);
+    ParserKeyword s_collection(Keyword::NAMED_COLLECTION);
+    ParserKeyword s_if_exists(Keyword::IF_EXISTS);
+    ParserKeyword s_on(Keyword::ON);
     ParserIdentifier name_p;
 
     String cluster_str;
diff --git a/src/Parsers/ParserDropQuery.cpp b/src/Parsers/ParserDropQuery.cpp
index 450c8a1afec..fab9938d0b0 100644
--- a/src/Parsers/ParserDropQuery.cpp
+++ b/src/Parsers/ParserDropQuery.cpp
@@ -12,18 +12,18 @@ namespace
 
 bool parseDropQuery(IParser::Pos & pos, ASTPtr & node, Expected & expected, const ASTDropQuery::Kind kind)
 {
-    ParserKeyword s_temporary("TEMPORARY");
-    ParserKeyword s_table("TABLE");
-    ParserKeyword s_dictionary("DICTIONARY");
-    ParserKeyword s_view("VIEW");
-    ParserKeyword s_database("DATABASE");
+    ParserKeyword s_temporary(Keyword::TEMPORARY);
+    ParserKeyword s_table(Keyword::TABLE);
+    ParserKeyword s_dictionary(Keyword::DICTIONARY);
+    ParserKeyword s_view(Keyword::VIEW);
+    ParserKeyword s_database(Keyword::DATABASE);
     ParserToken s_dot(TokenType::Dot);
-    ParserKeyword s_if_exists("IF EXISTS");
-    ParserKeyword s_if_empty("IF EMPTY");
+    ParserKeyword s_if_exists(Keyword::IF_EXISTS);
+    ParserKeyword s_if_empty(Keyword::IF_EMPTY);
     ParserIdentifier name_p(true);
-    ParserKeyword s_permanently("PERMANENTLY");
-    ParserKeyword s_no_delay("NO DELAY");
-    ParserKeyword s_sync("SYNC");
+    ParserKeyword s_permanently(Keyword::PERMANENTLY);
+    ParserKeyword s_no_delay(Keyword::NO_DELAY);
+    ParserKeyword s_sync(Keyword::SYNC);
 
     ASTPtr database;
     ASTPtr table;
@@ -80,7 +80,7 @@ bool parseDropQuery(IParser::Pos & pos, ASTPtr & node, Expected & expected, cons
     }
 
     /// common for tables / dictionaries / databases
-    if (ParserKeyword{"ON"}.ignore(pos, expected))
+    if (ParserKeyword{Keyword::ON}.ignore(pos, expected))
     {
         if (!ASTQueryWithOnCluster::parse(pos, cluster_str, expected))
             return false;
@@ -122,9 +122,9 @@ bool parseDropQuery(IParser::Pos & pos, ASTPtr & node, Expected & expected, cons
 
 bool ParserDropQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_drop("DROP");
-    ParserKeyword s_detach("DETACH");
-    ParserKeyword s_truncate("TRUNCATE");
+    ParserKeyword s_drop(Keyword::DROP);
+    ParserKeyword s_detach(Keyword::DETACH);
+    ParserKeyword s_truncate(Keyword::TRUNCATE);
 
     if (s_drop.ignore(pos, expected))
         return parseDropQuery(pos, node, expected, ASTDropQuery::Kind::Drop);
diff --git a/src/Parsers/ParserExplainQuery.cpp b/src/Parsers/ParserExplainQuery.cpp
index 148b671af71..16ba0701992 100644
--- a/src/Parsers/ParserExplainQuery.cpp
+++ b/src/Parsers/ParserExplainQuery.cpp
@@ -16,15 +16,15 @@ bool ParserExplainQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected
 {
     ASTExplainQuery::ExplainKind kind;
 
-    ParserKeyword s_ast("AST");
-    ParserKeyword s_explain("EXPLAIN");
-    ParserKeyword s_syntax("SYNTAX");
-    ParserKeyword s_query_tree("QUERY TREE");
-    ParserKeyword s_pipeline("PIPELINE");
-    ParserKeyword s_plan("PLAN");
-    ParserKeyword s_estimates("ESTIMATE");
-    ParserKeyword s_table_override("TABLE OVERRIDE");
-    ParserKeyword s_current_transaction("CURRENT TRANSACTION");
+    ParserKeyword s_ast(Keyword::AST);
+    ParserKeyword s_explain(Keyword::EXPLAIN);
+    ParserKeyword s_syntax(Keyword::SYNTAX);
+    ParserKeyword s_query_tree(Keyword::QUERY_TREE);
+    ParserKeyword s_pipeline(Keyword::PIPELINE);
+    ParserKeyword s_plan(Keyword::PLAN);
+    ParserKeyword s_estimates(Keyword::ESTIMATE);
+    ParserKeyword s_table_override(Keyword::TABLE_OVERRIDE);
+    ParserKeyword s_current_transaction(Keyword::CURRENT_TRANSACTION);
 
     if (s_explain.ignore(pos, expected))
     {
diff --git a/src/Parsers/ParserExternalDDLQuery.cpp b/src/Parsers/ParserExternalDDLQuery.cpp
index 01e8fff4f3c..ebe210a953a 100644
--- a/src/Parsers/ParserExternalDDLQuery.cpp
+++ b/src/Parsers/ParserExternalDDLQuery.cpp
@@ -27,7 +27,7 @@ namespace ErrorCodes
 bool ParserExternalDDLQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected & expected)
 {
     ParserFunction p_function;
-    ParserKeyword s_external("EXTERNAL DDL FROM");
+    ParserKeyword s_external(Keyword::EXTERNAL_DDL_FROM);
 
     ASTPtr from;
     auto external_ddl_query = std::make_shared<ASTExternalDDLQuery>();
@@ -61,16 +61,16 @@ bool ParserExternalDDLQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expect
         {
             /// Syntax error is ignored, so we need to convert the error code for parsing failure
 
-            if (ParserKeyword("ALTER TABLE").ignore(pos))
+            if (ParserKeyword(Keyword::ALTER_TABLE).ignore(pos))
                 throw Exception(ErrorCodes::MYSQL_SYNTAX_ERROR, "Cannot parse MySQL alter query.");
 
-            if (ParserKeyword("RENAME TABLE").ignore(pos))
+            if (ParserKeyword(Keyword::RENAME_TABLE).ignore(pos))
                 throw Exception(ErrorCodes::MYSQL_SYNTAX_ERROR, "Cannot parse MySQL rename query.");
 
-            if (ParserKeyword("DROP TABLE").ignore(pos) || ParserKeyword("TRUNCATE").ignore(pos))
+            if (ParserKeyword(Keyword::DROP_TABLE).ignore(pos) || ParserKeyword(Keyword::TRUNCATE).ignore(pos))
                 throw Exception(ErrorCodes::MYSQL_SYNTAX_ERROR, "Cannot parse MySQL drop query.");
 
-            if (ParserKeyword("CREATE TABLE").ignore(pos) || ParserKeyword("CREATE TEMPORARY TABLE").ignore(pos))
+            if (ParserKeyword(Keyword::CREATE_TABLE).ignore(pos) || ParserKeyword(Keyword::CREATE_TEMPORARY_TABLE).ignore(pos))
                 throw Exception(ErrorCodes::MYSQL_SYNTAX_ERROR, "Cannot parse MySQL create query.");
         }
 #endif
diff --git a/src/Parsers/ParserInsertQuery.cpp b/src/Parsers/ParserInsertQuery.cpp
index 8601e12ebcb..d1171dd4815 100644
--- a/src/Parsers/ParserInsertQuery.cpp
+++ b/src/Parsers/ParserInsertQuery.cpp
@@ -26,19 +26,19 @@ namespace ErrorCodes
 bool ParserInsertQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
     /// Create parsers
-    ParserKeyword s_insert_into("INSERT INTO");
-    ParserKeyword s_from_infile("FROM INFILE");
-    ParserKeyword s_compression("COMPRESSION");
-    ParserKeyword s_table("TABLE");
-    ParserKeyword s_function("FUNCTION");
+    ParserKeyword s_insert_into(Keyword::INSERT_INTO);
+    ParserKeyword s_from_infile(Keyword::FROM_INFILE);
+    ParserKeyword s_compression(Keyword::COMPRESSION);
+    ParserKeyword s_table(Keyword::TABLE);
+    ParserKeyword s_function(Keyword::FUNCTION);
     ParserToken s_dot(TokenType::Dot);
-    ParserKeyword s_values("VALUES");
-    ParserKeyword s_format("FORMAT");
-    ParserKeyword s_settings("SETTINGS");
-    ParserKeyword s_select("SELECT");
-    ParserKeyword s_watch("WATCH");
-    ParserKeyword s_partition_by("PARTITION BY");
-    ParserKeyword s_with("WITH");
+    ParserKeyword s_values(Keyword::VALUES);
+    ParserKeyword s_format(Keyword::FORMAT);
+    ParserKeyword s_settings(Keyword::SETTINGS);
+    ParserKeyword s_select(Keyword::SELECT);
+    ParserKeyword s_watch(Keyword::WATCH);
+    ParserKeyword s_partition_by(Keyword::PARTITION_BY);
+    ParserKeyword s_with(Keyword::WITH);
     ParserToken s_lparen(TokenType::OpeningRoundBracket);
     ParserToken s_rparen(TokenType::ClosingRoundBracket);
     ParserToken s_semicolon(TokenType::Semicolon);
diff --git a/src/Parsers/ParserKillQueryQuery.cpp b/src/Parsers/ParserKillQueryQuery.cpp
index 0b1b37e61bf..55bd5100009 100644
--- a/src/Parsers/ParserKillQueryQuery.cpp
+++ b/src/Parsers/ParserKillQueryQuery.cpp
@@ -14,16 +14,16 @@ bool ParserKillQueryQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expect
     String cluster_str;
     auto query = std::make_shared<ASTKillQueryQuery>();
 
-    ParserKeyword p_kill{"KILL"};
-    ParserKeyword p_query{"QUERY"};
-    ParserKeyword p_mutation{"MUTATION"};
-    ParserKeyword p_part_move_to_shard{"PART_MOVE_TO_SHARD"};
-    ParserKeyword p_transaction{"TRANSACTION"};
-    ParserKeyword p_on{"ON"};
-    ParserKeyword p_test{"TEST"};
-    ParserKeyword p_sync{"SYNC"};
-    ParserKeyword p_async{"ASYNC"};
-    ParserKeyword p_where{"WHERE"};
+    ParserKeyword p_kill{Keyword::KILL};
+    ParserKeyword p_query{Keyword::QUERY};
+    ParserKeyword p_mutation{Keyword::MUTATION};
+    ParserKeyword p_part_move_to_shard{Keyword::PART_MOVE_TO_SHARD};
+    ParserKeyword p_transaction{Keyword::TRANSACTION};
+    ParserKeyword p_on{Keyword::ON};
+    ParserKeyword p_test{Keyword::TEST};
+    ParserKeyword p_sync{Keyword::SYNC};
+    ParserKeyword p_async{Keyword::ASYNC};
+    ParserKeyword p_where{Keyword::WHERE};
     ParserExpression p_where_expression;
 
     if (!p_kill.ignore(pos, expected))
diff --git a/src/Parsers/ParserOptimizeQuery.cpp b/src/Parsers/ParserOptimizeQuery.cpp
index e887ff445d2..7162f2ac3da 100644
--- a/src/Parsers/ParserOptimizeQuery.cpp
+++ b/src/Parsers/ParserOptimizeQuery.cpp
@@ -24,12 +24,12 @@ bool ParserOptimizeQueryColumnsSpecification::parseImpl(Pos & pos, ASTPtr & node
 
 bool ParserOptimizeQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_optimize_table("OPTIMIZE TABLE");
-    ParserKeyword s_partition("PARTITION");
-    ParserKeyword s_final("FINAL");
-    ParserKeyword s_deduplicate("DEDUPLICATE");
-    ParserKeyword s_cleanup("CLEANUP");
-    ParserKeyword s_by("BY");
+    ParserKeyword s_optimize_table(Keyword::OPTIMIZE_TABLE);
+    ParserKeyword s_partition(Keyword::PARTITION);
+    ParserKeyword s_final(Keyword::FINAL);
+    ParserKeyword s_deduplicate(Keyword::DEDUPLICATE);
+    ParserKeyword s_cleanup(Keyword::CLEANUP);
+    ParserKeyword s_by(Keyword::BY);
     ParserToken s_dot(TokenType::Dot);
     ParserIdentifier name_p(true);
     ParserPartition partition_p;
@@ -55,7 +55,7 @@ bool ParserOptimizeQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expecte
             return false;
     }
 
