From 30f48e18938bbc5683d781f1cbfe7bfcf3fec8d9 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Sun, 28 Jan 2024 23:54:35 +0100
Subject: [PATCH 01/73] Use MergeTree as a default table engine

---
 src/Core/Settings.h               | 2 +-
 src/Core/SettingsChangesHistory.h | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/src/Core/Settings.h b/src/Core/Settings.h
index e0b3ca39899..4460a365846 100644
--- a/src/Core/Settings.h
+++ b/src/Core/Settings.h
@@ -558,7 +558,7 @@ class IColumn;
     M(UInt64, min_free_disk_space_for_temporary_data, 0, "The minimum disk space to keep while writing temporary data used in external sorting and aggregation.", 0) \
     \
     M(DefaultTableEngine, default_temporary_table_engine, DefaultTableEngine::Memory, "Default table engine used when ENGINE is not set in CREATE TEMPORARY statement.",0) \
-    M(DefaultTableEngine, default_table_engine, DefaultTableEngine::None, "Default table engine used when ENGINE is not set in CREATE statement.",0) \
+    M(DefaultTableEngine, default_table_engine, DefaultTableEngine::MergeTree, "Default table engine used when ENGINE is not set in CREATE statement.",0) \
     M(Bool, show_table_uuid_in_table_create_query_if_not_nil, false, "For tables in databases with Engine=Atomic show UUID of the table in its CREATE query.", 0) \
     M(Bool, database_atomic_wait_for_drop_and_detach_synchronously, false, "When executing DROP or DETACH TABLE in Atomic database, wait for table data to be finally dropped or detached.", 0) \
     M(Bool, enable_scalar_subquery_optimization, true, "If it is set to true, prevent scalar subqueries from (de)serializing large scalar values and possibly avoid running the same subquery more than once.", 0) \
diff --git a/src/Core/SettingsChangesHistory.h b/src/Core/SettingsChangesHistory.h
index dff0ebb759c..7bdab886934 100644
--- a/src/Core/SettingsChangesHistory.h
+++ b/src/Core/SettingsChangesHistory.h
@@ -100,6 +100,7 @@ static std::map<ClickHouseVersion, SettingsChangesHistory::SettingsChanges> sett
               {"function_visible_width_behavior", 0, 1, "We changed the default behavior of `visibleWidth` to be more precise"},
               {"max_estimated_execution_time", 0, 0, "Separate max_execution_time and max_estimated_execution_time"},
               {"iceberg_engine_ignore_schema_evolution", false, false, "Allow to ignore schema evolution in Iceberg table engine"},
+              {"default_table_engine", DefaultTableEngine::None, DefaultTableEngine::MergeTree, "Set default table engine to MergeTree for better usability"},
               {"optimize_injective_functions_in_group_by", false, true, "Replace injective functions by it's arguments in GROUP BY section in analyzer"}}},
     {"23.12", {{"allow_suspicious_ttl_expressions", true, false, "It is a new setting, and in previous versions the behavior was equivalent to allowing."},
               {"input_format_parquet_allow_missing_columns", false, true, "Allow missing columns in Parquet files by default"},

From 3f1ec9a9881949c7e676cf11f35fb75df3b95f78 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Tue, 30 Jan 2024 04:23:16 +0100
Subject: [PATCH 02/73] Fix error

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

diff --git a/src/Core/SettingsChangesHistory.h b/src/Core/SettingsChangesHistory.h
index 692d8fc6360..53b14ddc385 100644
--- a/src/Core/SettingsChangesHistory.h
+++ b/src/Core/SettingsChangesHistory.h
@@ -102,7 +102,7 @@ static std::map<ClickHouseVersion, SettingsChangesHistory::SettingsChanges> sett
               {"function_visible_width_behavior", 0, 1, "We changed the default behavior of `visibleWidth` to be more precise"},
               {"max_estimated_execution_time", 0, 0, "Separate max_execution_time and max_estimated_execution_time"},
               {"iceberg_engine_ignore_schema_evolution", false, false, "Allow to ignore schema evolution in Iceberg table engine"},
-              {"default_table_engine", DefaultTableEngine::None, DefaultTableEngine::MergeTree, "Set default table engine to MergeTree for better usability"},
+              {"default_table_engine", "None", "MergeTree", "Set default table engine to MergeTree for better usability"},
               {"optimize_injective_functions_in_group_by", false, true, "Replace injective functions by it's arguments in GROUP BY section in analyzer"},
               {"update_insert_deduplication_token_in_dependent_materialized_views", false, false, "Allow to update insert deduplication token with table identifier during insert in dependent materialized views"},
               {"azure_max_unexpected_write_error_retries", 4, 4, "The maximum number of retries in case of unexpected errors during Azure blob storage write"}}},

From 17ab2674f4c8ad7a09194659e0a0c86d4440f203 Mon Sep 17 00:00:00 2001
From: Nikita Taranov <nikita.taranov@clickhouse.com>
Date: Tue, 30 Jan 2024 20:35:10 +0100
Subject: [PATCH 03/73] impl

---
 src/Common/ElapsedTimeProfileEventIncrement.h |  3 +-
 src/Common/ProfileEvents.cpp                  |  7 +++
 .../MergeTreeDataPartWriterOnDisk.cpp         | 15 ++++++
 .../MergeTree/MergeTreeDataPartWriterOnDisk.h |  2 +
 .../MergeTree/MergeTreeDataWriter.cpp         | 47 ++++++++++++++-----
 5 files changed, 60 insertions(+), 14 deletions(-)

diff --git a/src/Common/ElapsedTimeProfileEventIncrement.h b/src/Common/ElapsedTimeProfileEventIncrement.h
index b30afd24a4c..731295a4cfd 100644
--- a/src/Common/ElapsedTimeProfileEventIncrement.h
+++ b/src/Common/ElapsedTimeProfileEventIncrement.h
@@ -14,12 +14,13 @@ enum Time
     Seconds,
 };
 
-template <Time time>
+template <Time unit>
 struct ProfileEventTimeIncrement
 {
     explicit ProfileEventTimeIncrement<time>(ProfileEvents::Event event_)
         : event(event_), watch(CLOCK_MONOTONIC) {}
 
