From 1910434174234ccdf5e30291453eb5ebda687b9b Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 30 Aug 2023 20:48:26 +0200
Subject: [PATCH 01/37] Remove CurrentStatusInfo

---
 src/Common/StatusInfo.cpp              | 57 --------------------------
 src/Common/StatusInfo.h                | 39 ------------------
 src/Interpreters/ExternalLoader.cpp    | 10 -----
 src/Server/PrometheusMetricsWriter.cpp | 38 -----------------
 src/Server/PrometheusMetricsWriter.h   |  2 -
 5 files changed, 146 deletions(-)
 delete mode 100644 src/Common/StatusInfo.cpp
 delete mode 100644 src/Common/StatusInfo.h

diff --git a/src/Common/StatusInfo.cpp b/src/Common/StatusInfo.cpp
deleted file mode 100644
index 1f9ddfaf4b9..00000000000
--- a/src/Common/StatusInfo.cpp
+++ /dev/null
@@ -1,57 +0,0 @@
-#include <Common/StatusInfo.h>
-#include <Common/ExternalLoaderStatus.h>
-
-/// Available status. Add something here as you wish.
-#define APPLY_FOR_STATUS(M) \
-    M(DictionaryStatus, "Dictionary Status.", DB::getStatusEnumAllPossibleValues()) \
-
-
-namespace CurrentStatusInfo
-{
-    #define M(NAME, DOCUMENTATION, ENUM) extern const Status NAME = Status(__COUNTER__);
-        APPLY_FOR_STATUS(M)
-    #undef M
-    constexpr Status END = Status(__COUNTER__);
-
-    std::mutex locks[END] {};
-    std::unordered_map<String, Int8> values[END] {};
-
-    const char * getName(Status event)
-    {
-        static const char * strings[] =
-        {
-        #define M(NAME, DOCUMENTATION, ENUM) #NAME,
-            APPLY_FOR_STATUS(M)
-        #undef M
-        };
-
-        return strings[event];
-    }
-
-    const char * getDocumentation(Status event)
-    {
-        static const char * strings[] =
-        {
-        #define M(NAME, DOCUMENTATION, ENUM) #DOCUMENTATION,
-            APPLY_FOR_STATUS(M)
-        #undef M
-        };
-
-        return strings[event];
-    }
-
-    const std::vector<std::pair<String, Int8>> & getAllPossibleValues(Status event)
-    {
-        static const std::vector<std::pair<String, Int8>> enum_values [] =
-        {
-        #define M(NAME, DOCUMENTATION, ENUM) ENUM,
-            APPLY_FOR_STATUS(M)
-        #undef M
-        };
-        return enum_values[event];
-    }
-
-    Status end() { return END; }
-}
-
-#undef APPLY_FOR_STATUS
diff --git a/src/Common/StatusInfo.h b/src/Common/StatusInfo.h
deleted file mode 100644
index 91e6d4d3b85..00000000000
--- a/src/Common/StatusInfo.h
+++ /dev/null
@@ -1,39 +0,0 @@
-#pragma once
-
-#include <stddef.h>
-#include <cstdint>
-#include <utility>
-#include <atomic>
-#include <vector>
-#include <base/types.h>
-#include <base/strong_typedef.h>
-#include <mutex>
-#include <unordered_map>
-
-
-namespace CurrentStatusInfo
-{
-    using Status = StrongTypedef<size_t, struct StatusTag>;
-    using Key = std::string;
-
-    const char * getName(Status event);
-    const char * getDocumentation(Status event);
-    const std::vector<std::pair<String, Int8>> & getAllPossibleValues(Status event);
-
-    extern std::unordered_map<String, Int8> values[];
-    extern std::mutex locks[];
-
-    Status end();
-
-    inline void set(Status status, Key key, Int8 value)
-    {
-        std::lock_guard lock(locks[status]);
-        values[status][key] = value;
-    }
-
-    inline void unset(Status status, Key key)
-    {
-        std::lock_guard lock(locks[status]);
-        values[status].erase(key);
-    }
-}
diff --git a/src/Interpreters/ExternalLoader.cpp b/src/Interpreters/ExternalLoader.cpp
index 5dee750889c..56d480d8735 100644
--- a/src/Interpreters/ExternalLoader.cpp
+++ b/src/Interpreters/ExternalLoader.cpp
@@ -1,7 +1,6 @@
 #include "ExternalLoader.h"
 
 #include <mutex>
-#include <pcg_random.hpp>
 #include <Common/MemoryTrackerBlockerInThread.h>
 #include <Common/Config/AbstractConfigurationComparison.h>
 #include <Common/Exception.h>
@@ -9,7 +8,6 @@
 #include <Common/ThreadPool.h>
 #include <Common/randomSeed.h>
 #include <Common/setThreadName.h>
-#include <Common/StatusInfo.h>
 #include <Common/scope_guard_safe.h>
 #include <Common/logger_useful.h>
 #include <base/chrono_io.h>
@@ -18,12 +16,6 @@
 #include <unordered_set>
 
 
-namespace CurrentStatusInfo
-{
-    extern const Status DictionaryStatus;
-}
-
-
 namespace DB
 {
 namespace ErrorCodes
@@ -1145,7 +1137,6 @@ private:
         if (info && (info->loading_id == loading_id))
         {
             info->loading_id = info->state_id;
-            CurrentStatusInfo::set(CurrentStatusInfo::DictionaryStatus, name, static_cast<Int8>(info->status()));
         }
         min_id_to_finish_loading_dependencies.erase(std::this_thread::get_id());
 
@@ -1307,7 +1298,6 @@ scope_guard ExternalLoader::addConfigRepository(std::unique_ptr<IExternalLoaderC
     return [this, ptr, name]()
     {
         config_files_reader->removeConfigRepository(ptr);
-        CurrentStatusInfo::unset(CurrentStatusInfo::DictionaryStatus, name);
         reloadConfig(name);
     };
 }
diff --git a/src/Server/PrometheusMetricsWriter.cpp b/src/Server/PrometheusMetricsWriter.cpp
index 2331e455225..a7d90b9985c 100644
--- a/src/Server/PrometheusMetricsWriter.cpp
+++ b/src/Server/PrometheusMetricsWriter.cpp
@@ -1,7 +1,6 @@
 #include "PrometheusMetricsWriter.h"
 
 #include <IO/WriteHelpers.h>
-#include <Common/StatusInfo.h>
 #include <regex>    /// TODO: this library is harmful.
 #include <algorithm>
 
@@ -51,7 +50,6 @@ PrometheusMetricsWriter::PrometheusMetricsWriter(
     , send_events(config.getBool(config_name + ".events", true))
     , send_metrics(config.getBool(config_name + ".metrics", true))
     , send_asynchronous_metrics(config.getBool(config_name + ".asynchronous_metrics", true))
-    , send_status_info(config.getBool(config_name + ".status_info", true))
 {
 }
 
@@ -120,42 +118,6 @@ void PrometheusMetricsWriter::write(WriteBuffer & wb) const
             writeOutLine(wb, key, value.value);
         }
     }
-
-    if (send_status_info)
-    {
-        for (size_t i = 0, end = CurrentStatusInfo::end(); i < end; ++i)
-        {
-            std::lock_guard lock(CurrentStatusInfo::locks[static_cast<CurrentStatusInfo::Status>(i)]);
-            std::string metric_name{CurrentStatusInfo::getName(static_cast<CurrentStatusInfo::Status>(i))};
-            std::string metric_doc{CurrentStatusInfo::getDocumentation(static_cast<CurrentStatusInfo::Status>(i))};
-
-            convertHelpToSingleLine(metric_doc);
-
-            if (!replaceInvalidChars(metric_name))
-                continue;
-            std::string key{current_status_prefix + metric_name};
-
-            writeOutLine(wb, "# HELP", key, metric_doc);
-            writeOutLine(wb, "# TYPE", key, "gauge");
-
-            for (const auto & value: CurrentStatusInfo::values[i])
-            {
-                for (const auto & enum_value: CurrentStatusInfo::getAllPossibleValues(static_cast<CurrentStatusInfo::Status>(i)))
-                {
-                    DB::writeText(key, wb);
-                    DB::writeChar('{', wb);
-                    DB::writeText(key, wb);
-                    DB::writeChar('=', wb);
-                    writeDoubleQuotedString(enum_value.first, wb);
-                    DB::writeText(",name=", wb);
-                    writeDoubleQuotedString(value.first, wb);
-                    DB::writeText("} ", wb);
-                    DB::writeText(value.second == enum_value.second, wb);
-                    DB::writeChar('\n', wb);
-                }
-            }
-        }
-    }
 }
 
 }
diff --git a/src/Server/PrometheusMetricsWriter.h b/src/Server/PrometheusMetricsWriter.h
index b4f6ab57def..b05eeaf0a3a 100644
--- a/src/Server/PrometheusMetricsWriter.h
+++ b/src/Server/PrometheusMetricsWriter.h
@@ -27,12 +27,10 @@ private:
     const bool send_events;
     const bool send_metrics;
     const bool send_asynchronous_metrics;
-    const bool send_status_info;
 
     static inline constexpr auto profile_events_prefix = "ClickHouseProfileEvents_";
     static inline constexpr auto current_metrics_prefix = "ClickHouseMetrics_";
     static inline constexpr auto asynchronous_metrics_prefix = "ClickHouseAsyncMetrics_";
-    static inline constexpr auto current_status_prefix = "ClickHouseStatusInfo_";
 };
 
 }

From 9d6dc1067f172187a739e0939e23acad0e631dc1 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 30 Aug 2023 20:49:02 +0200
Subject: [PATCH 02/37] Remove CurrentStatusInfo

---
 ...HouseStatusInfo_DictionaryStatus.reference | 18 ---------
 ...s_ClickHouseStatusInfo_DictionaryStatus.sh | 38 -------------------
 2 files changed, 56 deletions(-)
 delete mode 100644 tests/queries/0_stateless/02390_prometheus_ClickHouseStatusInfo_DictionaryStatus.reference
 delete mode 100755 tests/queries/0_stateless/02390_prometheus_ClickHouseStatusInfo_DictionaryStatus.sh

diff --git a/tests/queries/0_stateless/02390_prometheus_ClickHouseStatusInfo_DictionaryStatus.reference b/tests/queries/0_stateless/02390_prometheus_ClickHouseStatusInfo_DictionaryStatus.reference
deleted file mode 100644
index 50c91c3fa0c..00000000000
--- a/tests/queries/0_stateless/02390_prometheus_ClickHouseStatusInfo_DictionaryStatus.reference
+++ /dev/null
@@ -1,18 +0,0 @@
-status before reload
-status after reload
-NOT_LOADED 0
-LOADED 0
-FAILED 1
-LOADING 0
-FAILED_AND_RELOADING 0
-LOADED_AND_RELOADING 0
-NOT_EXIST 0
-status after reload, table exists
-NOT_LOADED 0
-LOADED 1
-FAILED 0
-LOADING 0
-FAILED_AND_RELOADING 0
-LOADED_AND_RELOADING 0
-NOT_EXIST 0
-status after drop
diff --git a/tests/queries/0_stateless/02390_prometheus_ClickHouseStatusInfo_DictionaryStatus.sh b/tests/queries/0_stateless/02390_prometheus_ClickHouseStatusInfo_DictionaryStatus.sh
deleted file mode 100755
index 65025858e20..00000000000
--- a/tests/queries/0_stateless/02390_prometheus_ClickHouseStatusInfo_DictionaryStatus.sh
+++ /dev/null
@@ -1,38 +0,0 @@
-#!/usr/bin/env bash
-# Tags: no-ordinary-database
-
-CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
-# shellcheck source=../shell_config.sh
-. "$CUR_DIR"/../shell_config.sh
-
-function get_dictionary_status()
-{
-    local name=$1 && shift
-    $CLICKHOUSE_CURL -sS "$CLICKHOUSE_URL_PROMETHEUS" | {
-        awk -F'[{}=," ]' -vname="$name" '/ClickHouseStatusInfo_DictionaryStatus{/ && $(NF-3) == name { print $4, $NF }'
-    }
-}
-
-$CLICKHOUSE_CLIENT -q "CREATE DICTIONARY dict (key Int, value String) PRIMARY KEY key SOURCE(CLICKHOUSE(TABLE data)) LAYOUT(HASHED()) LIFETIME(0)"
-uuid="$($CLICKHOUSE_CLIENT -q "SELECT uuid FROM system.dictionaries WHERE database = '$CLICKHOUSE_DATABASE' AND name = 'dict'")"
-
-echo 'status before reload'
-get_dictionary_status "$uuid"
-
-# source table does not exists
-# NOTE: when dictionary does not exist it produce BAD_ARGUMENTS error, so using UNKNOWN_TABLE is safe
-$CLICKHOUSE_CLIENT -n -q "SYSTEM RELOAD DICTIONARY dict -- { serverError UNKNOWN_TABLE }"
-echo 'status after reload'
-get_dictionary_status "$uuid"
-
-# create source
-$CLICKHOUSE_CLIENT -q "CREATE TABLE data (key Int, value String) Engine=Null"
-$CLICKHOUSE_CLIENT -q "SYSTEM RELOAD DICTIONARY dict"
-echo 'status after reload, table exists'
-get_dictionary_status "$uuid"
-
-# remove dictionary
-$CLICKHOUSE_CLIENT -q "DROP DICTIONARY dict"
-$CLICKHOUSE_CLIENT -q "DROP TABLE data"
-echo 'status after drop'
-get_dictionary_status "$uuid"

From bb47e2cc234baab26d018f081cb257ff1d0b9cf0 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 30 Aug 2023 20:56:14 +0200
Subject: [PATCH 03/37] Remove CurrentStatusInfo

---
 .../testdata/configs/xml/config.xml           | 21 -------------------
 .../testdata/configs/yandex_xml/config.xml    | 21 -------------------
 programs/server/config.xml                    |  2 --
 .../test_config_xml_full/configs/config.xml   | 21 -------------------
 4 files changed, 65 deletions(-)

diff --git a/programs/diagnostics/testdata/configs/xml/config.xml b/programs/diagnostics/testdata/configs/xml/config.xml
index c08b0b2970f..50352ca800e 100644
--- a/programs/diagnostics/testdata/configs/xml/config.xml
+++ b/programs/diagnostics/testdata/configs/xml/config.xml
@@ -760,27 +760,6 @@
     </graphite>
     -->
 
-    <!-- Serve endpoint for Prometheus monitoring. -->
-    <!--
-        endpoint - mertics path (relative to root, statring with "/")
-        port - port to setup server. If not defined or 0 than http_port used
-        metrics - send data from table system.metrics
-        events - send data from table system.events
-        asynchronous_metrics - send data from table system.asynchronous_metrics
-        status_info - send data from different component from CH, ex: Dictionaries status
-    -->
-    <!--
-    <prometheus>
-        <endpoint>/metrics</endpoint>
-        <port>9363</port>
-
-        <metrics>true</metrics>
-        <events>true</events>
-        <asynchronous_metrics>true</asynchronous_metrics>
-        <status_info>true</status_info>
-    </prometheus>
-    -->
-
     <!-- 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.
diff --git a/programs/diagnostics/testdata/configs/yandex_xml/config.xml b/programs/diagnostics/testdata/configs/yandex_xml/config.xml
index c65418c6363..181f52a92f8 100644
--- a/programs/diagnostics/testdata/configs/yandex_xml/config.xml
+++ b/programs/diagnostics/testdata/configs/yandex_xml/config.xml
@@ -760,27 +760,6 @@
     </graphite>
     -->
 
-    <!-- Serve endpoint for Prometheus monitoring. -->
-    <!--
-        endpoint - mertics path (relative to root, statring with "/")
-        port - port to setup server. If not defined or 0 than http_port used
-        metrics - send data from table system.metrics
-        events - send data from table system.events
-        asynchronous_metrics - send data from table system.asynchronous_metrics
-        status_info - send data from different component from CH, ex: Dictionaries status
-    -->
-    <!--
-    <prometheus>
-        <endpoint>/metrics</endpoint>
-        <port>9363</port>
-
-        <metrics>true</metrics>
-        <events>true</events>
-        <asynchronous_metrics>true</asynchronous_metrics>
-        <status_info>true</status_info>
-    </prometheus>
-    -->
-
     <!-- 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.
diff --git a/programs/server/config.xml b/programs/server/config.xml
index 07052441a01..7f1d395faa4 100644
--- a/programs/server/config.xml
+++ b/programs/server/config.xml
@@ -971,7 +971,6 @@
         metrics - send data from table system.metrics
         events - send data from table system.events
         asynchronous_metrics - send data from table system.asynchronous_metrics
-        status_info - send data from different component from CH, ex: Dictionaries status
     -->
     <!--
     <prometheus>
@@ -981,7 +980,6 @@
         <metrics>true</metrics>
         <events>true</events>
         <asynchronous_metrics>true</asynchronous_metrics>
-        <status_info>true</status_info>
     </prometheus>
     -->
 
diff --git a/tests/integration/test_config_xml_full/configs/config.xml b/tests/integration/test_config_xml_full/configs/config.xml
index d142df18af8..ac59b3428e8 100644
--- a/tests/integration/test_config_xml_full/configs/config.xml
+++ b/tests/integration/test_config_xml_full/configs/config.xml
@@ -674,27 +674,6 @@
     </graphite>
     -->
 
-    <!-- Serve endpoint for Prometheus monitoring. -->
-    <!--
-        endpoint - mertics path (relative to root, statring with "/")
-        port - port to setup server. If not defined or 0 than http_port used
-        metrics - send data from table system.metrics
-        events - send data from table system.events
-        asynchronous_metrics - send data from table system.asynchronous_metrics
-        status_info - send data from different component from CH, ex: Dictionaries status
-    -->
-    <!--
-    <prometheus>
-        <endpoint>/metrics</endpoint>
-        <port>9363</port>
-
-        <metrics>true</metrics>
-        <events>true</events>
-        <asynchronous_metrics>true</asynchronous_metrics>
-        <status_info>true</status_info>
-    </prometheus>
-    -->
-
     <!-- 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.

From 3d5e7aefe97b4a674223144425a082d20d371a60 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 24 Sep 2023 21:01:55 +0200
Subject: [PATCH 04/37] Remove status_info from testdata

---
 programs/diagnostics/testdata/configs/yaml/config.yaml | 2 --
 1 file changed, 2 deletions(-)

diff --git a/programs/diagnostics/testdata/configs/yaml/config.yaml b/programs/diagnostics/testdata/configs/yaml/config.yaml
index 80d65e6b187..354065a8a9b 100644
--- a/programs/diagnostics/testdata/configs/yaml/config.yaml
+++ b/programs/diagnostics/testdata/configs/yaml/config.yaml
@@ -638,7 +638,6 @@ default_session_timeout: 60
 # metrics - send data from table system.metrics
 # events - send data from table system.events
 # asynchronous_metrics - send data from table system.asynchronous_metrics
-# status_info - send data from different component from CH, ex: Dictionaries status
 
 # prometheus:
 #     endpoint: /metrics
@@ -647,7 +646,6 @@ default_session_timeout: 60
 #     metrics: true
 #     events: true
 #     asynchronous_metrics: true
-#     status_info: true
 
 # Query log. Used only for queries with setting log_queries = 1.
 query_log:

From 51ef0ec4bc5f41078cf54539b248ba3354fac0ca Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 24 Sep 2023 21:02:23 +0200
Subject: [PATCH 05/37] Remove status_info from example

---
 programs/server/config.yaml.example | 2 --
 1 file changed, 2 deletions(-)

diff --git a/programs/server/config.yaml.example b/programs/server/config.yaml.example
index a8f97ae5632..b472b6f4a45 100644
--- a/programs/server/config.yaml.example
+++ b/programs/server/config.yaml.example
@@ -639,7 +639,6 @@ default_session_timeout: 60
 # metrics - send data from table system.metrics
 # events - send data from table system.events
 # asynchronous_metrics - send data from table system.asynchronous_metrics
-# status_info - send data from different component from CH, ex: Dictionaries status
 
 # prometheus:
 #     endpoint: /metrics
@@ -648,7 +647,6 @@ default_session_timeout: 60
 #     metrics: true
 #     events: true
 #     asynchronous_metrics: true
-#     status_info: true
 
 # Query log. Used only for queries with setting log_queries = 1.
 query_log:

From 5e5761f6b3781edfccb2ca725dd74682feaad158 Mon Sep 17 00:00:00 2001
From: zvonand <azvonov@altinity.com>
Date: Sun, 24 Sep 2023 13:01:23 +0200
Subject: [PATCH 06/37] added docs

---
 .../functions/other-functions.md              | 29 +++++++++++++++----
 .../functions/other-functions.md              | 16 +++++++---
 2 files changed, 36 insertions(+), 9 deletions(-)

diff --git a/docs/en/sql-reference/functions/other-functions.md b/docs/en/sql-reference/functions/other-functions.md
index 566eff76584..378fc821df9 100644
--- a/docs/en/sql-reference/functions/other-functions.md
+++ b/docs/en/sql-reference/functions/other-functions.md
@@ -660,21 +660,26 @@ SELECT
 
 ## formatReadableTimeDelta
 
-Given a time interval (delta) in seconds, this function returns a time delta with year/month/day/hour/minute/second as string.
+Given a time interval (delta) in seconds, this function returns a time delta with year/month/day/hour/minute/second/millisecond/microsecond/nanosecond as string.
 
 **Syntax**
 
 ``` sql
-formatReadableTimeDelta(column[, maximum_unit])
+formatReadableTimeDelta(column[, maximum_unit, minimum_unit])
 ```
 
 **Arguments**
 
 - `column` — A column with a numeric time delta.
-- `maximum_unit` — Optional. Maximum unit to show. Acceptable values `seconds`, `minutes`, `hours`, `days`, `months`, `years`.
-
-Example:
+- `maximum_unit` — Optional. Maximum unit to show.
+  * 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.
 
+**Example**
 ``` sql
 SELECT
     arrayJoin([100, 12345, 432546534]) AS elapsed,
@@ -703,6 +708,20 @@ SELECT
 └────────────┴─────────────────────────────────────────────────────────────────┘
 ```
 
+```sql
+SELECT
+    arrayJoin([100, 12345, 432546534.00000006]) AS elapsed,
+    formatReadableTimeDelta(elapsed, 'minutes', 'nanoseconds') AS time_delta
+```
+
+```text
+┌────────────elapsed─┬─time_delta─────────────────────────────────────┐
+│                100 │ 1 minute and 40 seconds                        │
+│              12345 │ 205 minutes and 45 seconds                     │
+│ 432546534.00000006 │ 7209108 minutes, 54 seconds and 60 nanoseconds │
+└────────────────────┴────────────────────────────────────────────────┘
+```
+
 ## parseTimeDelta
 
 Parse a sequence of numbers followed by something resembling a time unit.
diff --git a/docs/ru/sql-reference/functions/other-functions.md b/docs/ru/sql-reference/functions/other-functions.md
index 78608aaf817..835aed934d5 100644
--- a/docs/ru/sql-reference/functions/other-functions.md
+++ b/docs/ru/sql-reference/functions/other-functions.md
@@ -659,9 +659,11 @@ SELECT
 └────────────────┴───────────────────┘
 ```
 
+
 ## formatReadableTimeDelta {#formatreadabletimedelta}
 
-Принимает дельту времени в секундах. Возвращает дельту времени с (год, месяц, день, час, минута, секунда) в виде строки.
+Принимает дельту времени в секундах. Возвращает дельту времени в виде строки.
+Есть возможность указать наиболее крупную единицу измерения для вывода (год, месяц, день, час, минута, секунда, милли/микро/наносекунда).
 
 **Синтаксис**
 
@@ -671,10 +673,16 @@ formatReadableTimeDelta(column[, maximum_unit])
 
 **Аргументы**
 
--   `column` — Столбец с числовой дельтой времени.
--   `maximum_unit` — Опциональный параметр. Максимальная единица измерения для отображения. Допустимые значения: секунды, минуты, часы, дни, месяцы, годы.
+- `column` — Столбец с числовой дельтой времени.
+- `maximum_unit` — Опциональный параметр. Максимальная единица измерения для отображения.
+  * Допустимые значения: `nanoseconds`, `microseconds`, `milliseconds`, `seconds`, `minutes`, `hours`, `days`, `months`, `years`.
+  * Значение по умолчанию: `years`.
+- `minimum_unit` — Опциональный параметр. Минимальная единица измерения для отображения. Более мелкие единицы будут отброшены.
+  * Допустимые значения: `nanoseconds`, `microseconds`, `milliseconds`, `seconds`, `minutes`, `hours`, `days`, `months`, `years`.
+  * Если минимальная единица задана явно и превышает максимальную единицу, будет выкинуто исключение.
+  * Значение по умолчанию: `seconds` если максимальная единица -- секунда или более крупный интервал, в противном случае -- `nanoseconds`.
 
-Пример:
+**Пример**
 
 ``` sql
 SELECT

From 77785925269390facbf31eb82e098bafc83b0b9b Mon Sep 17 00:00:00 2001
From: zvonand <azvonov@altinity.com>
Date: Sun, 24 Sep 2023 13:04:56 +0200
Subject: [PATCH 07/37] added subseconds to formatReadableTimeDelta and add
 tests

---
 src/Functions/formatReadableTimeDelta.cpp     | 250 +++++++++++++-----
 ...at_readable_timedelta_subseconds.reference |  24 ++
 ...7_format_readable_timedelta_subseconds.sql |  30 +++
 3 files changed, 243 insertions(+), 61 deletions(-)
 create mode 100644 tests/queries/0_stateless/02887_format_readable_timedelta_subseconds.reference
 create mode 100644 tests/queries/0_stateless/02887_format_readable_timedelta_subseconds.sql

diff --git a/src/Functions/formatReadableTimeDelta.cpp b/src/Functions/formatReadableTimeDelta.cpp
index 5fd48c01e8f..96d3a6fd966 100644
--- a/src/Functions/formatReadableTimeDelta.cpp
+++ b/src/Functions/formatReadableTimeDelta.cpp
@@ -2,11 +2,11 @@
 #include <Functions/IFunction.h>
 #include <Functions/FunctionHelpers.h>
 #include <Columns/ColumnString.h>
-#include <Columns/ColumnVector.h>
 #include <Common/NaNUtils.h>
 #include <DataTypes/DataTypeString.h>
 #include <IO/WriteBufferFromVector.h>
 #include <IO/WriteHelpers.h>
+#include <IO/DoubleConverter.h>
 
 
 namespace DB
@@ -14,9 +14,10 @@ namespace DB
 
 namespace ErrorCodes
 {
-    extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH;
-    extern const int ILLEGAL_TYPE_OF_ARGUMENT;
-    extern const int BAD_ARGUMENTS;
+extern const int CANNOT_PRINT_FLOAT_OR_DOUBLE_NUMBER;
+extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH;
+extern const int ILLEGAL_TYPE_OF_ARGUMENT;
+extern const int BAD_ARGUMENTS;
 }
 
 namespace
@@ -52,75 +53,88 @@ public:
     {
         if (arguments.empty())
             throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH,
-                "Number of arguments for function {} doesn't match: passed {}, should be at least 1.",
-                getName(), arguments.size());
+                            "Number of arguments for function {} doesn't match: passed {}, should be at least 1.",
+                            getName(), arguments.size());
 
-        if (arguments.size() > 2)
+        if (arguments.size() > 3)
             throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH,
-                "Number of arguments for function {} doesn't match: passed {}, should be at most 2.",
-                getName(), arguments.size());
+                            "Number of arguments for function {} doesn't match: passed {}, should be 1, 2 or 3.",
+                            getName(), arguments.size());
 
         const IDataType & type = *arguments[0];
 
         if (!isNativeNumber(type))
             throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Cannot format {} as time delta", type.getName());
 
-        if (arguments.size() == 2)
+        if (arguments.size() >= 2)
         {
             const auto * maximum_unit_arg = arguments[1].get();
             if (!isStringOrFixedString(maximum_unit_arg))
                 throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of argument maximum_unit of function {}",
                                 maximum_unit_arg->getName(), getName());
+
+            if (arguments.size() == 3)
+            {
+                const auto * minimum_unit_arg = arguments[2].get();
+                if (!isStringOrFixedString(minimum_unit_arg))
+                    throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of argument minimum_unit of function {}",
+                                    minimum_unit_arg->getName(), getName());
+            }
         }
 
         return std::make_shared<DataTypeString>();
     }
 
-    ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {1}; }
+    ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {1, 2}; }
 
     bool useDefaultImplementationForConstants() const override { return true; }
 
     enum Unit
     {
-        Seconds,
-        Minutes,
-        Hours,
-        Days,
-        Months,
-        Years
+        Nanoseconds = 1,
+        Microseconds = 2,
+        Milliseconds = 3,
+        Seconds = 4,
+        Minutes = 5,
+        Hours = 6,
+        Days = 7,
+        Months = 8,
+        Years = 9
     };
 
     ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override
     {
-        std::string_view maximum_unit_str;
-        if (arguments.size() == 2)
+        std::string_view maximum_unit_str, minimum_unit_str;
+        if (arguments.size() >= 2)
         {
             const ColumnPtr & maximum_unit_column = arguments[1].column;
             const ColumnConst * maximum_unit_const_col = checkAndGetColumnConstStringOrFixedString(maximum_unit_column.get());
             if (maximum_unit_const_col)
                 maximum_unit_str = maximum_unit_const_col->getDataColumn().getDataAt(0).toView();
+
+            if (arguments.size() == 3)
+            {
+                const ColumnPtr & minimum_unit_column = arguments[2].column;
+                const ColumnConst * minimum_unit_const_col = checkAndGetColumnConstStringOrFixedString(minimum_unit_column.get());
+                if (minimum_unit_const_col)
+                    minimum_unit_str = minimum_unit_const_col->getDataColumn().getDataAt(0).toView();
+            }
         }
+        /// Default means "use all available whole units".
+        Unit max_unit = dispatchUnit(maximum_unit_str, Years, "maximum");
+        /// Set seconds as min_unit by default not to ruin old use cases
+        Unit min_unit = dispatchUnit(minimum_unit_str, Seconds, "minimum");
 
-        Unit max_unit;
-
-        /// Default means "use all available units".
-        if (maximum_unit_str.empty() || maximum_unit_str == "years")
-            max_unit = Years;
-        else if (maximum_unit_str == "months")
-            max_unit = Months;
-        else if (maximum_unit_str == "days")
-            max_unit = Days;
-        else if (maximum_unit_str == "hours")
-            max_unit = Hours;
-        else if (maximum_unit_str == "minutes")
-            max_unit = Minutes;
-        else if (maximum_unit_str == "seconds")
-            max_unit = Seconds;
-        else
-            throw Exception(ErrorCodes::BAD_ARGUMENTS,
-                "Unexpected value of maximum unit argument ({}) for function {}, the only allowed values are:"
-                " 'seconds', 'minutes', 'hours', 'days', 'months', 'years'.",
-                maximum_unit_str, getName());
+        if (min_unit > max_unit)
+        {
+            if (minimum_unit_str.empty())
+                min_unit = Nanoseconds;   /// User wants sub-second max_unit. Show him all sub-second units unless other specified.
+            else
+                throw Exception(ErrorCodes::BAD_ARGUMENTS,
+                                "Explicitly specified value of minimum unit argument ({}) for function {} "
+                                "must not be greater than maximum unit value ({}).",
+                                minimum_unit_str, getName(), maximum_unit_str);
+        }
 
         auto col_to = ColumnString::create();
 
@@ -152,14 +166,61 @@ public:
                 /// To output separators between parts: ", " and " and ".
                 bool has_output = false;
 