-    if (ParserKeyword{"ON"}.ignore(pos, expected) && !ASTQueryWithOnCluster::parse(pos, cluster_str, expected))
+    if (ParserKeyword{Keyword::ON}.ignore(pos, expected) && !ASTQueryWithOnCluster::parse(pos, cluster_str, expected))
         return false;
 
     if (s_partition.ignore(pos, expected))
diff --git a/src/Parsers/ParserPartition.cpp b/src/Parsers/ParserPartition.cpp
index f7d972dd4af..80a28f4803e 100644
--- a/src/Parsers/ParserPartition.cpp
+++ b/src/Parsers/ParserPartition.cpp
@@ -14,8 +14,8 @@ namespace DB
 
 bool ParserPartition::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_id("ID");
-    ParserKeyword s_all("ALL");
+    ParserKeyword s_id(Keyword::ID);
+    ParserKeyword s_all(Keyword::ALL);
     ParserStringLiteral parser_string_literal;
     ParserSubstitution parser_substitution;
     ParserExpression parser_expr;
diff --git a/src/Parsers/ParserProjectionSelectQuery.cpp b/src/Parsers/ParserProjectionSelectQuery.cpp
index b2adb5cf154..9ddd536f847 100644
--- a/src/Parsers/ParserProjectionSelectQuery.cpp
+++ b/src/Parsers/ParserProjectionSelectQuery.cpp
@@ -15,10 +15,10 @@ bool ParserProjectionSelectQuery::parseImpl(Pos & pos, ASTPtr & node, Expected &
     auto select_query = std::make_shared<ASTProjectionSelectQuery>();
     node = select_query;
 
-    ParserKeyword s_with("WITH");
-    ParserKeyword s_select("SELECT");
-    ParserKeyword s_group_by("GROUP BY");
-    ParserKeyword s_order_by("ORDER BY");
+    ParserKeyword s_with(Keyword::WITH);
+    ParserKeyword s_select(Keyword::SELECT);
+    ParserKeyword s_group_by(Keyword::GROUP_BY);
+    ParserKeyword s_order_by(Keyword::ORDER_BY);
 
     ParserNotEmptyExpressionList exp_list_for_with_clause(false);
     ParserNotEmptyExpressionList exp_list_for_select_clause(true); /// Allows aliases without AS keyword.
diff --git a/src/Parsers/ParserQueryWithOutput.cpp b/src/Parsers/ParserQueryWithOutput.cpp
index 4a0b928608b..cb0c10cd1c9 100644
--- a/src/Parsers/ParserQueryWithOutput.cpp
+++ b/src/Parsers/ParserQueryWithOutput.cpp
@@ -105,32 +105,32 @@ bool ParserQueryWithOutput::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
     /// FIXME: try to prettify this cast using `as<>()`
     auto & query_with_output = dynamic_cast<ASTQueryWithOutput &>(*query);
 
-    ParserKeyword s_into_outfile("INTO OUTFILE");
+    ParserKeyword s_into_outfile(Keyword::INTO_OUTFILE);
     if (s_into_outfile.ignore(pos, expected))
     {
         ParserStringLiteral out_file_p;
         if (!out_file_p.parse(pos, query_with_output.out_file, expected))
             return false;
 
-        ParserKeyword s_append("APPEND");
+        ParserKeyword s_append(Keyword::APPEND);
         if (s_append.ignore(pos, expected))
         {
             query_with_output.is_outfile_append = true;
         }
 
-        ParserKeyword s_truncate("TRUNCATE");
+        ParserKeyword s_truncate(Keyword::TRUNCATE);
         if (s_truncate.ignore(pos, expected))
         {
             query_with_output.is_outfile_truncate = true;
         }
 
-        ParserKeyword s_stdout("AND STDOUT");
+        ParserKeyword s_stdout(Keyword::AND_STDOUT);
         if (s_stdout.ignore(pos, expected))
         {
             query_with_output.is_into_outfile_with_stdout = true;
         }
 
-        ParserKeyword s_compression_method("COMPRESSION");
+        ParserKeyword s_compression_method(Keyword::COMPRESSION);
         if (s_compression_method.ignore(pos, expected))
         {
             ParserStringLiteral compression;
@@ -138,7 +138,7 @@ bool ParserQueryWithOutput::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
                 return false;
             query_with_output.children.push_back(query_with_output.compression);
 
-            ParserKeyword s_compression_level("LEVEL");
+            ParserKeyword s_compression_level(Keyword::LEVEL);
             if (s_compression_level.ignore(pos, expected))
             {
                 ParserNumber compression_level;
@@ -152,7 +152,7 @@ bool ParserQueryWithOutput::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
 
     }
 
-    ParserKeyword s_format("FORMAT");
+    ParserKeyword s_format(Keyword::FORMAT);
 
     if (s_format.ignore(pos, expected))
     {
@@ -166,7 +166,7 @@ bool ParserQueryWithOutput::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
     }
 
     // SETTINGS key1 = value1, key2 = value2, ...
-    ParserKeyword s_settings("SETTINGS");
+    ParserKeyword s_settings(Keyword::SETTINGS);
     if (!query_with_output.settings_ast && s_settings.ignore(pos, expected))
     {
         ParserSetQuery parser_settings(true);
diff --git a/src/Parsers/ParserRefreshStrategy.cpp b/src/Parsers/ParserRefreshStrategy.cpp
index b41473e2aa5..e7912293d85 100644
--- a/src/Parsers/ParserRefreshStrategy.cpp
+++ b/src/Parsers/ParserRefreshStrategy.cpp
@@ -18,9 +18,16 @@ namespace ErrorCodes
 
 bool ParserRefreshStrategy::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
+    ParserKeyword s_after{Keyword::AFTER};
+    ParserKeyword s_every{Keyword::EVERY};
+    ParserKeyword s_offset{Keyword::OFFSET};
+    ParserKeyword s_randomize_for{Keyword::RANDOMIZE_FOR};
+    ParserKeyword s_depends_on{Keyword::DEPENDS_ON};
+    ParserKeyword s_settings{Keyword::SETTINGS};
+
     auto refresh = std::make_shared<ASTRefreshStrategy>();
 
-    if (ParserKeyword{"AFTER"}.ignore(pos, expected))
+    if (s_after.ignore(pos, expected))
     {
         refresh->schedule_kind = RefreshScheduleKind::AFTER;
         ASTPtr period;
@@ -29,14 +36,14 @@ bool ParserRefreshStrategy::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
 
         refresh->set(refresh->period, period);
     }
-    else if (ParserKeyword{"EVERY"}.ignore(pos, expected))
+    else if (s_every.ignore(pos, expected))
     {
         refresh->schedule_kind = RefreshScheduleKind::EVERY;
         ASTPtr period;
         if (!ParserTimeInterval{{.allow_mixing_calendar_and_clock_units = false}}.parse(pos, period, expected))
             return false;
         refresh->set(refresh->period, period);
-        if (ParserKeyword{"OFFSET"}.ignore(pos, expected))
+        if (s_offset.ignore(pos, expected))
         {
             ASTPtr periodic_offset;
             if (!ParserTimeInterval{{.allow_zero = true}}.parse(pos, periodic_offset, expected))
@@ -53,7 +60,7 @@ bool ParserRefreshStrategy::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
     if (refresh->schedule_kind == RefreshScheduleKind::UNKNOWN)
         return false;
 
-    if (ParserKeyword{"RANDOMIZE FOR"}.ignore(pos, expected))
+    if (s_randomize_for.ignore(pos, expected))
     {
         ASTPtr spread;
         if (!ParserTimeInterval{{.allow_zero = true}}.parse(pos, spread, expected))
@@ -62,7 +69,7 @@ bool ParserRefreshStrategy::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
         refresh->set(refresh->spread, spread);
     }
 
-    if (ParserKeyword{"DEPENDS ON"}.ignore(pos, expected))
+    if (s_depends_on.ignore(pos, expected))
     {
         if (refresh->schedule_kind == RefreshScheduleKind::AFTER)
             throw Exception(ErrorCodes::BAD_ARGUMENTS,
@@ -81,7 +88,7 @@ bool ParserRefreshStrategy::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
     }
 
     // Refresh SETTINGS
-    if (ParserKeyword{"SETTINGS"}.ignore(pos, expected))
+    if (s_settings.ignore(pos, expected))
     {
         /// Settings are written like SET query, so parse them with ParserSetQuery
         ASTPtr settings;
diff --git a/src/Parsers/ParserRenameQuery.cpp b/src/Parsers/ParserRenameQuery.cpp
index f9d29108ed6..6b652c27f6c 100644
--- a/src/Parsers/ParserRenameQuery.cpp
+++ b/src/Parsers/ParserRenameQuery.cpp
@@ -11,15 +11,15 @@ namespace DB
 
 bool ParserRenameQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_rename("RENAME");
-    ParserKeyword s_rename_table("RENAME TABLE");
-    ParserKeyword s_exchange_tables("EXCHANGE TABLES");
-    ParserKeyword s_rename_dictionary("RENAME DICTIONARY");
-    ParserKeyword s_exchange_dictionaries("EXCHANGE DICTIONARIES");
-    ParserKeyword s_rename_database("RENAME DATABASE");
-    ParserKeyword s_if_exists("IF EXISTS");
-    ParserKeyword s_to("TO");
-    ParserKeyword s_and("AND");
+    ParserKeyword s_rename(Keyword::RENAME);
+    ParserKeyword s_rename_table(Keyword::RENAME_TABLE);
+    ParserKeyword s_exchange_tables(Keyword::EXCHANGE_TABLES);
+    ParserKeyword s_rename_dictionary(Keyword::RENAME_DICTIONARY);
+    ParserKeyword s_exchange_dictionaries(Keyword::EXCHANGE_DICTIONARIES);
+    ParserKeyword s_rename_database(Keyword::RENAME_DATABASE);
+    ParserKeyword s_if_exists(Keyword::IF_EXISTS);
+    ParserKeyword s_to(Keyword::TO);
+    ParserKeyword s_and(Keyword::AND);
     ParserToken s_comma(TokenType::Comma);
 
     bool exchange = false;
@@ -39,7 +39,7 @@ bool ParserRenameQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
             return false;
 
         String cluster_str;
-        if (ParserKeyword{"ON"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::ON}.ignore(pos, expected))
         {
             if (!ASTQueryWithOnCluster::parse(pos, cluster_str, expected))
                 return false;
@@ -94,7 +94,7 @@ bool ParserRenameQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
     }
 