+    template <Time time = unit>
     UInt64 elapsed()
     {
         if constexpr (time == Time::Nanoseconds)
diff --git a/src/Common/ProfileEvents.cpp b/src/Common/ProfileEvents.cpp
index 8782f895f3f..ef0606162a2 100644
--- a/src/Common/ProfileEvents.cpp
+++ b/src/Common/ProfileEvents.cpp
@@ -197,6 +197,13 @@
     M(MergeTreeDataWriterBlocks, "Number of blocks INSERTed to MergeTree tables. Each block forms a data part of level zero.") \
     M(MergeTreeDataWriterBlocksAlreadySorted, "Number of blocks INSERTed to MergeTree tables that appeared to be already sorted.") \
     \
+    M(MergeTreeDataWriterSecondaryIndicesCalculationMicroseconds, "Time spent calculating secondary indices") \
+    M(MergeTreeDataWriterPrimaryKeyCalculationMicroseconds, "Time spent calculating primary key") \
+    M(MergeTreeDataWriterMergingBlocksMicroseconds, "Time spent merging input blocks (for special MergeTree engines)") \
+    M(MergeTreeDataWriterProjectionsCalculationMicroseconds, "Time spent calculating projections") \
+    M(MergeTreeDataProjectionWriterPrimaryKeyCalculationMicroseconds, "Time spent calculating primary key") \
+    M(MergeTreeDataProjectionWriterMergingBlocksMicroseconds, "Time spent merging input blocks") \
+    \
     M(InsertedWideParts, "Number of parts inserted in Wide format.") \
     M(InsertedCompactParts, "Number of parts inserted in Compact format.") \
     M(MergedIntoWideParts, "Number of parts merged into Wide format.") \
diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp
index 6e544b4a35a..eb9e2f52ad9 100644
--- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp
+++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp
@@ -1,6 +1,13 @@
 #include <Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h>
 #include <Storages/MergeTree/MergeTreeIndexInverted.h>
+#include <Common/ElapsedTimeProfileEventIncrement.h>
 #include <Common/MemoryTrackerBlockerInThread.h>
+#include <Common/logger_useful.h>
+
+namespace ProfileEvents
+{
+extern const Event MergeTreeDataWriterSecondaryIndicesCalculationMicroseconds;
+}
 
 namespace DB
 {
@@ -148,6 +155,7 @@ MergeTreeDataPartWriterOnDisk::MergeTreeDataPartWriterOnDisk(
     , default_codec(default_codec_)
     , compute_granularity(index_granularity.empty())
     , compress_primary_key(settings.compress_primary_key)
+    , log(getLogger(storage.getLogName() + " (DataPartWriter)"))
 {
     if (settings.blocks_are_granules_size && !index_granularity.empty())
         throw Exception(ErrorCodes::LOGICAL_ERROR,
@@ -354,6 +362,7 @@ void MergeTreeDataPartWriterOnDisk::calculateAndSerializeSkipIndices(const Block
             store = it->second;
         }
 
+        size_t index_build_us = 0;
         for (const auto & granule : granules_to_write)
         {
             if (skip_index_accumulated_marks[i] == index_helper->index.granularity)
@@ -378,11 +387,17 @@ void MergeTreeDataPartWriterOnDisk::calculateAndSerializeSkipIndices(const Block
                     writeBinaryLittleEndian(1UL, marks_out);
             }
 
+            ProfileEventTimeIncrement<Microseconds> watch(ProfileEvents::MergeTreeDataWriterSecondaryIndicesCalculationMicroseconds);
+
             size_t pos = granule.start_row;
             skip_indices_aggregators[i]->update(skip_indexes_block, &pos, granule.rows_to_write);
             if (granule.is_complete)
                 ++skip_index_accumulated_marks[i];
+
+            index_build_us += watch.elapsed<Microseconds>();
         }
+        // clang-format off
+        LOG_DEBUG(log, "Spent {} ms calculating index {} for the part {}", index_build_us / 1000, skip_indices[i]->index.name, data_part->name);
     }
 }
 
diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h
index 4d081778e68..5292c0d5590 100644
--- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h
+++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h
@@ -190,6 +190,8 @@ private:
     void initStatistics();
 
     virtual void fillIndexGranularity(size_t index_granularity_for_block, size_t rows_in_block) = 0;
+
+    LoggerPtr log;
 };
 
 }
diff --git a/src/Storages/MergeTree/MergeTreeDataWriter.cpp b/src/Storages/MergeTree/MergeTreeDataWriter.cpp
index ce3015c5dcb..cba1131bd3e 100644
--- a/src/Storages/MergeTree/MergeTreeDataWriter.cpp
+++ b/src/Storages/MergeTree/MergeTreeDataWriter.cpp
@@ -1,21 +1,22 @@
-#include <Storages/MergeTree/MergeTreeDataWriter.h>
-#include <Storages/MergeTree/MergedBlockOutputStream.h>
-#include <Storages/MergeTree/DataPartStorageOnDiskFull.h>
 #include <Columns/ColumnConst.h>
-#include <Common/OpenTelemetryTraceContext.h>
-#include <Common/HashTable/HashMap.h>
-#include <Common/Exception.h>
+#include <DataTypes/DataTypeDate.h>
+#include <DataTypes/DataTypeDateTime.h>
+#include <DataTypes/ObjectUtils.h>
 #include <Disks/createVolume.h>
+#include <IO/HashingWriteBuffer.h>
+#include <IO/WriteHelpers.h>
 #include <Interpreters/AggregationCommon.h>
 #include <Interpreters/Context.h>
 #include <Interpreters/MergeTreeTransaction.h>
-#include <IO/HashingWriteBuffer.h>
-#include <DataTypes/DataTypeDateTime.h>
-#include <DataTypes/DataTypeDate.h>
-#include <DataTypes/ObjectUtils.h>
-#include <IO/WriteHelpers.h>
-#include <Common/typeid_cast.h>
 #include <Processors/TTL/ITTLAlgorithm.h>
+#include <Storages/MergeTree/DataPartStorageOnDiskFull.h>
+#include <Storages/MergeTree/MergeTreeDataWriter.h>
+#include <Storages/MergeTree/MergedBlockOutputStream.h>
+#include <Common/ElapsedTimeProfileEventIncrement.h>
+#include <Common/Exception.h>
+#include <Common/HashTable/HashMap.h>
+#include <Common/OpenTelemetryTraceContext.h>
+#include <Common/typeid_cast.h>
 
 #include <Parsers/queryToString.h>
 
@@ -35,11 +36,16 @@ namespace ProfileEvents
     extern const Event MergeTreeDataWriterRows;
     extern const Event MergeTreeDataWriterUncompressedBytes;
     extern const Event MergeTreeDataWriterCompressedBytes;
+    extern const Event MergeTreeDataWriterPrimaryKeyCalculationMicroseconds;
+    extern const Event MergeTreeDataWriterMergingBlocksMicroseconds;
+    extern const Event MergeTreeDataWriterProjectionsCalculationMicroseconds;
     extern const Event MergeTreeDataProjectionWriterBlocks;
     extern const Event MergeTreeDataProjectionWriterBlocksAlreadySorted;
     extern const Event MergeTreeDataProjectionWriterRows;
     extern const Event MergeTreeDataProjectionWriterUncompressedBytes;
     extern const Event MergeTreeDataProjectionWriterCompressedBytes;
+    extern const Event MergeTreeDataProjectionWriterPrimaryKeyCalculationMicroseconds;
+    extern const Event MergeTreeDataProjectionWriterMergingBlocksMicroseconds;
     extern const Event RejectedInserts;
 }
 
@@ -472,6 +478,8 @@ MergeTreeDataWriter::TemporaryPart MergeTreeDataWriter::writeTempPartImpl(
     IColumn::Permutation perm;
     if (!sort_description.empty())
     {
+        ProfileEventTimeIncrement<Microseconds> watch(ProfileEvents::MergeTreeDataWriterPrimaryKeyCalculationMicroseconds);
+
         if (!isAlreadySorted(block, sort_description))
         {
             stableGetPermutation(block, sort_description, perm);
@@ -483,7 +491,10 @@ MergeTreeDataWriter::TemporaryPart MergeTreeDataWriter::writeTempPartImpl(
 