+                Float64 whole_part;
+                std::string fractional_str = getFractionalString(std::modf(value, &whole_part));
+
                 switch (max_unit) /// A kind of Duff Device.
                 {
-                    case Years:     processUnit(365 * 24 * 3600, " year", 5, value, buf_to, has_output); [[fallthrough]];
-                    case Months:    processUnit(static_cast<UInt64>(30.5 * 24 * 3600), " month", 6, value, buf_to, has_output); [[fallthrough]];
-                    case Days:      processUnit(24 * 3600, " day", 4, value, buf_to, has_output); [[fallthrough]];
-                    case Hours:     processUnit(3600, " hour", 5, value, buf_to, has_output); [[fallthrough]];
-                    case Minutes:   processUnit(60, " minute", 7, value, buf_to, has_output); [[fallthrough]];
-                    case Seconds:   processUnit(1, " second", 7, value, buf_to, has_output);
+                    case Years:
+                        processUnit(365 * 24 * 3600, 0, " year", 5, whole_part, fractional_str, buf_to, has_output, min_unit, min_unit == Years);
+                        if (min_unit == Years)
+                            break;
+                        [[fallthrough]];
+
+                    case Months:
+                        processUnit(static_cast<UInt64>(30.5 * 24 * 3600), 0, " month", 6, whole_part, fractional_str, buf_to, has_output, min_unit, min_unit == Months);
+                        if (min_unit == Months)
+                            break;
+                        [[fallthrough]];
+
+                    case Days:
+                        processUnit(24 * 3600, 0, " day", 4, whole_part, fractional_str, buf_to, has_output, min_unit, min_unit == Days);
+                        if (min_unit == Days)
+                            break;
+                        [[fallthrough]];
+
+                    case Hours:
+                        processUnit(3600, 0, " hour", 5, whole_part, fractional_str, buf_to, has_output, min_unit, min_unit == Hours);
+                        if (min_unit == Hours)
+                            break;
+                        [[fallthrough]];
+
+                    case Minutes:
+                        processUnit(60, 0, " minute", 7, whole_part, fractional_str, buf_to, has_output, min_unit, min_unit == Minutes);
+                        if (min_unit == Minutes)
+                            break;
+                        [[fallthrough]];
+
+                    case Seconds:
+                        processUnit(1, 0, " second", 7, whole_part, fractional_str, buf_to, has_output, min_unit, min_unit == Seconds);
+                        if (min_unit == Seconds)
+                            break;
+                        [[fallthrough]];
+
+                    case Milliseconds:
+                        processUnit(1, 3, " millisecond", 12, whole_part, fractional_str, buf_to, has_output, min_unit, min_unit == Milliseconds);
+                        if (min_unit == Milliseconds)
+                            break;
+                        [[fallthrough]];
+
+                    case Microseconds:
+                        processUnit(1, 6, " microsecond", 12, whole_part, fractional_str, buf_to, has_output, min_unit, min_unit == Microseconds);
+                        if (min_unit == Microseconds)
+                            break;
+                        [[fallthrough]];
+
+                    case Nanoseconds:
+                        processUnit(1, 9, " nanosecond", 11, whole_part, fractional_str, buf_to, has_output, min_unit, true);
                 }
             }
 
@@ -172,42 +233,68 @@ public:
     }
 
     static void processUnit(
-        UInt64 unit_size, const char * unit_name, size_t unit_name_size,
-        Float64 & value, WriteBuffer & buf_to, bool & has_output)
+        UInt64 unit_multiplier, UInt32 unit_scale, const char * unit_name, size_t unit_name_size, Float64 & whole_part,
+        String & fractional_str, WriteBuffer & buf_to, bool & has_output,  Unit min_unit, bool is_minimum_unit)
     {
-        if (unlikely(value + 1.0 == value))
+        if (unlikely(whole_part + 1.0 == whole_part))
         {
             /// The case when value is too large so exact representation for subsequent smaller units is not possible.
-            writeText(std::floor(value / unit_size), buf_to);
+            writeText(std::floor(whole_part * DecimalUtils::scaleMultiplier<Int64>(unit_scale) / unit_multiplier), buf_to);
             buf_to.write(unit_name, unit_name_size);
             writeChar('s', buf_to);
             has_output = true;
-            value = 0;
+            whole_part = 0;
             return;
         }
-
-        UInt64 num_units = static_cast<UInt64>(value / unit_size);
-
-        if (!num_units)
+        UInt64 num_units = 0;
+        if (unit_scale == 0)  /// dealing with whole number of seconds
         {
-            /// Zero units, no need to print. But if it's the last (seconds) and the only unit, print "0 seconds" nevertheless.
-            if (unit_size > 1 || has_output)
-                return;
+            num_units = static_cast<UInt64>(std::floor(whole_part / unit_multiplier));
+
+            if (!num_units)
+            {
+                /// Zero units, no need to print. But if it's the last (seconds) and the only unit, print "0 seconds" nevertheless.
+                if (unit_multiplier != 1 || has_output)
+                    return;
+            }
+
+            /// Remaining value to print on next iteration.
+            whole_part -= num_units * unit_multiplier;
+        }
+        else   /// dealing with sub-seconds, a bit more peculiar to avoid more precision issues
+        {
+            if (whole_part >= 1)  /// There were no whole units printed
+            {
+                num_units += static_cast<UInt64>(whole_part) * DecimalUtils::scaleMultiplier<Int64>(unit_scale);
+                whole_part = 0;
+            }
+
+            for (UInt32 i = 0; i < unit_scale; ++i)
+            {
+                num_units += (fractional_str[i] - '0') * DecimalUtils::scaleMultiplier<Int64>(unit_scale - i - 1);
+                fractional_str[i] = '0';
+            }
+
+            if (!num_units)
+            {
+                /// Zero units, no need to print. But if it's the last (nanoseconds) and the only unit, print "0 nanoseconds" nevertheless.
+                if (!is_minimum_unit || has_output)
+                    return;
+            }
         }
 
-        /// Remaining value to print on next iteration.
-        value -= num_units * unit_size;
-
+        /// Write number of units
         if (has_output)
         {
             /// Need delimiter between values. The last delimiter is " and ", all previous are comma.
-            if (value < 1)
+            if (is_minimum_unit || (whole_part < 1 && fractional_str.substr(0, (4 - min_unit) * 3) == std::string((4 - min_unit) * 3, '0')))
                 writeCString(" and ", buf_to);
             else
                 writeCString(", ", buf_to);
         }
 
         writeText(num_units, buf_to);
+
         buf_to.write(unit_name, unit_name_size); /// If we just leave strlen(unit_name) here, clang-11 fails to make it compile-time.
 
         /// How to pronounce: unit vs. units.
@@ -216,6 +303,47 @@ public:
 
         has_output = true;
     }
+
+private:
+    static std::string getFractionalString(const Float64 & fractional_part)
+    {
+        DB::DoubleConverter<true>::BufferType buffer;
+        double_conversion::StringBuilder builder{buffer, sizeof(buffer)};
+
+        if (!DB::DoubleConverter<false>::instance().ToFixed(fractional_part, 9, &builder))
+            throw DB::Exception(DB::ErrorCodes::CANNOT_PRINT_FLOAT_OR_DOUBLE_NUMBER, "Cannot print double number: {}", fractional_part);
+
+        return std::string(buffer, builder.position()).substr(2);   /// do not return `0.` -- we don't need it
+    }
+
+    Unit dispatchUnit(const std::string_view & unit_str, const Unit default_unit, const std::string & bound_name) const
+    {
+        if (unit_str.empty())
+            return default_unit;
+        else if (unit_str == "years")
+            return Years;
+        else if (unit_str == "months")
+            return Months;
+        else if (unit_str == "days")
+            return Days;
+        else if (unit_str == "hours")
+            return Hours;
+        else if (unit_str == "minutes")
+            return Minutes;
+        else if (unit_str == "seconds")
+            return Seconds;
+        else if (unit_str == "milliseconds")
+            return Milliseconds;
+        else if (unit_str == "microseconds")
+            return Microseconds;
+        else if (unit_str == "nanoseconds")
+            return Nanoseconds;
+        else
+            throw Exception(ErrorCodes::BAD_ARGUMENTS,
+                            "Unexpected value of {} unit argument ({}) for function {}, the only allowed values are:"
+                            " 'nanoseconds', 'microseconds', 'nanoseconds', 'seconds', 'minutes', 'hours', 'days', 'months', 'years'.",
+                            bound_name, unit_str, getName());
+    }
 };
 
 }
diff --git a/tests/queries/0_stateless/02887_format_readable_timedelta_subseconds.reference b/tests/queries/0_stateless/02887_format_readable_timedelta_subseconds.reference
new file mode 100644
index 00000000000..389479eb529
--- /dev/null
+++ b/tests/queries/0_stateless/02887_format_readable_timedelta_subseconds.reference
@@ -0,0 +1,24 @@
+1 second
+1 minute
+2 minutes and 3 seconds
+24 seconds
+35 seconds
+1 minute and 6 seconds
+1 minute and 7 seconds
+48 seconds
+1 minute and 39 seconds
+1 hour
+1120 milliseconds
+60200 milliseconds
+123330 milliseconds
+24450 milliseconds
+35570 milliseconds
+66640 milliseconds
+67797 milliseconds, 979 microseconds and 790 nanoseconds
+48888 milliseconds and 888 microseconds
+99960 milliseconds
+3600000 milliseconds
+0 seconds
+1 hour and 1 second
+0 nanoseconds
+1000 milliseconds and 500 microseconds
diff --git a/tests/queries/0_stateless/02887_format_readable_timedelta_subseconds.sql b/tests/queries/0_stateless/02887_format_readable_timedelta_subseconds.sql
new file mode 100644
index 00000000000..0a8a76d54fd
--- /dev/null
+++ b/tests/queries/0_stateless/02887_format_readable_timedelta_subseconds.sql
@@ -0,0 +1,30 @@
+-- max_unit bigger than second, min_unit omitted (and considered 'seconds')
+WITH
+    'hours' AS maximum_unit,
+    arrayJoin([1.12, 60.2, 123.33, 24.45, 35.57, 66.64, 67.79, 48.88, 99.96, 3600]) AS elapsed
+SELECT
+    formatReadableTimeDelta(elapsed, maximum_unit) AS time_delta;
+
+-- max_unit smaller than second, min_unit omitted (and considered 'nanoseconds')
+WITH
+    'milliseconds' AS maximum_unit,
+    arrayJoin([1.12, 60.2, 123.33, 24.45, 35.57, 66.64, 67.79797979, 48.888888, 99.96, 3600]) AS elapsed
+SELECT
+    formatReadableTimeDelta(elapsed, maximum_unit) AS time_delta;
+
+-- Check exception is thrown
+SELECT formatReadableTimeDelta(1.1, 'seconds', 'hours'); -- { serverError BAD_ARGUMENTS }
+
+-- Check empty units are omitted unless they are the only one
+WITH
+    'hours' AS maximum_unit,
+    'microseconds' as minimum_unit,
+    arrayJoin([0, 3601.000000003]) AS elapsed
+SELECT
+    formatReadableTimeDelta(elapsed, maximum_unit, minimum_unit);
+
+WITH
+    'milliseconds' AS maximum_unit,
+    arrayJoin([0, 1.0005]) AS elapsed
+SELECT
+    formatReadableTimeDelta(elapsed, maximum_unit);

From b8695525ad1c54f03ba29b74ea05f9a2c15f3e50 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 24 Sep 2023 19:17:33 +0200
Subject: [PATCH 08/37] Add a test for compatibility

---
 .../0_stateless/02887_tuple_element_distributed.reference   | 6 ++++++
 .../queries/0_stateless/02887_tuple_element_distributed.sql | 1 +
 2 files changed, 7 insertions(+)
 create mode 100644 tests/queries/0_stateless/02887_tuple_element_distributed.reference
 create mode 100644 tests/queries/0_stateless/02887_tuple_element_distributed.sql

diff --git a/tests/queries/0_stateless/02887_tuple_element_distributed.reference b/tests/queries/0_stateless/02887_tuple_element_distributed.reference
new file mode 100644
index 00000000000..a9e2f17562a
--- /dev/null
+++ b/tests/queries/0_stateless/02887_tuple_element_distributed.reference
@@ -0,0 +1,6 @@
+1
+1
+1
+1
+1
+1
diff --git a/tests/queries/0_stateless/02887_tuple_element_distributed.sql b/tests/queries/0_stateless/02887_tuple_element_distributed.sql
new file mode 100644
index 00000000000..b9c5c856d00
--- /dev/null
+++ b/tests/queries/0_stateless/02887_tuple_element_distributed.sql
@@ -0,0 +1 @@
+SELECT equals(tupleElement(tuple('a', 10) AS x, 1), 'a') FROM remote('127.0.0.{1,2,3}', numbers(2));

From bb0375d97015a030193db8ffec13dd4df6edc1d0 Mon Sep 17 00:00:00 2001
From: Nikolay Degterinsky <evillique@gmail.com>
Date: Fri, 22 Sep 2023 20:20:21 +0000
Subject: [PATCH 09/37] Fix data race in SYSTEM STOP LISTEN

---
 programs/server/Server.cpp | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/programs/server/Server.cpp b/programs/server/Server.cpp
index a968acb549f..78f3f749d9a 100644
--- a/programs/server/Server.cpp
+++ b/programs/server/Server.cpp
@@ -1516,11 +1516,13 @@ try
 
     global_context->setStopServersCallback([&](const ServerType & server_type)
     {
+        std::lock_guard lock(servers_lock);
         stopServers(servers, server_type);
     });
 
     global_context->setStartServersCallback([&](const ServerType & server_type)
     {
+        std::lock_guard lock(servers_lock);
         createServers(
             config(),
             listen_hosts,

From 4fb12f8703d68400cd7d1ea71f18e645ea889c49 Mon Sep 17 00:00:00 2001
From: avogar <avogar@clickhouse.com>
Date: Mon, 18 Sep 2023 19:35:08 +0000
Subject: [PATCH 10/37] Add retries to rests
 test_async_query_sending/test_async_connect

---
 .../integration/test_hedged_requests/test.py  | 37 +++++++++++++------
 1 file changed, 26 insertions(+), 11 deletions(-)

diff --git a/tests/integration/test_hedged_requests/test.py b/tests/integration/test_hedged_requests/test.py
index 6f5d5706da4..d3f22625f76 100644
--- a/tests/integration/test_hedged_requests/test.py
+++ b/tests/integration/test_hedged_requests/test.py
@@ -133,7 +133,7 @@ def check_if_query_sending_was_suspended():
         "SELECT value FROM system.events WHERE event='SuspendSendingQueryToShard'"
     )
 
-    assert int(result) >= 1
+    return len(result) != 0 and int(result) >= 1
 
 
 def check_if_query_sending_was_not_suspended():
@@ -380,11 +380,19 @@ def test_async_connect(started_cluster):
     # Restart server to reset connection pool state
     NODES["node"].restart_clickhouse()
 
-    NODES["node"].query(
-        "SELECT hostName(), id FROM distributed_connect ORDER BY id LIMIT 1 SETTINGS prefer_localhost_replica = 0, connect_timeout_with_failover_ms=5000, async_query_sending_for_remote=1, max_threads=1, max_distributed_connections=1"
-    )
-    check_changing_replica_events(2)
-    check_if_query_sending_was_suspended()
+    attempt = 0
+    while attempt < 1000:
+        NODES["node"].query(
+            "SELECT hostName(), id FROM distributed_connect ORDER BY id LIMIT 1 SETTINGS prefer_localhost_replica = 0, connect_timeout_with_failover_ms=5000, async_query_sending_for_remote=1, max_threads=1, max_distributed_connections=1"
+        )
+
+        check_changing_replica_events(2)
+        if check_if_query_sending_was_suspended():
+            break
+
+        attempt += 1
+
+    assert attempt < 1000
 
     NODES["node"].query("DROP TABLE distributed_connect")
 
@@ -418,10 +426,17 @@ def test_async_query_sending(started_cluster):
     )
     check_if_query_sending_was_not_suspended()
 
-    NODES["node"].query(
-        "SELECT hostName(), id FROM distributed_query_sending ORDER BY id LIMIT 1 SETTINGS"
-        " prefer_localhost_replica = 0, async_query_sending_for_remote=1, max_threads = 1, max_distributed_connections=1"
-    )
-    check_if_query_sending_was_suspended()
+    attempt = 0
+    while attempt < 1000:
+        NODES["node"].query(
+            "SELECT hostName(), id FROM distributed_query_sending ORDER BY id LIMIT 1 SETTINGS"
+            " prefer_localhost_replica = 0, async_query_sending_for_remote=1, max_threads = 1, max_distributed_connections=1"
+        )
 
+        if check_if_query_sending_was_suspended():
+            break
+
+        attempt += 1
+
+    assert attempt < 1000
     NODES["node"].query("DROP TABLE distributed_query_sending")

From 92de61ca5bfe68aa7e758ee70cf6a0d94c8fb7f1 Mon Sep 17 00:00:00 2001
From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com>
Date: Tue, 19 Sep 2023 13:32:49 +0200
Subject: [PATCH 11/37] Use less attempts

---
 tests/integration/test_hedged_requests/test.py | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/tests/integration/test_hedged_requests/test.py b/tests/integration/test_hedged_requests/test.py
index d3f22625f76..02ecf3c1367 100644
--- a/tests/integration/test_hedged_requests/test.py
+++ b/tests/integration/test_hedged_requests/test.py
@@ -381,7 +381,7 @@ def test_async_connect(started_cluster):
     NODES["node"].restart_clickhouse()
 
     attempt = 0
-    while attempt < 1000:
+    while attempt < 100:
         NODES["node"].query(
             "SELECT hostName(), id FROM distributed_connect ORDER BY id LIMIT 1 SETTINGS prefer_localhost_replica = 0, connect_timeout_with_failover_ms=5000, async_query_sending_for_remote=1, max_threads=1, max_distributed_connections=1"
         )
@@ -392,7 +392,7 @@ def test_async_connect(started_cluster):
 
         attempt += 1
 
-    assert attempt < 1000
+    assert attempt < 100
 
     NODES["node"].query("DROP TABLE distributed_connect")
 
@@ -427,7 +427,7 @@ def test_async_query_sending(started_cluster):
     check_if_query_sending_was_not_suspended()
 
     attempt = 0
-    while attempt < 1000:
+    while attempt < 100:
         NODES["node"].query(
             "SELECT hostName(), id FROM distributed_query_sending ORDER BY id LIMIT 1 SETTINGS"
             " prefer_localhost_replica = 0, async_query_sending_for_remote=1, max_threads = 1, max_distributed_connections=1"
@@ -438,5 +438,5 @@ def test_async_query_sending(started_cluster):
 
         attempt += 1
 
-    assert attempt < 1000
+    assert attempt < 100
     NODES["node"].query("DROP TABLE distributed_query_sending")

From ac384677cce8a59265d9a05cf968b355b769fc38 Mon Sep 17 00:00:00 2001
From: kssenii <sumarokovakseniia@mail.ru>
Date: Mon, 25 Sep 2023 13:10:25 +0200
Subject: [PATCH 12/37] Delete a test

---
 tests/queries/0_stateless/02344_describe_cache.reference | 1 -
 tests/queries/0_stateless/02344_describe_cache.sql       | 4 ----
 2 files changed, 5 deletions(-)
 delete mode 100644 tests/queries/0_stateless/02344_describe_cache.reference
 delete mode 100644 tests/queries/0_stateless/02344_describe_cache.sql

diff --git a/tests/queries/0_stateless/02344_describe_cache.reference b/tests/queries/0_stateless/02344_describe_cache.reference
deleted file mode 100644
index 5db1aa3e7f5..00000000000
--- a/tests/queries/0_stateless/02344_describe_cache.reference
+++ /dev/null
@@ -1 +0,0 @@
-134217728	10000000	33554432	4194304	1	0	0	0	/var/lib/clickhouse/filesystem_caches/s3_cache/	2	0
diff --git a/tests/queries/0_stateless/02344_describe_cache.sql b/tests/queries/0_stateless/02344_describe_cache.sql
deleted file mode 100644
index 9c5c5c10952..00000000000
--- a/tests/queries/0_stateless/02344_describe_cache.sql
+++ /dev/null
@@ -1,4 +0,0 @@
--- Tags: no-fasttest, no-parallel
-
-SYSTEM DROP FILESYSTEM CACHE 's3_cache';
-DESCRIBE FILESYSTEM CACHE 's3_cache';

From d2b787429dafd7b66a3fb21eab94e5cf8d524397 Mon Sep 17 00:00:00 2001
From: Azat Khuzhin <a3at.mail@gmail.com>
Date: Mon, 25 Sep 2023 14:23:46 +0200
Subject: [PATCH 13/37] Remove 02151_hash_table_sizes_stats_distributed (fixes
 broken CI) (#54969)

* Fix 02151_hash_table_sizes_stats_distributed (fixes broken CI)

Fixes: #54961
Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>

* Remove 02151_hash_table_sizes_stats_distributed

Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>

---------

Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>
---
 ...sh_table_sizes_stats_distributed.reference | 33 -------
 ...2151_hash_table_sizes_stats_distributed.sh | 95 -------------------
 2 files changed, 128 deletions(-)
 delete mode 100644 tests/queries/0_stateless/02151_hash_table_sizes_stats_distributed.reference
 delete mode 100755 tests/queries/0_stateless/02151_hash_table_sizes_stats_distributed.sh

diff --git a/tests/queries/0_stateless/02151_hash_table_sizes_stats_distributed.reference b/tests/queries/0_stateless/02151_hash_table_sizes_stats_distributed.reference
deleted file mode 100644
index 0d10114f4ff..00000000000
--- a/tests/queries/0_stateless/02151_hash_table_sizes_stats_distributed.reference
+++ /dev/null
@@ -1,33 +0,0 @@
-1
-1
---
-1
-1
---
-1
-1
---
-1
-1
---
-1
-1
-1
-1
---
-1
-1
---
-1
-1
-1
-1
---
-1
-1
---
-1
-1
-1
-1
---
diff --git a/tests/queries/0_stateless/02151_hash_table_sizes_stats_distributed.sh b/tests/queries/0_stateless/02151_hash_table_sizes_stats_distributed.sh
deleted file mode 100755
index 77b9b2942c5..00000000000
--- a/tests/queries/0_stateless/02151_hash_table_sizes_stats_distributed.sh
+++ /dev/null
@@ -1,95 +0,0 @@
-#!/usr/bin/env bash
-# Tags: long, distributed, no-tsan, no-debug
-
-# These tests don't use `current_database = currentDatabase()` condition, because database name isn't propagated during remote queries.
-
-# shellcheck disable=SC2154
-
-CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
-# shellcheck source=../shell_config.sh
-. "$CURDIR"/../shell_config.sh
-
-
-# tests rely on that all the rows are unique and max_threads divides table_size
-table_size=1000005
-max_threads=5
-
-
-prepare_table() {
-  table_name="t_hash_table_sizes_stats_$RANDOM$RANDOM"
-  $CLICKHOUSE_CLIENT -q "DROP TABLE IF EXISTS $table_name;"
-  if [ -z "$1" ]; then
-    $CLICKHOUSE_CLIENT -q "CREATE TABLE $table_name(number UInt64) Engine=MergeTree() ORDER BY tuple() SETTINGS index_granularity = 8192, index_granularity_bytes = '10Mi';"
-  else
-    $CLICKHOUSE_CLIENT -q "CREATE TABLE $table_name(number UInt64) Engine=MergeTree() ORDER BY $1 SETTINGS index_granularity = 8192, index_granularity_bytes = '10Mi';"
-  fi
-  $CLICKHOUSE_CLIENT -q "SYSTEM STOP MERGES $table_name;"
-  for ((i = 1; i <= max_threads; i++)); do
-    cnt=$((table_size / max_threads))
-    from=$(((i - 1) * cnt))
-    $CLICKHOUSE_CLIENT -q "INSERT INTO $table_name SELECT * FROM numbers($from, $cnt);"
-  done
-  $CLICKHOUSE_CLIENT -q "DROP TABLE IF EXISTS ${table_name}_d;"
-  $CLICKHOUSE_CLIENT -q "CREATE TABLE ${table_name}_d AS $table_name ENGINE = Distributed(test_cluster_two_shards, currentDatabase(), $table_name);"
-  table_name="${table_name}_d"
-}
-
-prepare_table_with_sorting_key() {
-  prepare_table "$1"
-}
-
-run_query() {
-  query_id="${CLICKHOUSE_DATABASE}_hash_table_sizes_stats_$RANDOM$RANDOM"
-  $CLICKHOUSE_CLIENT --query_id="$query_id" --multiquery -q "
-    SET max_block_size = $((table_size / 10));
-    SET merge_tree_min_rows_for_concurrent_read = 1;
-    SET max_untracked_memory = 0;
-    SET prefer_localhost_replica = 1;
-    $query"
-}
-
-check_preallocated_elements() {
-  $CLICKHOUSE_CLIENT -q "SYSTEM FLUSH LOGS"
-  # rows may be distributed in any way including "everything goes to the one particular thread"
-  min=$1
-  if [ -z "$2" ]; then
-    max=$1
-  else
-    max=$2
-  fi
-  $CLICKHOUSE_CLIENT --param_query_id="$query_id" -q "
-    SELECT COUNT(*)
-      FROM system.query_log
-     WHERE event_date >= yesterday() AND (query_id = {query_id:String} OR initial_query_id = {query_id:String})
-           AND ProfileEvents['AggregationPreallocatedElementsInHashTables'] BETWEEN $min AND $max
-  GROUP BY query_id"
-}
-
-check_convertion_to_two_level() {
-  $CLICKHOUSE_CLIENT -q "SYSTEM FLUSH LOGS"
-  # rows may be distributed in any way including "everything goes to the one particular thread"
-  $CLICKHOUSE_CLIENT --param_query_id="$query_id" -q "
-    SELECT SUM(ProfileEvents['AggregationHashTablesInitializedAsTwoLevel']) BETWEEN 1 AND $max_threads
-      FROM system.query_log
-     WHERE event_date >= yesterday() AND (query_id = {query_id:String} OR initial_query_id = {query_id:String})
-  GROUP BY query_id"
-}
-
-print_border() {
-  echo "--"
-}
-
-
-# shellcheck source=./02151_hash_table_sizes_stats.testcases
-source "$CURDIR"/02151_hash_table_sizes_stats.testcases
-
-
-test_one_thread_simple_group_by
-test_one_thread_simple_group_by_with_limit
-test_one_thread_simple_group_by_with_join_and_subquery
-test_several_threads_simple_group_by_with_limit_single_level_ht
-test_several_threads_simple_group_by_with_limit_two_level_ht
-test_several_threads_simple_group_by_with_limit_and_rollup_single_level_ht
-test_several_threads_simple_group_by_with_limit_and_rollup_two_level_ht
-test_several_threads_simple_group_by_with_limit_and_cube_single_level_ht
-test_several_threads_simple_group_by_with_limit_and_cube_two_level_ht

From 5a6d37d8e793299bf15c7cad2c366575ac5670d3 Mon Sep 17 00:00:00 2001
From: Alex Cheng <chengxin1985@gmail.com>
Date: Fri, 22 Sep 2023 19:55:44 +0800
Subject: [PATCH 14/37] fix exception message in Pipe::addTransform

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

diff --git a/src/QueryPipeline/Pipe.cpp b/src/QueryPipeline/Pipe.cpp
index 293d152ea65..ff842478712 100644
--- a/src/QueryPipeline/Pipe.cpp
+++ b/src/QueryPipeline/Pipe.cpp
@@ -556,7 +556,7 @@ void Pipe::addTransform(ProcessorPtr transform, InputPort * totals, InputPort *
         throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot add transform consuming totals to Pipe because Pipe does not have totals");
 
     if (extremes && !extremes_port)
-        throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot add transform consuming extremes to Pipe because it already has extremes");
+        throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot add transform consuming extremes to Pipe because Pipe does not have extremes");
 
     if (totals)
     {

From b7a3873265ae7afa68b379920c9d60cb2fbf7d82 Mon Sep 17 00:00:00 2001
From: flynn <fenglv15@mails.ucas.ac.cn>
Date: Fri, 22 Sep 2023 17:43:52 +0000
Subject: [PATCH 15/37] Add function hammingDistance

---
 src/Functions/FunctionsStringSimilarity.h     |  42 +++++--
 src/Functions/stringHammingDistance.cpp       | 106 ++++++++++++++++++
 .../02884_string_hamming_distance.reference   |  10 ++
 .../02884_string_hamming_distance.sql         |  17 +++
 4 files changed, 166 insertions(+), 9 deletions(-)
 create mode 100644 src/Functions/stringHammingDistance.cpp
 create mode 100644 tests/queries/0_stateless/02884_string_hamming_distance.reference
 create mode 100644 tests/queries/0_stateless/02884_string_hamming_distance.sql

diff --git a/src/Functions/FunctionsStringSimilarity.h b/src/Functions/FunctionsStringSimilarity.h
index a360442ea75..e148730054d 100644
--- a/src/Functions/FunctionsStringSimilarity.h
+++ b/src/Functions/FunctionsStringSimilarity.h
@@ -27,6 +27,9 @@ namespace ErrorCodes
     extern const int TOO_LARGE_STRING_SIZE;
 }
 