     String cluster_str;
-    if (ParserKeyword{"ON"}.ignore(pos, expected))
+    if (ParserKeyword{Keyword::ON}.ignore(pos, expected))
     {
         if (!ASTQueryWithOnCluster::parse(pos, cluster_str, expected))
             return false;
diff --git a/src/Parsers/ParserSelectQuery.cpp b/src/Parsers/ParserSelectQuery.cpp
index 641e74b5f18..ce15c7b03fd 100644
--- a/src/Parsers/ParserSelectQuery.cpp
+++ b/src/Parsers/ParserSelectQuery.cpp
@@ -37,35 +37,35 @@ bool ParserSelectQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
     auto select_query = std::make_shared<ASTSelectQuery>();
     node = select_query;
 
-    ParserKeyword s_select("SELECT");
-    ParserKeyword s_all("ALL");
-    ParserKeyword s_distinct("DISTINCT");
-    ParserKeyword s_distinct_on("DISTINCT ON");
-    ParserKeyword s_from("FROM");
-    ParserKeyword s_prewhere("PREWHERE");
-    ParserKeyword s_where("WHERE");
-    ParserKeyword s_group_by("GROUP BY");
-    ParserKeyword s_with("WITH");
-    ParserKeyword s_totals("TOTALS");
-    ParserKeyword s_having("HAVING");
-    ParserKeyword s_window("WINDOW");
-    ParserKeyword s_order_by("ORDER BY");
-    ParserKeyword s_limit("LIMIT");
-    ParserKeyword s_settings("SETTINGS");
-    ParserKeyword s_by("BY");
-    ParserKeyword s_rollup("ROLLUP");
-    ParserKeyword s_cube("CUBE");
-    ParserKeyword s_grouping_sets("GROUPING SETS");
-    ParserKeyword s_top("TOP");
-    ParserKeyword s_with_ties("WITH TIES");
-    ParserKeyword s_offset("OFFSET");
-    ParserKeyword s_fetch("FETCH");
-    ParserKeyword s_only("ONLY");
-    ParserKeyword s_row("ROW");
-    ParserKeyword s_rows("ROWS");
-    ParserKeyword s_first("FIRST");
-    ParserKeyword s_next("NEXT");
-    ParserKeyword s_interpolate("INTERPOLATE");
+    ParserKeyword s_select(Keyword::SELECT);
+    ParserKeyword s_all(Keyword::ALL);
+    ParserKeyword s_distinct(Keyword::DISTINCT);
+    ParserKeyword s_distinct_on(Keyword::DISTINCT_ON);
+    ParserKeyword s_from(Keyword::FROM);
+    ParserKeyword s_prewhere(Keyword::PREWHERE);
+    ParserKeyword s_where(Keyword::WHERE);
+    ParserKeyword s_group_by(Keyword::GROUP_BY);
+    ParserKeyword s_with(Keyword::WITH);
+    ParserKeyword s_totals(Keyword::TOTALS);
+    ParserKeyword s_having(Keyword::HAVING);
+    ParserKeyword s_window(Keyword::WINDOW);
+    ParserKeyword s_order_by(Keyword::ORDER_BY);
+    ParserKeyword s_limit(Keyword::LIMIT);
+    ParserKeyword s_settings(Keyword::SETTINGS);
+    ParserKeyword s_by(Keyword::BY);
+    ParserKeyword s_rollup(Keyword::ROLLUP);
+    ParserKeyword s_cube(Keyword::CUBE);
+    ParserKeyword s_grouping_sets(Keyword::GROUPING_SETS);
+    ParserKeyword s_top(Keyword::TOP);
+    ParserKeyword s_with_ties(Keyword::WITH_TIES);
+    ParserKeyword s_offset(Keyword::OFFSET);
+    ParserKeyword s_fetch(Keyword::FETCH);
+    ParserKeyword s_only(Keyword::ONLY);
+    ParserKeyword s_row(Keyword::ROW);
+    ParserKeyword s_rows(Keyword::ROWS);
+    ParserKeyword s_first(Keyword::FIRST);
+    ParserKeyword s_next(Keyword::NEXT);
+    ParserKeyword s_interpolate(Keyword::INTERPOLATE);
 
     ParserNotEmptyExpressionList exp_list(false);
     ParserNotEmptyExpressionList exp_list_for_with_clause(false);
diff --git a/src/Parsers/ParserSetQuery.cpp b/src/Parsers/ParserSetQuery.cpp
index 4ad715ba499..13b881635cd 100644
--- a/src/Parsers/ParserSetQuery.cpp
+++ b/src/Parsers/ParserSetQuery.cpp
@@ -210,9 +210,9 @@ bool ParserSetQuery::parseNameValuePair(SettingChange & change, IParser::Pos & p
     if (!s_eq.ignore(pos, expected))
         return false;
 
-    if (ParserKeyword("TRUE").ignore(pos, expected))
+    if (ParserKeyword(Keyword::TRUE_KEYWORD).ignore(pos, expected))
         value = std::make_shared<ASTLiteral>(Field(static_cast<UInt64>(1)));
-    else if (ParserKeyword("FALSE").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::FALSE_KEYWORD).ignore(pos, expected))
         value = std::make_shared<ASTLiteral>(Field(static_cast<UInt64>(0)));
     /// for SETTINGS disk=disk(type='s3', path='', ...)
     else if (function_p.parse(pos, function_ast, expected) && function_ast->as<ASTFunction>()->name == "disk")
@@ -269,16 +269,16 @@ bool ParserSetQuery::parseNameValuePairWithParameterOrDefault(
     }
 
     /// Default
-    if (ParserKeyword("DEFAULT").ignore(pos, expected))
+    if (ParserKeyword(Keyword::DEFAULT).ignore(pos, expected))
     {
         default_settings = name;
         return true;
     }
 
     /// Setting
-    if (ParserKeyword("TRUE").ignore(pos, expected))
+    if (ParserKeyword(Keyword::TRUE_KEYWORD).ignore(pos, expected))
         node = std::make_shared<ASTLiteral>(Field(static_cast<UInt64>(1)));
-    else if (ParserKeyword("FALSE").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::FALSE_KEYWORD).ignore(pos, expected))
         node = std::make_shared<ASTLiteral>(Field(static_cast<UInt64>(0)));
     else if (function_p.parse(pos, function_ast, expected) && function_ast->as<ASTFunction>()->name == "disk")
     {
@@ -303,13 +303,13 @@ bool ParserSetQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 
     if (!parse_only_internals)
     {
-        ParserKeyword s_set("SET");
+        ParserKeyword s_set(Keyword::SET);
 
         if (!s_set.ignore(pos, expected))
             return false;
 
         /// Parse SET TRANSACTION ... queries using ParserTransactionControl
-        if (ParserKeyword{"TRANSACTION"}.check(pos, expected))
+        if (ParserKeyword{Keyword::TRANSACTION}.check(pos, expected))
             return false;
     }
 
diff --git a/src/Parsers/ParserShowColumnsQuery.cpp b/src/Parsers/ParserShowColumnsQuery.cpp
index e5fc5c39701..5d26d7bf1d4 100644
--- a/src/Parsers/ParserShowColumnsQuery.cpp
+++ b/src/Parsers/ParserShowColumnsQuery.cpp
@@ -23,19 +23,19 @@ bool ParserShowColumnsQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe
 
     auto query = std::make_shared<ASTShowColumnsQuery>();
 
-    if (!ParserKeyword("SHOW").ignore(pos, expected))
+    if (!ParserKeyword(Keyword::SHOW).ignore(pos, expected))
         return false;
 
-    if (ParserKeyword("EXTENDED").ignore(pos, expected))
+    if (ParserKeyword(Keyword::EXTENDED).ignore(pos, expected))
         query->extended = true;
 
-    if (ParserKeyword("FULL").ignore(pos, expected))
+    if (ParserKeyword(Keyword::FULL).ignore(pos, expected))
         query->full = true;
 
-    if (!(ParserKeyword("COLUMNS").ignore(pos, expected) || ParserKeyword("FIELDS").ignore(pos, expected)))
+    if (!(ParserKeyword(Keyword::COLUMNS).ignore(pos, expected) || ParserKeyword(Keyword::FIELDS).ignore(pos, expected)))
         return false;
 
-    if (ParserKeyword("FROM").ignore(pos, expected) || ParserKeyword("IN").ignore(pos, expected))
+    if (ParserKeyword(Keyword::FROM).ignore(pos, expected) || ParserKeyword(Keyword::IN).ignore(pos, expected))
     {
         if (!ParserCompoundIdentifier().parse(pos, from1, expected))
             return false;
@@ -55,7 +55,7 @@ bool ParserShowColumnsQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe
     }
     else
     {
-        if (ParserKeyword("FROM").ignore(pos, expected) || ParserKeyword("IN").ignore(pos, expected))
+        if (ParserKeyword(Keyword::FROM).ignore(pos, expected) || ParserKeyword(Keyword::IN).ignore(pos, expected))
             if (!ParserIdentifier().parse(pos, from2, expected))
                 return false;
 
@@ -65,10 +65,10 @@ bool ParserShowColumnsQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe
         query->database = from2_str;
     }
 
-    if (ParserKeyword("NOT").ignore(pos, expected))
+    if (ParserKeyword(Keyword::NOT).ignore(pos, expected))
         query->not_like = true;
 
-    if (bool insensitive = ParserKeyword("ILIKE").ignore(pos, expected); insensitive || ParserKeyword("LIKE").ignore(pos, expected))
+    if (bool insensitive = ParserKeyword(Keyword::ILIKE).ignore(pos, expected); insensitive || ParserKeyword(Keyword::LIKE).ignore(pos, expected))
     {
         if (insensitive)
             query->case_insensitive_like = true;
@@ -78,11 +78,11 @@ bool ParserShowColumnsQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe
     }
     else if (query->not_like)
         return false;
-    else if (ParserKeyword("WHERE").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::WHERE).ignore(pos, expected))
         if (!ParserExpressionWithOptionalAlias(false).parse(pos, query->where_expression, expected))
             return false;
 
-    if (ParserKeyword("LIMIT").ignore(pos, expected))
+    if (ParserKeyword(Keyword::LIMIT).ignore(pos, expected))
         if (!ParserExpressionWithOptionalAlias(false).parse(pos, query->limit_length, expected))
             return false;
 
diff --git a/src/Parsers/ParserShowEngineQuery.h b/src/Parsers/ParserShowEngineQuery.h
index e06326436f1..0e9a742c0e9 100644
--- a/src/Parsers/ParserShowEngineQuery.h
+++ b/src/Parsers/ParserShowEngineQuery.h
@@ -20,7 +20,7 @@ protected:
     {
         auto query = std::make_shared<ASTShowEnginesQuery>();
 
-        if (!ParserKeyword("SHOW ENGINES").ignore(pos, expected))
+        if (!ParserKeyword(Keyword::SHOW_ENGINES).ignore(pos, expected))
             return false;
 
         node = query;
diff --git a/src/Parsers/ParserShowFunctionsQuery.cpp b/src/Parsers/ParserShowFunctionsQuery.cpp
index 524d936c4f1..37b68dcec08 100644
--- a/src/Parsers/ParserShowFunctionsQuery.cpp
+++ b/src/Parsers/ParserShowFunctionsQuery.cpp
@@ -13,10 +13,10 @@ bool ParserShowFunctionsQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & ex
     ASTPtr like;
 
     auto query = std::make_shared<ASTShowFunctionsQuery>();
-    if (!ParserKeyword("SHOW FUNCTIONS").ignore(pos, expected))
+    if (!ParserKeyword(Keyword::SHOW_FUNCTIONS).ignore(pos, expected))
         return false;
 