     Names partition_key_columns = metadata_snapshot->getPartitionKey().column_names;
     if (context->getSettingsRef().optimize_on_insert)
+    {
+        ProfileEventTimeIncrement<Microseconds> watch(ProfileEvents::MergeTreeDataWriterMergingBlocksMicroseconds);
         block = mergeBlock(block, sort_description, partition_key_columns, perm_ptr, data.merging_params);
+    }
 
     /// Size of part would not be greater than block.bytes() + epsilon
     size_t expected_size = block.bytes();
@@ -588,7 +599,13 @@ MergeTreeDataWriter::TemporaryPart MergeTreeDataWriter::writeTempPartImpl(
 
     for (const auto & projection : metadata_snapshot->getProjections())
     {
-        auto projection_block = projection.calculate(block, context);
+        Block projection_block;
+        {
+            ProfileEventTimeIncrement<Microseconds> watch(ProfileEvents::MergeTreeDataWriterProjectionsCalculationMicroseconds);
+            projection_block = projection.calculate(block, context);
+            LOG_DEBUG(log, "Spent {} ms calculating projection {} for the part {}", watch.elapsed(), projection.name, new_data_part->name);
+        }
+
         if (projection_block.rows())
         {
             auto proj_temp_part = writeProjectionPart(data, log, projection_block, projection, new_data_part.get());
@@ -685,6 +702,8 @@ MergeTreeDataWriter::TemporaryPart MergeTreeDataWriter::writeProjectionPartImpl(
     IColumn::Permutation perm;
     if (!sort_description.empty())
     {
+        ProfileEventTimeIncrement<Microseconds> watch(ProfileEvents::MergeTreeDataProjectionWriterPrimaryKeyCalculationMicroseconds);
+
         if (!isAlreadySorted(block, sort_description))
         {
             stableGetPermutation(block, sort_description, perm);
@@ -696,6 +715,8 @@ MergeTreeDataWriter::TemporaryPart MergeTreeDataWriter::writeProjectionPartImpl(
 
     if (projection.type == ProjectionDescription::Type::Aggregate)
     {
+        ProfileEventTimeIncrement<Microseconds> watch(ProfileEvents::MergeTreeDataProjectionWriterMergingBlocksMicroseconds);
+
         MergeTreeData::MergingParams projection_merging_params;
         projection_merging_params.mode = MergeTreeData::MergingParams::Aggregating;
         block = mergeBlock(block, sort_description, {}, perm_ptr, projection_merging_params);

From 1be8b61c5b0ca501588c618c986dfb32e1586a1b Mon Sep 17 00:00:00 2001
From: Nikita Taranov <nikita.taranov@clickhouse.com>
Date: Tue, 30 Jan 2024 20:50:03 +0100
Subject: [PATCH 04/73] add test

---
 .../MergeTreeDataPartWriterOnDisk.cpp         |  1 -
 .../MergeTree/MergeTreeDataWriter.cpp         |  2 +-
 ...2_perf_introspection_for_inserts.reference |  1 +
 .../02982_perf_introspection_for_inserts.sh   | 46 +++++++++++++++++++
 4 files changed, 48 insertions(+), 2 deletions(-)
 create mode 100644 tests/queries/0_stateless/02982_perf_introspection_for_inserts.reference
 create mode 100755 tests/queries/0_stateless/02982_perf_introspection_for_inserts.sh

diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp
index eb9e2f52ad9..153bcaa6320 100644
--- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp
+++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp
@@ -396,7 +396,6 @@ void MergeTreeDataPartWriterOnDisk::calculateAndSerializeSkipIndices(const Block
 
             index_build_us += watch.elapsed<Microseconds>();
         }
-        // clang-format off
         LOG_DEBUG(log, "Spent {} ms calculating index {} for the part {}", index_build_us / 1000, skip_indices[i]->index.name, data_part->name);
     }
 }
diff --git a/src/Storages/MergeTree/MergeTreeDataWriter.cpp b/src/Storages/MergeTree/MergeTreeDataWriter.cpp
index cba1131bd3e..8f522832bce 100644
--- a/src/Storages/MergeTree/MergeTreeDataWriter.cpp
+++ b/src/Storages/MergeTree/MergeTreeDataWriter.cpp
@@ -603,7 +603,7 @@ MergeTreeDataWriter::TemporaryPart MergeTreeDataWriter::writeTempPartImpl(
         {
             ProfileEventTimeIncrement<Microseconds> watch(ProfileEvents::MergeTreeDataWriterProjectionsCalculationMicroseconds);
             projection_block = projection.calculate(block, context);
-            LOG_DEBUG(log, "Spent {} ms calculating projection {} for the part {}", watch.elapsed(), projection.name, new_data_part->name);
+            LOG_DEBUG(log, "Spent {} ms calculating projection {} for the part {}", watch.elapsed() / 1000, projection.name, new_data_part->name);
         }
 
         if (projection_block.rows())
diff --git a/tests/queries/0_stateless/02982_perf_introspection_for_inserts.reference b/tests/queries/0_stateless/02982_perf_introspection_for_inserts.reference
new file mode 100644
index 00000000000..50d4d226b46
--- /dev/null
+++ b/tests/queries/0_stateless/02982_perf_introspection_for_inserts.reference
@@ -0,0 +1 @@
+1	1	1	1	1
diff --git a/tests/queries/0_stateless/02982_perf_introspection_for_inserts.sh b/tests/queries/0_stateless/02982_perf_introspection_for_inserts.sh
new file mode 100755
index 00000000000..a45f88fc10f
--- /dev/null
+++ b/tests/queries/0_stateless/02982_perf_introspection_for_inserts.sh
@@ -0,0 +1,46 @@
+#!/usr/bin/env bash
+
+CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
+# shellcheck source=../shell_config.sh
+. "$CURDIR"/../shell_config.sh
+
+
+$CLICKHOUSE_CLIENT -q """
+CREATE TABLE t02982
+(
+    n UInt64,
+    s Nullable(String),
+    INDEX idx1 n TYPE minmax GRANULARITY 2,
+    INDEX idx2 n * length(s) TYPE set(1000) GRANULARITY 2,
+    PROJECTION pr_sort
+    (
+        SELECT
+            n,
+            sum(length(s))
+        GROUP BY n
+    )
+)
+ENGINE = MergeTree
+ORDER BY n;
+"""
+
+query_id=$RANDOM
+
+$CLICKHOUSE_CLIENT --query_id $query_id -q """
+INSERT INTO t02982 SELECT
+    number,
+    'a'
+FROM numbers_mt(1000000);
+"""
+
+$CLICKHOUSE_CLIENT -q "SYSTEM FLUSH LOGS"
+$CLICKHOUSE_CLIENT -q """
+SELECT
+    ProfileEvents['MergeTreeDataProjectionWriterMergingBlocksMicroseconds'] > 0,
+    ProfileEvents['MergeTreeDataProjectionWriterPrimaryKeyCalculationMicroseconds'] > 0,
+    ProfileEvents['MergeTreeDataWriterPrimaryKeyCalculationMicroseconds'] > 0,
+    ProfileEvents['MergeTreeDataWriterProjectionsCalculationMicroseconds'] > 0,
+    ProfileEvents['MergeTreeDataWriterSecondaryIndicesCalculationMicroseconds'] > 0
+FROM system.query_log
+WHERE current_database = currentDatabase() AND query_id='$query_id' AND type = 'QueryFinish';
+"""