+template <typename T>
+concept has_max_string_size = requires { T::max_string_size; };
+
 template <typename Impl, typename Name>
 class FunctionsStringSimilarity : public IFunction
 {
@@ -68,10 +71,17 @@ public:
         {
             ResultType res{};
             const String & needle = col_needle_const->getValue<String>();
-            if (needle.size() > Impl::max_string_size)
+            if constexpr (has_max_string_size<Impl>)
             {
-                throw Exception(ErrorCodes::TOO_LARGE_STRING_SIZE, "String size of needle is too big for function {}. "
-                    "Should be at most {}", getName(), Impl::max_string_size);
+                if (needle.size() > Impl::max_string_size)
+                {
+                    throw Exception(
+                        ErrorCodes::TOO_LARGE_STRING_SIZE,
+                        "String size of needle is too big for function {}. "
+                        "Should be at most {}",
+                        getName(),
+                        Impl::max_string_size);
+                }
             }
             Impl::constantConstant(col_haystack_const->getValue<String>(), needle, res);
             return result_type->createColumnConst(col_haystack_const->size(), toField(res));
@@ -88,10 +98,17 @@ public:
         if (col_haystack_vector && col_needle_const)
         {
             const String & needle = col_needle_const->getValue<String>();
-            if (needle.size() > Impl::max_string_size)
+            if constexpr (has_max_string_size<Impl>)
             {
-                throw Exception(ErrorCodes::TOO_LARGE_STRING_SIZE, "String size of needle is too big for function {}. "
-                    "Should be at most {}", getName(), Impl::max_string_size);
+                if (needle.size() > Impl::max_string_size)
+                {
+                    throw Exception(
+                        ErrorCodes::TOO_LARGE_STRING_SIZE,
+                        "String size of needle is too big for function {}. "
+                        "Should be at most {}",
+                        getName(),
+                        Impl::max_string_size);
+                }
             }
             Impl::vectorConstant(col_haystack_vector->getChars(), col_haystack_vector->getOffsets(), needle, vec_res);
         }
@@ -107,10 +124,17 @@ public:
         else if (col_haystack_const && col_needle_vector)
         {
             const String & haystack = col_haystack_const->getValue<String>();
-            if (haystack.size() > Impl::max_string_size)
+            if constexpr (has_max_string_size<Impl>)
             {
-                throw Exception(ErrorCodes::TOO_LARGE_STRING_SIZE, "String size of haystack is too big for function {}. "
-                    "Should be at most {}", getName(), Impl::max_string_size);
+                if (haystack.size() > Impl::max_string_size)
+                {
+                    throw Exception(
+                        ErrorCodes::TOO_LARGE_STRING_SIZE,
+                        "String size of haystack is too big for function {}. "
+                        "Should be at most {}",
+                        getName(),
+                        Impl::max_string_size);
+                }
             }
             Impl::constantVector(haystack, col_needle_vector->getChars(), col_needle_vector->getOffsets(), vec_res);
         }
diff --git a/src/Functions/stringHammingDistance.cpp b/src/Functions/stringHammingDistance.cpp
new file mode 100644
index 00000000000..fd22ca96ba3
--- /dev/null
+++ b/src/Functions/stringHammingDistance.cpp
@@ -0,0 +1,106 @@
+#include <Columns/ColumnString.h>
+#include <Columns/ColumnsNumber.h>
+#include <DataTypes/DataTypeString.h>
+#include <DataTypes/DataTypesNumber.h>
+#include <Functions/FunctionFactory.h>
+#include <Functions/FunctionsStringSimilarity.h>
+
+#ifdef __SSE4_2__
+#    include <nmmintrin.h>
+#endif
+
+namespace DB
+{
+struct StringHammingDistanceImpl
+{
+    using ResultType = UInt64;
+
+    static void constantConstant(const std::string & haystack, const std::string & needle, UInt64 & res)
+    {
+        res = process(haystack.data(), haystack.size(), needle.data(), needle.size());
+    }
+
+    static void vectorVector(
+        const ColumnString::Chars & haystack_data,
+        const ColumnString::Offsets & haystack_offsets,
+        const ColumnString::Chars & needle_data,
+        const ColumnString::Offsets & needle_offsets,
+        PaddedPODArray<UInt64> & res)
+    {
+        size_t size = res.size();
+        const char * haystack = reinterpret_cast<const char *>(haystack_data.data());
+        const char * needle = reinterpret_cast<const char *>(needle_data.data());
+        for (size_t i = 0; i < size; ++i)
+        {
+            res[i] = process(
+                haystack + haystack_offsets[i - 1],
+                haystack_offsets[i] - haystack_offsets[i - 1] - 1,
+                needle + needle_offsets[i - 1],
+                needle_offsets[i] - needle_offsets[i - 1] - 1);
+        }
+    }
+
+    static void constantVector(
+        const std::string & haystack,
+        const ColumnString::Chars & needle_data,
+        const ColumnString::Offsets & needle_offsets,
+        PaddedPODArray<UInt64> & res)
+    {
+        const char * haystack_data = haystack.data();
+        size_t haystack_size = haystack.size();
+        const char * needle = reinterpret_cast<const char *>(needle_data.data());
+        size_t size = res.size();
+        for (size_t i = 0; i < size; ++i)
+        {
+            res[i] = process(haystack_data, haystack_size, needle + needle_offsets[i - 1], needle_offsets[i] - needle_offsets[i - 1] - 1);
+        }
+    }
+
+    static void vectorConstant(
+        const ColumnString::Chars & data, const ColumnString::Offsets & offsets, const std::string & needle, PaddedPODArray<UInt64> & res)
+    {
+        constantVector(needle, data, offsets, res);
+    }
+
+private:
+    static UInt64 inline process(const char * haystack, size_t haystack_size, const char * needle, size_t needle_size)
+    {
+        UInt64 res = 0;
+        const char * haystack_end = haystack + haystack_size;
+        const char * needle_end = needle + needle_size;
+
+#ifdef __SSE4_2__
+        static constexpr auto mode = _SIDD_UBYTE_OPS | _SIDD_CMP_EQUAL_EACH | _SIDD_NEGATIVE_POLARITY;
+
+        const char * haystack_end16 = haystack + haystack_size / 16 * 16;
+        const char * needle_end16 = needle + needle_size / 16 * 16;
+
+        for (; haystack < haystack_end16 && needle < needle_end16; haystack += 16, needle += 16)
+        {
+            __m128i s1 = _mm_loadu_si128(reinterpret_cast<const __m128i *>(haystack));
+            __m128i s2 = _mm_loadu_si128(reinterpret_cast<const __m128i *>(needle));
+            auto result_mask = _mm_cmpestrm(s1, 16, s2, 16, mode);
+            const __m128i mask_hi = _mm_unpackhi_epi64(result_mask, result_mask);
+            res += _mm_popcnt_u64(_mm_cvtsi128_si64(result_mask)) + _mm_popcnt_u64(_mm_cvtsi128_si64(mask_hi));
+        }
+#endif
+        for (; haystack != haystack_end && needle != needle_end; ++haystack, ++needle)
+            res += *haystack != *needle;
+
+        res = res + (haystack_end - haystack) + (needle_end - needle);
+        return res;
+    }
+};
+
+struct NameStringHammingDistance
+{
+    static constexpr auto name = "hammingDistance";
+};
+
+using FunctionStringHammingDistance = FunctionsStringSimilarity<StringHammingDistanceImpl, NameStringHammingDistance>;
+
+REGISTER_FUNCTION(StringHammingDistance)
+{
+    factory.registerFunction<FunctionStringHammingDistance>();
+}
+}
diff --git a/tests/queries/0_stateless/02884_string_hamming_distance.reference b/tests/queries/0_stateless/02884_string_hamming_distance.reference
new file mode 100644
index 00000000000..656cd356016
--- /dev/null
+++ b/tests/queries/0_stateless/02884_string_hamming_distance.reference
@@ -0,0 +1,10 @@
+0
+1
+7
+7
+3
+6
+3
+6
+6
+3
diff --git a/tests/queries/0_stateless/02884_string_hamming_distance.sql b/tests/queries/0_stateless/02884_string_hamming_distance.sql
new file mode 100644
index 00000000000..3f026d6f8d0
--- /dev/null
+++ b/tests/queries/0_stateless/02884_string_hamming_distance.sql
@@ -0,0 +1,17 @@
+select hammingDistance('abcd', 'abcd');
+drop table if exists t;
+create table t
+(
+	s1 String,
+	s2 String
+) engine = MergeTree order by s1;
+
+insert into t values ('abcdefg', 'abcdef') ('abcdefg', 'bcdefg') ('abcdefg', '');
+
+select hammingDistance(s1, s2) from t;
+
+select hammingDistance('abc', s2) from t;
+
+select hammingDistance(s2, 'def') from t;
+
+drop table t;

From 5a72de34ed26792d1a6a22ab2cb0287d99f40936 Mon Sep 17 00:00:00 2001
From: flynn <fenglv15@mails.ucas.ac.cn>
Date: Fri, 22 Sep 2023 17:48:17 +0000
Subject: [PATCH 16/37] add doc

---
 .../functions/string-search-functions.md      | 26 ++++++++++++++++++-
 1 file changed, 25 insertions(+), 1 deletion(-)

diff --git a/docs/en/sql-reference/functions/string-search-functions.md b/docs/en/sql-reference/functions/string-search-functions.md
index c10a1036677..cb4963c1091 100644
--- a/docs/en/sql-reference/functions/string-search-functions.md
+++ b/docs/en/sql-reference/functions/string-search-functions.md
@@ -680,4 +680,28 @@ Like [hasSubsequence](#hasSubsequence) but assumes `haystack` and `needle` are U
 
 ## hasSubsequenceCaseInsensitiveUTF8
 
-Like [hasSubsequenceUTF8](#hasSubsequenceUTF8) but searches case-insensitively.
\ No newline at end of file
+Like [hasSubsequenceUTF8](#hasSubsequenceUTF8) but searches case-insensitively.
+
+## hammingDistance
+
+Calculates the hamming distance between two strings.
+
+**Syntax**
+
+```sql
+hammingDistance(haystack, needle)
+```
+
+**Examples**
+
+``` sql
+SELECT hammingDistance('abc', 'ab') ;
+```
+
+Result:
+
+``` text
+┌─hammingDistance('abc', 'ab')─┐
+│                            1 │
+└──────────────────────────────┘
+```

From cbbabc4e1fa8715133a6f527c63da6c333879901 Mon Sep 17 00:00:00 2001
From: flynn <fenglv15@mails.ucas.ac.cn>
Date: Sat, 23 Sep 2023 04:11:20 +0000
Subject: [PATCH 17/37] fix

---
 .../functions/string-search-functions.md           | 14 +++++++-------
 ...HammingDistance.cpp => byteHammingDistance.cpp} | 11 ++++++-----
 2 files changed, 13 insertions(+), 12 deletions(-)
 rename src/Functions/{stringHammingDistance.cpp => byteHammingDistance.cpp} (90%)

diff --git a/docs/en/sql-reference/functions/string-search-functions.md b/docs/en/sql-reference/functions/string-search-functions.md
index cb4963c1091..c7e2bfa1cad 100644
--- a/docs/en/sql-reference/functions/string-search-functions.md
+++ b/docs/en/sql-reference/functions/string-search-functions.md
@@ -682,26 +682,26 @@ Like [hasSubsequence](#hasSubsequence) but assumes `haystack` and `needle` are U
 
 Like [hasSubsequenceUTF8](#hasSubsequenceUTF8) but searches case-insensitively.
 
-## hammingDistance
+## byteHammingDistance
 
-Calculates the hamming distance between two strings.
+Calculates the hamming distance between two byte strings.
 
 **Syntax**
 
 ```sql
-hammingDistance(haystack, needle)
+byteHammingDistance(haystack, needle)
 ```
 
 **Examples**
 
 ``` sql
-SELECT hammingDistance('abc', 'ab') ;
+SELECT byteHammingDistance('abc', 'ab') ;
 ```
 
 Result:
 
 ``` text
-┌─hammingDistance('abc', 'ab')─┐
-│                            1 │
-└──────────────────────────────┘
+┌─byteHammingDistance('abc', 'ab')─┐
+│                                1 │
+└──────────────────────────────────┘
 ```
diff --git a/src/Functions/stringHammingDistance.cpp b/src/Functions/byteHammingDistance.cpp
similarity index 90%
rename from src/Functions/stringHammingDistance.cpp
rename to src/Functions/byteHammingDistance.cpp
index fd22ca96ba3..80df4163c1c 100644
--- a/src/Functions/stringHammingDistance.cpp
+++ b/src/Functions/byteHammingDistance.cpp
@@ -11,7 +11,7 @@
 
 namespace DB
 {
-struct StringHammingDistanceImpl
+struct ByteHammingDistanceImpl
 {
     using ResultType = UInt64;
 
@@ -92,15 +92,16 @@ private:
     }
 };
 
-struct NameStringHammingDistance
+struct NameByteHammingDistance
 {
-    static constexpr auto name = "hammingDistance";
+    static constexpr auto name = "byteHammingDistance";
 };
 
-using FunctionStringHammingDistance = FunctionsStringSimilarity<StringHammingDistanceImpl, NameStringHammingDistance>;
+using FunctionByteHammingDistance = FunctionsStringSimilarity<ByteHammingDistanceImpl, NameByteHammingDistance>;
 
 REGISTER_FUNCTION(StringHammingDistance)
 {
-    factory.registerFunction<FunctionStringHammingDistance>();
+    factory.registerFunction<FunctionByteHammingDistance>(
+        FunctionDocumentation{.description = R"(Calculates the hamming distance between two bytes strings.)"});
 }
 }

From 53da7243ea362c8f3cfe211a49827d9606805709 Mon Sep 17 00:00:00 2001
From: flynn <fenglv15@mails.ucas.ac.cn>
Date: Sat, 23 Sep 2023 04:14:21 +0000
Subject: [PATCH 18/37] fix spell check

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

diff --git a/utils/check-style/aspell-ignore/en/aspell-dict.txt b/utils/check-style/aspell-ignore/en/aspell-dict.txt
index e5a031dabe3..ef3ad7d793c 100644
--- a/utils/check-style/aspell-ignore/en/aspell-dict.txt
+++ b/utils/check-style/aspell-ignore/en/aspell-dict.txt
@@ -1178,6 +1178,7 @@ builtins
 byteSize
 bytebase
 bytesToCutForIPv
+byteHammingDistance
 cLoki
 caConfig
 cacheSessions

From c1f4d911b8de9b7b025f0d97198bfde816fea74d Mon Sep 17 00:00:00 2001
From: flynn <fenglv15@mails.ucas.ac.cn>
Date: Sat, 23 Sep 2023 04:52:31 +0000
Subject: [PATCH 19/37] update test

---
 ...ce.reference => 02884_byte_hamming_distance.reference} | 0
 ...mming_distance.sql => 02884_byte_hamming_distance.sql} | 8 ++++----
 2 files changed, 4 insertions(+), 4 deletions(-)
 rename tests/queries/0_stateless/{02884_string_hamming_distance.reference => 02884_byte_hamming_distance.reference} (100%)
 rename tests/queries/0_stateless/{02884_string_hamming_distance.sql => 02884_byte_hamming_distance.sql} (52%)

diff --git a/tests/queries/0_stateless/02884_string_hamming_distance.reference b/tests/queries/0_stateless/02884_byte_hamming_distance.reference
similarity index 100%
rename from tests/queries/0_stateless/02884_string_hamming_distance.reference
rename to tests/queries/0_stateless/02884_byte_hamming_distance.reference
diff --git a/tests/queries/0_stateless/02884_string_hamming_distance.sql b/tests/queries/0_stateless/02884_byte_hamming_distance.sql
similarity index 52%
rename from tests/queries/0_stateless/02884_string_hamming_distance.sql
rename to tests/queries/0_stateless/02884_byte_hamming_distance.sql
index 3f026d6f8d0..62d4348449c 100644
--- a/tests/queries/0_stateless/02884_string_hamming_distance.sql
+++ b/tests/queries/0_stateless/02884_byte_hamming_distance.sql
@@ -1,4 +1,4 @@
-select hammingDistance('abcd', 'abcd');
+select byteHammingDistance('abcd', 'abcd');
 drop table if exists t;
 create table t
 (
@@ -8,10 +8,10 @@ create table t
 
 insert into t values ('abcdefg', 'abcdef') ('abcdefg', 'bcdefg') ('abcdefg', '');
 
-select hammingDistance(s1, s2) from t;
+select byteHammingDistance(s1, s2) from t;
 
-select hammingDistance('abc', s2) from t;
+select byteHammingDistance('abc', s2) from t;
 
-select hammingDistance(s2, 'def') from t;
+select byteHammingDistance(s2, 'def') from t;
 
 drop table t;

From 03b088a5710b6e458f2460bb067da8c43a5ca0b4 Mon Sep 17 00:00:00 2001
From: flynn <fenglv15@mails.ucas.ac.cn>
Date: Sat, 23 Sep 2023 17:17:35 +0000
Subject: [PATCH 20/37] add more functions

---
 .../functions/string-search-functions.md      |  53 +++++
 src/Functions/FunctionsStringDistance.cpp     | 218 ++++++++++++++++++
 src/Functions/byteHammingDistance.cpp         | 107 ---------
 .../02884_byte_hamming_distance.reference     |  10 -
 .../02884_string_distance_function.reference  |  37 +++
 ...sql => 02884_string_distance_function.sql} |  15 +-
 .../aspell-ignore/en/aspell-dict.txt          |   3 +
 7 files changed, 322 insertions(+), 121 deletions(-)
 create mode 100644 src/Functions/FunctionsStringDistance.cpp
 delete mode 100644 src/Functions/byteHammingDistance.cpp
 delete mode 100644 tests/queries/0_stateless/02884_byte_hamming_distance.reference
 create mode 100644 tests/queries/0_stateless/02884_string_distance_function.reference
 rename tests/queries/0_stateless/{02884_byte_hamming_distance.sql => 02884_string_distance_function.sql} (50%)