-    if (bool insensitive = ParserKeyword("ILIKE").ignore(pos, expected); insensitive || ParserKeyword("LIKE").ignore(pos, expected))
+    if (bool insensitive = ParserKeyword(Keyword::ILIKE).ignore(pos, expected); insensitive || ParserKeyword(Keyword::LIKE).ignore(pos, expected))
     {
         if (insensitive)
             query->case_insensitive_like = true;
diff --git a/src/Parsers/ParserShowIndexesQuery.cpp b/src/Parsers/ParserShowIndexesQuery.cpp
index 7852ed9806d..495dfc5101f 100644
--- a/src/Parsers/ParserShowIndexesQuery.cpp
+++ b/src/Parsers/ParserShowIndexesQuery.cpp
@@ -22,16 +22,16 @@ bool ParserShowIndexesQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe
 
     auto query = std::make_shared<ASTShowIndexesQuery>();
 
-    if (!ParserKeyword("SHOW").ignore(pos, expected))
+    if (!ParserKeyword(Keyword::SHOW).ignore(pos, expected))
         return false;
 
-    if (ParserKeyword("EXTENDED").ignore(pos, expected))
+    if (ParserKeyword(Keyword::EXTENDED).ignore(pos, expected))
         query->extended = true;
 
-    if (!(ParserKeyword("INDEX").ignore(pos, expected) || ParserKeyword("INDEXES").ignore(pos, expected) || ParserKeyword("INDICES").ignore(pos, expected) || ParserKeyword("KEYS").ignore(pos, expected)))
+    if (!(ParserKeyword(Keyword::INDEX).ignore(pos, expected) || ParserKeyword(Keyword::INDEXES).ignore(pos, expected) || ParserKeyword(Keyword::INDICES).ignore(pos, expected) || ParserKeyword(Keyword::KEYS).ignore(pos, expected)))
         return false;
 
-    if (ParserKeyword("FROM").ignore(pos, expected) || ParserKeyword("IN").ignore(pos, expected))
+    if (ParserKeyword(Keyword::FROM).ignore(pos, expected) || ParserKeyword(Keyword::IN).ignore(pos, expected))
     {
         if (!ParserCompoundIdentifier().parse(pos, from1, expected))
             return false;
@@ -51,7 +51,7 @@ bool ParserShowIndexesQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe
     }
     else
     {
-        if (ParserKeyword("FROM").ignore(pos, expected) || ParserKeyword("IN").ignore(pos, expected))
+        if (ParserKeyword(Keyword::FROM).ignore(pos, expected) || ParserKeyword(Keyword::IN).ignore(pos, expected))
             if (!ParserIdentifier().parse(pos, from2, expected))
                 return false;
 
@@ -61,7 +61,7 @@ bool ParserShowIndexesQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe
         query->database = from2_str;
     }
 
-    if (ParserKeyword("WHERE").ignore(pos, expected))
+    if (ParserKeyword(Keyword::WHERE).ignore(pos, expected))
         if (!ParserExpressionWithOptionalAlias(false).parse(pos, query->where_expression, expected))
             return false;
 
diff --git a/src/Parsers/ParserShowProcesslistQuery.h b/src/Parsers/ParserShowProcesslistQuery.h
index de08894e059..29c2970a8a1 100644
--- a/src/Parsers/ParserShowProcesslistQuery.h
+++ b/src/Parsers/ParserShowProcesslistQuery.h
@@ -20,7 +20,7 @@ protected:
     {
         auto query = std::make_shared<ASTShowProcesslistQuery>();
 
-        if (!ParserKeyword("SHOW PROCESSLIST").ignore(pos, expected))
+        if (!ParserKeyword(Keyword::SHOW_PROCESSLIST).ignore(pos, expected))
             return false;
 
         node = query;
diff --git a/src/Parsers/ParserShowSettingQuery.cpp b/src/Parsers/ParserShowSettingQuery.cpp
index 2586cbdfb43..9e4736a5eb7 100644
--- a/src/Parsers/ParserShowSettingQuery.cpp
+++ b/src/Parsers/ParserShowSettingQuery.cpp
@@ -11,7 +11,7 @@ namespace DB
 
 bool ParserShowSettingQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    if (!ParserKeyword("SHOW SETTING").ignore(pos, expected))
+    if (!ParserKeyword(Keyword::SHOW_SETTING).ignore(pos, expected))
         return false;
 
     ASTPtr setting_name_identifier;
diff --git a/src/Parsers/ParserShowTablesQuery.cpp b/src/Parsers/ParserShowTablesQuery.cpp
index 2b220dd0a0c..68b89bd0551 100644
--- a/src/Parsers/ParserShowTablesQuery.cpp
+++ b/src/Parsers/ParserShowTablesQuery.cpp
@@ -17,25 +17,25 @@ namespace DB
 
 bool ParserShowTablesQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_show("SHOW");
-    ParserKeyword s_full("FULL");
-    ParserKeyword s_temporary("TEMPORARY");
-    ParserKeyword s_tables("TABLES");
-    ParserKeyword s_databases("DATABASES");
-    ParserKeyword s_clusters("CLUSTERS");
-    ParserKeyword s_cluster("CLUSTER");
-    ParserKeyword s_dictionaries("DICTIONARIES");
-    ParserKeyword s_caches("FILESYSTEM CACHES");
-    ParserKeyword s_settings("SETTINGS");
-    ParserKeyword s_merges("MERGES");
-    ParserKeyword s_changed("CHANGED");
-    ParserKeyword s_from("FROM");
-    ParserKeyword s_in("IN");
-    ParserKeyword s_not("NOT");
-    ParserKeyword s_like("LIKE");
-    ParserKeyword s_ilike("ILIKE");
-    ParserKeyword s_where("WHERE");
-    ParserKeyword s_limit("LIMIT");
+    ParserKeyword s_show(Keyword::SHOW);
+    ParserKeyword s_full(Keyword::FULL);
+    ParserKeyword s_temporary(Keyword::TEMPORARY);
+    ParserKeyword s_tables(Keyword::TABLES);
+    ParserKeyword s_databases(Keyword::DATABASES);
+    ParserKeyword s_clusters(Keyword::CLUSTERS);
+    ParserKeyword s_cluster(Keyword::CLUSTER);
+    ParserKeyword s_dictionaries(Keyword::DICTIONARIES);
+    ParserKeyword s_caches(Keyword::FILESYSTEM_CACHES);
+    ParserKeyword s_settings(Keyword::SETTINGS);
+    ParserKeyword s_merges(Keyword::MERGES);
+    ParserKeyword s_changed(Keyword::CHANGED);
+    ParserKeyword s_from(Keyword::FROM);
+    ParserKeyword s_in(Keyword::IN);
+    ParserKeyword s_not(Keyword::NOT);
+    ParserKeyword s_like(Keyword::LIKE);
+    ParserKeyword s_ilike(Keyword::ILIKE);
+    ParserKeyword s_where(Keyword::WHERE);
+    ParserKeyword s_limit(Keyword::LIMIT);
     ParserStringLiteral like_p;
     ParserIdentifier name_p(true);
     ParserExpressionWithOptionalAlias exp_elem(false);
@@ -147,7 +147,7 @@ bool ParserShowTablesQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
                 return false;
         }
 
-        /// Not expected due to "SHOW SETTINGS PROFILES"
+        /// Not expected due to Keyword::SHOW SETTINGS PROFILES
         if (bool insensitive = s_ilike.ignore(pos, expected); insensitive || s_like.ignore(pos, expected))
         {
             if (insensitive)
diff --git a/src/Parsers/ParserSystemQuery.cpp b/src/Parsers/ParserSystemQuery.cpp
index a50e65aa134..df168e74772 100644
--- a/src/Parsers/ParserSystemQuery.cpp
+++ b/src/Parsers/ParserSystemQuery.cpp
@@ -28,7 +28,7 @@ namespace ErrorCodes
     String cluster;
     bool parsed_on_cluster = false;
 
-    if (ParserKeyword{"ON"}.ignore(pos, expected))
+    if (ParserKeyword{Keyword::ON}.ignore(pos, expected))
     {
         if (!ASTQueryWithOnCluster::parse(pos, cluster, expected))
             return false;
@@ -52,7 +52,7 @@ namespace ErrorCodes
     if (!parsed_table && require_table)
         return false;
 
-    if (!parsed_on_cluster && ParserKeyword{"ON"}.ignore(pos, expected))
+    if (!parsed_on_cluster && ParserKeyword{Keyword::ON}.ignore(pos, expected))
         if (!ASTQueryWithOnCluster::parse(pos, cluster, expected))
             return false;
 
@@ -82,7 +82,7 @@ enum class SystemQueryTargetType
     String cluster;
     bool parsed_on_cluster = false;
 
-    if (ParserKeyword{"ON"}.ignore(pos, expected))
+    if (ParserKeyword{Keyword::ON}.ignore(pos, expected))
     {
         if (!ASTQueryWithOnCluster::parse(pos, cluster, expected))
             return false;
@@ -108,7 +108,7 @@ enum class SystemQueryTargetType
             return false;
     }
 
-    if (!parsed_on_cluster && ParserKeyword{"ON"}.ignore(pos, expected))
+    if (!parsed_on_cluster && ParserKeyword{Keyword::ON}.ignore(pos, expected))
     {
         if (!ASTQueryWithOnCluster::parse(pos, cluster, expected))
             return false;
@@ -142,7 +142,7 @@ enum class SystemQueryTargetType
                                     Expected & expected)
 {
     String cluster_str;
-    if (ParserKeyword{"ON"}.ignore(pos, expected))
+    if (ParserKeyword{Keyword::ON}.ignore(pos, expected))
     {
         if (!ASTQueryWithOnCluster::parse(pos, cluster_str, expected))
             return false;
@@ -162,29 +162,29 @@ enum class SystemQueryTargetType
         return false;
     res->replica = ast->as<ASTLiteral &>().value.safeGet<String>();
 
-    if (ParserKeyword{"FROM SHARD"}.ignore(pos, expected))
+    if (ParserKeyword{Keyword::FROM_SHARD}.ignore(pos, expected))
     {
         if (!ParserStringLiteral{}.parse(pos, ast, expected))
             return false;
         res->shard = ast->as<ASTLiteral &>().value.safeGet<String>();
     }
 