From d2ce39387f9a3ef09582a7a914952d76c69e77ac Mon Sep 17 00:00:00 2001
From: Bhavna Jindal <bhavna.jindal@ibm.com>
Date: Tue, 5 Dec 2023 07:44:42 -0800
Subject: [PATCH 05/73] implemented series_outlier method

---
 .../functions/time-series-functions.md        |  51 ++++++
 src/Functions/seriesOutliersTukey.cpp         | 173 ++++++++++++++++++
 .../02813_seriesOutliersTukey.reference       |   3 +
 .../0_stateless/02813_seriesOutliersTukey.sql |  12 ++
 .../aspell-ignore/en/aspell-dict.txt          |   1 +
 5 files changed, 240 insertions(+)
 create mode 100644 src/Functions/seriesOutliersTukey.cpp
 create mode 100644 tests/queries/0_stateless/02813_seriesOutliersTukey.reference
 create mode 100644 tests/queries/0_stateless/02813_seriesOutliersTukey.sql

diff --git a/docs/en/sql-reference/functions/time-series-functions.md b/docs/en/sql-reference/functions/time-series-functions.md
index 016c3410944..4eb838721b2 100644
--- a/docs/en/sql-reference/functions/time-series-functions.md
+++ b/docs/en/sql-reference/functions/time-series-functions.md
@@ -8,6 +8,57 @@ sidebar_label: Time Series
 
 Below functions are used for time series analysis.
 
+## seriesOutliersTukey
+
+Detects a possible anomaly in series using [Tukey Fences](https://en.wikipedia.org/wiki/Outlier#Tukey%27s_fences).
+
+**Syntax**
+
+``` sql
+seriesOutliersTukey(series);
+```
+
+**Arguments**
+
+- `series` - An array of numeric values
+
+**Returned value**
+
+- Returns an array of the same length where each value represents a modified Z-score of possible anomaly of corresponding element in the series.
+- A value greater than 3 or lesser than -3 indicates a possible anomaly.
+
+Type: [Array](../../sql-reference/data-types/array.md).
+
+**Examples**
+
+Query:
+
+``` sql
+seriesOutliersTukey([-3,2.4,15,3.9,5,6,4.5,5.2,3,4,5,16,7,5,5,4]) AS print_0;
+```
+
+Result:
+
+``` text
+┌───────────print_0──────────────────────────────────────────────────────────────────┐
+│[-2.7121212121212137,0,4.196969696969699,0,0,0,0,0,0,0,0,4.803030303030305,0,0,0,0] │
+└────────────────────────────────────────────────────────────────────────────────────┘
+```
+
+Query:
+
+``` sql
+seriesOutliersTukey(arrayMap(x -> sin(x / 10), range(30))) AS print_0;
+```
+
+Result:
+
+``` text
+┌───────────print_0────────────────────────────────────────────┐
+│[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0] │
+└──────────────────────────────────────────────────────────────┘
+```
+
 ## seriesPeriodDetectFFT
 