diff --git a/docs/en/sql-reference/functions/string-search-functions.md b/docs/en/sql-reference/functions/string-search-functions.md
index c7e2bfa1cad..e75c4fa8252 100644
--- a/docs/en/sql-reference/functions/string-search-functions.md
+++ b/docs/en/sql-reference/functions/string-search-functions.md
@@ -705,3 +705,56 @@ Result:
 │                                1 │
 └──────────────────────────────────┘
 ```
+
+- Alias: mismatches
+
+## byteJaccardIndex
+
+Calculates the jaccard similarity index between two byte strings.
+
+**Syntax**
+
+```sql
+byteJaccardIndex(haystack, needle)
+```
+
+**Examples**
+
+``` sql
+SELECT byteJaccardIndex('clickhouse', 'mouse');
+```
+
+Result:
+
+``` text
+┌─byteJaccardIndex('clickhouse', 'mouse')─┐
+│                                     0.4 │
+└─────────────────────────────────────────┘
+```
+
+## byteEditDistance
+
+Calcultes the edit distance between two byte strings.
+
+**Syntax**
+
+```sql
+byteEidtDistance(haystack, needle)
+```
+
+**Examples**
+
+``` sql
+SELECT byteEditDistance('clickhouse', 'mouse');
+```
+
+Result:
+
+``` text
+┌─byteEditDistance('clickhouse', 'mouse')─┐
+│                                       6 │
+└─────────────────────────────────────────┘
+```
+
+- Alias: byteLevenshteinDistance
+
diff --git a/src/Functions/FunctionsStringDistance.cpp b/src/Functions/FunctionsStringDistance.cpp
new file mode 100644
index 00000000000..73ff379aae0
--- /dev/null
+++ b/src/Functions/FunctionsStringDistance.cpp
@@ -0,0 +1,218 @@
+#include <Columns/ColumnString.h>
+#include <Columns/ColumnsNumber.h>
+#include <DataTypes/DataTypeString.h>
+#include <DataTypes/DataTypesNumber.h>
+#include <Functions/FunctionFactory.h>
+#include <Functions/FunctionsStringSimilarity.h>
+#include <Common/PODArray.h>
+
+#ifdef __SSE4_2__
+#    include <nmmintrin.h>
+#endif
+
+namespace DB
+{
+namespace ErrorCodes
+{
+extern const int TOO_LARGE_STRING_SIZE;
+}
+
+template <typename Op>
+struct FunctionDistanceImpl
+{
+    using ResultType = typename Op::ResultType;
+
+    static void constantConstant(const std::string & haystack, const std::string & needle, ResultType & res)
+    {
+        res = Op::process(haystack.data(), haystack.size(), needle.data(), needle.size());
+    }
+
+    static void vectorVector(
+        const ColumnString::Chars & haystack_data,
+        const ColumnString::Offsets & haystack_offsets,
+        const ColumnString::Chars & needle_data,
+        const ColumnString::Offsets & needle_offsets,
+        PaddedPODArray<ResultType> & res)
+    {
+        size_t size = res.size();
+        const char * haystack = reinterpret_cast<const char *>(haystack_data.data());
+        const char * needle = reinterpret_cast<const char *>(needle_data.data());
+        for (size_t i = 0; i < size; ++i)
+        {
+            res[i] = Op::process(
+                haystack + haystack_offsets[i - 1],
+                haystack_offsets[i] - haystack_offsets[i - 1] - 1,
+                needle + needle_offsets[i - 1],
+                needle_offsets[i] - needle_offsets[i - 1] - 1);
+        }
+    }
+
+    static void constantVector(
+        const std::string & haystack,
+        const ColumnString::Chars & needle_data,
+        const ColumnString::Offsets & needle_offsets,
+        PaddedPODArray<ResultType> & res)
+    {
+        const char * haystack_data = haystack.data();
+        size_t haystack_size = haystack.size();
+        const char * needle = reinterpret_cast<const char *>(needle_data.data());
+        size_t size = res.size();
+        for (size_t i = 0; i < size; ++i)
+        {
+            res[i]
+                = Op::process(haystack_data, haystack_size, needle + needle_offsets[i - 1], needle_offsets[i] - needle_offsets[i - 1] - 1);
+        }
+    }
+
+    static void vectorConstant(
+        const ColumnString::Chars & data,
+        const ColumnString::Offsets & offsets,
+        const std::string & needle,
+        PaddedPODArray<ResultType> & res)
+    {
+        constantVector(needle, data, offsets, res);
+    }
+
+};
+
+struct ByteHammingDistanceImpl
+{
+    using ResultType = UInt64;
+    static ResultType inline process(
+        const char * __restrict haystack, size_t haystack_size, const char * __restrict needle, size_t needle_size)
+    {
+        UInt64 res = 0;
+        const char * haystack_end = haystack + haystack_size;
+        const char * needle_end = needle + needle_size;
+
+#ifdef __SSE4_2__
+        static constexpr auto mode = _SIDD_UBYTE_OPS | _SIDD_CMP_EQUAL_EACH | _SIDD_NEGATIVE_POLARITY;
+
+        const char * haystack_end16 = haystack + haystack_size / 16 * 16;
+        const char * needle_end16 = needle + needle_size / 16 * 16;
+
+        for (; haystack < haystack_end16 && needle < needle_end16; haystack += 16, needle += 16)
+        {
+            __m128i s1 = _mm_loadu_si128(reinterpret_cast<const __m128i *>(haystack));
+            __m128i s2 = _mm_loadu_si128(reinterpret_cast<const __m128i *>(needle));
+            auto result_mask = _mm_cmpestrm(s1, 16, s2, 16, mode);
+            const __m128i mask_hi = _mm_unpackhi_epi64(result_mask, result_mask);
+            res += _mm_popcnt_u64(_mm_cvtsi128_si64(result_mask)) + _mm_popcnt_u64(_mm_cvtsi128_si64(mask_hi));
+        }
+#endif
+        for (; haystack != haystack_end && needle != needle_end; ++haystack, ++needle)
+            res += *haystack != *needle;
+
+        res = res + (haystack_end - haystack) + (needle_end - needle);
+        return res;
+    }
+};
+
+struct ByteJaccardIndexImpl
+{
+    using ResultType = Float64;
+    static ResultType inline process(
+        const char * __restrict haystack, size_t haystack_size, const char * __restrict needle, size_t needle_size)
+    {
+        if (haystack_size == 0 || needle_size == 0)
+            return 0;
+
+        std::unordered_set<char> haystack_set(haystack, haystack + haystack_size);
+        std::unordered_set<char> needle_set(needle, needle + needle_size);
+
+        size_t intersect = 0;
+        for (auto elem : haystack_set)
+        {
+            intersect += needle_set.contains(elem);
+        }
+        return static_cast<Float64>(intersect) / (haystack_set.size() + needle_set.size() - intersect);
+    }
+};
+
+struct ByteEditDistanceImpl
+{
+    using ResultType = UInt64;
+    static constexpr size_t max_string_size = 1u << 16;
+
+    static ResultType inline process(
+        const char * __restrict haystack, size_t haystack_size, const char * __restrict needle, size_t needle_size)
+    {
+        if (haystack_size == 0 || needle_size == 0)
+            return haystack_size + needle_size;
+
+        /// Safety threshold against DoS, since we use two array to calculate the distance.
+        if (haystack_size > max_string_size || needle_size > max_string_size)
+            throw Exception(
+                ErrorCodes::TOO_LARGE_STRING_SIZE,
+                "The string size is too big for function byteEditDistance. "
+                "Should be at most {}",
+                max_string_size);
+
+        PaddedPODArray<ResultType> distances0(haystack_size + 1, 0);
+        PaddedPODArray<ResultType> distances1(haystack_size + 1, 0);
+
+        ResultType substitution = 0;
+        ResultType insertion = 0;
+        ResultType deletion = 0;
+
+        for (size_t i = 0; i <= haystack_size; ++i)
+            distances0[i] = i;
+
+        for (size_t pos_needle = 0; pos_needle < needle_size; ++pos_needle)
+        {
+            distances1[0] = pos_needle + 1;
+
+            for (size_t pos_haystack = 0; pos_haystack < haystack_size; pos_haystack++)
+            {
+                deletion = distances0[pos_haystack + 1] + 1;
+                insertion = distances1[pos_haystack] + 1;
+                substitution = distances0[pos_haystack];
+
+                if (*(needle + pos_needle) != *(haystack + pos_haystack))
+                    substitution += 1;
+
+                distances1[pos_haystack + 1] = std::min(deletion, std::min(substitution, insertion));
+            }
+            distances0.swap(distances1);
+        }
+
+        return distances0[haystack_size];
+    }
+};
+
+struct NameByteHammingDistance
+{
+    static constexpr auto name = "byteHammingDistance";
+};
+
+struct NameByteJaccardIndex
+{
+    static constexpr auto name = "byteJaccardIndex";
+};
+
+struct NameByteEditDistance
+{
+    static constexpr auto name = "byteEditDistance";
+};
+
+using FunctionByteHammingDistance
+    = FunctionsStringSimilarity<FunctionDistanceImpl<ByteHammingDistanceImpl>, NameByteHammingDistance>;
+
+using FunctionByteJaccardIndex = FunctionsStringSimilarity<FunctionDistanceImpl<ByteJaccardIndexImpl>, NameByteJaccardIndex>;
+
+using FunctionByteEditDistance = FunctionsStringSimilarity<FunctionDistanceImpl<ByteEditDistanceImpl>, NameByteEditDistance>;
+
+REGISTER_FUNCTION(StringHammingDistance)
+{
+    factory.registerFunction<FunctionByteHammingDistance>(
+        FunctionDocumentation{.description = R"(Calculates the hamming distance between two bytes strings.)"});
+    factory.registerAlias("mismatches", NameByteHammingDistance::name);
+
+    factory.registerFunction<FunctionByteJaccardIndex>(
+        FunctionDocumentation{.description = R"(Calculates the jaccard similarity index between two bytes strings.)"});
+
+    factory.registerFunction<FunctionByteEditDistance>(
+        FunctionDocumentation{.description = R"(Calculates the edit distance between two bytes strings.)"});
+    factory.registerAlias("byteLevenshteinDistance", NameByteEditDistance::name);
+}
+}
diff --git a/src/Functions/byteHammingDistance.cpp b/src/Functions/byteHammingDistance.cpp
deleted file mode 100644
index 80df4163c1c..00000000000
--- a/src/Functions/byteHammingDistance.cpp
+++ /dev/null
@@ -1,107 +0,0 @@
-#include <Columns/ColumnString.h>
-#include <Columns/ColumnsNumber.h>
-#include <DataTypes/DataTypeString.h>
-#include <DataTypes/DataTypesNumber.h>
-#include <Functions/FunctionFactory.h>
-#include <Functions/FunctionsStringSimilarity.h>
-
-#ifdef __SSE4_2__
-#    include <nmmintrin.h>
-#endif
-
-namespace DB
-{
-struct ByteHammingDistanceImpl
-{
-    using ResultType = UInt64;
-
-    static void constantConstant(const std::string & haystack, const std::string & needle, UInt64 & res)
-    {
-        res = process(haystack.data(), haystack.size(), needle.data(), needle.size());
-    }
-
-    static void vectorVector(
-        const ColumnString::Chars & haystack_data,
-        const ColumnString::Offsets & haystack_offsets,
-        const ColumnString::Chars & needle_data,
-        const ColumnString::Offsets & needle_offsets,
-        PaddedPODArray<UInt64> & res)
-    {
-        size_t size = res.size();
-        const char * haystack = reinterpret_cast<const char *>(haystack_data.data());
-        const char * needle = reinterpret_cast<const char *>(needle_data.data());
-        for (size_t i = 0; i < size; ++i)
-        {
-            res[i] = process(
-                haystack + haystack_offsets[i - 1],
-                haystack_offsets[i] - haystack_offsets[i - 1] - 1,
-                needle + needle_offsets[i - 1],
-                needle_offsets[i] - needle_offsets[i - 1] - 1);
-        }
-    }
-
-    static void constantVector(
-        const std::string & haystack,
-        const ColumnString::Chars & needle_data,
-        const ColumnString::Offsets & needle_offsets,
-        PaddedPODArray<UInt64> & res)
-    {
-        const char * haystack_data = haystack.data();
-        size_t haystack_size = haystack.size();
-        const char * needle = reinterpret_cast<const char *>(needle_data.data());
-        size_t size = res.size();
-        for (size_t i = 0; i < size; ++i)
-        {
-            res[i] = process(haystack_data, haystack_size, needle + needle_offsets[i - 1], needle_offsets[i] - needle_offsets[i - 1] - 1);
-        }
-    }
-
-    static void vectorConstant(
-        const ColumnString::Chars & data, const ColumnString::Offsets & offsets, const std::string & needle, PaddedPODArray<UInt64> & res)
-    {
-        constantVector(needle, data, offsets, res);
-    }
-
-private:
-    static UInt64 inline process(const char * haystack, size_t haystack_size, const char * needle, size_t needle_size)
-    {
-        UInt64 res = 0;
-        const char * haystack_end = haystack + haystack_size;
-        const char * needle_end = needle + needle_size;
-
-#ifdef __SSE4_2__
-        static constexpr auto mode = _SIDD_UBYTE_OPS | _SIDD_CMP_EQUAL_EACH | _SIDD_NEGATIVE_POLARITY;
-
-        const char * haystack_end16 = haystack + haystack_size / 16 * 16;
-        const char * needle_end16 = needle + needle_size / 16 * 16;
-
-        for (; haystack < haystack_end16 && needle < needle_end16; haystack += 16, needle += 16)
-        {
-            __m128i s1 = _mm_loadu_si128(reinterpret_cast<const __m128i *>(haystack));
-            __m128i s2 = _mm_loadu_si128(reinterpret_cast<const __m128i *>(needle));
-            auto result_mask = _mm_cmpestrm(s1, 16, s2, 16, mode);
-            const __m128i mask_hi = _mm_unpackhi_epi64(result_mask, result_mask);
-            res += _mm_popcnt_u64(_mm_cvtsi128_si64(result_mask)) + _mm_popcnt_u64(_mm_cvtsi128_si64(mask_hi));
-        }
-#endif
-        for (; haystack != haystack_end && needle != needle_end; ++haystack, ++needle)
-            res += *haystack != *needle;
-
-        res = res + (haystack_end - haystack) + (needle_end - needle);
-        return res;
-    }
-};
-
-struct NameByteHammingDistance
-{
-    static constexpr auto name = "byteHammingDistance";
-};
-
-using FunctionByteHammingDistance = FunctionsStringSimilarity<ByteHammingDistanceImpl, NameByteHammingDistance>;
-
-REGISTER_FUNCTION(StringHammingDistance)
-{
-    factory.registerFunction<FunctionByteHammingDistance>(
-        FunctionDocumentation{.description = R"(Calculates the hamming distance between two bytes strings.)"});
-}
-}
diff --git a/tests/queries/0_stateless/02884_byte_hamming_distance.reference b/tests/queries/0_stateless/02884_byte_hamming_distance.reference
deleted file mode 100644
index 656cd356016..00000000000
--- a/tests/queries/0_stateless/02884_byte_hamming_distance.reference
+++ /dev/null
@@ -1,10 +0,0 @@
-0
-1
-7
-7
-3
-6
-3
-6
-6
-3
diff --git a/tests/queries/0_stateless/02884_string_distance_function.reference b/tests/queries/0_stateless/02884_string_distance_function.reference
new file mode 100644
index 00000000000..d50ffa0a48e
--- /dev/null
+++ b/tests/queries/0_stateless/02884_string_distance_function.reference
@@ -0,0 +1,37 @@
+0
+1
+7
+7
+10
+3
+6
+3
+10
+6
+6
+3
+10
+1
+7
+7
+10
+3
+6
+3
+10
+6
+6
+3
+10
+0.8571428571428571
+0.8571428571428571
+0
+0.4
+1
+1
+7
+6
+1
+1
+7
+6
diff --git a/tests/queries/0_stateless/02884_byte_hamming_distance.sql b/tests/queries/0_stateless/02884_string_distance_function.sql
similarity index 50%
rename from tests/queries/0_stateless/02884_byte_hamming_distance.sql
rename to tests/queries/0_stateless/02884_string_distance_function.sql
index 62d4348449c..bab22f24bf2 100644
--- a/tests/queries/0_stateless/02884_byte_hamming_distance.sql
+++ b/tests/queries/0_stateless/02884_string_distance_function.sql
@@ -6,12 +6,19 @@ create table t
 	s2 String
 ) engine = MergeTree order by s1;
 
-insert into t values ('abcdefg', 'abcdef') ('abcdefg', 'bcdefg') ('abcdefg', '');
-
+insert into t values ('abcdefg', 'abcdef') ('abcdefg', 'bcdefg') ('abcdefg', '') ('mouse', 'clickhouse');
 select byteHammingDistance(s1, s2) from t;
-
 select byteHammingDistance('abc', s2) from t;
-
 select byteHammingDistance(s2, 'def') from t;
 
+select mismatches(s1, s2) from t;
+select mismatches('abc', s2) from t;
+select mismatches(s2, 'def') from t;
+
+select byteJaccardIndex(s1, s2) from t;
+select byteEditDistance(s1, s2) from t;
+select byteLevenshteinDistance(s1, s2) from t;
+
+SELECT byteEditDistance(randomString(power(2, 17)), 'abc'); -- { serverError 131 }
+
 drop table t;
diff --git a/utils/check-style/aspell-ignore/en/aspell-dict.txt b/utils/check-style/aspell-ignore/en/aspell-dict.txt
index ef3ad7d793c..560b9130e10 100644
--- a/utils/check-style/aspell-ignore/en/aspell-dict.txt
+++ b/utils/check-style/aspell-ignore/en/aspell-dict.txt
@@ -1178,7 +1178,10 @@ builtins
 byteSize
 bytebase
 bytesToCutForIPv
+byteEditDistance
 byteHammingDistance
+byteJaccardIndex
+byteLevenshteinDistance
 cLoki
 caConfig
 cacheSessions

From 9b40ca9d5fe7a8ab2d47f10ea4d66c55ec24e35b Mon Sep 17 00:00:00 2001
From: flynn <fenglv15@mails.ucas.ac.cn>
Date: Sat, 23 Sep 2023 17:50:02 +0000
Subject: [PATCH 21/37] fix spell

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

diff --git a/docs/en/sql-reference/functions/string-search-functions.md b/docs/en/sql-reference/functions/string-search-functions.md
index e75c4fa8252..8a2e257ed34 100644
--- a/docs/en/sql-reference/functions/string-search-functions.md
+++ b/docs/en/sql-reference/functions/string-search-functions.md
@@ -734,7 +734,7 @@ Result:
 
 ## byteEditDistance
 
-Calcultes the edit distance between two byte strings.
+Calculates the edit distance between two byte strings.
 
 **Syntax**
 

From 601f904275834f6750638142083c4d9e531a4348 Mon Sep 17 00:00:00 2001
From: avogar <avogar@clickhouse.com>
Date: Wed, 20 Sep 2023 16:00:28 +0000
Subject: [PATCH 22/37] Move obsolete format settings to separate section

---
 src/Core/Settings.cpp                    |  2 +-
 src/Core/Settings.h                      | 26 ++++++++++++++++--------
 src/Storages/FileLog/FileLogSettings.h   |  2 +-
 src/Storages/Hive/HiveSettings.h         |  2 +-
 src/Storages/Kafka/KafkaSettings.h       |  2 +-
 src/Storages/NATS/NATSSettings.h         |  2 +-
 src/Storages/RabbitMQ/RabbitMQSettings.h |  2 +-
 src/Storages/S3Queue/S3QueueSettings.h   |  2 +-
 src/Storages/SetSettings.h               |  2 +-
 9 files changed, 25 insertions(+), 17 deletions(-)

diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp
index ed0868eaa4d..33db7660abd 100644
--- a/src/Core/Settings.cpp
+++ b/src/Core/Settings.cpp
@@ -158,6 +158,6 @@ void Settings::applyCompatibilitySetting(const String & compatibility_value)
     }
 }
 
-IMPLEMENT_SETTINGS_TRAITS(FormatFactorySettingsTraits, FORMAT_FACTORY_SETTINGS)
+IMPLEMENT_SETTINGS_TRAITS(FormatFactorySettingsTraits, LIST_OF_ALL_FORMAT_SETTINGS)
 
 }
diff --git a/src/Core/Settings.h b/src/Core/Settings.h
index f2b55fbcd0b..8338e5c7d59 100644
--- a/src/Core/Settings.h
+++ b/src/Core/Settings.h
@@ -806,7 +806,7 @@ class IColumn;
     M(Bool, create_index_ignore_unique, false, "Ignore UNIQUE keyword in CREATE UNIQUE INDEX. Made for SQL compatibility tests.", 0)                                                                                                  \
 
 // End of COMMON_SETTINGS
-    // Please add settings related to formats into the FORMAT_FACTORY_SETTINGS and move obsolete settings to OBSOLETE_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.
 
 #define MAKE_OBSOLETE(M, TYPE, NAME, DEFAULT) \
     M(TYPE, NAME, DEFAULT, "Obsolete setting, does nothing.", BaseSettingsHelpers::Flags::OBSOLETE)
@@ -859,9 +859,6 @@ class IColumn;
     MAKE_OBSOLETE(M, Seconds, drain_timeout, 3) \
     MAKE_OBSOLETE(M, UInt64, backup_threads, 16) \
     MAKE_OBSOLETE(M, UInt64, restore_threads, 16) \
-    MAKE_OBSOLETE(M, Bool, input_format_arrow_import_nested, false) \
-    MAKE_OBSOLETE(M, Bool, input_format_parquet_import_nested, false) \
-    MAKE_OBSOLETE(M, Bool, input_format_orc_import_nested, false) \
     MAKE_OBSOLETE(M, Bool, optimize_duplicate_order_by_and_distinct, false) \
 
     /** The section above is for obsolete settings. Do not add anything there. */
@@ -1068,10 +1065,17 @@ class IColumn;
 // End of FORMAT_FACTORY_SETTINGS
 // Please add settings non-related to formats into the COMMON_SETTINGS above.
 
-#define LIST_OF_SETTINGS(M, ALIAS)    \
-    COMMON_SETTINGS(M, ALIAS)         \
-    OBSOLETE_SETTINGS(M, ALIAS)       \
-    FORMAT_FACTORY_SETTINGS(M, ALIAS)
+#define OBSOLETE_FORMAT_SETTINGS(M, ALIAS) \
+    /** Obsolete format settings that do nothing but left for compatibility reasons. Remove each one after half a year of obsolescence. */ \
+    MAKE_OBSOLETE(M, Bool, input_format_arrow_import_nested, false) \
+    MAKE_OBSOLETE(M, Bool, input_format_parquet_import_nested, false) \
+    MAKE_OBSOLETE(M, Bool, input_format_orc_import_nested, false) \
+
+#define LIST_OF_SETTINGS(M, ALIAS)     \
+    COMMON_SETTINGS(M, ALIAS)          \
+    OBSOLETE_SETTINGS(M, ALIAS)        \
+    FORMAT_FACTORY_SETTINGS(M, ALIAS)  \
+    OBSOLETE_FORMAT_SETTINGS(M, ALIAS) \
 
 DECLARE_SETTINGS_TRAITS_ALLOW_CUSTOM_SETTINGS(SettingsTraits, LIST_OF_SETTINGS)
 
@@ -1110,10 +1114,14 @@ private:
     std::unordered_set<std::string_view> settings_changed_by_compatibility_setting;
 };
 
+#define LIST_OF_ALL_FORMAT_SETTINGS(M, ALIAS) \
+    FORMAT_FACTORY_SETTINGS(M, ALIAS)         \
+    OBSOLETE_FORMAT_SETTINGS(M, ALIAS)        \
+
 /*
  * User-specified file format settings for File and URL engines.
  */
-DECLARE_SETTINGS_TRAITS(FormatFactorySettingsTraits, FORMAT_FACTORY_SETTINGS)
+DECLARE_SETTINGS_TRAITS(FormatFactorySettingsTraits, LIST_OF_ALL_FORMAT_SETTINGS)
 
 struct FormatFactorySettings : public BaseSettings<FormatFactorySettingsTraits>
 {
diff --git a/src/Storages/FileLog/FileLogSettings.h b/src/Storages/FileLog/FileLogSettings.h
index 00c774dba08..4f3157424db 100644
--- a/src/Storages/FileLog/FileLogSettings.h
+++ b/src/Storages/FileLog/FileLogSettings.h
@@ -21,7 +21,7 @@ class ASTStorage;
 
 #define LIST_OF_FILELOG_SETTINGS(M, ALIAS) \
     FILELOG_RELATED_SETTINGS(M, ALIAS) \
-    FORMAT_FACTORY_SETTINGS(M, ALIAS)
+    LIST_OF_ALL_FORMAT_SETTINGS(M, ALIAS)
 
 DECLARE_SETTINGS_TRAITS(FileLogSettingsTraits, LIST_OF_FILELOG_SETTINGS)
 
diff --git a/src/Storages/Hive/HiveSettings.h b/src/Storages/Hive/HiveSettings.h
index aaaa3306cf7..cbef19b3fa7 100644
--- a/src/Storages/Hive/HiveSettings.h
+++ b/src/Storages/Hive/HiveSettings.h
@@ -20,7 +20,7 @@ class ASTStorage;
 
 #define LIST_OF_HIVE_SETTINGS(M, ALIAS) \
     HIVE_RELATED_SETTINGS(M, ALIAS) \
-    FORMAT_FACTORY_SETTINGS(M, ALIAS)
+    LIST_OF_ALL_FORMAT_SETTINGS(M, ALIAS)
 
 DECLARE_SETTINGS_TRAITS(HiveSettingsTraits, LIST_OF_HIVE_SETTINGS)
 
diff --git a/src/Storages/Kafka/KafkaSettings.h b/src/Storages/Kafka/KafkaSettings.h
index 03c415b80e1..cfe9f16c019 100644
--- a/src/Storages/Kafka/KafkaSettings.h
+++ b/src/Storages/Kafka/KafkaSettings.h
@@ -39,7 +39,7 @@ class ASTStorage;
 
 #define LIST_OF_KAFKA_SETTINGS(M, ALIAS) \
     KAFKA_RELATED_SETTINGS(M, ALIAS) \
-    FORMAT_FACTORY_SETTINGS(M, ALIAS)
+    LIST_OF_ALL_FORMAT_SETTINGS(M, ALIAS)
 
 DECLARE_SETTINGS_TRAITS(KafkaSettingsTraits, LIST_OF_KAFKA_SETTINGS)
 
diff --git a/src/Storages/NATS/NATSSettings.h b/src/Storages/NATS/NATSSettings.h
index e82f351dd42..2482f811e50 100644
--- a/src/Storages/NATS/NATSSettings.h
+++ b/src/Storages/NATS/NATSSettings.h
@@ -30,7 +30,7 @@ class ASTStorage;
 
 #define LIST_OF_NATS_SETTINGS(M, ALIAS) \
     NATS_RELATED_SETTINGS(M, ALIAS) \
-    FORMAT_FACTORY_SETTINGS(M, ALIAS)
+    LIST_OF_ALL_FORMAT_SETTINGS(M, ALIAS)
 
 DECLARE_SETTINGS_TRAITS(NATSSettingsTraits, LIST_OF_NATS_SETTINGS)
 
diff --git a/src/Storages/RabbitMQ/RabbitMQSettings.h b/src/Storages/RabbitMQ/RabbitMQSettings.h
index 58d80c6ec9d..78cc8bf4df1 100644
--- a/src/Storages/RabbitMQ/RabbitMQSettings.h
+++ b/src/Storages/RabbitMQ/RabbitMQSettings.h
@@ -38,7 +38,7 @@ namespace DB
 
 #define LIST_OF_RABBITMQ_SETTINGS(M, ALIAS) \
     RABBITMQ_RELATED_SETTINGS(M, ALIAS) \
-    FORMAT_FACTORY_SETTINGS(M, ALIAS)
+    LIST_OF_ALL_FORMAT_SETTINGS(M, ALIAS)
 
 DECLARE_SETTINGS_TRAITS(RabbitMQSettingsTraits, LIST_OF_RABBITMQ_SETTINGS)
 
diff --git a/src/Storages/S3Queue/S3QueueSettings.h b/src/Storages/S3Queue/S3QueueSettings.h
index 75defc4a57f..328980f46d8 100644
--- a/src/Storages/S3Queue/S3QueueSettings.h
+++ b/src/Storages/S3Queue/S3QueueSettings.h
@@ -33,7 +33,7 @@ class ASTStorage;
 
 #define LIST_OF_S3QUEUE_SETTINGS(M, ALIAS) \
     S3QUEUE_RELATED_SETTINGS(M, ALIAS) \
-    FORMAT_FACTORY_SETTINGS(M, ALIAS)
+    LIST_OF_ALL_FORMAT_SETTINGS(M, ALIAS)
 
 DECLARE_SETTINGS_TRAITS(S3QueueSettingsTraits, LIST_OF_S3QUEUE_SETTINGS)
 
diff --git a/src/Storages/SetSettings.h b/src/Storages/SetSettings.h
index cc1cc642f91..98c1f3d97f5 100644
--- a/src/Storages/SetSettings.h
+++ b/src/Storages/SetSettings.h
@@ -15,7 +15,7 @@ class ASTStorage;
 
 #define LIST_OF_SET_SETTINGS(M, ALIAS) \
     SET_RELATED_SETTINGS(M, ALIAS) \
-    FORMAT_FACTORY_SETTINGS(M, ALIAS)
+    LIST_OF_ALL_FORMAT_SETTINGS(M, ALIAS)
 
 DECLARE_SETTINGS_TRAITS(setSettingsTraits, LIST_OF_SET_SETTINGS)
 

From 5b4033e829d5e5d56b9e70e40d41617513cbb064 Mon Sep 17 00:00:00 2001
From: Tyler Hannan <tyler@clickhouse.com>
Date: Mon, 25 Sep 2023 17:11:20 +0200
Subject: [PATCH 23/37] Update README.md

Updating README
---
 README.md | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/README.md b/README.md
index 6d253342a15..6736f5c67f3 100644
--- a/README.md
+++ b/README.md
@@ -22,9 +22,11 @@ curl https://clickhouse.com/ | sh
 
 ## Upcoming Events
 
-* [**v23.8 Community Call**](https://clickhouse.com/company/events/v23-8-community-release-call?utm_source=github&utm_medium=social&utm_campaign=release-webinar-2023-08) - Aug 31 - 23.8 is rapidly approaching. Original creator, co-founder, and CTO of ClickHouse Alexey Milovidov will walk us through the highlights of the release.
-* [**ClickHouse & AI - A Meetup in San Francisco**](https://www.meetup.com/clickhouse-silicon-valley-meetup-group/events/294472987) - Aug 8
-* [**ClickHouse Meetup in Paris**](https://www.meetup.com/clickhouse-france-user-group/events/294283460) - Sep 12
+* [**v23.9 Community Call**]([https://clickhouse.com/company/events/v23-8-community-release-call](https://clickhouse.com/company/events/v23-9-community-release-call)?utm_source=github&utm_medium=social&utm_campaign=release-webinar-2023-08) - Sep 28 - 23.9 is rapidly approaching. Original creator, co-founder, and CTO of ClickHouse Alexey Milovidov will walk us through the highlights of the release.
+* [**ClickHouse Meetup in Amsterdam**](https://www.meetup.com/clickhouse-netherlands-user-group/events/296334590/)) - Oct 31
+* [**ClickHouse Meetup in Beijing**](https://www.meetup.com/clickhouse-beijing-user-group/events/296334856/) - Nov 4
+* [**ClickHouse Meetup in San Francisco**](https://www.meetup.com/clickhouse-silicon-valley-meetup-group/events/296334923/) - Nov 14
+* [**ClickHouse Meetup in Singapore**](https://www.meetup.com/clickhouse-singapore-meetup-group/events/296334976/) - Nov 15
 
 Also, keep an eye out for upcoming meetups around the world. Somewhere else you want us to be? Please feel free to reach out to tyler <at> clickhouse <dot> com.
 

From 098efb261e38bfe8c46962f7a24292e4e3b77f5b Mon Sep 17 00:00:00 2001
From: Antonio Andelic <antonio@clickhouse.com>
Date: Tue, 26 Sep 2023 09:03:38 +0000
Subject: [PATCH 24/37] Throw proper exception if promise was not set

---
 .../Transforms/CreatingSetsTransform.cpp      | 23 +++++++++++++++++--
 1 file changed, 21 insertions(+), 2 deletions(-)

diff --git a/src/Processors/Transforms/CreatingSetsTransform.cpp b/src/Processors/Transforms/CreatingSetsTransform.cpp
index b2e459ee487..bc5da769dd6 100644
--- a/src/Processors/Transforms/CreatingSetsTransform.cpp
+++ b/src/Processors/Transforms/CreatingSetsTransform.cpp
@@ -17,9 +17,25 @@ namespace ErrorCodes
 {
     extern const int LOGICAL_ERROR;
     extern const int SET_SIZE_LIMIT_EXCEEDED;
+    extern const int UNKNOWN_EXCEPTION;
 }
 
-CreatingSetsTransform::~CreatingSetsTransform() = default;
+CreatingSetsTransform::~CreatingSetsTransform()
+{
+    if (promise_to_build)
+    {
+        /// set_exception can also throw
+        try
+        {
+            promise_to_build->set_exception(std::make_exception_ptr(
+                Exception(ErrorCodes::UNKNOWN_EXCEPTION, "Failed to build set, most likely pipeline executor was stopped")));
+        }
+        catch (...)
+        {
+            tryLogCurrentException(log, "Failed to set_exception for promise");
+        }
+    }
+}
 
 CreatingSetsTransform::CreatingSetsTransform(
     Block in_header_,
@@ -86,7 +102,7 @@ void CreatingSetsTransform::startSubquery()
             }
             else
             {
-                LOG_TRACE(log, "Waiting for set to be build by another thread, key: {}", set_and_key->key);
+                LOG_TRACE(log, "Waiting for set to be built by another thread, key: {}", set_and_key->key);
                 SharedSet set_built_by_another_thread = std::move(std::get<1>(from_cache));
                 const SetPtr & ready_set = set_built_by_another_thread.get();
                 if (!ready_set)
@@ -188,7 +204,10 @@ Chunk CreatingSetsTransform::generate()
     {
         set_and_key->set->finishInsert();
         if (promise_to_build)
+        {
             promise_to_build->set_value(set_and_key->set);
+            promise_to_build.reset();
+        }
     }
 
     if (table_out.initialized())

From d5de17f6aa1bde2823df4e329a1d516b7f807833 Mon Sep 17 00:00:00 2001
From: avogar <avogar@clickhouse.com>
Date: Tue, 26 Sep 2023 12:19:25 +0000
Subject: [PATCH 25/37] Fix bad cast to ColumnVector<Int128> in function if

---
 src/Functions/if.cpp                          | 19 ++++++++-------
 ...er_type_inference_in_if_function.reference | 24 +++++++++++++++++++
 ..._integer_type_inference_in_if_function.sql | 13 ++++++++++
 3 files changed, 48 insertions(+), 8 deletions(-)
 create mode 100644 tests/queries/0_stateless/02888_integer_type_inference_in_if_function.reference
 create mode 100644 tests/queries/0_stateless/02888_integer_type_inference_in_if_function.sql

diff --git a/src/Functions/if.cpp b/src/Functions/if.cpp
index a955230f3d3..eba1733c683 100644
--- a/src/Functions/if.cpp
+++ b/src/Functions/if.cpp
@@ -1096,22 +1096,25 @@ public:
             return res != nullptr;
         };
 
-        TypeIndex left_id = arg_then.type->getTypeId();
-        TypeIndex right_id = arg_else.type->getTypeId();
+        DataTypePtr left_type = arg_then.type;
+        DataTypePtr right_type = arg_else.type;
 
         if (const auto * left_array = checkAndGetDataType<DataTypeArray>(arg_then.type.get()))
-            left_id = left_array->getNestedType()->getTypeId();
+            left_type = left_array->getNestedType();
 
         if (const auto * right_array = checkAndGetDataType<DataTypeArray>(arg_else.type.get()))
-            right_id = right_array->getNestedType()->getTypeId();
+            right_type = right_array->getNestedType();
 
         /// Special case when one column is Integer and another is UInt64 that can be actually Int64.
         /// The result type for this case is Int64 and we need to change UInt64 type to Int64
         /// so the NumberTraits::ResultOfIf will return Int64 instead if Int128.
-        if (isNativeInteger(arg_then.type) && isUInt64ThatCanBeInt64(arg_else.type))
-            right_id = TypeIndex::Int64;
-        else if (isNativeInteger(arg_else.type) && isUInt64ThatCanBeInt64(arg_then.type))
-            left_id = TypeIndex::Int64;
+        if (isNativeInteger(left_type) && isUInt64ThatCanBeInt64(right_type))
+            right_type = std::make_shared<DataTypeInt64>();
+        else if (isNativeInteger(right_type) && isUInt64ThatCanBeInt64(left_type))
+            left_type = std::make_shared<DataTypeInt64>();
+
+        TypeIndex left_id = left_type->getTypeId();
+        TypeIndex right_id = right_type->getTypeId();
 
         if (!(callOnBasicTypes<true, true, true, false>(left_id, right_id, call)
             || (res = executeTyped<UUID, UUID>(cond_col, arguments, result_type, input_rows_count))
diff --git a/tests/queries/0_stateless/02888_integer_type_inference_in_if_function.reference b/tests/queries/0_stateless/02888_integer_type_inference_in_if_function.reference
new file mode 100644
index 00000000000..b5909a82df3
--- /dev/null
+++ b/tests/queries/0_stateless/02888_integer_type_inference_in_if_function.reference
@@ -0,0 +1,24 @@
+-9223372036854775808
+9223372036854775806
+-9223372036854775808
+9223372036854775806
+-9223372036854775808
+9223372036854775806
+-9223372036854775808
+9223372036854775806
+[2,65537,-9223372036854775808]
+[9223372036854775806]
+[2,65537,-9223372036854775808]
+[9223372036854775806]
+[2,65537,-9223372036854775808]
+[9223372036854775806]
+[2,65537,-9223372036854775808]
+[9223372036854775806]
+[[2,65537,-9223372036854775808]]
+[[9223372036854775806]]
+[[2,65537,-9223372036854775808]]
+[[9223372036854775806]]
+[[2,65537,-9223372036854775808]]
+[[9223372036854775806]]
+[[2,65537,-9223372036854775808]]
+[[9223372036854775806]]
diff --git a/tests/queries/0_stateless/02888_integer_type_inference_in_if_function.sql b/tests/queries/0_stateless/02888_integer_type_inference_in_if_function.sql
new file mode 100644
index 00000000000..065536e0e6f
--- /dev/null
+++ b/tests/queries/0_stateless/02888_integer_type_inference_in_if_function.sql
@@ -0,0 +1,13 @@
+SELECT if(number % 2, 9223372036854775806, -9223372036854775808) AS res FROM numbers(2);
+SELECT if(number % 2, materialize(9223372036854775806), -9223372036854775808) AS res FROM numbers(2);
+SELECT if(number % 2, 9223372036854775806, materialize(-9223372036854775808)) AS res FROM numbers(2);
+SELECT if(number % 2, materialize(9223372036854775806), materialize(-9223372036854775808)) AS res FROM numbers(2);
+SELECT if(number % 2, [9223372036854775806], [2, 65537, -9223372036854775808]) AS res FROM numbers(2);
+SELECT if(number % 2, materialize([9223372036854775806]), [2, 65537, -9223372036854775808]) AS res FROM numbers(2);
+SELECT if(number % 2, [9223372036854775806], materialize([2, 65537, -9223372036854775808])) AS res FROM numbers(2);
+SELECT if(number % 2, materialize([9223372036854775806]), materialize([2, 65537, -9223372036854775808])) AS res FROM numbers(2);
+SELECT if(number % 2, [[9223372036854775806]], [[2, 65537, -9223372036854775808]]) AS res FROM numbers(2);
+SELECT if(number % 2, materialize([[9223372036854775806]]), [[2, 65537, -9223372036854775808]]) AS res FROM numbers(2);
+SELECT if(number % 2, [[9223372036854775806]], materialize([[2, 65537, -9223372036854775808]])) AS res FROM numbers(2);
+SELECT if(number % 2, materialize([[9223372036854775806]]), materialize([[2, 65537, -9223372036854775808]])) AS res FROM numbers(2);
+

From d39bd7154fe304409889e0c51aa24113f24269e3 Mon Sep 17 00:00:00 2001
From: Vitaly Baranov <vitlibar@clickhouse.com>
Date: Sat, 23 Sep 2023 21:13:13 +0200
Subject: [PATCH 26/37] Use pregenerated gRPC protocol pb2 files to fix test
 flakyness.

---
 tests/integration/runner                      |  26 +--
 tests/integration/test_grpc_protocol/pb2      |   1 +
 .../protos/clickhouse_grpc.proto              |   1 -
 tests/integration/test_grpc_protocol/test.py  |  28 +--
 tests/integration/test_grpc_protocol_ssl/pb2  |   1 +
 .../protos/clickhouse_grpc.proto              |   1 -
 .../test_grpc_protocol_ssl/test.py            |  28 +--
 .../grpc_protocol_pb2                         |   1 +
 .../protos/clickhouse_grpc.proto              |   1 -
 .../test.py                                   |  22 +--
 .../test_server_reload/grpc_protocol_pb2      |   1 +
 .../protos/clickhouse_grpc.proto              |   1 -
 tests/integration/test_server_reload/test.py  |  23 +--
 .../test_session_log/grpc_protocol_pb2        |   1 +
 .../protos/clickhouse_grpc.proto              |   1 -
 tests/integration/test_session_log/test.py    |  22 +--
 utils/check-style/check-black                 |  12 +-
 utils/grpc-client/clickhouse-grpc-client.py   |  50 ++----
 utils/grpc-client/pb2/clickhouse_grpc_pb2.py  | 163 +++++++++++++++++
 .../pb2/clickhouse_grpc_pb2_grpc.py           | 165 ++++++++++++++++++
 utils/grpc-client/pb2/generate.py             |  29 +++
 .../grpc-client/protos/clickhouse_grpc.proto  |   1 -
 22 files changed, 432 insertions(+), 147 deletions(-)
 create mode 120000 tests/integration/test_grpc_protocol/pb2
 delete mode 120000 tests/integration/test_grpc_protocol/protos/clickhouse_grpc.proto
 create mode 120000 tests/integration/test_grpc_protocol_ssl/pb2
 delete mode 120000 tests/integration/test_grpc_protocol_ssl/protos/clickhouse_grpc.proto
 create mode 120000 tests/integration/test_profile_max_sessions_for_user/grpc_protocol_pb2
 delete mode 120000 tests/integration/test_profile_max_sessions_for_user/protos/clickhouse_grpc.proto
 create mode 120000 tests/integration/test_server_reload/grpc_protocol_pb2
 delete mode 120000 tests/integration/test_server_reload/protos/clickhouse_grpc.proto
 create mode 120000 tests/integration/test_session_log/grpc_protocol_pb2
 delete mode 120000 tests/integration/test_session_log/protos/clickhouse_grpc.proto
 create mode 100644 utils/grpc-client/pb2/clickhouse_grpc_pb2.py
 create mode 100644 utils/grpc-client/pb2/clickhouse_grpc_pb2_grpc.py
 create mode 100755 utils/grpc-client/pb2/generate.py
 delete mode 120000 utils/grpc-client/protos/clickhouse_grpc.proto

diff --git a/tests/integration/runner b/tests/integration/runner
index 4c2b1054538..7be491a9a57 100755
--- a/tests/integration/runner
+++ b/tests/integration/runner
@@ -27,7 +27,7 @@ CONTAINER_NAME = f"{VOLUME_NAME}_{random_str()}"
 
 CONFIG_DIR_IN_REPO = "programs/server"
 INTEGRATION_DIR_IN_REPO = "tests/integration"
-SRC_DIR_IN_REPO = "src"
+UTILS_DIR_IN_REPO = "utils"
 
 DIND_INTEGRATION_TESTS_IMAGE_NAME = "clickhouse/integration-tests-runner"
 
@@ -89,12 +89,16 @@ def check_args_and_update_paths(args):
         )
         logging.info("Cases dir is not set. Will use %s" % (args.cases_dir))
 
-    if args.src_dir:
-        if not os.path.isabs(args.src_dir):
-            args.src_dir = os.path.abspath(os.path.join(CURRENT_WORK_DIR, args.src_dir))
+    if args.utils_dir:
+        if not os.path.isabs(args.utils_dir):
+            args.utils_dir = os.path.abspath(
+                os.path.join(CURRENT_WORK_DIR, args.utils_dir)
+            )
     else:
-        args.src_dir = os.path.abspath(os.path.join(CLICKHOUSE_ROOT, SRC_DIR_IN_REPO))
-        logging.info("src dir is not set. Will use %s" % (args.src_dir))
+        args.utils_dir = os.path.abspath(
+            os.path.join(CLICKHOUSE_ROOT, UTILS_DIR_IN_REPO)
+        )
+        logging.info("utils dir is not set. Will use %s" % (args.utils_dir))
 
     logging.info(
         "base_configs_dir: {}, binary: {}, cases_dir: {} ".format(
@@ -115,7 +119,7 @@ def check_args_and_update_paths(args):
 
     if args.dockerd_volume:
         if not os.path.isabs(args.dockerd_volume):
-            args.src_dir = os.path.abspath(
+            args.dockerd_volume = os.path.abspath(
                 os.path.join(CURRENT_WORK_DIR, args.dockerd_volume)
             )
 
@@ -197,9 +201,9 @@ if __name__ == "__main__":
     )
 
     parser.add_argument(
-        "--src-dir",
-        default=os.environ.get("CLICKHOUSE_SRC_DIR"),
-        help="Path to the 'src' directory in repository. Used to provide schemas (e.g. *.proto) for some tests when those schemas are located in the 'src' directory",
+        "--utils-dir",
+        default=os.environ.get("CLICKHOUSE_UTILS_DIR"),
+        help="Path to the 'utils' directory in repository. Used to provide Python modules for grpc protocol schemas which are located in the 'utils' directory",
     )
 
     parser.add_argument(
@@ -424,7 +428,7 @@ if __name__ == "__main__":
         f"--volume={args.library_bridge_binary}:/clickhouse-library-bridge "
         f"--volume={args.base_configs_dir}:/clickhouse-config "
         f"--volume={args.cases_dir}:/ClickHouse/tests/integration "
-        f"--volume={args.src_dir}/Server/grpc_protos:/ClickHouse/src/Server/grpc_protos "
+        f"--volume={args.utils_dir}/grpc-client/pb2:/ClickHouse/utils/grpc-client/pb2 "
         f"--volume=/run:/run/host:ro {dockerd_internal_volume} {env_tags} {env_cleanup} "
         f"-e DOCKER_CLIENT_TIMEOUT=300 -e COMPOSE_HTTP_TIMEOUT=600 {use_analyzer} -e PYTHONUNBUFFERED=1 "
         f'-e PYTEST_ADDOPTS="{parallel_args} {pytest_opts} {tests_list} {rand_args} -vvv"'
diff --git a/tests/integration/test_grpc_protocol/pb2 b/tests/integration/test_grpc_protocol/pb2
new file mode 120000
index 00000000000..27cc9a4f321
--- /dev/null
+++ b/tests/integration/test_grpc_protocol/pb2
@@ -0,0 +1 @@
+../../../utils/grpc-client/pb2
\ No newline at end of file
diff --git a/tests/integration/test_grpc_protocol/protos/clickhouse_grpc.proto b/tests/integration/test_grpc_protocol/protos/clickhouse_grpc.proto
deleted file mode 120000
index 25d15f11e3b..00000000000
--- a/tests/integration/test_grpc_protocol/protos/clickhouse_grpc.proto
+++ /dev/null
@@ -1 +0,0 @@
-../../../../src/Server/grpc_protos/clickhouse_grpc.proto
\ No newline at end of file
diff --git a/tests/integration/test_grpc_protocol/test.py b/tests/integration/test_grpc_protocol/test.py
index efc7d98e820..0dc08630b77 100644
--- a/tests/integration/test_grpc_protocol/test.py
+++ b/tests/integration/test_grpc_protocol/test.py
@@ -10,32 +10,20 @@ from threading import Thread
 import gzip
 import lz4.frame
 
+script_dir = os.path.dirname(os.path.realpath(__file__))
+pb2_dir = os.path.join(script_dir, "pb2")
+if pb2_dir not in sys.path:
+    sys.path.append(pb2_dir)
+import clickhouse_grpc_pb2, clickhouse_grpc_pb2_grpc  # Execute pb2/generate.py to generate these modules.
+
+
 GRPC_PORT = 9100
-SCRIPT_DIR = os.path.dirname(os.path.realpath(__file__))
 DEFAULT_ENCODING = "utf-8"
 
 
-# Use grpcio-tools to generate *pb2.py files from *.proto.
-
-proto_dir = os.path.join(SCRIPT_DIR, "./protos")
-gen_dir = os.path.join(SCRIPT_DIR, "./_gen")
-os.makedirs(gen_dir, exist_ok=True)
-run_and_check(
-    "python3 -m grpc_tools.protoc -I{proto_dir} --python_out={gen_dir} --grpc_python_out={gen_dir} \
-    {proto_dir}/clickhouse_grpc.proto".format(
-        proto_dir=proto_dir, gen_dir=gen_dir
-    ),
-    shell=True,
-)
-
-sys.path.append(gen_dir)
-import clickhouse_grpc_pb2
-import clickhouse_grpc_pb2_grpc
-
-
 # Utilities
 
-config_dir = os.path.join(SCRIPT_DIR, "./configs")
+config_dir = os.path.join(script_dir, "./configs")
 cluster = ClickHouseCluster(__file__)
 node = cluster.add_instance(
     "node",
diff --git a/tests/integration/test_grpc_protocol_ssl/pb2 b/tests/integration/test_grpc_protocol_ssl/pb2
new file mode 120000
index 00000000000..27cc9a4f321
--- /dev/null
+++ b/tests/integration/test_grpc_protocol_ssl/pb2
@@ -0,0 +1 @@
+../../../utils/grpc-client/pb2
\ No newline at end of file
diff --git a/tests/integration/test_grpc_protocol_ssl/protos/clickhouse_grpc.proto b/tests/integration/test_grpc_protocol_ssl/protos/clickhouse_grpc.proto
deleted file mode 120000
index 25d15f11e3b..00000000000
--- a/tests/integration/test_grpc_protocol_ssl/protos/clickhouse_grpc.proto
+++ /dev/null
@@ -1 +0,0 @@
-../../../../src/Server/grpc_protos/clickhouse_grpc.proto
\ No newline at end of file
diff --git a/tests/integration/test_grpc_protocol_ssl/test.py b/tests/integration/test_grpc_protocol_ssl/test.py
index 4c7fe99f49d..2bc835e9c4c 100644
--- a/tests/integration/test_grpc_protocol_ssl/test.py
+++ b/tests/integration/test_grpc_protocol_ssl/test.py
@@ -4,35 +4,23 @@ import sys
 import grpc
 from helpers.cluster import ClickHouseCluster, run_and_check
 
+script_dir = os.path.dirname(os.path.realpath(__file__))
+pb2_dir = os.path.join(script_dir, "pb2")
+if pb2_dir not in sys.path:
+    sys.path.append(pb2_dir)
+import clickhouse_grpc_pb2, clickhouse_grpc_pb2_grpc  # Execute pb2/generate.py to generate these modules.
+
+
 # The test cluster is configured with certificate for that host name, see 'server-ext.cnf'.
 # The client have to verify server certificate against that name. Client uses SNI
 SSL_HOST = "integration-tests.clickhouse.com"
 GRPC_PORT = 9100
-SCRIPT_DIR = os.path.dirname(os.path.realpath(__file__))
 DEFAULT_ENCODING = "utf-8"
 
 
-# Use grpcio-tools to generate *pb2.py files from *.proto.
-
-proto_dir = os.path.join(SCRIPT_DIR, "./protos")
-gen_dir = os.path.join(SCRIPT_DIR, "./_gen")
-os.makedirs(gen_dir, exist_ok=True)
-run_and_check(
-    "python3 -m grpc_tools.protoc -I{proto_dir} --python_out={gen_dir} --grpc_python_out={gen_dir} \
-    {proto_dir}/clickhouse_grpc.proto".format(
-        proto_dir=proto_dir, gen_dir=gen_dir
-    ),
-    shell=True,
-)
-
-sys.path.append(gen_dir)
-import clickhouse_grpc_pb2
-import clickhouse_grpc_pb2_grpc
-
-
 # Utilities
 
-config_dir = os.path.join(SCRIPT_DIR, "./configs")
+config_dir = os.path.join(script_dir, "./configs")
 cluster = ClickHouseCluster(__file__)
 node = cluster.add_instance(
     "node",
diff --git a/tests/integration/test_profile_max_sessions_for_user/grpc_protocol_pb2 b/tests/integration/test_profile_max_sessions_for_user/grpc_protocol_pb2
new file mode 120000
index 00000000000..27cc9a4f321
--- /dev/null
+++ b/tests/integration/test_profile_max_sessions_for_user/grpc_protocol_pb2
@@ -0,0 +1 @@
+../../../utils/grpc-client/pb2
\ No newline at end of file
diff --git a/tests/integration/test_profile_max_sessions_for_user/protos/clickhouse_grpc.proto b/tests/integration/test_profile_max_sessions_for_user/protos/clickhouse_grpc.proto
deleted file mode 120000
index 25d15f11e3b..00000000000
--- a/tests/integration/test_profile_max_sessions_for_user/protos/clickhouse_grpc.proto
+++ /dev/null
@@ -1 +0,0 @@
-../../../../src/Server/grpc_protos/clickhouse_grpc.proto
\ No newline at end of file
diff --git a/tests/integration/test_profile_max_sessions_for_user/test.py b/tests/integration/test_profile_max_sessions_for_user/test.py
index 28c2597e52d..133991fed7a 100755
--- a/tests/integration/test_profile_max_sessions_for_user/test.py
+++ b/tests/integration/test_profile_max_sessions_for_user/test.py
@@ -9,9 +9,15 @@ import threading
 
 from helpers.cluster import ClickHouseCluster, run_and_check
 from helpers.test_tools import assert_logs_contain_with_retry
-
 from helpers.uclient import client, prompt
 
+script_dir = os.path.dirname(os.path.realpath(__file__))
+grpc_protocol_pb2_dir = os.path.join(script_dir, "grpc_protocol_pb2")
+if grpc_protocol_pb2_dir not in sys.path:
+    sys.path.append(grpc_protocol_pb2_dir)
+import clickhouse_grpc_pb2, clickhouse_grpc_pb2_grpc  # Execute grpc_protocol_pb2/generate.py to generate these modules.
+
+
 MAX_SESSIONS_FOR_USER = 2
 POSTGRES_SERVER_PORT = 5433
 MYSQL_SERVER_PORT = 9001
@@ -20,22 +26,8 @@ GRPC_PORT = 9100
 TEST_USER = "test_user"
 TEST_PASSWORD = "123"
 
-SCRIPT_DIR = os.path.dirname(os.path.realpath(__file__))
 DEFAULT_ENCODING = "utf-8"
 
-# Use grpcio-tools to generate *pb2.py files from *.proto.
-proto_dir = os.path.join(SCRIPT_DIR, "./protos")
-gen_dir = os.path.join(SCRIPT_DIR, "./_gen")
-os.makedirs(gen_dir, exist_ok=True)
-run_and_check(
-    f"python3 -m grpc_tools.protoc -I{proto_dir} --python_out={gen_dir} --grpc_python_out={gen_dir} {proto_dir}/clickhouse_grpc.proto",
-    shell=True,
-)
-
-sys.path.append(gen_dir)
-
-import clickhouse_grpc_pb2
-import clickhouse_grpc_pb2_grpc
 
 cluster = ClickHouseCluster(__file__)
 instance = cluster.add_instance(
diff --git a/tests/integration/test_server_reload/grpc_protocol_pb2 b/tests/integration/test_server_reload/grpc_protocol_pb2
new file mode 120000
index 00000000000..27cc9a4f321
--- /dev/null
+++ b/tests/integration/test_server_reload/grpc_protocol_pb2
@@ -0,0 +1 @@
+../../../utils/grpc-client/pb2
\ No newline at end of file
diff --git a/tests/integration/test_server_reload/protos/clickhouse_grpc.proto b/tests/integration/test_server_reload/protos/clickhouse_grpc.proto
deleted file mode 120000
index 25d15f11e3b..00000000000
--- a/tests/integration/test_server_reload/protos/clickhouse_grpc.proto
+++ /dev/null
@@ -1 +0,0 @@
-../../../../src/Server/grpc_protos/clickhouse_grpc.proto
\ No newline at end of file
diff --git a/tests/integration/test_server_reload/test.py b/tests/integration/test_server_reload/test.py
index da5208799ac..655c83dd81f 100644
--- a/tests/integration/test_server_reload/test.py
+++ b/tests/integration/test_server_reload/test.py
@@ -22,6 +22,13 @@ from pathlib import Path
 from requests.exceptions import ConnectionError
 from urllib3.util.retry import Retry
 
+script_dir = os.path.dirname(os.path.realpath(__file__))
+grpc_protocol_pb2_dir = os.path.join(script_dir, "grpc_protocol_pb2")
+if grpc_protocol_pb2_dir not in sys.path:
+    sys.path.append(grpc_protocol_pb2_dir)
+import clickhouse_grpc_pb2, clickhouse_grpc_pb2_grpc  # Execute grpc_protocol_pb2/generate.py to generate these modules.
+
+
 cluster = ClickHouseCluster(__file__)
 instance = cluster.add_instance(
     "instance",
@@ -44,22 +51,6 @@ instance = cluster.add_instance(
 LOADS_QUERY = "SELECT value FROM system.events WHERE event = 'MainConfigLoads'"
 
 
-# Use grpcio-tools to generate *pb2.py files from *.proto.
-
-proto_dir = Path(__file__).parent / "protos"
-gen_dir = Path(__file__).parent / "_gen"
-gen_dir.mkdir(exist_ok=True)
-run_and_check(
-    f"python3 -m grpc_tools.protoc -I{proto_dir!s} --python_out={gen_dir!s} --grpc_python_out={gen_dir!s} \
-    {proto_dir!s}/clickhouse_grpc.proto",
-    shell=True,
-)
-
-sys.path.append(str(gen_dir))
-import clickhouse_grpc_pb2
-import clickhouse_grpc_pb2_grpc
-
-
 @pytest.fixture(name="cluster", scope="module")
 def fixture_cluster():
     try:
diff --git a/tests/integration/test_session_log/grpc_protocol_pb2 b/tests/integration/test_session_log/grpc_protocol_pb2
new file mode 120000
index 00000000000..27cc9a4f321
--- /dev/null
+++ b/tests/integration/test_session_log/grpc_protocol_pb2
@@ -0,0 +1 @@
+../../../utils/grpc-client/pb2
\ No newline at end of file
diff --git a/tests/integration/test_session_log/protos/clickhouse_grpc.proto b/tests/integration/test_session_log/protos/clickhouse_grpc.proto
deleted file mode 120000
index 25d15f11e3b..00000000000
--- a/tests/integration/test_session_log/protos/clickhouse_grpc.proto
+++ /dev/null
@@ -1 +0,0 @@
-../../../../src/Server/grpc_protos/clickhouse_grpc.proto
\ No newline at end of file
diff --git a/tests/integration/test_session_log/test.py b/tests/integration/test_session_log/test.py
index 0c350e6c2c9..922e2557c50 100644
--- a/tests/integration/test_session_log/test.py
+++ b/tests/integration/test_session_log/test.py
@@ -8,27 +8,19 @@ import threading
 
 from helpers.cluster import ClickHouseCluster, run_and_check
 
+script_dir = os.path.dirname(os.path.realpath(__file__))
+grpc_protocol_pb2_dir = os.path.join(script_dir, "grpc_protocol_pb2")
+if grpc_protocol_pb2_dir not in sys.path:
+    sys.path.append(grpc_protocol_pb2_dir)
+import clickhouse_grpc_pb2, clickhouse_grpc_pb2_grpc  # Execute grpc_protocol_pb2/generate.py to generate these modules.
+
+
 POSTGRES_SERVER_PORT = 5433
 MYSQL_SERVER_PORT = 9001
 GRPC_PORT = 9100
 SESSION_LOG_MATCHING_FIELDS = "auth_id, auth_type, client_version_major, client_version_minor, client_version_patch, interface"
-
-SCRIPT_DIR = os.path.dirname(os.path.realpath(__file__))
 DEFAULT_ENCODING = "utf-8"
 
-# Use grpcio-tools to generate *pb2.py files from *.proto.
-proto_dir = os.path.join(SCRIPT_DIR, "./protos")
-gen_dir = os.path.join(SCRIPT_DIR, "./_gen")
-os.makedirs(gen_dir, exist_ok=True)
-run_and_check(
-    f"python3 -m grpc_tools.protoc -I{proto_dir} --python_out={gen_dir} --grpc_python_out={gen_dir} {proto_dir}/clickhouse_grpc.proto",
-    shell=True,
-)
-
-sys.path.append(gen_dir)
-
-import clickhouse_grpc_pb2
-import clickhouse_grpc_pb2_grpc
 
 cluster = ClickHouseCluster(__file__)
 instance = cluster.add_instance(
diff --git a/utils/check-style/check-black b/utils/check-style/check-black
index 33c463d1668..f179ce68f50 100755
--- a/utils/check-style/check-black
+++ b/utils/check-style/check-black
@@ -13,10 +13,14 @@ tmp=$(mktemp)
 find_cmd=(
   find "$GIT_ROOT" -type f -not -path "${GIT_ROOT}contrib/*"
     \(
-      -name '*.py' -or -name "*.python" -or
-        \(
-          -executable -not -name "*.*" -exec sh -c 'file {} | grep -q "Python script"' \;
-        \)
+      \(
+        -name '*.py' -or -name "*.python" -or
+          \(
+            -executable -not -name "*.*" -exec sh -c 'file {} | grep -q "Python script"' \;
+          \)
+      \)
+      # We skip modules generated by the protocol buffer compiler from *.proto files.
+      -and -not -name '*_pb2.py' -and -not -name '*_pb2_grpc.py'
     \)
 )
 
diff --git a/utils/grpc-client/clickhouse-grpc-client.py b/utils/grpc-client/clickhouse-grpc-client.py
index 0caa9e6fdca..e89b6282741 100755
--- a/utils/grpc-client/clickhouse-grpc-client.py
+++ b/utils/grpc-client/clickhouse-grpc-client.py
@@ -10,10 +10,6 @@
 # Most of the command line options are the same, for more information type
 # ./clickhouse_grpc_client.py --help
 
-import grpc  # pip3 install grpcio
-import grpc_tools  # pip3 install grpcio-tools
-import argparse, cmd, os, signal, subprocess, sys, threading, time, uuid
-
 DEFAULT_HOST = "localhost"
 DEFAULT_PORT = 9100
 DEFAULT_USER_NAME = "default"
@@ -24,6 +20,16 @@ STDIN_BUFFER_SIZE = 1048576
 DEFAULT_ENCODING = "utf-8"
 
 
+import grpc  # pip3 install grpcio
+import argparse, cmd, os, signal, sys, threading, time, uuid
+
+script_dir = os.path.dirname(os.path.realpath(__file__))
+pb2_dir = os.path.join(script_dir, "pb2")
+if pb2_dir not in sys.path:
+    sys.path.append(pb2_dir)
+import clickhouse_grpc_pb2, clickhouse_grpc_pb2_grpc  # Execute pb2/generate.py to generate these modules.
+
+
 class ClickHouseGRPCError(Exception):
     pass
 
@@ -80,8 +86,6 @@ class ClickHouseGRPCClient(cmd.Cmd):
         self.session_id = None
 
     def __enter__(self):
-        ClickHouseGRPCClient.__generate_pb2()
-        ClickHouseGRPCClient.__import_pb2()
         self.__connect()
         return self
 
@@ -229,40 +233,6 @@ class ClickHouseGRPCClient(cmd.Cmd):
         if result.HasField("exception"):
             raise ClickHouseGRPCError(result.exception.display_text)
 
-    # Use grpcio-tools to generate *pb2.py files from *.proto.
-    @staticmethod
-    def __generate_pb2():
-        script_dir = os.path.dirname(os.path.realpath(__file__))
-        proto_dir = os.path.join(script_dir, "./protos")
-        gen_dir = os.path.join(script_dir, "./_gen")
-        if os.path.exists(os.path.join(gen_dir, "clickhouse_grpc_pb2_grpc.py")):
-            return
-        os.makedirs(gen_dir, exist_ok=True)
-        cmd = [
-            "python3",
-            "-m",
-            "grpc_tools.protoc",
-            "-I" + proto_dir,
-            "--python_out=" + gen_dir,
-            "--grpc_python_out=" + gen_dir,
-            proto_dir + "/clickhouse_grpc.proto",
-        ]
-        p = subprocess.Popen(cmd, stderr=subprocess.PIPE)
-        # We don't want to show grpc_tools warnings.
-        errors = p.stderr.read().decode().strip("\n").split("\n")
-        only_warnings = all(("Warning" in error) for error in errors)
-        if not only_warnings:
-            error_print("\n".join(errors))
-
-    # Import the generated *pb2.py files.
-    @staticmethod
-    def __import_pb2():
-        script_dir = os.path.dirname(os.path.realpath(__file__))
-        gen_dir = os.path.join(script_dir, "./_gen")
-        sys.path.append(gen_dir)
-        global clickhouse_grpc_pb2, clickhouse_grpc_pb2_grpc
-        import clickhouse_grpc_pb2, clickhouse_grpc_pb2_grpc
-
     # Prints only if interactive mode is activated.
     def verbatim_print(self, *args, **kwargs):
         if self.verbatim:
diff --git a/utils/grpc-client/pb2/clickhouse_grpc_pb2.py b/utils/grpc-client/pb2/clickhouse_grpc_pb2.py
new file mode 100644
index 00000000000..6218047af3c
--- /dev/null
+++ b/utils/grpc-client/pb2/clickhouse_grpc_pb2.py
@@ -0,0 +1,163 @@
+# -*- coding: utf-8 -*-
+# Generated by the protocol buffer compiler.  DO NOT EDIT!
+# source: clickhouse_grpc.proto
+"""Generated protocol buffer code."""
+from google.protobuf.internal import enum_type_wrapper
+from google.protobuf import descriptor as _descriptor
+from google.protobuf import descriptor_pool as _descriptor_pool
+from google.protobuf import message as _message
+from google.protobuf import reflection as _reflection
+from google.protobuf import symbol_database as _symbol_database
+# @@protoc_insertion_point(imports)
+
+_sym_db = _symbol_database.Default()
+
+
+
+
+DESCRIPTOR = _descriptor_pool.Default().AddSerializedFile(b'\n\x15\x63lickhouse_grpc.proto\x12\x0f\x63lickhouse.grpc\")\n\x0bNameAndType\x12\x0c\n\x04name\x18\x01 \x01(\t\x12\x0c\n\x04type\x18\x02 \x01(\t\"\xf5\x01\n\rExternalTable\x12\x0c\n\x04name\x18\x01 \x01(\t\x12-\n\x07\x63olumns\x18\x02 \x03(\x0b\x32\x1c.clickhouse.grpc.NameAndType\x12\x0c\n\x04\x64\x61ta\x18\x03 \x01(\x0c\x12\x0e\n\x06\x66ormat\x18\x04 \x01(\t\x12\x18\n\x10\x63ompression_type\x18\x06 \x01(\t\x12>\n\x08settings\x18\x05 \x03(\x0b\x32,.clickhouse.grpc.ExternalTable.SettingsEntry\x1a/\n\rSettingsEntry\x12\x0b\n\x03key\x18\x01 \x01(\t\x12\r\n\x05value\x18\x02 \x01(\t:\x02\x38\x01\"\x85\x03\n\x1cObsoleteTransportCompression\x12U\n\talgorithm\x18\x01 \x01(\x0e\x32\x42.clickhouse.grpc.ObsoleteTransportCompression.CompressionAlgorithm\x12M\n\x05level\x18\x02 \x01(\x0e\x32>.clickhouse.grpc.ObsoleteTransportCompression.CompressionLevel\"R\n\x14\x43ompressionAlgorithm\x12\x12\n\x0eNO_COMPRESSION\x10\x00\x12\x0b\n\x07\x44\x45\x46LATE\x10\x01\x12\x08\n\x04GZIP\x10\x02\x12\x0f\n\x0bSTREAM_GZIP\x10\x03\"k\n\x10\x43ompressionLevel\x12\x14\n\x10\x43OMPRESSION_NONE\x10\x00\x12\x13\n\x0f\x43OMPRESSION_LOW\x10\x01\x12\x16\n\x12\x43OMPRESSION_MEDIUM\x10\x02\x12\x14\n\x10\x43OMPRESSION_HIGH\x10\x03\"\x8e\x06\n\tQueryInfo\x12\r\n\x05query\x18\x01 \x01(\t\x12\x10\n\x08query_id\x18\x02 \x01(\t\x12:\n\x08settings\x18\x03 \x03(\x0b\x32(.clickhouse.grpc.QueryInfo.SettingsEntry\x12\x10\n\x08\x64\x61tabase\x18\x04 \x01(\t\x12\x12\n\ninput_data\x18\x05 \x01(\x0c\x12\x1c\n\x14input_data_delimiter\x18\x06 \x01(\x0c\x12\x15\n\routput_format\x18\x07 \x01(\t\x12\x1b\n\x13send_output_columns\x18\x18 \x01(\x08\x12\x37\n\x0f\x65xternal_tables\x18\x08 \x03(\x0b\x32\x1e.clickhouse.grpc.ExternalTable\x12\x11\n\tuser_name\x18\t \x01(\t\x12\x10\n\x08password\x18\n \x01(\t\x12\r\n\x05quota\x18\x0b \x01(\t\x12\x12\n\nsession_id\x18\x0c \x01(\t\x12\x15\n\rsession_check\x18\r \x01(\x08\x12\x17\n\x0fsession_timeout\x18\x0e \x01(\r\x12\x0e\n\x06\x63\x61ncel\x18\x0f \x01(\x08\x12\x17\n\x0fnext_query_info\x18\x10 \x01(\x08\x12\x1e\n\x16input_compression_type\x18\x14 \x01(\t\x12\x1f\n\x17output_compression_type\x18\x15 \x01(\t\x12 \n\x18output_compression_level\x18\x13 \x01(\x05\x12\"\n\x1atransport_compression_type\x18\x16 \x01(\t\x12#\n\x1btransport_compression_level\x18\x17 \x01(\x05\x12R\n\x1bobsolete_result_compression\x18\x11 \x01(\x0b\x32-.clickhouse.grpc.ObsoleteTransportCompression\x12!\n\x19obsolete_compression_type\x18\x12 \x01(\t\x1a/\n\rSettingsEntry\x12\x0b\n\x03key\x18\x01 \x01(\t\x12\r\n\x05value\x18\x02 \x01(\t:\x02\x38\x01\"\xa1\x01\n\x08LogEntry\x12\x0c\n\x04time\x18\x01 \x01(\r\x12\x19\n\x11time_microseconds\x18\x02 \x01(\r\x12\x11\n\tthread_id\x18\x03 \x01(\x04\x12\x10\n\x08query_id\x18\x04 \x01(\t\x12)\n\x05level\x18\x05 \x01(\x0e\x32\x1a.clickhouse.grpc.LogsLevel\x12\x0e\n\x06source\x18\x06 \x01(\t\x12\x0c\n\x04text\x18\x07 \x01(\t\"z\n\x08Progress\x12\x11\n\tread_rows\x18\x01 \x01(\x04\x12\x12\n\nread_bytes\x18\x02 \x01(\x04\x12\x1a\n\x12total_rows_to_read\x18\x03 \x01(\x04\x12\x14\n\x0cwritten_rows\x18\x04 \x01(\x04\x12\x15\n\rwritten_bytes\x18\x05 \x01(\x04\"p\n\x05Stats\x12\x0c\n\x04rows\x18\x01 \x01(\x04\x12\x0e\n\x06\x62locks\x18\x02 \x01(\x04\x12\x17\n\x0f\x61llocated_bytes\x18\x03 \x01(\x04\x12\x15\n\rapplied_limit\x18\x04 \x01(\x08\x12\x19\n\x11rows_before_limit\x18\x05 \x01(\x04\"R\n\tException\x12\x0c\n\x04\x63ode\x18\x01 \x01(\x05\x12\x0c\n\x04name\x18\x02 \x01(\t\x12\x14\n\x0c\x64isplay_text\x18\x03 \x01(\t\x12\x13\n\x0bstack_trace\x18\x04 \x01(\t\"\xeb\x02\n\x06Result\x12\x10\n\x08query_id\x18\t \x01(\t\x12\x11\n\ttime_zone\x18\n \x01(\t\x12\x15\n\routput_format\x18\x0b \x01(\t\x12\x34\n\x0eoutput_columns\x18\x0c \x03(\x0b\x32\x1c.clickhouse.grpc.NameAndType\x12\x0e\n\x06output\x18\x01 \x01(\x0c\x12\x0e\n\x06totals\x18\x02 \x01(\x0c\x12\x10\n\x08\x65xtremes\x18\x03 \x01(\x0c\x12\'\n\x04logs\x18\x04 \x03(\x0b\x32\x19.clickhouse.grpc.LogEntry\x12+\n\x08progress\x18\x05 \x01(\x0b\x32\x19.clickhouse.grpc.Progress\x12%\n\x05stats\x18\x06 \x01(\x0b\x32\x16.clickhouse.grpc.Stats\x12-\n\texception\x18\x07 \x01(\x0b\x32\x1a.clickhouse.grpc.Exception\x12\x11\n\tcancelled\x18\x08 \x01(\x08*\x9d\x01\n\tLogsLevel\x12\x0c\n\x08LOG_NONE\x10\x00\x12\r\n\tLOG_FATAL\x10\x01\x12\x10\n\x0cLOG_CRITICAL\x10\x02\x12\r\n\tLOG_ERROR\x10\x03\x12\x0f\n\x0bLOG_WARNING\x10\x04\x12\x0e\n\nLOG_NOTICE\x10\x05\x12\x13\n\x0fLOG_INFORMATION\x10\x06\x12\r\n\tLOG_DEBUG\x10\x07\x12\r\n\tLOG_TRACE\x10\x08\x32\xdb\x02\n\nClickHouse\x12\x45\n\x0c\x45xecuteQuery\x12\x1a.clickhouse.grpc.QueryInfo\x1a\x17.clickhouse.grpc.Result\"\x00\x12V\n\x1b\x45xecuteQueryWithStreamInput\x12\x1a.clickhouse.grpc.QueryInfo\x1a\x17.clickhouse.grpc.Result\"\x00(\x01\x12W\n\x1c\x45xecuteQueryWithStreamOutput\x12\x1a.clickhouse.grpc.QueryInfo\x1a\x17.clickhouse.grpc.Result\"\x00\x30\x01\x12U\n\x18\x45xecuteQueryWithStreamIO\x12\x1a.clickhouse.grpc.QueryInfo\x1a\x17.clickhouse.grpc.Result\"\x00(\x01\x30\x01\x62\x06proto3')
+
+_LOGSLEVEL = DESCRIPTOR.enum_types_by_name['LogsLevel']
+LogsLevel = enum_type_wrapper.EnumTypeWrapper(_LOGSLEVEL)
+LOG_NONE = 0
+LOG_FATAL = 1
+LOG_CRITICAL = 2
+LOG_ERROR = 3
+LOG_WARNING = 4
+LOG_NOTICE = 5
+LOG_INFORMATION = 6
+LOG_DEBUG = 7
+LOG_TRACE = 8
+
+
+_NAMEANDTYPE = DESCRIPTOR.message_types_by_name['NameAndType']
+_EXTERNALTABLE = DESCRIPTOR.message_types_by_name['ExternalTable']
+_EXTERNALTABLE_SETTINGSENTRY = _EXTERNALTABLE.nested_types_by_name['SettingsEntry']
+_OBSOLETETRANSPORTCOMPRESSION = DESCRIPTOR.message_types_by_name['ObsoleteTransportCompression']
+_QUERYINFO = DESCRIPTOR.message_types_by_name['QueryInfo']
+_QUERYINFO_SETTINGSENTRY = _QUERYINFO.nested_types_by_name['SettingsEntry']
+_LOGENTRY = DESCRIPTOR.message_types_by_name['LogEntry']
+_PROGRESS = DESCRIPTOR.message_types_by_name['Progress']
+_STATS = DESCRIPTOR.message_types_by_name['Stats']
+_EXCEPTION = DESCRIPTOR.message_types_by_name['Exception']
+_RESULT = DESCRIPTOR.message_types_by_name['Result']
+_OBSOLETETRANSPORTCOMPRESSION_COMPRESSIONALGORITHM = _OBSOLETETRANSPORTCOMPRESSION.enum_types_by_name['CompressionAlgorithm']
+_OBSOLETETRANSPORTCOMPRESSION_COMPRESSIONLEVEL = _OBSOLETETRANSPORTCOMPRESSION.enum_types_by_name['CompressionLevel']
+NameAndType = _reflection.GeneratedProtocolMessageType('NameAndType', (_message.Message,), {
+  'DESCRIPTOR' : _NAMEANDTYPE,
+  '__module__' : 'clickhouse_grpc_pb2'
+  # @@protoc_insertion_point(class_scope:clickhouse.grpc.NameAndType)
+  })
+_sym_db.RegisterMessage(NameAndType)
+
+ExternalTable = _reflection.GeneratedProtocolMessageType('ExternalTable', (_message.Message,), {
+
+  'SettingsEntry' : _reflection.GeneratedProtocolMessageType('SettingsEntry', (_message.Message,), {
+    'DESCRIPTOR' : _EXTERNALTABLE_SETTINGSENTRY,
+    '__module__' : 'clickhouse_grpc_pb2'
+    # @@protoc_insertion_point(class_scope:clickhouse.grpc.ExternalTable.SettingsEntry)
+    })
+  ,
+  'DESCRIPTOR' : _EXTERNALTABLE,
+  '__module__' : 'clickhouse_grpc_pb2'
+  # @@protoc_insertion_point(class_scope:clickhouse.grpc.ExternalTable)
+  })
+_sym_db.RegisterMessage(ExternalTable)
+_sym_db.RegisterMessage(ExternalTable.SettingsEntry)
+
+ObsoleteTransportCompression = _reflection.GeneratedProtocolMessageType('ObsoleteTransportCompression', (_message.Message,), {
+  'DESCRIPTOR' : _OBSOLETETRANSPORTCOMPRESSION,
+  '__module__' : 'clickhouse_grpc_pb2'
+  # @@protoc_insertion_point(class_scope:clickhouse.grpc.ObsoleteTransportCompression)
+  })
+_sym_db.RegisterMessage(ObsoleteTransportCompression)
+
+QueryInfo = _reflection.GeneratedProtocolMessageType('QueryInfo', (_message.Message,), {
+
+  'SettingsEntry' : _reflection.GeneratedProtocolMessageType('SettingsEntry', (_message.Message,), {
+    'DESCRIPTOR' : _QUERYINFO_SETTINGSENTRY,
+    '__module__' : 'clickhouse_grpc_pb2'
+    # @@protoc_insertion_point(class_scope:clickhouse.grpc.QueryInfo.SettingsEntry)
+    })
+  ,
+  'DESCRIPTOR' : _QUERYINFO,
+  '__module__' : 'clickhouse_grpc_pb2'
+  # @@protoc_insertion_point(class_scope:clickhouse.grpc.QueryInfo)
+  })
+_sym_db.RegisterMessage(QueryInfo)
+_sym_db.RegisterMessage(QueryInfo.SettingsEntry)
+
+LogEntry = _reflection.GeneratedProtocolMessageType('LogEntry', (_message.Message,), {
+  'DESCRIPTOR' : _LOGENTRY,
+  '__module__' : 'clickhouse_grpc_pb2'
+  # @@protoc_insertion_point(class_scope:clickhouse.grpc.LogEntry)
+  })
+_sym_db.RegisterMessage(LogEntry)
+
+Progress = _reflection.GeneratedProtocolMessageType('Progress', (_message.Message,), {
+  'DESCRIPTOR' : _PROGRESS,
+  '__module__' : 'clickhouse_grpc_pb2'
+  # @@protoc_insertion_point(class_scope:clickhouse.grpc.Progress)
+  })
+_sym_db.RegisterMessage(Progress)
+
+Stats = _reflection.GeneratedProtocolMessageType('Stats', (_message.Message,), {
+  'DESCRIPTOR' : _STATS,
+  '__module__' : 'clickhouse_grpc_pb2'
+  # @@protoc_insertion_point(class_scope:clickhouse.grpc.Stats)
+  })
+_sym_db.RegisterMessage(Stats)
+
+Exception = _reflection.GeneratedProtocolMessageType('Exception', (_message.Message,), {
+  'DESCRIPTOR' : _EXCEPTION,
+  '__module__' : 'clickhouse_grpc_pb2'
+  # @@protoc_insertion_point(class_scope:clickhouse.grpc.Exception)
+  })
+_sym_db.RegisterMessage(Exception)
+
+Result = _reflection.GeneratedProtocolMessageType('Result', (_message.Message,), {
+  'DESCRIPTOR' : _RESULT,
+  '__module__' : 'clickhouse_grpc_pb2'
+  # @@protoc_insertion_point(class_scope:clickhouse.grpc.Result)
+  })
+_sym_db.RegisterMessage(Result)
+
+_CLICKHOUSE = DESCRIPTOR.services_by_name['ClickHouse']
+if _descriptor._USE_C_DESCRIPTORS == False:
+
+  DESCRIPTOR._options = None
+  _EXTERNALTABLE_SETTINGSENTRY._options = None
+  _EXTERNALTABLE_SETTINGSENTRY._serialized_options = b'8\001'
+  _QUERYINFO_SETTINGSENTRY._options = None
+  _QUERYINFO_SETTINGSENTRY._serialized_options = b'8\001'
+  _LOGSLEVEL._serialized_start=2363
+  _LOGSLEVEL._serialized_end=2520
+  _NAMEANDTYPE._serialized_start=42
+  _NAMEANDTYPE._serialized_end=83
+  _EXTERNALTABLE._serialized_start=86
+  _EXTERNALTABLE._serialized_end=331
+  _EXTERNALTABLE_SETTINGSENTRY._serialized_start=284
+  _EXTERNALTABLE_SETTINGSENTRY._serialized_end=331
+  _OBSOLETETRANSPORTCOMPRESSION._serialized_start=334
+  _OBSOLETETRANSPORTCOMPRESSION._serialized_end=723
+  _OBSOLETETRANSPORTCOMPRESSION_COMPRESSIONALGORITHM._serialized_start=532
+  _OBSOLETETRANSPORTCOMPRESSION_COMPRESSIONALGORITHM._serialized_end=614
+  _OBSOLETETRANSPORTCOMPRESSION_COMPRESSIONLEVEL._serialized_start=616
+  _OBSOLETETRANSPORTCOMPRESSION_COMPRESSIONLEVEL._serialized_end=723
+  _QUERYINFO._serialized_start=726
+  _QUERYINFO._serialized_end=1508
+  _QUERYINFO_SETTINGSENTRY._serialized_start=284
+  _QUERYINFO_SETTINGSENTRY._serialized_end=331
+  _LOGENTRY._serialized_start=1511
+  _LOGENTRY._serialized_end=1672
+  _PROGRESS._serialized_start=1674
+  _PROGRESS._serialized_end=1796
+  _STATS._serialized_start=1798
+  _STATS._serialized_end=1910
+  _EXCEPTION._serialized_start=1912
+  _EXCEPTION._serialized_end=1994
+  _RESULT._serialized_start=1997
+  _RESULT._serialized_end=2360
+  _CLICKHOUSE._serialized_start=2523
+  _CLICKHOUSE._serialized_end=2870
+# @@protoc_insertion_point(module_scope)
diff --git a/utils/grpc-client/pb2/clickhouse_grpc_pb2_grpc.py b/utils/grpc-client/pb2/clickhouse_grpc_pb2_grpc.py
new file mode 100644
index 00000000000..1c71218bbe5
--- /dev/null
+++ b/utils/grpc-client/pb2/clickhouse_grpc_pb2_grpc.py
@@ -0,0 +1,165 @@
+# Generated by the gRPC Python protocol compiler plugin. DO NOT EDIT!
+"""Client and server classes corresponding to protobuf-defined services."""
+import grpc
+
+import clickhouse_grpc_pb2 as clickhouse__grpc__pb2
+
+
+class ClickHouseStub(object):
+    """Missing associated documentation comment in .proto file."""
+
+    def __init__(self, channel):
+        """Constructor.
+
+        Args:
+            channel: A grpc.Channel.
+        """
+        self.ExecuteQuery = channel.unary_unary(
+                '/clickhouse.grpc.ClickHouse/ExecuteQuery',
+                request_serializer=clickhouse__grpc__pb2.QueryInfo.SerializeToString,
+                response_deserializer=clickhouse__grpc__pb2.Result.FromString,
+                )
+        self.ExecuteQueryWithStreamInput = channel.stream_unary(
+                '/clickhouse.grpc.ClickHouse/ExecuteQueryWithStreamInput',
+                request_serializer=clickhouse__grpc__pb2.QueryInfo.SerializeToString,
+                response_deserializer=clickhouse__grpc__pb2.Result.FromString,
+                )
+        self.ExecuteQueryWithStreamOutput = channel.unary_stream(
+                '/clickhouse.grpc.ClickHouse/ExecuteQueryWithStreamOutput',
+                request_serializer=clickhouse__grpc__pb2.QueryInfo.SerializeToString,
+                response_deserializer=clickhouse__grpc__pb2.Result.FromString,
+                )
+        self.ExecuteQueryWithStreamIO = channel.stream_stream(
+                '/clickhouse.grpc.ClickHouse/ExecuteQueryWithStreamIO',
+                request_serializer=clickhouse__grpc__pb2.QueryInfo.SerializeToString,
+                response_deserializer=clickhouse__grpc__pb2.Result.FromString,
+                )
+
+
+class ClickHouseServicer(object):
+    """Missing associated documentation comment in .proto file."""
+
+    def ExecuteQuery(self, request, context):
+        """Missing associated documentation comment in .proto file."""
+        context.set_code(grpc.StatusCode.UNIMPLEMENTED)
+        context.set_details('Method not implemented!')
+        raise NotImplementedError('Method not implemented!')
+
+    def ExecuteQueryWithStreamInput(self, request_iterator, context):
+        """Missing associated documentation comment in .proto file."""
+        context.set_code(grpc.StatusCode.UNIMPLEMENTED)
+        context.set_details('Method not implemented!')
+        raise NotImplementedError('Method not implemented!')
+
+    def ExecuteQueryWithStreamOutput(self, request, context):
+        """Missing associated documentation comment in .proto file."""
+        context.set_code(grpc.StatusCode.UNIMPLEMENTED)
+        context.set_details('Method not implemented!')
+        raise NotImplementedError('Method not implemented!')
+
+    def ExecuteQueryWithStreamIO(self, request_iterator, context):
+        """Missing associated documentation comment in .proto file."""
+        context.set_code(grpc.StatusCode.UNIMPLEMENTED)
+        context.set_details('Method not implemented!')
+        raise NotImplementedError('Method not implemented!')
+
+
+def add_ClickHouseServicer_to_server(servicer, server):
+    rpc_method_handlers = {
+            'ExecuteQuery': grpc.unary_unary_rpc_method_handler(
+                    servicer.ExecuteQuery,
+                    request_deserializer=clickhouse__grpc__pb2.QueryInfo.FromString,
+                    response_serializer=clickhouse__grpc__pb2.Result.SerializeToString,
+            ),
+            'ExecuteQueryWithStreamInput': grpc.stream_unary_rpc_method_handler(
+                    servicer.ExecuteQueryWithStreamInput,
+                    request_deserializer=clickhouse__grpc__pb2.QueryInfo.FromString,
+                    response_serializer=clickhouse__grpc__pb2.Result.SerializeToString,
+            ),
+            'ExecuteQueryWithStreamOutput': grpc.unary_stream_rpc_method_handler(
+                    servicer.ExecuteQueryWithStreamOutput,
+                    request_deserializer=clickhouse__grpc__pb2.QueryInfo.FromString,
+                    response_serializer=clickhouse__grpc__pb2.Result.SerializeToString,
+            ),
+            'ExecuteQueryWithStreamIO': grpc.stream_stream_rpc_method_handler(
+                    servicer.ExecuteQueryWithStreamIO,
+                    request_deserializer=clickhouse__grpc__pb2.QueryInfo.FromString,
+                    response_serializer=clickhouse__grpc__pb2.Result.SerializeToString,
+            ),
+    }
+    generic_handler = grpc.method_handlers_generic_handler(
+            'clickhouse.grpc.ClickHouse', rpc_method_handlers)
+    server.add_generic_rpc_handlers((generic_handler,))
+
+
+ # This class is part of an EXPERIMENTAL API.
+class ClickHouse(object):
+    """Missing associated documentation comment in .proto file."""
+
+    @staticmethod
+    def ExecuteQuery(request,
+            target,
+            options=(),
+            channel_credentials=None,
+            call_credentials=None,
+            insecure=False,
+            compression=None,
+            wait_for_ready=None,
+            timeout=None,
+            metadata=None):
+        return grpc.experimental.unary_unary(request, target, '/clickhouse.grpc.ClickHouse/ExecuteQuery',
+            clickhouse__grpc__pb2.QueryInfo.SerializeToString,
+            clickhouse__grpc__pb2.Result.FromString,
+            options, channel_credentials,
+            insecure, call_credentials, compression, wait_for_ready, timeout, metadata)
+
+    @staticmethod
+    def ExecuteQueryWithStreamInput(request_iterator,
+            target,
+            options=(),
+            channel_credentials=None,
+            call_credentials=None,
+            insecure=False,
+            compression=None,
+            wait_for_ready=None,
+            timeout=None,
+            metadata=None):
+        return grpc.experimental.stream_unary(request_iterator, target, '/clickhouse.grpc.ClickHouse/ExecuteQueryWithStreamInput',
+            clickhouse__grpc__pb2.QueryInfo.SerializeToString,
+            clickhouse__grpc__pb2.Result.FromString,
+            options, channel_credentials,
+            insecure, call_credentials, compression, wait_for_ready, timeout, metadata)
+
+    @staticmethod
+    def ExecuteQueryWithStreamOutput(request,
+            target,
+            options=(),
+            channel_credentials=None,
+            call_credentials=None,
+            insecure=False,
+            compression=None,
+            wait_for_ready=None,
+            timeout=None,
+            metadata=None):
+        return grpc.experimental.unary_stream(request, target, '/clickhouse.grpc.ClickHouse/ExecuteQueryWithStreamOutput',
+            clickhouse__grpc__pb2.QueryInfo.SerializeToString,
+            clickhouse__grpc__pb2.Result.FromString,
+            options, channel_credentials,
+            insecure, call_credentials, compression, wait_for_ready, timeout, metadata)
+
+    @staticmethod
+    def ExecuteQueryWithStreamIO(request_iterator,
+            target,
+            options=(),
+            channel_credentials=None,
+            call_credentials=None,
+            insecure=False,
+            compression=None,
+            wait_for_ready=None,
+            timeout=None,
+            metadata=None):
+        return grpc.experimental.stream_stream(request_iterator, target, '/clickhouse.grpc.ClickHouse/ExecuteQueryWithStreamIO',
+            clickhouse__grpc__pb2.QueryInfo.SerializeToString,
+            clickhouse__grpc__pb2.Result.FromString,
+            options, channel_credentials,
+            insecure, call_credentials, compression, wait_for_ready, timeout, metadata)
diff --git a/utils/grpc-client/pb2/generate.py b/utils/grpc-client/pb2/generate.py
new file mode 100755
index 00000000000..2f4b3bf5af7
--- /dev/null
+++ b/utils/grpc-client/pb2/generate.py
@@ -0,0 +1,29 @@
+#!/usr/bin/env python3
+
+import grpc_tools  # pip3 install grpcio-tools
+
+import os
+import subprocess
+
+
+script_dir = os.path.dirname(os.path.realpath(__file__))
+dest_dir = script_dir
+src_dir = os.path.abspath(os.path.join(script_dir, "../../../src/Server/grpc_protos"))
+src_filename = "clickhouse_grpc.proto"
+
+
+def generate():
+    cmd = [
+        "python3",
+        "-m",
+        "grpc_tools.protoc",
+        "-I" + src_dir,
+        "--python_out=" + dest_dir,
+        "--grpc_python_out=" + dest_dir,
+        os.path.join(src_dir, src_filename),
+    ]
+    subprocess.run(cmd)
+
+
+if __name__ == "__main__":
+    generate()
diff --git a/utils/grpc-client/protos/clickhouse_grpc.proto b/utils/grpc-client/protos/clickhouse_grpc.proto
deleted file mode 120000
index 14d51a404bd..00000000000
--- a/utils/grpc-client/protos/clickhouse_grpc.proto
+++ /dev/null
@@ -1 +0,0 @@
-../../../src/Server/grpc_protos/clickhouse_grpc.proto
\ No newline at end of file

From eb1b3e6a3ada95bfd3e9258946e19efc68b1b013 Mon Sep 17 00:00:00 2001
From: Maksim Kita <kitaetoya@gmail.com>
Date: Tue, 26 Sep 2023 19:08:22 +0300
Subject: [PATCH 27/37] ProfileEvents added ContextLockWaitMicroseconds

---
 src/Common/ProfileEvents.cpp | 1 +
 src/Interpreters/Context.cpp | 6 +++++-
 2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/src/Common/ProfileEvents.cpp b/src/Common/ProfileEvents.cpp
index fd7b47b4f87..0df0bc89b38 100644
--- a/src/Common/ProfileEvents.cpp
+++ b/src/Common/ProfileEvents.cpp
@@ -211,6 +211,7 @@
     \
     M(RegexpCreated, "Compiled regular expressions. Identical regular expressions compiled just once and cached forever.") \
     M(ContextLock, "Number of times the lock of Context was acquired or tried to acquire. This is global lock.") \
+    M(ContextLockWaitMicroseconds, "Context lock wait time in microseconds") \
     \
     M(StorageBufferFlush, "Number of times a buffer in a 'Buffer' table was flushed.") \
     M(StorageBufferErrorOnFlush, "Number of times a buffer in the 'Buffer' table has not been able to flush due to error writing in the destination table.") \
diff --git a/src/Interpreters/Context.cpp b/src/Interpreters/Context.cpp
index 86651f31e7c..edec3d6bb29 100644
--- a/src/Interpreters/Context.cpp
+++ b/src/Interpreters/Context.cpp
@@ -108,6 +108,7 @@ namespace fs = std::filesystem;
 namespace ProfileEvents
 {
     extern const Event ContextLock;
+    extern const Event ContextLockWaitMicroseconds;
 }
 
 namespace CurrentMetrics
@@ -704,7 +705,10 @@ std::unique_lock<std::recursive_mutex> Context::getLock() const
 {
     ProfileEvents::increment(ProfileEvents::ContextLock);
     CurrentMetrics::Increment increment{CurrentMetrics::ContextLockWait};
-    return std::unique_lock(shared->mutex);
+    Stopwatch watch;
+    auto lock = std::unique_lock(shared->mutex);
+    ProfileEvents::increment(ProfileEvents::ContextLockWaitMicroseconds, watch.elapsedMicroseconds());
+    return lock;
 }
 
 ProcessList & Context::getProcessList() { return shared->process_list; }

From c12c114ade6939b5929cf2e6af0c445053bf4e39 Mon Sep 17 00:00:00 2001
From: Maksim Kita <kitaetoya@gmail.com>
Date: Tue, 26 Sep 2023 15:58:08 +0300
Subject: [PATCH 28/37] Setting compile_aggregate_expressions comment fix

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

diff --git a/src/Core/Settings.h b/src/Core/Settings.h
index 71356d6fb36..901ac6a2860 100644
--- a/src/Core/Settings.h
+++ b/src/Core/Settings.h
@@ -149,7 +149,7 @@ class IColumn;
     M(Bool, allow_suspicious_indices, false, "Reject primary/secondary indexes and sorting keys with identical expressions", 0) \
     M(Bool, compile_expressions, false, "Compile some scalar functions and operators to native code.", 0) \
     M(UInt64, min_count_to_compile_expression, 3, "The number of identical expressions before they are JIT-compiled", 0) \
-    M(Bool, compile_aggregate_expressions, true, "Compile aggregate functions to native code. This feature has a bug and should not be used.", 0) \
+    M(Bool, compile_aggregate_expressions, true, "Compile aggregate functions to native code.", 0) \
     M(UInt64, min_count_to_compile_aggregate_expression, 3, "The number of identical aggregate expressions before they are JIT-compiled", 0) \
     M(Bool, compile_sort_description, true, "Compile sort description to native code.", 0) \
     M(UInt64, min_count_to_compile_sort_description, 3, "The number of identical sort descriptions before they are JIT-compiled", 0) \

From dcdc8bf0e94a971dd7814cb2f0a52beaab33bc00 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Mon, 18 Sep 2023 15:26:54 +0000
Subject: [PATCH 29/37] Added field "is_deterministic" to system.functions

---
 docs/en/operations/system-tables/functions.md | 38 ++++++----
 .../System/StorageSystemFunctions.cpp         | 75 ++++++++++++++-----
 .../02117_show_create_table_system.reference  |  1 +
 .../02815_no_throw_in_simple_queries.sh       |  7 +-
 4 files changed, 85 insertions(+), 36 deletions(-)

diff --git a/docs/en/operations/system-tables/functions.md b/docs/en/operations/system-tables/functions.md
index 60bfa08975b..d52bf24f289 100644
--- a/docs/en/operations/system-tables/functions.md
+++ b/docs/en/operations/system-tables/functions.md
@@ -7,28 +7,34 @@ Contains information about normal and aggregate functions.
 
 Columns:
 
-- `name`(`String`) – The name of the function.
-- `is_aggregate`(`UInt8`) — Whether the function is aggregate.
+- `name` ([String](../../sql-reference/data-types/string.md)) – The name of the function.
+- `is_aggregate` ([UInt8](../../sql-reference/data-types/int-uint.md)) — Whether the function is an aggregate function.
+- `is_deterministic` ([Nullable](../../sql-reference/data-types/nullable.md)([UInt8](../../sql-reference/data-types/int-uint.md))) - Whether the function is deterministic.
+- `case_insensitive`, ([UInt8](../../sql-reference/data-types/int-uint.md)) - Whether the function name can be used case-insensitively.
+- `alias_to`, ([String](../../sql-reference/data-types/string.md)) - The original function name, if the function name is an alias.
+- `create_query`, ([String](../../sql-reference/data-types/enum.md)) - Unused.
+- `origin`, ([Enum8](../../sql-reference/data-types/string.md)) - Unused.
+- `description`, ([String](../../sql-reference/data-types/string.md)) - A high-level description what the function does.
+- `syntax`, ([String](../../sql-reference/data-types/string.md)) - Signature of the function.
+- `arguments`, ([String](../../sql-reference/data-types/string.md)) - What arguments does the function take.
+- `returned_value`, ([String](../../sql-reference/data-types/string.md)) - What does the function return.
+- `examples`, ([String](../../sql-reference/data-types/string.md)) - Example usage of the function.
+- `categories`, ([String](../../sql-reference/data-types/string.md)) - The category of the function.
 
 **Example**
 