-    if (ParserKeyword{"FROM"}.ignore(pos, expected))
+    if (ParserKeyword{Keyword::FROM}.ignore(pos, expected))
     {
         // way 1. parse replica database
         // way 2. parse replica table
         // way 3. parse replica zkpath
-        if (ParserKeyword{"DATABASE"}.ignore(pos, expected))
+        if (ParserKeyword{Keyword::DATABASE}.ignore(pos, expected))
         {
             ParserIdentifier database_parser;
             if (!database_parser.parse(pos, res->database, expected))
                 return false;
         }
-        else if (!database && ParserKeyword{"TABLE"}.ignore(pos, expected))
+        else if (!database && ParserKeyword{Keyword::TABLE}.ignore(pos, expected))
         {
             parseDatabaseAndTableAsAST(pos, expected, res->database, res->table);
         }
-        else if (ParserKeyword{"ZKPATH"}.ignore(pos, expected))
+        else if (ParserKeyword{Keyword::ZKPATH}.ignore(pos, expected))
         {
             ASTPtr path_ast;
             if (!ParserStringLiteral{}.parse(pos, path_ast, expected))
@@ -205,7 +205,7 @@ enum class SystemQueryTargetType
 
 bool ParserSystemQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected & expected)
 {
-    if (!ParserKeyword{"SYSTEM"}.ignore(pos, expected))
+    if (!ParserKeyword{Keyword::SYSTEM}.ignore(pos, expected))
         return false;
 
     using Type = ASTSystemQuery::Type;
@@ -216,7 +216,7 @@ bool ParserSystemQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected &
 
     for (const auto & type : magic_enum::enum_values<Type>())
     {
-        if (ParserKeyword{ASTSystemQuery::typeToString(type)}.ignore(pos, expected))
+        if (ParserKeyword::createDeprecated(ASTSystemQuery::typeToString(type)).ignore(pos, expected))
         {
             res->type = type;
             found = true;
@@ -281,12 +281,12 @@ bool ParserSystemQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected &
                 return false;
             if (res->type == Type::SYNC_REPLICA)
             {
-                if (ParserKeyword{"STRICT"}.ignore(pos, expected))
+                if (ParserKeyword{Keyword::STRICT}.ignore(pos, expected))
                     res->sync_replica_mode = SyncReplicaMode::STRICT;
-                else if (ParserKeyword{"LIGHTWEIGHT"}.ignore(pos, expected))
+                else if (ParserKeyword{Keyword::LIGHTWEIGHT}.ignore(pos, expected))
                 {
                     res->sync_replica_mode = SyncReplicaMode::LIGHTWEIGHT;
-                    if (ParserKeyword{"FROM"}.ignore(pos, expected))
+                    if (ParserKeyword{Keyword::FROM}.ignore(pos, expected))
                     {
                         do
                         {
@@ -297,7 +297,7 @@ bool ParserSystemQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected &
                         } while (ParserToken{TokenType::Comma}.ignore(pos, expected));
                     }
                 }
-                else if (ParserKeyword{"PULL"}.ignore(pos, expected))
+                else if (ParserKeyword{Keyword::PULL}.ignore(pos, expected))
                     res->sync_replica_mode = SyncReplicaMode::PULL;
             }
             break;
@@ -360,7 +360,7 @@ bool ParserSystemQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected &
                 return true;
             };
 
-            if (ParserKeyword{"ON VOLUME"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::ON_VOLUME}.ignore(pos, expected))
             {
                 if (!parse_on_volume())
                     return false;
@@ -369,7 +369,7 @@ bool ParserSystemQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected &
             {
                 if (!parseQueryWithOnCluster(res, pos, expected))
                     return false;
-                if (ParserKeyword{"ON VOLUME"}.ignore(pos, expected))
+                if (ParserKeyword{Keyword::ON_VOLUME}.ignore(pos, expected))
                 {
                     if (!parse_on_volume())
                         return false;
@@ -419,10 +419,10 @@ bool ParserSystemQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected &
             if (!parseDatabaseAndTableAsAST(pos, expected, res->database, res->table))
                 return false;
 
-            if (ParserKeyword{"UNSET FAKE TIME"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::UNSET_FAKE_TIME}.ignore(pos, expected))
                 break;
 
-            if (!ParserKeyword{"SET FAKE TIME"}.ignore(pos, expected))
+            if (!ParserKeyword{Keyword::SET_FAKE_TIME}.ignore(pos, expected))
                 return false;
             ASTPtr ast;
             if (!ParserStringLiteral{}.parse(pos, ast, expected))
@@ -442,9 +442,9 @@ bool ParserSystemQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected &
                 return false;
 
             ASTPtr seconds;
-            if (!(ParserKeyword{"FOR"}.ignore(pos, expected)
+            if (!(ParserKeyword{Keyword::FOR}.ignore(pos, expected)
                 && ParserUnsignedInteger().parse(pos, seconds, expected)
-                && ParserKeyword{"SECOND"}.ignore(pos, expected)))   /// SECOND, not SECONDS to be consistent with INTERVAL parsing in SQL
+                && ParserKeyword{Keyword::SECOND}.ignore(pos, expected)))   /// SECOND, not SECONDS to be consistent with INTERVAL parsing in SQL
             {
                 return false;
             }
@@ -459,10 +459,10 @@ bool ParserSystemQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected &
             if (path_parser.parse(pos, ast, expected))
             {
                 res->filesystem_cache_name = ast->as<ASTLiteral>()->value.safeGet<String>();
-                if (ParserKeyword{"KEY"}.ignore(pos, expected) && ParserIdentifier().parse(pos, ast, expected))
+                if (ParserKeyword{Keyword::KEY}.ignore(pos, expected) && ParserIdentifier().parse(pos, ast, expected))
                 {
                     res->key_to_drop = ast->as<ASTIdentifier>()->name();
-                    if (ParserKeyword{"OFFSET"}.ignore(pos, expected) && ParserLiteral().parse(pos, ast, expected))
+                    if (ParserKeyword{Keyword::OFFSET}.ignore(pos, expected) && ParserLiteral().parse(pos, ast, expected))
                         res->offset_to_drop = ast->as<ASTLiteral>()->value.safeGet<UInt64>();
                 }
             }
@@ -486,18 +486,18 @@ bool ParserSystemQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected &
         }
         case Type::DROP_SCHEMA_CACHE:
         {
-            if (ParserKeyword{"FOR"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::FOR}.ignore(pos, expected))
             {
-                if (ParserKeyword{"FILE"}.ignore(pos, expected))
-                    res->schema_cache_storage = "FILE";
-                else if (ParserKeyword{"S3"}.ignore(pos, expected))
-                    res->schema_cache_storage = "S3";
-                else if (ParserKeyword{"HDFS"}.ignore(pos, expected))
-                    res->schema_cache_storage = "HDFS";
-                else if (ParserKeyword{"URL"}.ignore(pos, expected))
-                    res->schema_cache_storage = "URL";
-                else if (ParserKeyword{"AZURE"}.ignore(pos, expected))
-                    res->schema_cache_storage = "AZURE";
+                if (ParserKeyword{Keyword::FILE}.ignore(pos, expected))
+                    res->schema_cache_storage = toStringView(Keyword::FILE);
+                else if (ParserKeyword{Keyword::S3}.ignore(pos, expected))
+                    res->schema_cache_storage = toStringView(Keyword::S3);
+                else if (ParserKeyword{Keyword::HDFS}.ignore(pos, expected))
+                    res->schema_cache_storage = toStringView(Keyword::HDFS);
+                else if (ParserKeyword{Keyword::URL}.ignore(pos, expected))
+                    res->schema_cache_storage = toStringView(Keyword::URL);
+                else if (ParserKeyword{Keyword::AZURE}.ignore(pos, expected))
+                    res->schema_cache_storage = toStringView(Keyword::AZURE);
                 else
                     return false;
             }
@@ -505,10 +505,11 @@ bool ParserSystemQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected &
         }
         case Type::DROP_FORMAT_SCHEMA_CACHE:
         {
-            if (ParserKeyword{"FOR"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::FOR}.ignore(pos, expected))
             {
-                if (ParserKeyword{"Protobuf"}.ignore(pos, expected))
-                    res->schema_cache_format = "Protobuf";
+                if (ParserKeyword{Keyword::PROTOBUF}.ignore(pos, expected))
+                    res->schema_cache_format = toStringView(Keyword::PROTOBUF);
+
                 else
                     return false;
             }
@@ -517,7 +518,7 @@ bool ParserSystemQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected &
         case Type::UNFREEZE:
         {
             ASTPtr ast;
-            if (ParserKeyword{"WITH NAME"}.ignore(pos, expected) && ParserStringLiteral{}.parse(pos, ast, expected))
+            if (ParserKeyword{Keyword::WITH_NAME}.ignore(pos, expected) && ParserStringLiteral{}.parse(pos, ast, expected))
             {
                 res->backup_name = ast->as<ASTLiteral &>().value.get<const String &>();
             }
@@ -541,7 +542,7 @@ bool ParserSystemQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected &
 
                 for (const auto & cur_type : magic_enum::enum_values<ServerType::Type>())
                 {
-                    if (ParserKeyword{ServerType::serverTypeToString(cur_type)}.ignore(pos, expected))
+                    if (ParserKeyword::createDeprecated(ServerType::serverTypeToString(cur_type)).ignore(pos, expected))
                     {
                         type = cur_type;
                         break;
@@ -573,7 +574,7 @@ bool ParserSystemQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected &
             if (!parse_server_type(base_type, base_custom_name))
                 return false;
 
-            if (ParserKeyword{"EXCEPT"}.ignore(pos, expected))
+            if (ParserKeyword{Keyword::EXCEPT}.ignore(pos, expected))
             {
                 if (base_type != ServerType::Type::QUERIES_ALL &&
                     base_type != ServerType::Type::QUERIES_DEFAULT &&
diff --git a/src/Parsers/ParserTablePropertiesQuery.cpp b/src/Parsers/ParserTablePropertiesQuery.cpp
index 94f264fcc89..48a5c0ebde0 100644
--- a/src/Parsers/ParserTablePropertiesQuery.cpp
+++ b/src/Parsers/ParserTablePropertiesQuery.cpp
@@ -12,14 +12,14 @@ namespace DB
 
 bool ParserTablePropertiesQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_exists("EXISTS");
-    ParserKeyword s_temporary("TEMPORARY");
-    ParserKeyword s_show("SHOW");
-    ParserKeyword s_create("CREATE");
-    ParserKeyword s_database("DATABASE");
-    ParserKeyword s_table("TABLE");
-    ParserKeyword s_view("VIEW");
-    ParserKeyword s_dictionary("DICTIONARY");
+    ParserKeyword s_exists(Keyword::EXISTS);
+    ParserKeyword s_temporary(Keyword::TEMPORARY);
+    ParserKeyword s_show(Keyword::SHOW);
+    ParserKeyword s_create(Keyword::CREATE);
+    ParserKeyword s_database(Keyword::DATABASE);
+    ParserKeyword s_table(Keyword::TABLE);
+    ParserKeyword s_view(Keyword::VIEW);
+    ParserKeyword s_dictionary(Keyword::DICTIONARY);
     ParserToken s_dot(TokenType::Dot);
     ParserIdentifier name_p(true);
 
diff --git a/src/Parsers/ParserTablesInSelectQuery.cpp b/src/Parsers/ParserTablesInSelectQuery.cpp
index ca209739dad..ab35141403a 100644
--- a/src/Parsers/ParserTablesInSelectQuery.cpp
+++ b/src/Parsers/ParserTablesInSelectQuery.cpp
@@ -31,11 +31,11 @@ bool ParserTableExpression::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
         return false;
 
     /// FINAL
-    if (ParserKeyword("FINAL").ignore(pos, expected))
+    if (ParserKeyword(Keyword::FINAL).ignore(pos, expected))
         res->final = true;
 
     /// SAMPLE number
-    if (ParserKeyword("SAMPLE").ignore(pos, expected))
+    if (ParserKeyword(Keyword::SAMPLE).ignore(pos, expected))
     {
         ParserSampleRatio ratio;
 
@@ -43,7 +43,7 @@ bool ParserTableExpression::parseImpl(Pos & pos, ASTPtr & node, Expected & expec
             return false;
 
         /// OFFSET number
-        if (ParserKeyword("OFFSET").ignore(pos, expected))
+        if (ParserKeyword(Keyword::OFFSET).ignore(pos, expected))
         {
             if (!ratio.parse(pos, res->sample_offset, expected))
                 return false;
@@ -76,7 +76,7 @@ bool ParserArrayJoin::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
     Pos saved_pos = pos;
     bool has_array_join = false;
 