 Finds the period of the given time series data using FFT
diff --git a/src/Functions/seriesOutliersTukey.cpp b/src/Functions/seriesOutliersTukey.cpp
new file mode 100644
index 00000000000..1291d5bf09c
--- /dev/null
+++ b/src/Functions/seriesOutliersTukey.cpp
@@ -0,0 +1,173 @@
+#include <cmath>
+#include <Columns/ColumnArray.h>
+#include <Columns/ColumnsNumber.h>
+#include <DataTypes/DataTypeArray.h>
+#include <DataTypes/DataTypesNumber.h>
+#include <Functions/FunctionFactory.h>
+#include <Functions/FunctionHelpers.h>
+#include <Functions/IFunction.h>
+
+namespace DB
+{
+namespace ErrorCodes
+{
+extern const int BAD_ARGUMENTS;
+extern const int ILLEGAL_COLUMN;
+}
+
+//Detects a possible anomaly in series using [Tukey Fences](https://en.wikipedia.org/wiki/Outlier#Tukey%27s_fences)
+class FunctionSeriesOutliersTukey : public IFunction
+{
+public:
+    static constexpr auto name = "seriesOutliersTukey";
+
+    static FunctionPtr create(ContextPtr) { return std::make_shared<FunctionSeriesOutliersTukey>(); }
+
+    std::string getName() const override { return name; }
+
+    size_t getNumberOfArguments() const override { return 1; }
+
+    bool useDefaultImplementationForConstants() const override { return true; }
+
+    bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return true; }
+
+    DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override
+    {
+        FunctionArgumentDescriptors args{{"time_series", &isArray<IDataType>, nullptr, "Array"}};
+        validateFunctionArgumentTypes(*this, arguments, args);
+
+        return std::make_shared<DataTypeArray>(std::make_shared<DataTypeFloat64>());
+    }
+
+    ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t) const override
+    {
+        ColumnPtr array_ptr = arguments[0].column;
+        const ColumnArray * array = checkAndGetColumn<ColumnArray>(array_ptr.get());
+
+        const IColumn & src_data = array->getData();
+        const ColumnArray::Offsets & src_offsets = array->getOffsets();
+
+        ColumnPtr res;
+
+        if (executeNumber<UInt8>(src_data, src_offsets, res) || executeNumber<UInt16>(src_data, src_offsets, res)
+            || executeNumber<UInt32>(src_data, src_offsets, res) || executeNumber<UInt64>(src_data, src_offsets, res)
+            || executeNumber<Int8>(src_data, src_offsets, res) || executeNumber<Int16>(src_data, src_offsets, res)
+            || executeNumber<Int32>(src_data, src_offsets, res) || executeNumber<Int64>(src_data, src_offsets, res)
+            || executeNumber<Float32>(src_data, src_offsets, res) || executeNumber<Float64>(src_data, src_offsets, res))
+        {
+            return res;
+        }
+        else
+            throw Exception(
+                ErrorCodes::ILLEGAL_COLUMN,
+                "Illegal column {} of first argument of function {}",
+                arguments[0].column->getName(),
+                getName());
+    }
+
+    template <typename T>
+    bool executeNumber(const IColumn & src_data, const ColumnArray::Offsets & src_offsets, ColumnPtr & res_ptr) const
+    {
+        const ColumnVector<T> * src_data_concrete = checkAndGetColumn<ColumnVector<T>>(&src_data);
+        if (!src_data_concrete)
+            return false;
+
+        const PaddedPODArray<T> & src_vec = src_data_concrete->getData();
+
+        auto outliers = ColumnFloat64::create();
+        auto & outlier_data = outliers->getData();
+
+        ColumnArray::ColumnOffsets::MutablePtr res_offsets = ColumnArray::ColumnOffsets::create();
+        auto & res_offsets_data = res_offsets->getData();
+
+        ColumnArray::Offset prev_src_offset = 0;
+        for (auto curr_src_offset : src_offsets)
+        {
+            chassert(prev_src_offset < curr_src_offset);
+            size_t len = curr_src_offset - prev_src_offset;
+            if (len < 4)
+                throw Exception(ErrorCodes::BAD_ARGUMENTS, "At least four data points are needed for function {}", getName());
+
+            std::vector<Float64> src_sorted(src_vec.begin() + prev_src_offset, src_vec.begin() + curr_src_offset);
+            std::sort(src_sorted.begin(), src_sorted.end());
+
+            size_t q1_index = len / 4;
+            size_t q3_index = (len * 3) / 4;
+
+            Float64 q1 = (len % 2 != 0) ? src_sorted[q1_index] : (src_sorted[q1_index - 1] + src_sorted[q1_index]) / 2;
+            Float64 q3 = (len % 2 != 0) ? src_sorted[q3_index] : (src_sorted[q3_index - 1] + src_sorted[q3_index]) / 2;
+
+            Float64 iqr = q3 - q1;
+
+            Float64 lower_fence = q1 - 1.5 * iqr;
+            Float64 upper_fence = q3 + 1.5 * iqr;
+
+            for (auto elem : src_vec)
+            {
+                auto score = std::min((elem - lower_fence) / iqr, 0.0) + std::max((elem - upper_fence) / iqr, 0.0);
+                outlier_data.push_back(score);
+            }
+            res_offsets_data.push_back(outlier_data.size());
+            prev_src_offset = curr_src_offset;
+        }
+
+        res_ptr = ColumnArray::create(std::move(outliers), std::move(res_offsets));
+        return true;
+    }
+};
+
+REGISTER_FUNCTION(SeriesOutliersTukey)
+{
+    factory.registerFunction<FunctionSeriesOutliersTukey>(FunctionDocumentation{
+        .description = R"(
+Detects a possible anomaly in series using [Tukey Fences](https://en.wikipedia.org/wiki/Outlier#Tukey%27s_fences).
+
+**Syntax**
+
+``` sql
+seriesOutliersTukey(series);
+```
+
+**Arguments**
+
+- `series` - An array of numeric values
+
+**Returned value**
+
+- Returns an array of the same length where each value represents a modified Z-score of possible anomaly of corresponding element in the series.
+- A value greater than 3 or lesser than -3 indicates a possible anomaly.
+
+Type: [Array](../../sql-reference/data-types/array.md).
+
+**Examples**
+
+Query:
+
+``` sql
+seriesOutliersTukey([-3,2.4,15,3.9,5,6,4.5,5.2,3,4,5,16,7,5,5,4]) AS print_0;
+```
+
+Result:
+
+``` text
+┌───────────print_0──────────────────────────────────────────────────────────────────┐
+│[-2.7121212121212137,0,4.196969696969699,0,0,0,0,0,0,0,0,4.803030303030305,0,0,0,0] │
+└────────────────────────────────────────────────────────────────────────────────────┘
+```
+
+Query:
+
+``` sql
+seriesOutliersTukey(arrayMap(x -> sin(x / 10), range(30))) AS print_0;
+```
+
+Result:
+
+``` text
+┌───────────print_0────────────────────────────────────────────┐
+│[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0] │
+└──────────────────────────────────────────────────────────────┘
+```)",
+        .categories{"Time series analysis"}});
+}
+}
diff --git a/tests/queries/0_stateless/02813_seriesOutliersTukey.reference b/tests/queries/0_stateless/02813_seriesOutliersTukey.reference
new file mode 100644
index 00000000000..b370b4b5fe2
--- /dev/null
+++ b/tests/queries/0_stateless/02813_seriesOutliersTukey.reference
@@ -0,0 +1,3 @@
+[-2.7121212121212137,0,4.196969696969699,0,0,0,0,0,0,0,0,4.803030303030305,0,0,0,0]
+[0,0,0,0,0,0,0,0,0,5.228971962616823,0,0,0,0,0,0]
+[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]
diff --git a/tests/queries/0_stateless/02813_seriesOutliersTukey.sql b/tests/queries/0_stateless/02813_seriesOutliersTukey.sql
new file mode 100644
index 00000000000..4b24f1bffbb
--- /dev/null
+++ b/tests/queries/0_stateless/02813_seriesOutliersTukey.sql
@@ -0,0 +1,12 @@
+DROP TABLE IF EXISTS tb1;
+
+CREATE TABLE tb1 (n UInt32, a Array(Float64)) engine=Memory;
+INSERT INTO tb1 VALUES (1, [-3,2.4,15,3.9,5,6,4.5,5.2,3,4,5,16,7,5,5,4]);
+INSERT INTO tb1 VALUES (2, [-3,2.4,15,3.9,5,6,4.5,5.2,12,45,12,3.4,3,4,5,6]);
+
+SELECT seriesOutliersTukey(a) FROM tb1 ORDER BY n;
+DROP TABLE IF EXISTS tb1;
+SELECT seriesOutliersTukey(arrayMap(x -> sin(x / 10), range(30)));
+SELECT seriesOutliersTukey([-3, 2.4, 15, NULL]); -- { serverError ILLEGAL_COLUMN}
+SELECT seriesOutliersTukey([]); -- { serverError ILLEGAL_COLUMN}
+SELECT seriesOutliersTukey([-3, 2.4, 15]); -- { serverError BAD_ARGUMENTS}
\ No newline at end of file
diff --git a/utils/check-style/aspell-ignore/en/aspell-dict.txt b/utils/check-style/aspell-ignore/en/aspell-dict.txt
index 9f87255c9fa..64327aba2d1 100644
--- a/utils/check-style/aspell-ignore/en/aspell-dict.txt
+++ b/utils/check-style/aspell-ignore/en/aspell-dict.txt
@@ -939,6 +939,7 @@ TotalTemporaryFiles
 Tradeoff
 Transactional
 TwoColumnList
+Tukey
 UBSan
 UDFs
 UInt

From f4c14e15a2c4010b16240dfa4990a2f45c6d2f37 Mon Sep 17 00:00:00 2001
From: Bhavna Jindal <bhavna.jindal@ibm.com>
Date: Wed, 10 Jan 2024 09:20:49 -0800
Subject: [PATCH 06/73] added a fix for failing testcase

---
 src/Functions/seriesOutliersTukey.cpp                   | 4 ++--
 tests/queries/0_stateless/02813_seriesOutliersTukey.sql | 3 +--
 2 files changed, 3 insertions(+), 4 deletions(-)

diff --git a/src/Functions/seriesOutliersTukey.cpp b/src/Functions/seriesOutliersTukey.cpp
index 1291d5bf09c..4640258acc1 100644
--- a/src/Functions/seriesOutliersTukey.cpp
+++ b/src/Functions/seriesOutliersTukey.cpp
@@ -102,9 +102,9 @@ public:
             Float64 lower_fence = q1 - 1.5 * iqr;
             Float64 upper_fence = q3 + 1.5 * iqr;
 