 ```sql
- SELECT * FROM system.functions LIMIT 10;
+ SELECT name, is_aggregate, is_deterministic, case_insensitive, alias_to FROM system.functions LIMIT 5;
 ```
 
 ```text
-┌─name──────────────────┬─is_aggregate─┬─case_insensitive─┬─alias_to─┬─create_query─┬─origin─┐
-│ logTrace              │            0 │                0 │          │              │ System │
-│ aes_decrypt_mysql     │            0 │                0 │          │              │ System │
-│ aes_encrypt_mysql     │            0 │                0 │          │              │ System │
-│ decrypt               │            0 │                0 │          │              │ System │
-│ encrypt               │            0 │                0 │          │              │ System │
-│ toBool                │            0 │                0 │          │              │ System │
-│ windowID              │            0 │                0 │          │              │ System │
-│ hopStart              │            0 │                0 │          │              │ System │
-│ hop                   │            0 │                0 │          │              │ System │
-│ snowflakeToDateTime64 │            0 │                0 │          │              │ System │
-└───────────────────────┴──────────────┴──────────────────┴──────────┴──────────────┴────────┘
+┌─name─────────────────────┬─is_aggregate─┬─is_deterministic─┬─case_insensitive─┬─alias_to─┐
+│ BLAKE3                   │            0 │                1 │                0 │          │
+│ sipHash128Reference      │            0 │                1 │                0 │          │
+│ mapExtractKeyLike        │            0 │                1 │                0 │          │
+│ sipHash128ReferenceKeyed │            0 │                1 │                0 │          │
+│ mapPartialSort           │            0 │                1 │                0 │          │
+└──────────────────────────┴──────────────┴──────────────────┴──────────────────┴──────────┘
 