-    if (ParserKeyword("LEFT ARRAY JOIN").ignore(pos, expected))
+    if (ParserKeyword(Keyword::LEFT_ARRAY_JOIN).ignore(pos, expected))
     {
         res->kind = ASTArrayJoin::Kind::Left;
         has_array_join = true;
@@ -86,9 +86,9 @@ bool ParserArrayJoin::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
         pos = saved_pos;
 
         /// INNER may be specified explicitly, otherwise it is assumed as default.
-        ParserKeyword("INNER").ignore(pos, expected);
+        ParserKeyword(Keyword::INNER).ignore(pos, expected);
 
-        if (ParserKeyword("ARRAY JOIN").ignore(pos, expected))
+        if (ParserKeyword(Keyword::ARRAY_JOIN).ignore(pos, expected))
         {
             res->kind = ASTArrayJoin::Kind::Inner;
             has_array_join = true;
@@ -111,15 +111,15 @@ bool ParserArrayJoin::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 
 void ParserTablesInSelectQueryElement::parseJoinStrictness(Pos & pos, ASTTableJoin & table_join)
 {
-    if (ParserKeyword("ANY").ignore(pos))
+    if (ParserKeyword(Keyword::ANY).ignore(pos))
         table_join.strictness = JoinStrictness::Any;
-    else if (ParserKeyword("ALL").ignore(pos))
+    else if (ParserKeyword(Keyword::ALL).ignore(pos))
         table_join.strictness = JoinStrictness::All;
-    else if (ParserKeyword("ASOF").ignore(pos))
+    else if (ParserKeyword(Keyword::ASOF).ignore(pos))
         table_join.strictness = JoinStrictness::Asof;
-    else if (ParserKeyword("SEMI").ignore(pos))
+    else if (ParserKeyword(Keyword::SEMI).ignore(pos))
         table_join.strictness = JoinStrictness::Semi;
-    else if (ParserKeyword("ANTI").ignore(pos) || ParserKeyword("ONLY").ignore(pos))
+    else if (ParserKeyword(Keyword::ANTI).ignore(pos) || ParserKeyword(Keyword::ONLY).ignore(pos))
         table_join.strictness = JoinStrictness::Anti;
 }
 
@@ -146,9 +146,9 @@ bool ParserTablesInSelectQueryElement::parseImpl(Pos & pos, ASTPtr & node, Expec
         }
         else
         {
-            if (ParserKeyword("GLOBAL").ignore(pos))
+            if (ParserKeyword(Keyword::GLOBAL).ignore(pos))
                 table_join->locality = JoinLocality::Global;
-            else if (ParserKeyword("LOCAL").ignore(pos))
+            else if (ParserKeyword(Keyword::LOCAL).ignore(pos))
                 table_join->locality = JoinLocality::Local;
 
             table_join->strictness = JoinStrictness::Unspecified;
@@ -157,17 +157,17 @@ bool ParserTablesInSelectQueryElement::parseImpl(Pos & pos, ASTPtr & node, Expec
             parseJoinStrictness(pos, *table_join);
 
             bool no_kind = false;
-            if (ParserKeyword("INNER").ignore(pos))
+            if (ParserKeyword(Keyword::INNER).ignore(pos))
                 table_join->kind = JoinKind::Inner;
-            else if (ParserKeyword("LEFT").ignore(pos))
+            else if (ParserKeyword(Keyword::LEFT).ignore(pos))
                 table_join->kind = JoinKind::Left;
-            else if (ParserKeyword("RIGHT").ignore(pos))
+            else if (ParserKeyword(Keyword::RIGHT).ignore(pos))
                 table_join->kind = JoinKind::Right;
-            else if (ParserKeyword("FULL").ignore(pos))
+            else if (ParserKeyword(Keyword::FULL).ignore(pos))
                 table_join->kind = JoinKind::Full;
-            else if (ParserKeyword("CROSS").ignore(pos))
+            else if (ParserKeyword(Keyword::CROSS).ignore(pos))
                 table_join->kind = JoinKind::Cross;
-            else if (ParserKeyword("PASTE").ignore(pos))
+            else if (ParserKeyword(Keyword::PASTE).ignore(pos))
                 table_join->kind = JoinKind::Paste;
             else
                 no_kind = true;
@@ -180,7 +180,7 @@ bool ParserTablesInSelectQueryElement::parseImpl(Pos & pos, ASTPtr & node, Expec
                 || table_join->kind == JoinKind::Right
                 || table_join->kind == JoinKind::Full)
             {
-                ParserKeyword("OUTER").ignore(pos);
+                ParserKeyword(Keyword::OUTER).ignore(pos);
             }
 
             if (no_kind)
@@ -201,7 +201,7 @@ bool ParserTablesInSelectQueryElement::parseImpl(Pos & pos, ASTPtr & node, Expec
                 (table_join->kind != JoinKind::Left && table_join->kind != JoinKind::Right))
                 throw Exception(ErrorCodes::SYNTAX_ERROR, "SEMI|ANTI JOIN should be LEFT or RIGHT.");
 
-            if (!ParserKeyword("JOIN").ignore(pos, expected))
+            if (!ParserKeyword(Keyword::JOIN).ignore(pos, expected))
                 return false;
         }
 
@@ -211,7 +211,7 @@ bool ParserTablesInSelectQueryElement::parseImpl(Pos & pos, ASTPtr & node, Expec
         if (table_join->kind != JoinKind::Comma
             && table_join->kind != JoinKind::Cross && table_join->kind != JoinKind::Paste)
         {
-            if (ParserKeyword("USING").ignore(pos, expected))
+            if (ParserKeyword(Keyword::USING).ignore(pos, expected))
             {
                 /// Expression for USING could be in parentheses or not.
                 bool in_parens = pos->type == TokenType::OpeningRoundBracket;
@@ -228,7 +228,7 @@ bool ParserTablesInSelectQueryElement::parseImpl(Pos & pos, ASTPtr & node, Expec
                     ++pos;
                 }
             }
-            else if (ParserKeyword("ON").ignore(pos, expected))
+            else if (ParserKeyword(Keyword::ON).ignore(pos, expected))
             {
                 if (!ParserExpression().parse(pos, table_join->on_expression, expected))
                     return false;
diff --git a/src/Parsers/ParserTransactionControl.cpp b/src/Parsers/ParserTransactionControl.cpp
index da593170002..43d28850246 100644
--- a/src/Parsers/ParserTransactionControl.cpp
+++ b/src/Parsers/ParserTransactionControl.cpp
@@ -12,13 +12,13 @@ bool ParserTransactionControl::parseImpl(Pos & pos, ASTPtr & node, Expected & ex
     ASTTransactionControl::QueryType action;
     UInt64 snapshot = 0;
 
-    if (ParserKeyword("BEGIN TRANSACTION").ignore(pos, expected))
+    if (ParserKeyword(Keyword::BEGIN_TRANSACTION).ignore(pos, expected))
         action = ASTTransactionControl::BEGIN;
-    else if (ParserKeyword("COMMIT").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::COMMIT).ignore(pos, expected))
         action = ASTTransactionControl::COMMIT;
-    else if (ParserKeyword("ROLLBACK").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::ROLLBACK).ignore(pos, expected))
         action = ASTTransactionControl::ROLLBACK;
-    else if (ParserKeyword("SET TRANSACTION SNAPSHOT").ignore(pos, expected))
+    else if (ParserKeyword(Keyword::SET_TRANSACTION_SNAPSHOT).ignore(pos, expected))
     {
         action = ASTTransactionControl::SET_SNAPSHOT;
         ASTPtr ast;
diff --git a/src/Parsers/ParserUndropQuery.cpp b/src/Parsers/ParserUndropQuery.cpp
index e67fd7f5f01..07ca8a3b5fd 100644
--- a/src/Parsers/ParserUndropQuery.cpp
+++ b/src/Parsers/ParserUndropQuery.cpp
@@ -13,7 +13,7 @@ namespace
 
 bool parseUndropQuery(IParser::Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_table("TABLE");
+    ParserKeyword s_table(Keyword::TABLE);
     ParserToken s_dot(TokenType::Dot);
     ParserIdentifier name_p(true);
 
@@ -35,7 +35,7 @@ bool parseUndropQuery(IParser::Pos & pos, ASTPtr & node, Expected & expected)
         if (!name_p.parse(pos, table, expected))
             return false;
     }
-    if (ParserKeyword("UUID").ignore(pos, expected))
+    if (ParserKeyword(Keyword::UUID).ignore(pos, expected))
     {
         ParserStringLiteral uuid_p;
         ASTPtr ast_uuid;
@@ -43,7 +43,7 @@ bool parseUndropQuery(IParser::Pos & pos, ASTPtr & node, Expected & expected)
             return false;
         uuid = parseFromString<UUID>(ast_uuid->as<ASTLiteral>()->value.get<String>());
     }
-    if (ParserKeyword{"ON"}.ignore(pos, expected))
+    if (ParserKeyword{Keyword::ON}.ignore(pos, expected))
     {
         if (!ASTQueryWithOnCluster::parse(pos, cluster_str, expected))
             return false;
@@ -70,7 +70,7 @@ bool parseUndropQuery(IParser::Pos & pos, ASTPtr & node, Expected & expected)
 
 bool ParserUndropQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_undrop("UNDROP");
+    ParserKeyword s_undrop(Keyword::UNDROP);
 
     if (s_undrop.ignore(pos, expected))
         return parseUndropQuery(pos, node, expected);
diff --git a/src/Parsers/ParserUseQuery.cpp b/src/Parsers/ParserUseQuery.cpp
index 350e573281c..b510ab9e59b 100644
--- a/src/Parsers/ParserUseQuery.cpp
+++ b/src/Parsers/ParserUseQuery.cpp
@@ -10,7 +10,7 @@ namespace DB
 
 bool ParserUseQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_use("USE");
+    ParserKeyword s_use(Keyword::USE);
     ParserIdentifier name_p{/*allow_query_parameter*/ true};
 
     if (!s_use.ignore(pos, expected))
diff --git a/src/Parsers/ParserWatchQuery.cpp b/src/Parsers/ParserWatchQuery.cpp
index 913d9c4ff3f..ad3aeb6385d 100644
--- a/src/Parsers/ParserWatchQuery.cpp
+++ b/src/Parsers/ParserWatchQuery.cpp
@@ -20,11 +20,11 @@ namespace DB
 
 bool ParserWatchQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
-    ParserKeyword s_watch("WATCH");
+    ParserKeyword s_watch(Keyword::WATCH);
     ParserToken s_dot(TokenType::Dot);
     ParserIdentifier name_p(true);
-    ParserKeyword s_events("EVENTS");
-    ParserKeyword s_limit("LIMIT");
+    ParserKeyword s_events(Keyword::EVENTS);
+    ParserKeyword s_limit(Keyword::LIMIT);
 