-            for (auto elem : src_vec)
+            for (ColumnArray::Offset j = prev_src_offset; j < curr_src_offset; ++j)
             {
-                auto score = std::min((elem - lower_fence) / iqr, 0.0) + std::max((elem - upper_fence) / iqr, 0.0);
+                auto score = std::min((src_vec[j] - lower_fence) / iqr, 0.0) + std::max((src_vec[j] - upper_fence) / iqr, 0.0);
                 outlier_data.push_back(score);
             }
             res_offsets_data.push_back(outlier_data.size());
diff --git a/tests/queries/0_stateless/02813_seriesOutliersTukey.sql b/tests/queries/0_stateless/02813_seriesOutliersTukey.sql
index 4b24f1bffbb..baf8a229403 100644
--- a/tests/queries/0_stateless/02813_seriesOutliersTukey.sql
+++ b/tests/queries/0_stateless/02813_seriesOutliersTukey.sql
@@ -1,8 +1,7 @@
 DROP TABLE IF EXISTS tb1;
 
 CREATE TABLE tb1 (n UInt32, a Array(Float64)) engine=Memory;
-INSERT INTO tb1 VALUES (1, [-3,2.4,15,3.9,5,6,4.5,5.2,3,4,5,16,7,5,5,4]);
-INSERT INTO tb1 VALUES (2, [-3,2.4,15,3.9,5,6,4.5,5.2,12,45,12,3.4,3,4,5,6]);
+INSERT INTO tb1 VALUES (1, [-3,2.4,15,3.9,5,6,4.5,5.2,3,4,5,16,7,5,5,4]), (2, [-3,2.4,15,3.9,5,6,4.5,5.2,12,45,12,3.4,3,4,5,6]);
 
 SELECT seriesOutliersTukey(a) FROM tb1 ORDER BY n;
 DROP TABLE IF EXISTS tb1;

From 90dfacd5a67e0addde4973484a4bc1944a6ca4b1 Mon Sep 17 00:00:00 2001
From: Bhavna Jindal <bhavna.jindal@ibm.com>
Date: Wed, 10 Jan 2024 09:20:49 -0800
Subject: [PATCH 07/73] added a fix for failing testcase

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

diff --git a/src/Functions/seriesOutliersTukey.cpp b/src/Functions/seriesOutliersTukey.cpp
index 4640258acc1..697f32f2046 100644
--- a/src/Functions/seriesOutliersTukey.cpp
+++ b/src/Functions/seriesOutliersTukey.cpp
@@ -83,7 +83,7 @@ public:
         ColumnArray::Offset prev_src_offset = 0;
         for (auto curr_src_offset : src_offsets)
         {
-            chassert(prev_src_offset < curr_src_offset);
+            chassert(prev_src_offset <= curr_src_offset);
             size_t len = curr_src_offset - prev_src_offset;
             if (len < 4)
                 throw Exception(ErrorCodes::BAD_ARGUMENTS, "At least four data points are needed for function {}", getName());

From 487ef67e9759394de5ebfcc91749db5624d2fd03 Mon Sep 17 00:00:00 2001
From: Bhavna Jindal <bhavna.jindal@ibm.com>
Date: Thu, 18 Jan 2024 07:26:32 -0800
Subject: [PATCH 08/73] Added custom tukey

---
 .../functions/time-series-functions.md        |   8 +-
 src/Functions/seriesOutliersTukey.cpp         | 120 ++++++++++++++----
 .../02813_seriesOutliersTukey.reference       |   9 ++
 .../0_stateless/02813_seriesOutliersTukey.sql |  19 ++-
 4 files changed, 124 insertions(+), 32 deletions(-)

diff --git a/docs/en/sql-reference/functions/time-series-functions.md b/docs/en/sql-reference/functions/time-series-functions.md
index 4eb838721b2..9eea5a8eb1f 100644
--- a/docs/en/sql-reference/functions/time-series-functions.md
+++ b/docs/en/sql-reference/functions/time-series-functions.md
@@ -8,14 +8,14 @@ sidebar_label: Time Series
 
 Below functions are used for time series analysis.
 
-## seriesOutliersTukey
+## seriesOutliersDetectTukey
 
 Detects a possible anomaly in series using [Tukey Fences](https://en.wikipedia.org/wiki/Outlier#Tukey%27s_fences).
 
 **Syntax**
 
 ``` sql
-seriesOutliersTukey(series);
+seriesOutliersDetectTukey(series);
 ```
 
 **Arguments**
@@ -34,7 +34,7 @@ Type: [Array](../../sql-reference/data-types/array.md).
 Query:
 
 ``` sql
-seriesOutliersTukey([-3,2.4,15,3.9,5,6,4.5,5.2,3,4,5,16,7,5,5,4]) AS print_0;
+seriesOutliersDetectTukey([-3,2.4,15,3.9,5,6,4.5,5.2,3,4,5,16,7,5,5,4]) AS print_0;
 ```
 
 Result:
@@ -48,7 +48,7 @@ Result:
 Query:
 
 ``` sql
-seriesOutliersTukey(arrayMap(x -> sin(x / 10), range(30))) AS print_0;
+seriesOutliersDetectTukey(arrayMap(x -> sin(x / 10), range(30))) AS print_0;
 ```
 
 Result:
diff --git a/src/Functions/seriesOutliersTukey.cpp b/src/Functions/seriesOutliersTukey.cpp
index 697f32f2046..72a16949605 100644
--- a/src/Functions/seriesOutliersTukey.cpp
+++ b/src/Functions/seriesOutliersTukey.cpp
@@ -1,5 +1,6 @@
 #include <cmath>
 #include <Columns/ColumnArray.h>
+#include <Columns/ColumnString.h>
 #include <Columns/ColumnsNumber.h>
 #include <DataTypes/DataTypeArray.h>
 #include <DataTypes/DataTypesNumber.h>
@@ -16,16 +17,18 @@ extern const int ILLEGAL_COLUMN;
 }
 
 //Detects a possible anomaly in series using [Tukey Fences](https://en.wikipedia.org/wiki/Outlier#Tukey%27s_fences)
-class FunctionSeriesOutliersTukey : public IFunction
+class FunctionSeriesOutliersDetectTukey : public IFunction
 {
 public:
-    static constexpr auto name = "seriesOutliersTukey";
+    static constexpr auto name = "seriesOutliersDetectTukey";
 
-    static FunctionPtr create(ContextPtr) { return std::make_shared<FunctionSeriesOutliersTukey>(); }
+    static FunctionPtr create(ContextPtr) { return std::make_shared<FunctionSeriesOutliersDetectTukey>(); }
 
     std::string getName() const override { return name; }
 
-    size_t getNumberOfArguments() const override { return 1; }
+    bool isVariadic() const override { return true; }
+
+    size_t getNumberOfArguments() const override { return 0; }
 
     bool useDefaultImplementationForConstants() const override { return true; }
 
@@ -33,12 +36,20 @@ public:
 
     DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override
     {
-        FunctionArgumentDescriptors args{{"time_series", &isArray<IDataType>, nullptr, "Array"}};
-        validateFunctionArgumentTypes(*this, arguments, args);
+        FunctionArgumentDescriptors mandatory_args{{"time_series", &isArray<IDataType>, nullptr, "Array"}};
+        FunctionArgumentDescriptors optional_args{
+            {"kind", &isString<IDataType>, isColumnConst, "const String"},
+            {"min_percentile", &isNativeNumber<IDataType>, isColumnConst, "Number"},
+            {"max_percentile", &isNativeNumber<IDataType>, isColumnConst, "Number"}
+        };
+
+        validateFunctionArgumentTypes(*this, arguments, mandatory_args, optional_args);
 
         return std::make_shared<DataTypeArray>(std::make_shared<DataTypeFloat64>());
     }
 
+    ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {1,2,3}; }
+
     ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t) const override
     {
         ColumnPtr array_ptr = arguments[0].column;
@@ -47,13 +58,54 @@ public:
         const IColumn & src_data = array->getData();
         const ColumnArray::Offsets & src_offsets = array->getOffsets();
 
+        Float64 min_percentile = 0.25;
+        Float64 max_percentile = 0.75;
+
+        if(arguments.size() > 1)
+        {
+            //const IColumn * arg_column = arguments[1].column.get();
+            const ColumnConst * arg_string = checkAndGetColumnConstStringOrFixedString(arguments[1].column.get());
+
+            if (!arg_string)
+                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "The second argument of function {} must be constant String", getName());
+
+            String kind = arg_string->getValue<String>();
+            if(kind == "ctukey"){
+                min_percentile = 0.10; //default 10th percentile
+                max_percentile = 0.90; //default 90th percentile
+
+                if(arguments.size() > 2)
+                {
+                    Float64 p_min = arguments[2].column->getFloat64(0);
+                    if(p_min >= 2.0 && p_min <= 98.0)
+                        min_percentile = p_min/100;
+                    else
+                      throw Exception(ErrorCodes::BAD_ARGUMENTS, "The third argumet of function {} must be in range [2.0, 98.0]", getName());  
+                }
+
+                if(arguments.size() == 4)
+                {
+                    Float64 p_max = arguments[3].column->getFloat64(0);
+                    if(p_max >= 2.0 && p_max <= 98.0 && p_max > min_percentile*100)
+                        max_percentile = p_max/100;
+                    else
+                      throw Exception(ErrorCodes::BAD_ARGUMENTS, "The fourth argumet of function {} must be in range [2.0, 98.0]", getName());  
+                }
+            } 
+            else
+            {
+                if(kind != "tukey")
+                    throw Exception(ErrorCodes::BAD_ARGUMENTS, "The second argument of function {} can only be 'tukey' or 'ctukey'.", getName());
+            }
+        }
+
         ColumnPtr res;
 
-        if (executeNumber<UInt8>(src_data, src_offsets, res) || executeNumber<UInt16>(src_data, src_offsets, res)
-            || executeNumber<UInt32>(src_data, src_offsets, res) || executeNumber<UInt64>(src_data, src_offsets, res)
-            || executeNumber<Int8>(src_data, src_offsets, res) || executeNumber<Int16>(src_data, src_offsets, res)
-            || executeNumber<Int32>(src_data, src_offsets, res) || executeNumber<Int64>(src_data, src_offsets, res)
-            || executeNumber<Float32>(src_data, src_offsets, res) || executeNumber<Float64>(src_data, src_offsets, res))
+        if (executeNumber<UInt8>(src_data, src_offsets, min_percentile, max_percentile, res) || executeNumber<UInt16>(src_data, src_offsets,min_percentile, max_percentile, res)
+            || executeNumber<UInt32>(src_data, src_offsets, min_percentile, max_percentile,res) || executeNumber<UInt64>(src_data, src_offsets,min_percentile, max_percentile, res)
+            || executeNumber<Int8>(src_data, src_offsets,min_percentile, max_percentile, res) || executeNumber<Int16>(src_data, src_offsets,min_percentile, max_percentile, res)
+            || executeNumber<Int32>(src_data, src_offsets,min_percentile, max_percentile, res) || executeNumber<Int64>(src_data, src_offsets, min_percentile, max_percentile,res)
+            || executeNumber<Float32>(src_data, src_offsets,min_percentile, max_percentile, res) || executeNumber<Float64>(src_data, src_offsets,min_percentile, max_percentile, res))
         {
             return res;
         }
@@ -66,7 +118,11 @@ public:
     }
 
     template <typename T>
-    bool executeNumber(const IColumn & src_data, const ColumnArray::Offsets & src_offsets, ColumnPtr & res_ptr) const
+    bool executeNumber(const IColumn & src_data, 
+    const ColumnArray::Offsets & src_offsets, 
+    Float64 min_percentile,
+    Float64 max_percentile,
+    ColumnPtr & res_ptr) const
     {
         const ColumnVector<T> * src_data_concrete = checkAndGetColumn<ColumnVector<T>>(&src_data);
         if (!src_data_concrete)
@@ -91,16 +147,34 @@ public:
             std::vector<Float64> src_sorted(src_vec.begin() + prev_src_offset, src_vec.begin() + curr_src_offset);
             std::sort(src_sorted.begin(), src_sorted.end());
 
-            size_t q1_index = len / 4;
-            size_t q3_index = (len * 3) / 4;
+            Float64 q1, q2;
 
-            Float64 q1 = (len % 2 != 0) ? src_sorted[q1_index] : (src_sorted[q1_index - 1] + src_sorted[q1_index]) / 2;
-            Float64 q3 = (len % 2 != 0) ? src_sorted[q3_index] : (src_sorted[q3_index - 1] + src_sorted[q3_index]) / 2;
+            auto p1 = len * min_percentile;
+            if(p1 == static_cast<Int64>(p1)){
+                size_t index = static_cast<size_t>(p1)-1;
+                q1 = (src_sorted[index] + src_sorted[index+1])/2;
+            }
+            else
+            {   
+                size_t index = static_cast<size_t>(std::ceil(p1))-1;
+                q1 = src_sorted[index];
+            }
 
-            Float64 iqr = q3 - q1;
+            auto p2 = len * max_percentile;
+            if(p2 == static_cast<Int64>(p2)){
+                size_t index = static_cast<size_t>(p2)-1;
+                q2 = (src_sorted[index] + src_sorted[index+1])/2;
+            }
+            else
+            {   
+                size_t index = static_cast<size_t>(std::ceil(p2))-1;
+                q2 = src_sorted[index];
+            }
+
+            Float64 iqr = q2 - q1;
 
             Float64 lower_fence = q1 - 1.5 * iqr;
-            Float64 upper_fence = q3 + 1.5 * iqr;
+            Float64 upper_fence = q2 + 1.5 * iqr;
 
             for (ColumnArray::Offset j = prev_src_offset; j < curr_src_offset; ++j)
             {
@@ -116,16 +190,16 @@ public:
     }
 };
 
-REGISTER_FUNCTION(SeriesOutliersTukey)
+REGISTER_FUNCTION(SeriesOutliersDetectTukey)
 {
-    factory.registerFunction<FunctionSeriesOutliersTukey>(FunctionDocumentation{
+    factory.registerFunction<FunctionSeriesOutliersDetectTukey>(FunctionDocumentation{
         .description = R"(
 Detects a possible anomaly in series using [Tukey Fences](https://en.wikipedia.org/wiki/Outlier#Tukey%27s_fences).
 