-10 rows in set. Elapsed: 0.002 sec.
+5 rows in set. Elapsed: 0.002 sec.
 ```
diff --git a/src/Storages/System/StorageSystemFunctions.cpp b/src/Storages/System/StorageSystemFunctions.cpp
index 0c609d84da3..776371a3835 100644
--- a/src/Storages/System/StorageSystemFunctions.cpp
+++ b/src/Storages/System/StorageSystemFunctions.cpp
@@ -1,5 +1,6 @@
 #include <AggregateFunctions/AggregateFunctionFactory.h>
 #include <DataTypes/DataTypeMap.h>
+#include <DataTypes/DataTypeNullable.h>
 #include <DataTypes/DataTypeString.h>
 #include <DataTypes/DataTypesNumber.h>
 #include <DataTypes/DataTypeEnum.h>
@@ -15,6 +16,15 @@
 namespace DB
 {
 
+namespace ErrorCodes
+{
+    extern const int DICTIONARIES_WAS_NOT_LOADED;
+    extern const int FUNCTION_NOT_ALLOWED;
+    extern const int LOGICAL_ERROR;
+    extern const int NOT_IMPLEMENTED;
+    extern const int SUPPORT_IS_DISABLED;
+};
+
 enum class FunctionOrigin : Int8
 {
     SYSTEM = 0,
@@ -29,6 +39,7 @@ namespace
         MutableColumns & res_columns,
         const String & name,
         UInt64 is_aggregate,
+        std::optional<UInt64> is_deterministic,
         const String & create_query,
         FunctionOrigin function_origin,
         const Factory & factory)
@@ -36,53 +47,58 @@ namespace
         res_columns[0]->insert(name);
         res_columns[1]->insert(is_aggregate);
 
+        if (!is_deterministic.has_value())
+            res_columns[2]->insertDefault();
+        else
+            res_columns[2]->insert(*is_deterministic);
+
         if constexpr (std::is_same_v<Factory, UserDefinedSQLFunctionFactory> || std::is_same_v<Factory, UserDefinedExecutableFunctionFactory>)
         {
-            res_columns[2]->insert(false);
-            res_columns[3]->insertDefault();
+            res_columns[3]->insert(false);
+            res_columns[4]->insertDefault();
         }
         else
         {
-            res_columns[2]->insert(factory.isCaseInsensitive(name));
+            res_columns[3]->insert(factory.isCaseInsensitive(name));
             if (factory.isAlias(name))
-                res_columns[3]->insert(factory.aliasTo(name));
+                res_columns[4]->insert(factory.aliasTo(name));
             else
-                res_columns[3]->insertDefault();
+                res_columns[4]->insertDefault();
         }
 
-        res_columns[4]->insert(create_query);
-        res_columns[5]->insert(static_cast<Int8>(function_origin));
+        res_columns[5]->insert(create_query);
+        res_columns[6]->insert(static_cast<Int8>(function_origin));
 
         if constexpr (std::is_same_v<Factory, FunctionFactory>)
         {
             if (factory.isAlias(name))
             {
-                res_columns[6]->insertDefault();
                 res_columns[7]->insertDefault();
                 res_columns[8]->insertDefault();
                 res_columns[9]->insertDefault();
                 res_columns[10]->insertDefault();
                 res_columns[11]->insertDefault();
+                res_columns[12]->insertDefault();
             }
             else
             {
                 auto documentation = factory.getDocumentation(name);
-                res_columns[6]->insert(documentation.description);
-                res_columns[7]->insert(documentation.syntax);
-                res_columns[8]->insert(documentation.argumentsAsString());
-                res_columns[9]->insert(documentation.returned_value);
-                res_columns[10]->insert(documentation.examplesAsString());
-                res_columns[11]->insert(documentation.categoriesAsString());
+                res_columns[7]->insert(documentation.description);
+                res_columns[8]->insert(documentation.syntax);
+                res_columns[9]->insert(documentation.argumentsAsString());
+                res_columns[10]->insert(documentation.returned_value);
+                res_columns[11]->insert(documentation.examplesAsString());
+                res_columns[12]->insert(documentation.categoriesAsString());
             }
         }
         else
         {
-            res_columns[6]->insertDefault();
             res_columns[7]->insertDefault();
             res_columns[8]->insertDefault();
             res_columns[9]->insertDefault();
             res_columns[10]->insertDefault();
             res_columns[11]->insertDefault();
+            res_columns[12]->insertDefault();
         }
     }
 }
@@ -102,6 +118,7 @@ NamesAndTypesList StorageSystemFunctions::getNamesAndTypes()
     return {
         {"name", std::make_shared<DataTypeString>()},
         {"is_aggregate", std::make_shared<DataTypeUInt8>()},
+        {"is_deterministic", std::make_shared<DataTypeNullable>(std::make_shared<DataTypeUInt8>())},
         {"case_insensitive", std::make_shared<DataTypeUInt8>()},
         {"alias_to", std::make_shared<DataTypeString>()},
         {"create_query", std::make_shared<DataTypeString>()},
@@ -121,14 +138,34 @@ void StorageSystemFunctions::fillData(MutableColumns & res_columns, ContextPtr c
     const auto & function_names = functions_factory.getAllRegisteredNames();
     for (const auto & function_name : function_names)
     {
-        fillRow(res_columns, function_name, UInt64(0), "", FunctionOrigin::SYSTEM, functions_factory);
+        std::optional<UInt64> is_deterministic;
+        try
+        {
+            is_deterministic = functions_factory.tryGet(function_name, context)->isDeterministic();
+        }
+        catch (const Exception & e)
+        {
+            /// Some functions throw because they need special configuration or setup before use.
+            if (e.code() == ErrorCodes::DICTIONARIES_WAS_NOT_LOADED
+                || e.code() == ErrorCodes::FUNCTION_NOT_ALLOWED
+                || e.code() == ErrorCodes::LOGICAL_ERROR
+                || e.code() == ErrorCodes::NOT_IMPLEMENTED
+                || e.code() == ErrorCodes::SUPPORT_IS_DISABLED)
+            {
+                /// Ignore exception, show is_deterministic = NULL.
+            }
+            else
+                throw;
+        }
+
+        fillRow(res_columns, function_name, 0, is_deterministic, "", FunctionOrigin::SYSTEM, functions_factory);
     }
 
     const auto & aggregate_functions_factory = AggregateFunctionFactory::instance();
     const auto & aggregate_function_names = aggregate_functions_factory.getAllRegisteredNames();
     for (const auto & function_name : aggregate_function_names)
     {
-        fillRow(res_columns, function_name, UInt64(1), "", FunctionOrigin::SYSTEM, aggregate_functions_factory);
+        fillRow(res_columns, function_name, 1, {1}, "", FunctionOrigin::SYSTEM, aggregate_functions_factory);
     }
 
     const auto & user_defined_sql_functions_factory = UserDefinedSQLFunctionFactory::instance();
@@ -136,14 +173,14 @@ void StorageSystemFunctions::fillData(MutableColumns & res_columns, ContextPtr c
     for (const auto & function_name : user_defined_sql_functions_names)
     {
         auto create_query = queryToString(user_defined_sql_functions_factory.get(function_name));
-        fillRow(res_columns, function_name, UInt64(0), create_query, FunctionOrigin::SQL_USER_DEFINED, user_defined_sql_functions_factory);
+        fillRow(res_columns, function_name, 0, {0}, create_query, FunctionOrigin::SQL_USER_DEFINED, user_defined_sql_functions_factory);
     }
 
     const auto & user_defined_executable_functions_factory = UserDefinedExecutableFunctionFactory::instance();
     const auto & user_defined_executable_functions_names = user_defined_executable_functions_factory.getRegisteredNames(context);
     for (const auto & function_name : user_defined_executable_functions_names)
     {
-        fillRow(res_columns, function_name, UInt64(0), "", FunctionOrigin::EXECUTABLE_USER_DEFINED, user_defined_executable_functions_factory);
+        fillRow(res_columns, function_name, 0, {0}, "", FunctionOrigin::EXECUTABLE_USER_DEFINED, user_defined_executable_functions_factory);
     }
 }
 
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 c7523ede607..92a38eaf35d 100644
--- a/tests/queries/0_stateless/02117_show_create_table_system.reference
+++ b/tests/queries/0_stateless/02117_show_create_table_system.reference
@@ -283,6 +283,7 @@ CREATE TABLE system.functions
 (
     `name` String,
     `is_aggregate` UInt8,
+    `is_deterministic` Nullable(UInt8),
     `case_insensitive` UInt8,
     `alias_to` String,
     `create_query` String,
diff --git a/tests/queries/0_stateless/02815_no_throw_in_simple_queries.sh b/tests/queries/0_stateless/02815_no_throw_in_simple_queries.sh
index a5c6de3ce58..e2e4f3f8433 100755
--- a/tests/queries/0_stateless/02815_no_throw_in_simple_queries.sh
+++ b/tests/queries/0_stateless/02815_no_throw_in_simple_queries.sh
@@ -53,4 +53,9 @@ expect eof
 EOF
 }
 
-run "$CLICKHOUSE_LOCAL"
+run "$CLICKHOUSE_LOCAL --disable_suggestion"
+# Suggestions are off because the suggestion feature initializes itself by reading all available function
+# names from "system.functions". Getting the value for field "is_obsolete" occasionally throws (e.g. for
+# certain dictionary functions when dictionaries are not set up yet). Exceptions are properly handled, but
+# they exist for a short time. This, in combination with CLICKHOUSE_TERMINATE_ON_ANY_EXCEPTION, terminates
+# clickhouse-local and clickhouse-client when run in interactive mode *with* suggestions.

From 252627ea8c76b8214520738bf6787bf9ea8e64f8 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Tue, 26 Sep 2023 21:34:32 +0200
Subject: [PATCH 30/37] Refuse anything other than LLD

---
 cmake/tools.cmake | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/cmake/tools.cmake b/cmake/tools.cmake
index 7fb45bc15d8..fcba3465b35 100644
--- a/cmake/tools.cmake
+++ b/cmake/tools.cmake
@@ -76,7 +76,7 @@ endif ()
 if (LINKER_NAME)
     message(STATUS "Using linker: ${LINKER_NAME}")
 else()
-    message(STATUS "Using linker: <default>")
+    message (FATAL_ERROR "The only supported linker is LLVM's LLD, but we cannot find it.")
 endif()
 
 # Archiver

From f0b570d0c3a76d4423b7d73a1c083ad94f5ffd92 Mon Sep 17 00:00:00 2001
From: Nikita Mikhaylov <mikhaylovnikitka@gmail.com>
Date: Tue, 26 Sep 2023 21:59:51 +0200
Subject: [PATCH 31/37] Fix pre-build scripts for old branches (#55032)

---
 docker/packager/binary/build.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docker/packager/binary/build.sh b/docker/packager/binary/build.sh
index 93de6eeb374..4875cb596af 100755
--- a/docker/packager/binary/build.sh
+++ b/docker/packager/binary/build.sh
@@ -56,7 +56,7 @@ ccache_status
 ccache --zero-stats ||:
 
 # Check whether the directory with pre-build scripts exists and not empty.
-if [ -d "/build/packages/pre-build" ] && [ -z "$(ls -A /build/packages/pre-build)" ]; then
+if [ ! -d "/build/packages/pre-build" ] || [ -n "$(ls -A /build/packages/pre-build)" ]; then
    echo "There are no subcommands to execute :)"
 else
   # Execute all commands

From e22de584e9d6925bc25b7c143046b5a7b7e7b6df Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Tue, 26 Sep 2023 19:34:13 +0000
Subject: [PATCH 32/37] information_schema.tables.data_length

---
 docs/en/operations/system-tables/information_schema.md | 4 +++-
 src/Storages/System/attachInformationSchemaTables.cpp  | 4 ++++
 2 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/docs/en/operations/system-tables/information_schema.md b/docs/en/operations/system-tables/information_schema.md
index bca37b47312..8edbc506411 100644
--- a/docs/en/operations/system-tables/information_schema.md
+++ b/docs/en/operations/system-tables/information_schema.md
@@ -198,8 +198,10 @@ Columns:
     - `FOREIGN TABLE`
     - `LOCAL TEMPORARY`
     - `SYSTEM VIEW`
+- `data_length` ([Nullable](../../sql-reference/data-types/nullable.md)([UInt64](../../sql-reference/data-types/int-uint.md))) — The size of
+  the data on-disk. NULL if it could not be determined.
+- `table_collation` ([Nullable](../../sql-reference/data-types/nullable.md)([String](../../sql-reference/data-types/string.md))) — The table default collation. Always `utf8mb4_0900_ai_ci`.
 - `table_comment` ([Nullable](../../sql-reference/data-types/nullable.md)([String](../../sql-reference/data-types/string.md))) — The comment used when creating the table.
-- `table_collation` ([Nullable](../../sql-reference/data-types/nullable.md)([String](../../sql-reference/data-types/string.md))) — The table default collation. Always `utf8mb4`.
 
 **Example**
 
diff --git a/src/Storages/System/attachInformationSchemaTables.cpp b/src/Storages/System/attachInformationSchemaTables.cpp
index 48cf8c0a69f..809dcca9f18 100644
--- a/src/Storages/System/attachInformationSchemaTables.cpp
+++ b/src/Storages/System/attachInformationSchemaTables.cpp
@@ -61,12 +61,14 @@ static constexpr std::string_view tables = R"(
         `table_schema` String,
         `table_name` String,
         `table_type` String,
+        `data_length` Nullable(UInt64),
         `table_collation` Nullable(String),
         `table_comment` Nullable(String),
         `TABLE_CATALOG` String,
         `TABLE_SCHEMA` String,
         `TABLE_NAME` String,
         `TABLE_TYPE` String,
+        `DATA_LENGTH` Nullable(UInt64),
         `TABLE_COLLATION` Nullable(String),
         `TABLE_COMMENT` Nullable(String)
     ) AS
@@ -80,12 +82,14 @@ static constexpr std::string_view tables = R"(
                 has_own_data = 0,      'FOREIGN TABLE',
                 'BASE TABLE'
                 )            AS table_type,
+        total_bytes AS data_length,
         'utf8mb4_0900_ai_ci' AS table_collation,
         comment              AS table_comment,
         table_catalog        AS TABLE_CATALOG,
         table_schema         AS TABLE_SCHEMA,
         table_name           AS TABLE_NAME,
         table_type           AS TABLE_TYPE,
+        data_length          AS DATA_LENGTH,
         table_collation      AS TABLE_COLLATION,
         table_comment        AS TABLE_COMMENT
     FROM system.tables

From 23bc2862806ccf6f755d5f2b7a62d8f42ebfdebd Mon Sep 17 00:00:00 2001
From: Michael Kolupaev <michael.kolupaev@clickhouse.com>
Date: Tue, 26 Sep 2023 13:23:26 -0700
Subject: [PATCH 33/37] Print more information about one logical error in
 MergeTreeDataWriter (#55001)

---
 src/Processors/Merges/Algorithms/AggregatingSortedAlgorithm.h   | 1 +
 src/Processors/Merges/Algorithms/CollapsingSortedAlgorithm.h    | 1 +
 .../Merges/Algorithms/FinishAggregatingInOrderAlgorithm.h       | 1 +
 .../Merges/Algorithms/GraphiteRollupSortedAlgorithm.h           | 1 +
 src/Processors/Merges/Algorithms/IMergingAlgorithm.h            | 1 +
 src/Processors/Merges/Algorithms/MergingSortedAlgorithm.h       | 1 +
 src/Processors/Merges/Algorithms/ReplacingSortedAlgorithm.h     | 1 +
 src/Processors/Merges/Algorithms/SummingSortedAlgorithm.h       | 1 +
 src/Processors/Merges/Algorithms/VersionedCollapsingAlgorithm.h | 1 +
 src/Processors/Transforms/ColumnGathererTransform.h             | 1 +
 src/Processors/Transforms/MergeJoinTransform.h                  | 1 +
 src/Storages/MergeTree/MergeTreeDataWriter.cpp                  | 2 +-
 12 files changed, 12 insertions(+), 1 deletion(-)

diff --git a/src/Processors/Merges/Algorithms/AggregatingSortedAlgorithm.h b/src/Processors/Merges/Algorithms/AggregatingSortedAlgorithm.h
index 456b94c69ce..72a1f4871d8 100644
--- a/src/Processors/Merges/Algorithms/AggregatingSortedAlgorithm.h
+++ b/src/Processors/Merges/Algorithms/AggregatingSortedAlgorithm.h
@@ -24,6 +24,7 @@ public:
         size_t max_block_size_rows_,
         size_t max_block_size_bytes_);
 
+    const char * getName() const override { return "AggregatingSortedAlgorithm"; }
     void initialize(Inputs inputs) override;
     void consume(Input & input, size_t source_num) override;
     Status merge() override;
diff --git a/src/Processors/Merges/Algorithms/CollapsingSortedAlgorithm.h b/src/Processors/Merges/Algorithms/CollapsingSortedAlgorithm.h
index c90ddbbab5d..28bb87cb394 100644
--- a/src/Processors/Merges/Algorithms/CollapsingSortedAlgorithm.h
+++ b/src/Processors/Merges/Algorithms/CollapsingSortedAlgorithm.h
@@ -38,6 +38,7 @@ public:
         WriteBuffer * out_row_sources_buf_ = nullptr,
         bool use_average_block_sizes = false);
 
+    const char * getName() const override { return "CollapsingSortedAlgorithm"; }
     Status merge() override;
 
 private:
diff --git a/src/Processors/Merges/Algorithms/FinishAggregatingInOrderAlgorithm.h b/src/Processors/Merges/Algorithms/FinishAggregatingInOrderAlgorithm.h
index 13522b70834..cc6578e79be 100644
--- a/src/Processors/Merges/Algorithms/FinishAggregatingInOrderAlgorithm.h
+++ b/src/Processors/Merges/Algorithms/FinishAggregatingInOrderAlgorithm.h
@@ -45,6 +45,7 @@ public:
         size_t max_block_size_rows_,
         size_t max_block_size_bytes_);
 
+    const char * getName() const override { return "FinishAggregatingInOrderAlgorithm"; }
     void initialize(Inputs inputs) override;
     void consume(Input & input, size_t source_num) override;
     Status merge() override;
diff --git a/src/Processors/Merges/Algorithms/GraphiteRollupSortedAlgorithm.h b/src/Processors/Merges/Algorithms/GraphiteRollupSortedAlgorithm.h
index f920d623b1f..a20a6eaf11f 100644
--- a/src/Processors/Merges/Algorithms/GraphiteRollupSortedAlgorithm.h
+++ b/src/Processors/Merges/Algorithms/GraphiteRollupSortedAlgorithm.h
@@ -30,6 +30,7 @@ public:
         Graphite::Params params_,
         time_t time_of_merge_);
 
+    const char * getName() const override { return "GraphiteRollupSortedAlgorithm"; }
     Status merge() override;
 
     struct ColumnsDefinition
diff --git a/src/Processors/Merges/Algorithms/IMergingAlgorithm.h b/src/Processors/Merges/Algorithms/IMergingAlgorithm.h
index c56324c7d9a..d55ceb78af4 100644
--- a/src/Processors/Merges/Algorithms/IMergingAlgorithm.h
+++ b/src/Processors/Merges/Algorithms/IMergingAlgorithm.h
@@ -47,6 +47,7 @@ public:
 
     using Inputs = std::vector<Input>;
 
+    virtual const char * getName() const = 0;
     virtual void initialize(Inputs inputs) = 0;
     virtual void consume(Input & input, size_t source_num) = 0;
     virtual Status merge() = 0;
diff --git a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.h b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.h
index 1357e58f0f1..bcb111baadf 100644
--- a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.h
+++ b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.h
@@ -26,6 +26,7 @@ public:
 
     void addInput();
 
+    const char * getName() const override { return "MergingSortedAlgorithm"; }
     void initialize(Inputs inputs) override;
     void consume(Input & input, size_t source_num) override;
     Status merge() override;
diff --git a/src/Processors/Merges/Algorithms/ReplacingSortedAlgorithm.h b/src/Processors/Merges/Algorithms/ReplacingSortedAlgorithm.h
index 6ee138aca88..b2258918fde 100644
--- a/src/Processors/Merges/Algorithms/ReplacingSortedAlgorithm.h
+++ b/src/Processors/Merges/Algorithms/ReplacingSortedAlgorithm.h
@@ -30,6 +30,7 @@ public:
         bool cleanup = false,
         size_t * cleanedup_rows_count = nullptr);
 
+    const char * getName() const override { return "ReplacingSortedAlgorithm"; }
     Status merge() override;
 
 private:
diff --git a/src/Processors/Merges/Algorithms/SummingSortedAlgorithm.h b/src/Processors/Merges/Algorithms/SummingSortedAlgorithm.h
index 8943e235729..dbbe4e53a5f 100644
--- a/src/Processors/Merges/Algorithms/SummingSortedAlgorithm.h
+++ b/src/Processors/Merges/Algorithms/SummingSortedAlgorithm.h
@@ -25,6 +25,7 @@ public:
         size_t max_block_size_rows,
         size_t max_block_size_bytes);
 
+    const char * getName() const override { return "SummingSortedAlgorithm"; }
     void initialize(Inputs inputs) override;
     void consume(Input & input, size_t source_num) override;
     Status merge() override;
diff --git a/src/Processors/Merges/Algorithms/VersionedCollapsingAlgorithm.h b/src/Processors/Merges/Algorithms/VersionedCollapsingAlgorithm.h
index 578100f080d..d98529b301c 100644
--- a/src/Processors/Merges/Algorithms/VersionedCollapsingAlgorithm.h
+++ b/src/Processors/Merges/Algorithms/VersionedCollapsingAlgorithm.h
@@ -25,6 +25,7 @@ public:
         WriteBuffer * out_row_sources_buf_ = nullptr,
         bool use_average_block_sizes = false);
 
+    const char * getName() const override { return "VersionedCollapsingAlgorithm"; }
     Status merge() override;
 
 private:
diff --git a/src/Processors/Transforms/ColumnGathererTransform.h b/src/Processors/Transforms/ColumnGathererTransform.h
index da6dc877abf..b5bbbff9aca 100644
--- a/src/Processors/Transforms/ColumnGathererTransform.h
+++ b/src/Processors/Transforms/ColumnGathererTransform.h
@@ -59,6 +59,7 @@ class ColumnGathererStream final : public IMergingAlgorithm
 public:
     ColumnGathererStream(size_t num_inputs, ReadBuffer & row_sources_buf_, size_t block_preferred_size_ = DEFAULT_BLOCK_SIZE);
 
+    const char * getName() const override { return "ColumnGathererStream"; }
     void initialize(Inputs inputs) override;
     void consume(Input & input, size_t source_num) override;
     Status merge() override;
diff --git a/src/Processors/Transforms/MergeJoinTransform.h b/src/Processors/Transforms/MergeJoinTransform.h
index 6bf4484df24..0dd4e24c922 100644
--- a/src/Processors/Transforms/MergeJoinTransform.h
+++ b/src/Processors/Transforms/MergeJoinTransform.h
@@ -229,6 +229,7 @@ class MergeJoinAlgorithm final : public IMergingAlgorithm
 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;
diff --git a/src/Storages/MergeTree/MergeTreeDataWriter.cpp b/src/Storages/MergeTree/MergeTreeDataWriter.cpp
index d6ba7cb3093..0f4a9c28884 100644
--- a/src/Storages/MergeTree/MergeTreeDataWriter.cpp
+++ b/src/Storages/MergeTree/MergeTreeDataWriter.cpp
@@ -365,7 +365,7 @@ Block MergeTreeDataWriter::mergeBlock(
 
     /// Check that after first merge merging_algorithm is waiting for data from input 0.
     if (status.required_source != 0)
-        throw Exception(ErrorCodes::LOGICAL_ERROR, "Logical error: required source after the first merge is not 0.");
+        throw Exception(ErrorCodes::LOGICAL_ERROR, "Logical error: required source after the first merge is not 0. Chunk rows: {}, is_finished: {}, required_source: {}, algorithm: {}", status.chunk.getNumRows(), status.is_finished, status.required_source, merging_algorithm->getName());
 
     status = merging_algorithm->merge();
 

From 1ac2247b3cc2aa020bc7b7fadb89edb1981493f6 Mon Sep 17 00:00:00 2001
From: Michael Kolupaev <michael.kolupaev@clickhouse.com>
Date: Tue, 26 Sep 2023 13:24:06 -0700
Subject: [PATCH 34/37] Add desperate instrumentation for debugging deadlock in
 MultiplexedConnections (#54940)

---
 base/base/hex.h                       | 10 ++++++++
 src/Client/MultiplexedConnections.cpp | 37 ++++++++++++++++++++++++++-
 src/Client/MultiplexedConnections.h   | 12 ++++++---
 3 files changed, 55 insertions(+), 4 deletions(-)

diff --git a/base/base/hex.h b/base/base/hex.h
index 937218fec5a..931f220aa08 100644
--- a/base/base/hex.h
+++ b/base/base/hex.h
@@ -289,3 +289,13 @@ inline void writeBinByte(UInt8 byte, void * out)
 {
     memcpy(out, &impl::bin_byte_to_char_table[static_cast<size_t>(byte) * 8], 8);
 }
+
+/// Converts byte array to a hex string. Useful for debug logging.
+inline std::string hexString(const void * data, size_t size)
+{
+    const char * p = reinterpret_cast<const char *>(data);
+    std::string s(size * 2, '\0');
+    for (size_t i = 0; i < size; ++i)
+        writeHexByteLowercase(p[i], s.data() + i * 2);
+    return s;
+}
diff --git a/src/Client/MultiplexedConnections.cpp b/src/Client/MultiplexedConnections.cpp
index c0218568d67..80d35b7dc1a 100644
--- a/src/Client/MultiplexedConnections.cpp
+++ b/src/Client/MultiplexedConnections.cpp
@@ -1,10 +1,13 @@
 #include <Client/MultiplexedConnections.h>
 
 #include <Common/thread_local_rng.h>
+#include <Common/logger_useful.h>
 #include <Core/Protocol.h>
 #include <IO/ConnectionTimeouts.h>
 #include <IO/Operators.h>
 #include <Interpreters/ClientInfo.h>
+#include <base/getThreadId.h>
+#include <base/hex.h>
 
 namespace DB
 {
@@ -19,6 +22,13 @@ namespace ErrorCodes
 }
 
 
+#define MUTEX_LOCK_TEMPORARY_DEBUG_INSTRUMENTATION \
+    mutex_last_locked_by.store((getThreadId() << 32) | __LINE__); \
+    memcpy(mutex_memory_dump.data(), &cancel_mutex, mutex_memory_dump.size()); \
+    mutex_locked += 1; \
+    SCOPE_EXIT({ mutex_locked -= 1; });
+
+
 MultiplexedConnections::MultiplexedConnections(Connection & connection, const Settings & settings_, const ThrottlerPtr & throttler)
     : settings(settings_)
 {
@@ -73,6 +83,7 @@ MultiplexedConnections::MultiplexedConnections(
 void MultiplexedConnections::sendScalarsData(Scalars & data)
 {
     std::lock_guard lock(cancel_mutex);
+    MUTEX_LOCK_TEMPORARY_DEBUG_INSTRUMENTATION
 
     if (!sent_query)
         throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot send scalars data: query not yet sent.");
@@ -88,6 +99,7 @@ void MultiplexedConnections::sendScalarsData(Scalars & data)
 void MultiplexedConnections::sendExternalTablesData(std::vector<ExternalTablesData> & data)
 {
     std::lock_guard lock(cancel_mutex);
+    MUTEX_LOCK_TEMPORARY_DEBUG_INSTRUMENTATION
 
     if (!sent_query)
         throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot send external tables data: query not yet sent.");
@@ -116,6 +128,7 @@ void MultiplexedConnections::sendQuery(
     bool with_pending_data)
 {
     std::lock_guard lock(cancel_mutex);
+    MUTEX_LOCK_TEMPORARY_DEBUG_INSTRUMENTATION
 
     if (sent_query)
         throw Exception(ErrorCodes::LOGICAL_ERROR, "Query already sent.");
@@ -173,6 +186,7 @@ void MultiplexedConnections::sendQuery(
 void MultiplexedConnections::sendIgnoredPartUUIDs(const std::vector<UUID> & uuids)
 {
     std::lock_guard lock(cancel_mutex);
+    MUTEX_LOCK_TEMPORARY_DEBUG_INSTRUMENTATION
 
     if (sent_query)
         throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot send uuids after query is sent.");
@@ -189,6 +203,7 @@ void MultiplexedConnections::sendIgnoredPartUUIDs(const std::vector<UUID> & uuid
 void MultiplexedConnections::sendReadTaskResponse(const String & response)
 {
     std::lock_guard lock(cancel_mutex);
+    MUTEX_LOCK_TEMPORARY_DEBUG_INSTRUMENTATION
     if (cancelled)
         return;
     current_connection->sendReadTaskResponse(response);
@@ -198,6 +213,7 @@ void MultiplexedConnections::sendReadTaskResponse(const String & response)
 void MultiplexedConnections::sendMergeTreeReadTaskResponse(const ParallelReadResponse & response)
 {
     std::lock_guard lock(cancel_mutex);
+    MUTEX_LOCK_TEMPORARY_DEBUG_INSTRUMENTATION
     if (cancelled)
         return;
     current_connection->sendMergeTreeReadTaskResponse(response);
@@ -207,13 +223,29 @@ void MultiplexedConnections::sendMergeTreeReadTaskResponse(const ParallelReadRes
 Packet MultiplexedConnections::receivePacket()
 {
     std::lock_guard lock(cancel_mutex);
+    MUTEX_LOCK_TEMPORARY_DEBUG_INSTRUMENTATION
     Packet packet = receivePacketUnlocked({});
     return packet;
 }
 
 void MultiplexedConnections::disconnect()
 {
-    std::lock_guard lock(cancel_mutex);
+    /// We've seen this lock mysteriously get stuck forever, without any other thread seeming to
+    /// hold the mutex. This is temporary code to print some extra information next time it happens.
+    /// std::lock_guard lock(cancel_mutex);
+    if (!cancel_mutex.try_lock_for(std::chrono::hours(1)))
+    {
+        UInt64 last_locked = mutex_last_locked_by.load();
+        std::array<UInt8, sizeof(std::timed_mutex)> new_memory_dump;
+        memcpy(new_memory_dump.data(), &cancel_mutex, new_memory_dump.size());
+        LOG_ERROR(&Poco::Logger::get("MultiplexedConnections"), "Deadlock in MultiplexedConnections::disconnect()! Mutex was last (instrumentedly) locked by thread {} on line {}, lock balance: {}, mutex memory when last locked: {}, mutex memory now: {}", last_locked >> 32, last_locked & 0xffffffff, mutex_locked.load(), hexString(mutex_memory_dump.data(), mutex_memory_dump.size()), hexString(new_memory_dump.data(), new_memory_dump.size()));
+        throw Exception(ErrorCodes::LOGICAL_ERROR, "Deadlock in MultiplexedConnections::disconnect()");
+    }
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wthread-safety-analysis"
+    std::lock_guard lock(cancel_mutex, std::adopt_lock);
+#pragma clang diagnostic pop
+    MUTEX_LOCK_TEMPORARY_DEBUG_INSTRUMENTATION
 
     for (ReplicaState & state : replica_states)
     {
@@ -229,6 +261,7 @@ void MultiplexedConnections::disconnect()
 void MultiplexedConnections::sendCancel()
 {
     std::lock_guard lock(cancel_mutex);
+    MUTEX_LOCK_TEMPORARY_DEBUG_INSTRUMENTATION
 
     if (!sent_query || cancelled)
         throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot cancel. Either no query sent or already cancelled.");
@@ -246,6 +279,7 @@ void MultiplexedConnections::sendCancel()
 Packet MultiplexedConnections::drain()
 {
     std::lock_guard lock(cancel_mutex);
+    MUTEX_LOCK_TEMPORARY_DEBUG_INSTRUMENTATION
 
     if (!cancelled)
         throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot drain connections: cancel first.");
@@ -286,6 +320,7 @@ Packet MultiplexedConnections::drain()
 std::string MultiplexedConnections::dumpAddresses() const
 {
     std::lock_guard lock(cancel_mutex);
+    MUTEX_LOCK_TEMPORARY_DEBUG_INSTRUMENTATION
     return dumpAddressesUnlocked();
 }
 
diff --git a/src/Client/MultiplexedConnections.h b/src/Client/MultiplexedConnections.h
index 371639d14c4..4c776e6e168 100644
--- a/src/Client/MultiplexedConnections.h
+++ b/src/Client/MultiplexedConnections.h
@@ -105,9 +105,15 @@ private:
     /// std::nullopt if parallel reading from replicas is not used
     std::optional<ReplicaInfo> replica_info;
 
-    /// A mutex for the sendCancel function to execute safely
-    /// in separate thread.
-    mutable std::mutex cancel_mutex;
+    /// A mutex for the sendCancel function to execute safely in separate thread.
+    mutable std::timed_mutex cancel_mutex;
+
+    /// Temporary instrumentation to debug a weird deadlock on cancel_mutex.
+    /// TODO: Once the investigation is done, get rid of these, and of INSTRUMENTED_LOCK_MUTEX, and
+    ///       change cancel_mutex to std::mutex.
+    mutable std::atomic<UInt64> mutex_last_locked_by{0};
+    mutable std::atomic<Int64> mutex_locked{0};
+    mutable std::array<UInt8, sizeof(std::timed_mutex)> mutex_memory_dump;
 
     friend struct RemoteQueryExecutorRoutine;
 };

From b11e8763231364a95b9d121f88e0bc9cc9c8bd23 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 27 Sep 2023 00:37:17 +0300
Subject: [PATCH 35/37] Update tools.cmake

---
 cmake/tools.cmake | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/cmake/tools.cmake b/cmake/tools.cmake
index fcba3465b35..e03b1d9d72a 100644
--- a/cmake/tools.cmake
+++ b/cmake/tools.cmake
@@ -75,9 +75,11 @@ endif ()
 
 if (LINKER_NAME)
     message(STATUS "Using linker: ${LINKER_NAME}")
-else()
+elseif (NOT ARCH_S390X AND NOT OS_FREEBSD)
     message (FATAL_ERROR "The only supported linker is LLVM's LLD, but we cannot find it.")
-endif()
+else ()
+    message(STATUS "Using linker: <default>")
+endif ()
 
 # Archiver
 

From 3d299d5f9df1819409e39beba7d60bdb3250f645 Mon Sep 17 00:00:00 2001
From: flynn <fenglv15@mails.ucas.ac.cn>
Date: Wed, 27 Sep 2023 03:16:52 +0000
Subject: [PATCH 36/37] Review fix

---
 .../functions/string-search-functions.md      | 12 +++----
 src/Functions/FunctionsStringDistance.cpp     |  9 +++--
 .../02884_string_distance_function.reference  | 36 ++++++++++++-------
 .../02884_string_distance_function.sql        | 15 +++++++-
 4 files changed, 48 insertions(+), 24 deletions(-)

diff --git a/docs/en/sql-reference/functions/string-search-functions.md b/docs/en/sql-reference/functions/string-search-functions.md
index 8a2e257ed34..55a3c3a013a 100644
--- a/docs/en/sql-reference/functions/string-search-functions.md
+++ b/docs/en/sql-reference/functions/string-search-functions.md
@@ -684,12 +684,12 @@ Like [hasSubsequenceUTF8](#hasSubsequenceUTF8) but searches case-insensitively.
 
 ## byteHammingDistance
 
-Calculates the hamming distance between two byte strings.
+Calculates the [hamming distance](https://en.wikipedia.org/wiki/Hamming_distance) between two byte strings.
 
 **Syntax**
 
 ```sql
-byteHammingDistance(haystack, needle)
+byteHammingDistance(string2, string2)
 ```
 
 **Examples**
@@ -710,12 +710,12 @@ Result:
 
 ## byteJaccardIndex
 
-Calculates the jaccard similarity index between two byte strings.
+Calculates the [Jaccard similarity index](https://en.wikipedia.org/wiki/Jaccard_index) between two byte strings.
 
 **Syntax**
 
 ```sql
-byteJaccardIndex(haystack, needle)
+byteJaccardIndex(string1, string2)
 ```
 
 **Examples**
@@ -734,12 +734,12 @@ Result:
 
 ## byteEditDistance
 
-Calculates the edit distance between two byte strings.
+Calculates the [edit distance](https://en.wikipedia.org/wiki/Edit_distance) between two byte strings.
 
 **Syntax**
 
 ```sql
-byteEidtDistance(haystack, needle)
+byteEditDistance(string1, string2)
 ```
 
 **Examples**
diff --git a/src/Functions/FunctionsStringDistance.cpp b/src/Functions/FunctionsStringDistance.cpp
index 73ff379aae0..5845dd33fb4 100644
--- a/src/Functions/FunctionsStringDistance.cpp
+++ b/src/Functions/FunctionsStringDistance.cpp
@@ -18,7 +18,7 @@ extern const int TOO_LARGE_STRING_SIZE;
 }
 
 template <typename Op>
-struct FunctionDistanceImpl
+struct FunctionStringDistanceImpl
 {
     using ResultType = typename Op::ResultType;
 
@@ -195,12 +195,11 @@ struct NameByteEditDistance
     static constexpr auto name = "byteEditDistance";
 };
 
-using FunctionByteHammingDistance
-    = FunctionsStringSimilarity<FunctionDistanceImpl<ByteHammingDistanceImpl>, NameByteHammingDistance>;
+using FunctionByteHammingDistance = FunctionsStringSimilarity<FunctionStringDistanceImpl<ByteHammingDistanceImpl>, NameByteHammingDistance>;
 
-using FunctionByteJaccardIndex = FunctionsStringSimilarity<FunctionDistanceImpl<ByteJaccardIndexImpl>, NameByteJaccardIndex>;
+using FunctionByteJaccardIndex = FunctionsStringSimilarity<FunctionStringDistanceImpl<ByteJaccardIndexImpl>, NameByteJaccardIndex>;
 
-using FunctionByteEditDistance = FunctionsStringSimilarity<FunctionDistanceImpl<ByteEditDistanceImpl>, NameByteEditDistance>;
+using FunctionByteEditDistance = FunctionsStringSimilarity<FunctionStringDistanceImpl<ByteEditDistanceImpl>, NameByteEditDistance>;
 
 REGISTER_FUNCTION(StringHammingDistance)
 {
diff --git a/tests/queries/0_stateless/02884_string_distance_function.reference b/tests/queries/0_stateless/02884_string_distance_function.reference
index d50ffa0a48e..e8353d89481 100644
--- a/tests/queries/0_stateless/02884_string_distance_function.reference
+++ b/tests/queries/0_stateless/02884_string_distance_function.reference
@@ -1,4 +1,25 @@
+const arguments byteHammingDistance
 0
+const arguments byteEditDistance
+6
+const arguments byteJaccardIndex
+0.4
+byteHammingDistance
+1
+7
+7
+10
+byteHammingDistance(const, non const)
+3
+6
+3
+10
+byteHammingDistance(non const, const)
+6
+6
+3
+10
+mismatches(alias)
 1
 7
 7
@@ -11,26 +32,17 @@
 6
 3
 10
-1
-7
-7
-10
-3
-6
-3
-10
-6
-6
-3
-10
+byteJaccardIndex
 0.8571428571428571
 0.8571428571428571
 0
 0.4
+byteEditDistance
 1
 1
 7
 6
+byteLevenshteinDistance
 1
 1
 7
diff --git a/tests/queries/0_stateless/02884_string_distance_function.sql b/tests/queries/0_stateless/02884_string_distance_function.sql
index bab22f24bf2..d19f8a9f43c 100644
--- a/tests/queries/0_stateless/02884_string_distance_function.sql
+++ b/tests/queries/0_stateless/02884_string_distance_function.sql
@@ -1,4 +1,10 @@
+select 'const arguments byteHammingDistance';
 select byteHammingDistance('abcd', 'abcd');
+select 'const arguments byteEditDistance';
+select byteEditDistance('clickhouse', 'mouse');
+select 'const arguments byteJaccardIndex';
+select byteJaccardIndex('clickhouse', 'mouse');
+
 drop table if exists t;
 create table t
 (
@@ -7,18 +13,25 @@ create table t
 ) engine = MergeTree order by s1;
 
 insert into t values ('abcdefg', 'abcdef') ('abcdefg', 'bcdefg') ('abcdefg', '') ('mouse', 'clickhouse');
+select 'byteHammingDistance';
 select byteHammingDistance(s1, s2) from t;
+select 'byteHammingDistance(const, non const)';
 select byteHammingDistance('abc', s2) from t;
+select 'byteHammingDistance(non const, const)';
 select byteHammingDistance(s2, 'def') from t;
 
+select 'mismatches(alias)';
 select mismatches(s1, s2) from t;
 select mismatches('abc', s2) from t;
 select mismatches(s2, 'def') from t;
 
+select 'byteJaccardIndex';
 select byteJaccardIndex(s1, s2) from t;
+select 'byteEditDistance';
 select byteEditDistance(s1, s2) from t;
+select 'byteLevenshteinDistance';
 select byteLevenshteinDistance(s1, s2) from t;
 
-SELECT byteEditDistance(randomString(power(2, 17)), 'abc'); -- { serverError 131 }
+SELECT byteEditDistance(randomString(power(2, 17)), 'abc'); -- { serverError TOO_LARGE_STRING_SIZE}
 
 drop table t;

From 886afd11b6b0181b8f2821d183abc08300e45556 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Wed, 27 Sep 2023 08:41:50 +0000
Subject: [PATCH 37/37] Fix expected results

---
 .../system-tables/information_schema.md       |  4 +-
 .../01161_information_schema.reference        | 42 +++++++-------
 .../0_stateless/01161_information_schema.sql  | 57 +++++++++++--------
 ...information_schema_show_database.reference |  8 +--
 4 files changed, 62 insertions(+), 49 deletions(-)

diff --git a/docs/en/operations/system-tables/information_schema.md b/docs/en/operations/system-tables/information_schema.md
index 8edbc506411..85f58efb240 100644
--- a/docs/en/operations/system-tables/information_schema.md
+++ b/docs/en/operations/system-tables/information_schema.md
@@ -293,7 +293,7 @@ is_trigger_deletable:       NO
 is_trigger_insertable_into: NO
 ```
 
-## KEY_COLUMN_USAGE (#key_column_usage)
+## KEY_COLUMN_USAGE {#key_column_usage}
 
 Contains columns from the [system.tables](../../operations/system-tables/tables.md) system table which are restricted by constraints.
 