     ASTPtr database;
     ASTPtr table;
diff --git a/src/Parsers/ParserWithElement.cpp b/src/Parsers/ParserWithElement.cpp
index a882af19dd9..5b31c929685 100644
--- a/src/Parsers/ParserWithElement.cpp
+++ b/src/Parsers/ParserWithElement.cpp
@@ -13,7 +13,7 @@ namespace DB
 bool ParserWithElement::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
 {
     ParserIdentifier s_ident;
-    ParserKeyword s_as("AS");
+    ParserKeyword s_as(Keyword::AS);
     ParserSubquery s_subquery;
 
     auto old_pos = pos;
diff --git a/src/Parsers/parseIntervalKind.cpp b/src/Parsers/parseIntervalKind.cpp
index 124246f2950..bbcff8826be 100644
--- a/src/Parsers/parseIntervalKind.cpp
+++ b/src/Parsers/parseIntervalKind.cpp
@@ -7,89 +7,88 @@ namespace DB
 {
 bool parseIntervalKind(IParser::Pos & pos, Expected & expected, IntervalKind & result)
 {
-    if (ParserKeyword("NANOSECOND").ignore(pos, expected) || ParserKeyword("NANOSECONDS").ignore(pos, expected)
-        || ParserKeyword("SQL_TSI_NANOSECOND").ignore(pos, expected)
-        || ParserKeyword("NS").ignore(pos, expected))
+    if (ParserKeyword(Keyword::NANOSECOND).ignore(pos, expected) || ParserKeyword(Keyword::NANOSECONDS).ignore(pos, expected)
+        || ParserKeyword(Keyword::SQL_TSI_NANOSECOND).ignore(pos, expected)
+        || ParserKeyword(Keyword::NS).ignore(pos, expected))
     {
         result = IntervalKind::Kind::Nanosecond;
         return true;
     }
 
-    if (ParserKeyword("MICROSECOND").ignore(pos, expected) || ParserKeyword("MICROSECONDS").ignore(pos, expected)
-        || ParserKeyword("SQL_TSI_MICROSECOND").ignore(pos, expected)
-        || ParserKeyword("MCS").ignore(pos, expected))
+    if (ParserKeyword(Keyword::MICROSECOND).ignore(pos, expected) || ParserKeyword(Keyword::MICROSECONDS).ignore(pos, expected)
+        || ParserKeyword(Keyword::SQL_TSI_MICROSECOND).ignore(pos, expected))
     {
         result = IntervalKind::Kind::Microsecond;
         return true;
     }
 
-    if (ParserKeyword("MILLISECOND").ignore(pos, expected) || ParserKeyword("MILLISECONDS").ignore(pos, expected)
-        || ParserKeyword("SQL_TSI_MILLISECOND").ignore(pos, expected)
-        || ParserKeyword("MS").ignore(pos, expected))
+    if (ParserKeyword(Keyword::MILLISECOND).ignore(pos, expected) || ParserKeyword(Keyword::MILLISECONDS).ignore(pos, expected)
+        || ParserKeyword(Keyword::SQL_TSI_MILLISECOND).ignore(pos, expected)
+        || ParserKeyword(Keyword::MS).ignore(pos, expected))
     {
         result = IntervalKind::Kind::Millisecond;
         return true;
     }
 
-    if (ParserKeyword("SECOND").ignore(pos, expected) || ParserKeyword("SECONDS").ignore(pos, expected)
-        || ParserKeyword("SQL_TSI_SECOND").ignore(pos, expected)
-        || ParserKeyword("SS").ignore(pos, expected) || ParserKeyword("S").ignore(pos, expected))
+    if (ParserKeyword(Keyword::SECOND).ignore(pos, expected) || ParserKeyword(Keyword::SECONDS).ignore(pos, expected)
+        || ParserKeyword(Keyword::SQL_TSI_SECOND).ignore(pos, expected)
+        || ParserKeyword(Keyword::SS).ignore(pos, expected) || ParserKeyword(Keyword::S).ignore(pos, expected))
     {
         result = IntervalKind::Kind::Second;
         return true;
     }
 
-    if (ParserKeyword("MINUTE").ignore(pos, expected) || ParserKeyword("MINUTES").ignore(pos, expected)
-        || ParserKeyword("SQL_TSI_MINUTE").ignore(pos, expected)
-        || ParserKeyword("MI").ignore(pos, expected) || ParserKeyword("N").ignore(pos, expected))
+    if (ParserKeyword(Keyword::MINUTE).ignore(pos, expected) || ParserKeyword(Keyword::MINUTES).ignore(pos, expected)
+        || ParserKeyword(Keyword::SQL_TSI_MINUTE).ignore(pos, expected)
+        || ParserKeyword(Keyword::MI).ignore(pos, expected) || ParserKeyword(Keyword::N).ignore(pos, expected))
     {
         result = IntervalKind::Kind::Minute;
         return true;
     }
 
-    if (ParserKeyword("HOUR").ignore(pos, expected) || ParserKeyword("HOURS").ignore(pos, expected)
-        || ParserKeyword("SQL_TSI_HOUR").ignore(pos, expected)
-        || ParserKeyword("HH").ignore(pos, expected) || ParserKeyword("H").ignore(pos, expected))
+    if (ParserKeyword(Keyword::HOUR).ignore(pos, expected) || ParserKeyword(Keyword::HOURS).ignore(pos, expected)
+        || ParserKeyword(Keyword::SQL_TSI_HOUR).ignore(pos, expected)
+        || ParserKeyword(Keyword::HH).ignore(pos, expected) || ParserKeyword(Keyword::H).ignore(pos, expected))
     {
         result = IntervalKind::Kind::Hour;
         return true;
     }
 
-    if (ParserKeyword("DAY").ignore(pos, expected) || ParserKeyword("DAYS").ignore(pos, expected)
-        || ParserKeyword("SQL_TSI_DAY").ignore(pos, expected)
-        || ParserKeyword("DD").ignore(pos, expected) || ParserKeyword("D").ignore(pos, expected))
+    if (ParserKeyword(Keyword::DAY).ignore(pos, expected) || ParserKeyword(Keyword::DAYS).ignore(pos, expected)
+        || ParserKeyword(Keyword::SQL_TSI_DAY).ignore(pos, expected)
+        || ParserKeyword(Keyword::DD).ignore(pos, expected) || ParserKeyword(Keyword::D).ignore(pos, expected))
     {
         result = IntervalKind::Kind::Day;
         return true;
     }
 
-    if (ParserKeyword("WEEK").ignore(pos, expected) || ParserKeyword("WEEKS").ignore(pos, expected)
-        || ParserKeyword("SQL_TSI_WEEK").ignore(pos, expected)
-        || ParserKeyword("WK").ignore(pos, expected) || ParserKeyword("WW").ignore(pos, expected))
+    if (ParserKeyword(Keyword::WEEK).ignore(pos, expected) || ParserKeyword(Keyword::WEEKS).ignore(pos, expected)
+        || ParserKeyword(Keyword::SQL_TSI_WEEK).ignore(pos, expected)
+        || ParserKeyword(Keyword::WK).ignore(pos, expected) || ParserKeyword(Keyword::WW).ignore(pos, expected))
     {
         result = IntervalKind::Kind::Week;
         return true;
     }
 
-    if (ParserKeyword("MONTH").ignore(pos, expected) || ParserKeyword("MONTHS").ignore(pos, expected)
-        || ParserKeyword("SQL_TSI_MONTH").ignore(pos, expected)
-        || ParserKeyword("MM").ignore(pos, expected) || ParserKeyword("M").ignore(pos, expected))
+    if (ParserKeyword(Keyword::MONTH).ignore(pos, expected) || ParserKeyword(Keyword::MONTHS).ignore(pos, expected)
+        || ParserKeyword(Keyword::SQL_TSI_MONTH).ignore(pos, expected)
+        || ParserKeyword(Keyword::MM).ignore(pos, expected) || ParserKeyword(Keyword::M).ignore(pos, expected))
     {
         result = IntervalKind::Kind::Month;
         return true;
     }
 
-    if (ParserKeyword("QUARTER").ignore(pos, expected) || ParserKeyword("QUARTERS").ignore(pos, expected)
-        || ParserKeyword("SQL_TSI_QUARTER").ignore(pos, expected)
-        || ParserKeyword("QQ").ignore(pos, expected) || ParserKeyword("Q").ignore(pos, expected))
+    if (ParserKeyword(Keyword::QUARTER).ignore(pos, expected) || ParserKeyword(Keyword::QUARTERS).ignore(pos, expected)
+        || ParserKeyword(Keyword::SQL_TSI_QUARTER).ignore(pos, expected)
+        || ParserKeyword(Keyword::QQ).ignore(pos, expected) || ParserKeyword(Keyword::Q).ignore(pos, expected))
     {
         result = IntervalKind::Kind::Quarter;
         return true;
     }
 
-    if (ParserKeyword("YEAR").ignore(pos, expected) || ParserKeyword("YEARS").ignore(pos, expected)
-        || ParserKeyword("SQL_TSI_YEAR").ignore(pos, expected)
-        || ParserKeyword("YYYY").ignore(pos, expected) || ParserKeyword("YY").ignore(pos, expected))
+    if (ParserKeyword(Keyword::YEAR).ignore(pos, expected) || ParserKeyword(Keyword::YEARS).ignore(pos, expected)
+        || ParserKeyword(Keyword::SQL_TSI_YEAR).ignore(pos, expected)
+        || ParserKeyword(Keyword::YYYY).ignore(pos, expected) || ParserKeyword(Keyword::YY).ignore(pos, expected))
     {
         result = IntervalKind::Kind::Year;
         return true;
diff --git a/src/Storages/System/StorageSystemKeywords.cpp b/src/Storages/System/StorageSystemKeywords.cpp
new file mode 100644
index 00000000000..b9d256bf4b2
--- /dev/null
+++ b/src/Storages/System/StorageSystemKeywords.cpp
@@ -0,0 +1,29 @@
+#include <Common/Macros.h>
+#include <Interpreters/Context.h>
+#include <Storages/System/StorageSystemKeywords.h>
+
+#include <Parsers/CommonParsers.h>
+
+
+namespace DB
+{
+
+ColumnsDescription StorageSystemKeywords::getColumnsDescription()
+{
+    return ColumnsDescription
+    {
+        {"keyword", std::make_shared<DataTypeString>(), "The keyword used in ClickHouse parser."},
+    };
+}
+
+void StorageSystemKeywords::fillData(MutableColumns & res_columns, ContextPtr context, const ActionsDAG::Node *, std::vector<UInt8>) const
+{
+    auto macros = context->getMacros();
+
+    for (const auto & keyword : getAllKeyWords())
+    {
+        res_columns[0]->insert(keyword);
+    }
+}
+
+}
diff --git a/src/Storages/System/StorageSystemKeywords.h b/src/Storages/System/StorageSystemKeywords.h
new file mode 100644
index 00000000000..5574c4a3abc
--- /dev/null
+++ b/src/Storages/System/StorageSystemKeywords.h
@@ -0,0 +1,28 @@
+#pragma once
+
+#include <DataTypes/DataTypeString.h>
+#include <Storages/System/IStorageSystemOneBlock.h>
+
+
+namespace DB
+{
+
+class Context;
+
+
+/** Information about macros for introspection.
+  */
+class StorageSystemKeywords final : public IStorageSystemOneBlock
+{
+public:
+    std::string getName() const override { return "SystemKeywords"; }
+
+    static ColumnsDescription getColumnsDescription();
+
+protected:
+    using IStorageSystemOneBlock::IStorageSystemOneBlock;
+
+    void fillData(MutableColumns & res_columns, ContextPtr, const ActionsDAG::Node *, std::vector<UInt8>) const override;
+};
+
+}
diff --git a/src/Storages/System/attachSystemTables.cpp b/src/Storages/System/attachSystemTables.cpp
index 79475b9695d..0de92c8f2f0 100644
--- a/src/Storages/System/attachSystemTables.cpp
+++ b/src/Storages/System/attachSystemTables.cpp
@@ -1,3 +1,4 @@
+#include "Storages/System/StorageSystemKeywords.h"
 #include "config.h"
 