@@ -352,7 +352,7 @@ referenced_table_name:         ᴺᵁᴸᴸ
 referenced_column_name:        ᴺᵁᴸᴸ
 ```
 
-## REFERENTIAL_CONSTRAINTS (#referential_constraints)
+## REFERENTIAL_CONSTRAINTS {#referential_constraints}
 
 Contains information about foreign keys. Currently returns an empty result (no rows) which is just enough to provide compatibility with 3rd party tools like Tableau Online.
 
diff --git a/tests/queries/0_stateless/01161_information_schema.reference b/tests/queries/0_stateless/01161_information_schema.reference
index 7531af6a365..a5ffe47a951 100644
--- a/tests/queries/0_stateless/01161_information_schema.reference
+++ b/tests/queries/0_stateless/01161_information_schema.reference
@@ -22,34 +22,36 @@ referential_constraints
 schemata
 tables
 views
+-- information_schema.schemata
 INFORMATION_SCHEMA	INFORMATION_SCHEMA	default	\N	\N	\N	\N	INFORMATION_SCHEMA	INFORMATION_SCHEMA	default	\N	\N	\N	\N
 information_schema	information_schema	default	\N	\N	\N	\N	information_schema	information_schema	default	\N	\N	\N	\N
-default	default	kcu	BASE TABLE	utf8mb4_0900_ai_ci		default	default	kcu	BASE TABLE	utf8mb4_0900_ai_ci	
-default	default	kcu2	BASE TABLE	utf8mb4_0900_ai_ci		default	default	kcu2	BASE TABLE	utf8mb4_0900_ai_ci	
-default	default	mv	VIEW	utf8mb4_0900_ai_ci		default	default	mv	VIEW	utf8mb4_0900_ai_ci	
-default	default	t	FOREIGN TABLE	utf8mb4_0900_ai_ci		default	default	t	FOREIGN TABLE	utf8mb4_0900_ai_ci	
-default	default	v	VIEW	utf8mb4_0900_ai_ci		default	default	v	VIEW	utf8mb4_0900_ai_ci	
-		tmp	LOCAL TEMPORARY	utf8mb4_0900_ai_ci				tmp	LOCAL TEMPORARY	utf8mb4_0900_ai_ci	
+-- information_schema.tables
+default	default	kcu1	BASE TABLE	0	utf8mb4_0900_ai_ci		default	default	kcu1	BASE TABLE	0	utf8mb4_0900_ai_ci	
+default	default	kcu2	BASE TABLE	0	utf8mb4_0900_ai_ci		default	default	kcu2	BASE TABLE	0	utf8mb4_0900_ai_ci	
+default	default	mv	VIEW	0	utf8mb4_0900_ai_ci		default	default	mv	VIEW	0	utf8mb4_0900_ai_ci	
+default	default	t	FOREIGN TABLE	0	utf8mb4_0900_ai_ci		default	default	t	FOREIGN TABLE	0	utf8mb4_0900_ai_ci	
+default	default	v	VIEW	\N	utf8mb4_0900_ai_ci		default	default	v	VIEW	\N	utf8mb4_0900_ai_ci	
+-- information_schema.views
 default	default	mv	SELECT * FROM system.one	NONE	NO	YES	NO	NO	NO	default	default	mv	SELECT * FROM system.one	NONE	NO	YES	NO	NO	NO
 default	default	v	SELECT n, f FROM default.t	NONE	NO	NO	NO	NO	NO	default	default	v	SELECT n, f FROM default.t	NONE	NO	NO	NO	NO	NO
-default	default	kcu	i	1		0	UInt32	\N	\N	32	2	0	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		UInt32	default	default	kcu	i	1		0	UInt32	\N	\N	32	2	0	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		UInt32
-default	default	kcu	s	2		0	String	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		String	default	default	kcu	s	2		0	String	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		String
-default	default	kcu2	i	1		0	UInt32	\N	\N	32	2	0	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		UInt32	default	default	kcu2	i	1		0	UInt32	\N	\N	32	2	0	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		UInt32
+-- information_schema.columns
+default	default	kcu1	i	1		0	UInt32	\N	\N	32	2	0	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		UInt32	default	default	kcu1	i	1		0	UInt32	\N	\N	32	2	0	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		UInt32
+default	default	kcu1	s	2		0	String	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		String	default	default	kcu1	s	2		0	String	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		String
 default	default	kcu2	d	2		0	Date	\N	\N	\N	\N	\N	0	\N	\N	\N	\N	\N	\N	\N	\N	\N		Date	default	default	kcu2	d	2		0	Date	\N	\N	\N	\N	\N	0	\N	\N	\N	\N	\N	\N	\N	\N	\N		Date
+default	default	kcu2	i	1		0	UInt32	\N	\N	32	2	0	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		UInt32	default	default	kcu2	i	1		0	UInt32	\N	\N	32	2	0	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		UInt32
 default	default	kcu2	u	3		0	UUID	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		UUID	default	default	kcu2	u	3		0	UUID	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		UUID
 default	default	mv	dummy	1		0	UInt8	\N	\N	8	2	0	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		UInt8	default	default	mv	dummy	1		0	UInt8	\N	\N	8	2	0	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		UInt8
-default	default	t	n	1		0	UInt64	\N	\N	64	2	0	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		UInt64	default	default	t	n	1		0	UInt64	\N	\N	64	2	0	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		UInt64
-default	default	t	f	2		0	Float32	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		Float32	default	default	t	f	2		0	Float32	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		Float32
-default	default	t	s	3		0	String	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		String	default	default	t	s	3		0	String	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		String
-default	default	t	fs	4		0	FixedString(42)	42	42	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		FixedString(42)	default	default	t	fs	4		0	FixedString(42)	42	42	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		FixedString(42)
 default	default	t	d	5		0	Decimal(9, 6)	\N	\N	9	10	6	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		Decimal(9, 6)	default	default	t	d	5		0	Decimal(9, 6)	\N	\N	9	10	6	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		Decimal(9, 6)
-default	default	v	n	1		1	Nullable(Int32)	\N	\N	32	2	0	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		Nullable(Int32)	default	default	v	n	1		1	Nullable(Int32)	\N	\N	32	2	0	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		Nullable(Int32)
+default	default	t	f	2		0	Float32	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		Float32	default	default	t	f	2		0	Float32	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		Float32
+default	default	t	fs	4		0	FixedString(42)	42	42	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		FixedString(42)	default	default	t	fs	4		0	FixedString(42)	42	42	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		FixedString(42)
+default	default	t	n	1		0	UInt64	\N	\N	64	2	0	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		UInt64	default	default	t	n	1		0	UInt64	\N	\N	64	2	0	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		UInt64
+default	default	t	s	3		0	String	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		String	default	default	t	s	3		0	String	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		String
 default	default	v	f	2		0	Float64	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		Float64	default	default	v	f	2		0	Float64	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		Float64
-		tmp	d	1		0	Date	\N	\N	\N	\N	\N	0	\N	\N	\N	\N	\N	\N	\N	\N	\N		Date			tmp	d	1		0	Date	\N	\N	\N	\N	\N	0	\N	\N	\N	\N	\N	\N	\N	\N	\N		Date
-		tmp	dt	2		0	DateTime	\N	\N	\N	\N	\N	0	\N	\N	\N	\N	\N	\N	\N	\N	\N		DateTime			tmp	dt	2		0	DateTime	\N	\N	\N	\N	\N	0	\N	\N	\N	\N	\N	\N	\N	\N	\N		DateTime
-		tmp	dtms	3		0	DateTime64(3)	\N	\N	\N	\N	\N	3	\N	\N	\N	\N	\N	\N	\N	\N	\N		DateTime64(3)			tmp	dtms	3		0	DateTime64(3)	\N	\N	\N	\N	\N	3	\N	\N	\N	\N	\N	\N	\N	\N	\N		DateTime64(3)
-1
-1
-def	default	PRIMARY	def	default	kcu	i	1	\N	\N	\N	\N	def	default	PRIMARY	def	default	kcu	i	1	\N	\N	\N	\N
+default	default	v	n	1		1	Nullable(Int32)	\N	\N	32	2	0	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		Nullable(Int32)	default	default	v	n	1		1	Nullable(Int32)	\N	\N	32	2	0	\N	\N	\N	\N	\N	\N	\N	\N	\N	\N		Nullable(Int32)
+-- information_schema.key_column_usage
+def	default	PRIMARY	def	default	kcu1	i	1	\N	\N	\N	\N	def	default	PRIMARY	def	default	kcu1	i	1	\N	\N	\N	\N
 def	default	PRIMARY	def	default	kcu2	d	1	\N	\N	\N	\N	def	default	PRIMARY	def	default	kcu2	d	1	\N	\N	\N	\N
 def	default	PRIMARY	def	default	kcu2	u	1	\N	\N	\N	\N	def	default	PRIMARY	def	default	kcu2	u	1	\N	\N	\N	\N
+-- information_schema.referential_constraints
+1
+1
diff --git a/tests/queries/0_stateless/01161_information_schema.sql b/tests/queries/0_stateless/01161_information_schema.sql
index 98a1113a786..c6a5ff4bd33 100644
--- a/tests/queries/0_stateless/01161_information_schema.sql
+++ b/tests/queries/0_stateless/01161_information_schema.sql
@@ -1,41 +1,52 @@
+-- Uppercase/lowercase are okay, mixed case isn't
 SHOW TABLES FROM information_schema;
 SHOW TABLES FROM INFORMATION_SCHEMA;
+SHOW TABLES FROM INFORMATION_schema; -- { serverError UNKNOWN_DATABASE }
 
-DROP TABLE IF EXISTS t;
 DROP VIEW IF EXISTS v;
+DROP TABLE IF EXISTS t;
 DROP VIEW IF EXISTS mv;
 DROP TABLE IF EXISTS tmp;
-DROP TABLE IF EXISTS kcu;
+DROP TABLE IF EXISTS kcu1;
 DROP TABLE IF EXISTS kcu2;
 
-CREATE TABLE t (n UInt64, f Float32, s String, fs FixedString(42), d Decimal(9, 6)) ENGINE=Memory;
+CREATE TABLE t (n UInt64, f Float32, s String, fs FixedString(42), d Decimal(9, 6)) ENGINE = Memory;
 CREATE VIEW v (n Nullable(Int32), f Float64) AS SELECT n, f FROM t;
-CREATE MATERIALIZED VIEW mv ENGINE=Null AS SELECT * FROM system.one;
+CREATE MATERIALIZED VIEW mv ENGINE = Null AS SELECT * FROM system.one;
 CREATE TEMPORARY TABLE tmp (d Date, dt DateTime, dtms DateTime64(3));
-CREATE TABLE kcu (i UInt32, s String) ENGINE MergeTree ORDER BY i;
+CREATE TABLE kcu1 (i UInt32, s String) ENGINE MergeTree ORDER BY i;
 CREATE TABLE kcu2 (i UInt32, d Date, u UUID) ENGINE MergeTree ORDER BY (u, d);
 
--- FIXME #28687
-SELECT * FROM information_schema.schemata WHERE schema_name ilike 'information_schema';
--- SELECT * FROM INFORMATION_SCHEMA.TABLES WHERE (TABLE_SCHEMA=currentDatabase() OR TABLE_SCHEMA='') AND TABLE_NAME NOT LIKE '%inner%';
-SELECT * FROM INFORMATION_SCHEMA.TABLES WHERE (table_schema = currentDatabase() OR table_schema = '') AND table_name NOT LIKE '%inner%';
-SELECT * FROM information_schema.views WHERE table_schema = currentDatabase();
--- SELECT * FROM INFORMATION_SCHEMA.COLUMNS WHERE (TABLE_SCHEMA=currentDatabase() OR TABLE_SCHEMA='') AND TABLE_NAME NOT LIKE '%inner%'
-SELECT * FROM INFORMATION_SCHEMA.COLUMNS WHERE (table_schema = currentDatabase() OR table_schema = '') AND table_name NOT LIKE '%inner%';
 
+SELECT '-- information_schema.schemata';
+SELECT * FROM information_schema.schemata WHERE schema_name ilike 'information_schema' ORDER BY schema_name;
+
+SELECT '-- information_schema.tables';
+SELECT * FROM information_schema.tables WHERE table_schema = currentDatabase() AND table_name NOT LIKE '%inner%' ORDER BY table_name;
+-- SELECT * FROM INFORMATION_SCHEMA.TABLES WHERE TABLE_SCHEMA=currentDatabase() ORDER BY schema_name;
+-- FIXME #28687
+
+SELECT '-- information_schema.views';
+SELECT * FROM information_schema.views WHERE table_schema = currentDatabase() ORDER BY table_name;
+-- SELECT * FROM INFORMATION_SCHEMA.COLUMNS WHERE (TABLE_SCHEMA=currentDatabase() OR TABLE_SCHEMA='') AND TABLE_NAME NOT LIKE '%inner%'
+
+SELECT '-- information_schema.columns';
+SELECT * FROM information_schema.columns WHERE table_schema = currentDatabase() AND table_name NOT LIKE '%inner%' ORDER BY table_name, column_name;
+
+SELECT '-- information_schema.key_column_usage';
+SELECT * FROM information_schema.key_column_usage WHERE table_schema = currentDatabase() AND table_name = 'kcu1' ORDER BY table_schema, column_name;
+SELECT * FROM information_schema.key_column_usage WHERE table_schema = currentDatabase() AND table_name = 'kcu2' ORDER BY table_schema, column_name;
+
+SELECT '-- information_schema.referential_constraints';
+SELECT * FROM information_schema.referential_constraints;
+--
 -- mixed upper/lowercase schema and table name:
 SELECT count() FROM information_schema.TABLES WHERE table_schema = currentDatabase() AND table_name = 't';
 SELECT count() FROM INFORMATION_SCHEMA.tables WHERE table_schema = currentDatabase() AND table_name = 't';
-SELECT count() FROM INFORMATION_schema.tables WHERE table_schema = currentDatabase() AND table_name = 't'; -- { serverError UNKNOWN_DATABASE }
 SELECT count() FROM information_schema.taBLES WHERE table_schema  =currentDatabase() AND table_name = 't'; -- { serverError UNKNOWN_TABLE }
 
-SELECT * FROM information_schema.key_column_usage WHERE table_schema = currentDatabase() AND table_name = 'kcu';
-SELECT * FROM information_schema.key_column_usage WHERE table_schema = currentDatabase() AND table_name = 'kcu2';
-
-SELECT * FROM information_schema.referential_constraints;
-
-drop view mv;
-drop view v;
-drop table t;
-drop table kcu;
-drop table kcu2;
+DROP VIEW mv;
+DROP VIEW v;
+DROP TABLE t;
+DROP TABLE kcu1;
+DROP TABLE kcu2;
diff --git a/tests/queries/0_stateless/02206_information_schema_show_database.reference b/tests/queries/0_stateless/02206_information_schema_show_database.reference
index b0eb08e00f5..3e46e6395ad 100644
--- a/tests/queries/0_stateless/02206_information_schema_show_database.reference
+++ b/tests/queries/0_stateless/02206_information_schema_show_database.reference
@@ -1,6 +1,6 @@
 CREATE DATABASE INFORMATION_SCHEMA\nENGINE = Memory
 CREATE VIEW INFORMATION_SCHEMA.COLUMNS\n(\n    `table_catalog` String,\n    `table_schema` String,\n    `table_name` String,\n    `column_name` String,\n    `ordinal_position` UInt64,\n    `column_default` String,\n    `is_nullable` String,\n    `data_type` String,\n    `character_maximum_length` Nullable(UInt64),\n    `character_octet_length` Nullable(UInt64),\n    `numeric_precision` Nullable(UInt64),\n    `numeric_precision_radix` Nullable(UInt64),\n    `numeric_scale` Nullable(UInt64),\n    `datetime_precision` Nullable(UInt64),\n    `character_set_catalog` Nullable(String),\n    `character_set_schema` Nullable(String),\n    `character_set_name` Nullable(String),\n    `collation_catalog` Nullable(String),\n    `collation_schema` Nullable(String),\n    `collation_name` Nullable(String),\n    `domain_catalog` Nullable(String),\n    `domain_schema` Nullable(String),\n    `domain_name` Nullable(String),\n    `column_comment` String,\n    `column_type` String,\n    `TABLE_CATALOG` String,\n    `TABLE_SCHEMA` String,\n    `TABLE_NAME` String,\n    `COLUMN_NAME` String,\n    `ORDINAL_POSITION` UInt64,\n    `COLUMN_DEFAULT` String,\n    `IS_NULLABLE` String,\n    `DATA_TYPE` String,\n    `CHARACTER_MAXIMUM_LENGTH` Nullable(UInt64),\n    `CHARACTER_OCTET_LENGTH` Nullable(UInt64),\n    `NUMERIC_PRECISION` Nullable(UInt64),\n    `NUMERIC_PRECISION_RADIX` Nullable(UInt64),\n    `NUMERIC_SCALE` Nullable(UInt64),\n    `DATETIME_PRECISION` Nullable(UInt64),\n    `CHARACTER_SET_CATALOG` Nullable(String),\n    `CHARACTER_SET_SCHEMA` Nullable(String),\n    `CHARACTER_SET_NAME` Nullable(String),\n    `COLLATION_CATALOG` Nullable(String),\n    `COLLATION_SCHEMA` Nullable(String),\n    `COLLATION_NAME` Nullable(String),\n    `DOMAIN_CATALOG` Nullable(String),\n    `DOMAIN_SCHEMA` Nullable(String),\n    `DOMAIN_NAME` Nullable(String),\n    `COLUMN_COMMENT` String,\n    `COLUMN_TYPE` String\n) AS\nSELECT\n    database AS table_catalog,\n    database AS table_schema,\n    table AS table_name,\n    name AS column_name,\n    position AS ordinal_position,\n    default_expression AS column_default,\n    type LIKE \'Nullable(%)\' AS is_nullable,\n    type AS data_type,\n    character_octet_length AS character_maximum_length,\n    character_octet_length,\n    numeric_precision,\n    numeric_precision_radix,\n    numeric_scale,\n    datetime_precision,\n    NULL AS character_set_catalog,\n    NULL AS character_set_schema,\n    NULL AS character_set_name,\n    NULL AS collation_catalog,\n    NULL AS collation_schema,\n    NULL AS collation_name,\n    NULL AS domain_catalog,\n    NULL AS domain_schema,\n    NULL AS domain_name,\n    comment AS column_comment,\n    type AS column_type,\n    table_catalog AS TABLE_CATALOG,\n    table_schema AS TABLE_SCHEMA,\n    table_name AS TABLE_NAME,\n    column_name AS COLUMN_NAME,\n    ordinal_position AS ORDINAL_POSITION,\n    column_default AS COLUMN_DEFAULT,\n    is_nullable AS IS_NULLABLE,\n    data_type AS DATA_TYPE,\n    character_maximum_length AS CHARACTER_MAXIMUM_LENGTH,\n    character_octet_length AS CHARACTER_OCTET_LENGTH,\n    numeric_precision AS NUMERIC_PRECISION,\n    numeric_precision_radix AS NUMERIC_PRECISION_RADIX,\n    numeric_scale AS NUMERIC_SCALE,\n    datetime_precision AS DATETIME_PRECISION,\n    character_set_catalog AS CHARACTER_SET_CATALOG,\n    character_set_schema AS CHARACTER_SET_SCHEMA,\n    character_set_name AS CHARACTER_SET_NAME,\n    collation_catalog AS COLLATION_CATALOG,\n    collation_schema AS COLLATION_SCHEMA,\n    collation_name AS COLLATION_NAME,\n    domain_catalog AS DOMAIN_CATALOG,\n    domain_schema AS DOMAIN_SCHEMA,\n    domain_name AS DOMAIN_NAME,\n    column_comment AS COLUMN_COMMENT,\n    column_type AS COLUMN_TYPE\nFROM system.columns
-CREATE VIEW INFORMATION_SCHEMA.TABLES (`table_catalog` String, `table_schema` String, `table_name` String, `table_type` String, `table_collation` Nullable(String), `table_comment` Nullable(String), `TABLE_CATALOG` String, `TABLE_SCHEMA` String, `TABLE_NAME` String, `TABLE_TYPE` String, `TABLE_COLLATION` Nullable(String), `TABLE_COMMENT` Nullable(String)) AS SELECT database AS table_catalog, database AS table_schema, name AS table_name, multiIf(is_temporary, \'LOCAL TEMPORARY\', engine LIKE \'%View\', \'VIEW\', engine LIKE \'System%\', \'SYSTEM VIEW\', has_own_data = 0, \'FOREIGN TABLE\', \'BASE TABLE\') AS table_type, \'utf8mb4_0900_ai_ci\' AS table_collation, comment AS table_comment, table_catalog AS TABLE_CATALOG, table_schema AS TABLE_SCHEMA, table_name AS TABLE_NAME, table_type AS TABLE_TYPE, table_collation AS TABLE_COLLATION, table_comment AS TABLE_COMMENT FROM system.tables
-CREATE VIEW INFORMATION_SCHEMA.tables (`table_catalog` String, `table_schema` String, `table_name` String, `table_type` String, `table_collation` Nullable(String), `table_comment` Nullable(String), `TABLE_CATALOG` String, `TABLE_SCHEMA` String, `TABLE_NAME` String, `TABLE_TYPE` String, `TABLE_COLLATION` Nullable(String), `TABLE_COMMENT` Nullable(String)) AS SELECT database AS table_catalog, database AS table_schema, name AS table_name, multiIf(is_temporary, \'LOCAL TEMPORARY\', engine LIKE \'%View\', \'VIEW\', engine LIKE \'System%\', \'SYSTEM VIEW\', has_own_data = 0, \'FOREIGN TABLE\', \'BASE TABLE\') AS table_type, \'utf8mb4_0900_ai_ci\' AS table_collation, comment AS table_comment, table_catalog AS TABLE_CATALOG, table_schema AS TABLE_SCHEMA, table_name AS TABLE_NAME, table_type AS TABLE_TYPE, table_collation AS TABLE_COLLATION, table_comment AS TABLE_COMMENT FROM system.tables
-CREATE VIEW information_schema.TABLES (`table_catalog` String, `table_schema` String, `table_name` String, `table_type` String, `table_collation` Nullable(String), `table_comment` Nullable(String), `TABLE_CATALOG` String, `TABLE_SCHEMA` String, `TABLE_NAME` String, `TABLE_TYPE` String, `TABLE_COLLATION` Nullable(String), `TABLE_COMMENT` Nullable(String)) AS SELECT database AS table_catalog, database AS table_schema, name AS table_name, multiIf(is_temporary, \'LOCAL TEMPORARY\', engine LIKE \'%View\', \'VIEW\', engine LIKE \'System%\', \'SYSTEM VIEW\', has_own_data = 0, \'FOREIGN TABLE\', \'BASE TABLE\') AS table_type, \'utf8mb4_0900_ai_ci\' AS table_collation, comment AS table_comment, table_catalog AS TABLE_CATALOG, table_schema AS TABLE_SCHEMA, table_name AS TABLE_NAME, table_type AS TABLE_TYPE, table_collation AS TABLE_COLLATION, table_comment AS TABLE_COMMENT FROM system.tables
-CREATE VIEW information_schema.tables (`table_catalog` String, `table_schema` String, `table_name` String, `table_type` String, `table_collation` Nullable(String), `table_comment` Nullable(String), `TABLE_CATALOG` String, `TABLE_SCHEMA` String, `TABLE_NAME` String, `TABLE_TYPE` String, `TABLE_COLLATION` Nullable(String), `TABLE_COMMENT` Nullable(String)) AS SELECT database AS table_catalog, database AS table_schema, name AS table_name, multiIf(is_temporary, \'LOCAL TEMPORARY\', engine LIKE \'%View\', \'VIEW\', engine LIKE \'System%\', \'SYSTEM VIEW\', has_own_data = 0, \'FOREIGN TABLE\', \'BASE TABLE\') AS table_type, \'utf8mb4_0900_ai_ci\' AS table_collation, comment AS table_comment, table_catalog AS TABLE_CATALOG, table_schema AS TABLE_SCHEMA, table_name AS TABLE_NAME, table_type AS TABLE_TYPE, table_collation AS TABLE_COLLATION, table_comment AS TABLE_COMMENT FROM system.tables
+CREATE VIEW INFORMATION_SCHEMA.TABLES (`table_catalog` String, `table_schema` String, `table_name` String, `table_type` String, `data_length` Nullable(UInt64), `table_collation` Nullable(String), `table_comment` Nullable(String), `TABLE_CATALOG` String, `TABLE_SCHEMA` String, `TABLE_NAME` String, `TABLE_TYPE` String, `DATA_LENGTH` Nullable(UInt64), `TABLE_COLLATION` Nullable(String), `TABLE_COMMENT` Nullable(String)) AS SELECT database AS table_catalog, database AS table_schema, name AS table_name, multiIf(is_temporary, \'LOCAL TEMPORARY\', engine LIKE \'%View\', \'VIEW\', engine LIKE \'System%\', \'SYSTEM VIEW\', has_own_data = 0, \'FOREIGN TABLE\', \'BASE TABLE\') AS table_type, total_bytes AS data_length, \'utf8mb4_0900_ai_ci\' AS table_collation, comment AS table_comment, table_catalog AS TABLE_CATALOG, table_schema AS TABLE_SCHEMA, table_name AS TABLE_NAME, table_type AS TABLE_TYPE, data_length AS DATA_LENGTH, table_collation AS TABLE_COLLATION, table_comment AS TABLE_COMMENT FROM system.tables
+CREATE VIEW INFORMATION_SCHEMA.tables (`table_catalog` String, `table_schema` String, `table_name` String, `table_type` String, `data_length` Nullable(UInt64), `table_collation` Nullable(String), `table_comment` Nullable(String), `TABLE_CATALOG` String, `TABLE_SCHEMA` String, `TABLE_NAME` String, `TABLE_TYPE` String, `DATA_LENGTH` Nullable(UInt64), `TABLE_COLLATION` Nullable(String), `TABLE_COMMENT` Nullable(String)) AS SELECT database AS table_catalog, database AS table_schema, name AS table_name, multiIf(is_temporary, \'LOCAL TEMPORARY\', engine LIKE \'%View\', \'VIEW\', engine LIKE \'System%\', \'SYSTEM VIEW\', has_own_data = 0, \'FOREIGN TABLE\', \'BASE TABLE\') AS table_type, total_bytes AS data_length, \'utf8mb4_0900_ai_ci\' AS table_collation, comment AS table_comment, table_catalog AS TABLE_CATALOG, table_schema AS TABLE_SCHEMA, table_name AS TABLE_NAME, table_type AS TABLE_TYPE, data_length AS DATA_LENGTH, table_collation AS TABLE_COLLATION, table_comment AS TABLE_COMMENT FROM system.tables
+CREATE VIEW information_schema.TABLES (`table_catalog` String, `table_schema` String, `table_name` String, `table_type` String, `data_length` Nullable(UInt64), `table_collation` Nullable(String), `table_comment` Nullable(String), `TABLE_CATALOG` String, `TABLE_SCHEMA` String, `TABLE_NAME` String, `TABLE_TYPE` String, `DATA_LENGTH` Nullable(UInt64), `TABLE_COLLATION` Nullable(String), `TABLE_COMMENT` Nullable(String)) AS SELECT database AS table_catalog, database AS table_schema, name AS table_name, multiIf(is_temporary, \'LOCAL TEMPORARY\', engine LIKE \'%View\', \'VIEW\', engine LIKE \'System%\', \'SYSTEM VIEW\', has_own_data = 0, \'FOREIGN TABLE\', \'BASE TABLE\') AS table_type, total_bytes AS data_length, \'utf8mb4_0900_ai_ci\' AS table_collation, comment AS table_comment, table_catalog AS TABLE_CATALOG, table_schema AS TABLE_SCHEMA, table_name AS TABLE_NAME, table_type AS TABLE_TYPE, data_length AS DATA_LENGTH, table_collation AS TABLE_COLLATION, table_comment AS TABLE_COMMENT FROM system.tables
+CREATE VIEW information_schema.tables (`table_catalog` String, `table_schema` String, `table_name` String, `table_type` String, `data_length` Nullable(UInt64), `table_collation` Nullable(String), `table_comment` Nullable(String), `TABLE_CATALOG` String, `TABLE_SCHEMA` String, `TABLE_NAME` String, `TABLE_TYPE` String, `DATA_LENGTH` Nullable(UInt64), `TABLE_COLLATION` Nullable(String), `TABLE_COMMENT` Nullable(String)) AS SELECT database AS table_catalog, database AS table_schema, name AS table_name, multiIf(is_temporary, \'LOCAL TEMPORARY\', engine LIKE \'%View\', \'VIEW\', engine LIKE \'System%\', \'SYSTEM VIEW\', has_own_data = 0, \'FOREIGN TABLE\', \'BASE TABLE\') AS table_type, total_bytes AS data_length, \'utf8mb4_0900_ai_ci\' AS table_collation, comment AS table_comment, table_catalog AS TABLE_CATALOG, table_schema AS TABLE_SCHEMA, table_name AS TABLE_NAME, table_type AS TABLE_TYPE, data_length AS DATA_LENGTH, table_collation AS TABLE_COLLATION, table_comment AS TABLE_COMMENT FROM system.tables