 #include <Databases/IDatabase.h>
@@ -184,6 +185,7 @@ void attachSystemTablesServer(ContextPtr context, IDatabase & system_database, b
     attachNoDescription<StorageSystemMySQLBinlogs>(context, system_database, "mysql_binlogs", "Shows a list of active binlogs for MaterializedMySQL.");
 #endif
 
+    attach<StorageSystemKeywords>(context, system_database, "keywords", "Contains a list of all keywords used in ClickHouse parser.");
     attachNoDescription<StorageSystemParts>(context, system_database, "parts", "Contains a list of currently existing (both active and inactive) parts of all *-MergeTree tables. Each part is represented by a single row.");
     attachNoDescription<StorageSystemProjectionParts>(context, system_database, "projection_parts", "Contains a list of currently existing projection parts (a copy of some part containing aggregated data or just sorted in different order) created for all the projections for all tables within a cluster.");
     attachNoDescription<StorageSystemDetachedParts>(context, system_database, "detached_parts", "Contains a list of all parts which are being found in /detached directory along with a reason why it was detached. ClickHouse server doesn't use such parts anyhow.");

From c67c04b89a319cc3d0bcbcc7c3eea169b49d96f5 Mon Sep 17 00:00:00 2001
From: Arthur Passos <arthur.ti@outlook.com>
Date: Mon, 18 Mar 2024 13:44:36 -0300
Subject: [PATCH 441/443] fix typo

---
 src/Storages/MergeTree/SimpleMergeSelector.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Storages/MergeTree/SimpleMergeSelector.h b/src/Storages/MergeTree/SimpleMergeSelector.h
index c20eaa6e8de..a12b8ba48cb 100644
--- a/src/Storages/MergeTree/SimpleMergeSelector.h
+++ b/src/Storages/MergeTree/SimpleMergeSelector.h
@@ -26,7 +26,7 @@ Then we need some balance between optimization of these two metrics.
 But some optimizations may improve both metrics.
 
 For example, we can look at the "merge tree" - the tree of data parts that were merged.
-If the tree is perfectly balanced then its depth is proportonal to the log(data size),
+If the tree is perfectly balanced then its depth is proportional to the log(data size),
 the total amount of work is proportional to data_size * log(data_size)
 and the write amplification is proportional to log(data_size).
 If it's not balanced (e.g. every new data part is always merged with existing data parts),

From ebaee8e5f62551b7c11d76ccdd7abac0bf4c23a0 Mon Sep 17 00:00:00 2001
From: Kirill Nikiforov <allmazz@allmazz.me>
Date: Mon, 18 Mar 2024 22:00:27 +0300
Subject: [PATCH 442/443] Add support for `ATTACH PARTITION ALL` (#61107)

---
 .../statements/alter/partition.md             |  1 +
 src/Storages/MergeTree/MergeTreeData.cpp      | 18 +++++---
 .../0_stateless/00753_alter_attach.reference  | 16 +++++++
 .../0_stateless/00753_alter_attach.sql        | 43 +++++++++++++++++++
 4 files changed, 73 insertions(+), 5 deletions(-)

diff --git a/docs/en/sql-reference/statements/alter/partition.md b/docs/en/sql-reference/statements/alter/partition.md
index a21ef4f4af5..ce5cecf6fd6 100644
--- a/docs/en/sql-reference/statements/alter/partition.md
+++ b/docs/en/sql-reference/statements/alter/partition.md
@@ -350,6 +350,7 @@ ALTER TABLE mt DELETE IN PARTITION ID '2' WHERE p = 2;
 You can specify the partition expression in `ALTER ... PARTITION` queries in different ways:
 
 - As a value from the `partition` column of the `system.parts` table. For example, `ALTER TABLE visits DETACH PARTITION 201901`.
+- Using the keyword `ALL`. It can be used only with DROP/DETACH/ATTACH. For example, `ALTER TABLE visits ATTACH PARTITION ALL`.
 - As a tuple of expressions or constants that matches (in types) the table partitioning keys tuple. In the case of a single element partitioning key, the expression should be wrapped in the `tuple (...)` function. For example, `ALTER TABLE visits DETACH PARTITION tuple(toYYYYMM(toDate('2019-01-25')))`.
 - Using the partition ID. Partition ID is a string identifier of the partition (human-readable, if possible) that is used as the names of partitions in the file system and in ZooKeeper. The partition ID must be specified in the `PARTITION ID` clause, in a single quotes. For example, `ALTER TABLE visits DETACH PARTITION ID '201901'`.
 - In the [ALTER ATTACH PART](#alter_attach-partition) and [DROP DETACHED PART](#alter_drop-detached) query, to specify the name of a part, use string literal with a value from the `name` column of the [system.detached_parts](/docs/en/operations/system-tables/detached_parts.md/#system_tables-detached_parts) table. For example, `ALTER TABLE visits ATTACH PART '201901_1_1_0'`.
diff --git a/src/Storages/MergeTree/MergeTreeData.cpp b/src/Storages/MergeTree/MergeTreeData.cpp
index a22a1904345..a5e309ae793 100644
--- a/src/Storages/MergeTree/MergeTreeData.cpp
+++ b/src/Storages/MergeTree/MergeTreeData.cpp
@@ -4875,8 +4875,8 @@ void MergeTreeData::checkAlterPartitionIsPossible(
                 const auto * partition_ast = command.partition->as<ASTPartition>();
                 if (partition_ast && partition_ast->all)
                 {
-                    if (command.type != PartitionCommand::DROP_PARTITION)
-                        throw DB::Exception(ErrorCodes::SUPPORT_IS_DISABLED, "Only support DROP/DETACH PARTITION ALL currently");
+                    if (command.type != PartitionCommand::DROP_PARTITION && command.type != PartitionCommand::ATTACH_PARTITION)
+                        throw DB::Exception(ErrorCodes::SUPPORT_IS_DISABLED, "Only support DROP/DETACH/ATTACH PARTITION ALL currently");
                 }
                 else
                 {
@@ -6109,15 +6109,23 @@ MergeTreeData::MutableDataPartsVector MergeTreeData::tryLoadPartsToAttach(const
     }
     else
     {
-        String partition_id = getPartitionIDFromQuery(partition, local_context);
-        LOG_DEBUG(log, "Looking for parts for partition {} in {}", partition_id, source_dir);
+        String partition_id;
+        if (partition->as<ASTPartition>()->all)
+        {
+            LOG_DEBUG(log, "Looking for parts for all partitions in {}", source_dir);
+        }
+        else
+        {
+            partition_id = getPartitionIDFromQuery(partition, local_context);
+            LOG_DEBUG(log, "Looking for parts for partition {} in {}", partition_id, source_dir);
+        }
 
         ActiveDataPartSet active_parts(format_version);
 
         auto detached_parts = getDetachedParts();
         std::erase_if(detached_parts, [&partition_id](const DetachedPartInfo & part_info)
         {
-            return !part_info.valid_name || !part_info.prefix.empty() || part_info.partition_id != partition_id;
+            return !part_info.valid_name || !part_info.prefix.empty() || (!partition_id.empty() && part_info.partition_id != partition_id);
         });
 
         for (const auto & part_info : detached_parts)
diff --git a/tests/queries/0_stateless/00753_alter_attach.reference b/tests/queries/0_stateless/00753_alter_attach.reference
index b0d2a3d031c..301d5ebf451 100644
--- a/tests/queries/0_stateless/00753_alter_attach.reference
+++ b/tests/queries/0_stateless/00753_alter_attach.reference
@@ -22,3 +22,19 @@
 2	2
 1	1
 1	1
+1	1
+2	2
+3	3
+2	2
+3	3
+1	1
+2	2
+3	3
+1	1
+2	2
+3	3
+2	2
+3	3
+1	1
+2	2
+3	3
diff --git a/tests/queries/0_stateless/00753_alter_attach.sql b/tests/queries/0_stateless/00753_alter_attach.sql
index 299d0dd995f..7f2b1214613 100644
--- a/tests/queries/0_stateless/00753_alter_attach.sql
+++ b/tests/queries/0_stateless/00753_alter_attach.sql
@@ -90,3 +90,46 @@ OPTIMIZE TABLE partition_all2 PARTITION ALL; -- { serverError 344 }
 
 DROP TABLE partition_all;
 DROP TABLE partition_all2;
+
+-- test ATTACH ALL
+CREATE TABLE partition_attach_all (x UInt64, p UInt8) ENGINE = MergeTree ORDER BY x PARTITION BY p;
+INSERT INTO partition_attach_all VALUES (1, 1), (2, 2), (3, 3);
+
+ALTER TABLE partition_attach_all DETACH PARTITION ALL;
+SELECT * FROM partition_attach_all ORDER BY x;
+ALTER TABLE partition_attach_all ATTACH PARTITION ALL;
+SELECT * FROM partition_attach_all ORDER BY x;
+
+ALTER TABLE partition_attach_all DETACH PARTITION 1;
+SELECT * FROM partition_attach_all ORDER BY x;
+ALTER TABLE partition_attach_all ATTACH PARTITION ALL;
+SELECT * FROM partition_attach_all ORDER BY x;
+
+ALTER TABLE partition_attach_all DROP PARTITION ALL;
+SELECT * FROM partition_attach_all ORDER BY x;
+ALTER TABLE partition_attach_all ATTACH PARTITION ALL;
+SELECT * FROM partition_attach_all ORDER BY x;
+
+CREATE TABLE replicated_partition_attach_all (x UInt64, p UInt8)
+    ENGINE = ReplicatedMergeTree('/clickhouse/tables/test_00753_{database}/replicated_partition_attach_all', '1')
+        ORDER BY x
+        PARTITION BY p;
+INSERT INTO replicated_partition_attach_all VALUES (1, 1), (2, 2), (3, 3);
+
+ALTER TABLE replicated_partition_attach_all DETACH PARTITION ALL;
+SELECT * FROM replicated_partition_attach_all ORDER BY x;
+ALTER TABLE replicated_partition_attach_all ATTACH PARTITION ALL;
+SELECT * FROM replicated_partition_attach_all ORDER BY x;
+
+ALTER TABLE replicated_partition_attach_all DETACH PARTITION 1;
+SELECT * FROM replicated_partition_attach_all ORDER BY x;
+ALTER TABLE replicated_partition_attach_all ATTACH PARTITION ALL;
+SELECT * FROM replicated_partition_attach_all ORDER BY x;
+
+ALTER TABLE replicated_partition_attach_all DROP PARTITION ALL;
+SELECT * FROM replicated_partition_attach_all ORDER BY x;
+ALTER TABLE replicated_partition_attach_all ATTACH PARTITION ALL;
+SELECT * FROM replicated_partition_attach_all ORDER BY x;
+
+DROP TABLE partition_attach_all;
+DROP TABLE replicated_partition_attach_all;

From 94a347cc419c2af101933d211d028771efd9e966 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Mon, 18 Mar 2024 20:34:15 +0000
Subject: [PATCH 443/443] Suppress bugprone-forward-declaration-namespace

---
 .clang-tidy | 1 +
 1 file changed, 1 insertion(+)

diff --git a/.clang-tidy b/.clang-tidy
index e7900cbe2aa..13c1b116ead 100644
--- a/.clang-tidy
+++ b/.clang-tidy
@@ -20,6 +20,7 @@ Checks: [
   '-bugprone-branch-clone',
   '-bugprone-easily-swappable-parameters',
   '-bugprone-exception-escape',
+  '-bugprone-forward-declaration-namespace',
   '-bugprone-implicit-widening-of-multiplication-result',
   '-bugprone-narrowing-conversions',
   '-bugprone-not-null-terminated-result',