 **Syntax**
 
 ``` sql
-seriesOutliersTukey(series);
+seriesOutliersDetectTukey(series);
 ```
 
 **Arguments**
@@ -144,7 +218,7 @@ Type: [Array](../../sql-reference/data-types/array.md).
 Query:
 
 ``` sql
-seriesOutliersTukey([-3,2.4,15,3.9,5,6,4.5,5.2,3,4,5,16,7,5,5,4]) AS print_0;
+seriesOutliersDetectTukey([-3,2.4,15,3.9,5,6,4.5,5.2,3,4,5,16,7,5,5,4]) AS print_0;
 ```
 
 Result:
@@ -158,7 +232,7 @@ Result:
 Query:
 
 ``` sql
-seriesOutliersTukey(arrayMap(x -> sin(x / 10), range(30))) AS print_0;
+seriesOutliersDetectTukey(arrayMap(x -> sin(x / 10), range(30))) AS print_0;
 ```
 
 Result:
diff --git a/tests/queries/0_stateless/02813_seriesOutliersTukey.reference b/tests/queries/0_stateless/02813_seriesOutliersTukey.reference
index b370b4b5fe2..990c8c11e9e 100644
--- a/tests/queries/0_stateless/02813_seriesOutliersTukey.reference
+++ b/tests/queries/0_stateless/02813_seriesOutliersTukey.reference
@@ -1,3 +1,12 @@
 [-2.7121212121212137,0,4.196969696969699,0,0,0,0,0,0,0,0,4.803030303030305,0,0,0,0]
 [0,0,0,0,0,0,0,0,0,5.228971962616823,0,0,0,0,0,0]
+[-2.7121212121212137,0,4.196969696969699,0,0,0,0,0,0,0,0,4.803030303030305,0,0,0,0]
+[0,0,0,0,0,0,0,0,0,5.228971962616823,0,0,0,0,0,0]
+[0,0,0,0,0,0,0,0,0,4.706896551724138,0,0,0,0,0,0]
+[-0.9615384615384613,0,1.9615384615384612,0,0,0,0,0,0.8076923076923075,19.26923076923077,0.8076923076923075,0,0,0,0,0,0,0]
+[0,0,0,0,0,0,0,0,0,4.706896551724138,0,0,0,0,0,0]
+[0,0,0,0,0,0,0,0,0,0.8076923076923077,0,0,0,0,0,0]
+[0,0,0,0,0,0,0,0,0,4.706896551724138,0,0,0,0,0,0]
+[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]
+[0,0,0,0]
 [0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]
diff --git a/tests/queries/0_stateless/02813_seriesOutliersTukey.sql b/tests/queries/0_stateless/02813_seriesOutliersTukey.sql
index baf8a229403..f8debc7b6db 100644
--- a/tests/queries/0_stateless/02813_seriesOutliersTukey.sql
+++ b/tests/queries/0_stateless/02813_seriesOutliersTukey.sql
@@ -3,9 +3,18 @@ DROP TABLE IF EXISTS tb1;
 CREATE TABLE tb1 (n UInt32, a Array(Float64)) engine=Memory;
 INSERT INTO tb1 VALUES (1, [-3,2.4,15,3.9,5,6,4.5,5.2,3,4,5,16,7,5,5,4]), (2, [-3,2.4,15,3.9,5,6,4.5,5.2,12,45,12,3.4,3,4,5,6]);
 
-SELECT seriesOutliersTukey(a) FROM tb1 ORDER BY n;
+SELECT seriesOutliersDetectTukey(a) FROM tb1 ORDER BY n;
+SELECT seriesOutliersDetectTukey(a,'ctukey', 25,75) FROM tb1 ORDER BY n;
 DROP TABLE IF EXISTS tb1;
-SELECT seriesOutliersTukey(arrayMap(x -> sin(x / 10), range(30)));
-SELECT seriesOutliersTukey([-3, 2.4, 15, NULL]); -- { serverError ILLEGAL_COLUMN}
-SELECT seriesOutliersTukey([]); -- { serverError ILLEGAL_COLUMN}
-SELECT seriesOutliersTukey([-3, 2.4, 15]); -- { serverError BAD_ARGUMENTS}
\ No newline at end of file
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.5, 5, 12, 45, 12, 3.4, 3, 4, 5, 6]);
+SELECT seriesOutliersDetectTukey([-3, 2.4, 15, 3.9, 5, 6, 4.5, 5.2, 12, 60, 12, 3.4, 3, 4, 5, 6, 3.4, 2.7]);
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.5, 5, 12, 45, 12, 3.4, 3, 4, 5, 6], 'ctukey', 25, 75);
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.5, 5, 12, 45, 12, 3.4, 3, 4, 5, 6], 'ctukey', 10, 90);
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.5, 5, 12, 45, 12, 3.4, 3, 4, 5, 6], 'tukey', 10, 90);
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.5, 5, 12, 45, 12, 3.4, 3, 4, 5, 6], 'ctukey', 2, 98)
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3], 'ctukey', 2, 98);
+SELECT seriesOutliersDetectTukey(arrayMap(x -> sin(x / 10), range(30)));
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3], 'xyz', 33, 53); -- { serverError BAD_ARGUMENTS}
+SELECT seriesOutliersDetectTukey([-3, 2.4, 15, NULL]); -- { serverError ILLEGAL_COLUMN}
+SELECT seriesOutliersDetectTukey([]); -- { serverError ILLEGAL_COLUMN}
+SELECT seriesOutliersDetectTukey([-3, 2.4, 15]); -- { serverError BAD_ARGUMENTS}
\ No newline at end of file

From 4bc26fe45fa78fbfd1394d22f73a86cf1f83e586 Mon Sep 17 00:00:00 2001
From: Bhavna Jindal <bhavna.jindal@ibm.com>
Date: Wed, 31 Jan 2024 10:35:08 -0800
Subject: [PATCH 09/73] Added support for custom percentiles and K

---
 .../functions/time-series-functions.md        |  41 ++--
 src/Functions/seriesOutliersTukey.cpp         | 184 +++++++++++-------
 .../02813_seriesOutliersTukey.reference       |  20 +-
 .../0_stateless/02813_seriesOutliersTukey.sql |  26 ++-
 .../aspell-ignore/en/aspell-dict.txt          |   3 +
 5 files changed, 168 insertions(+), 106 deletions(-)

diff --git a/docs/en/sql-reference/functions/time-series-functions.md b/docs/en/sql-reference/functions/time-series-functions.md
index 9eea5a8eb1f..bd50ef556f7 100644
--- a/docs/en/sql-reference/functions/time-series-functions.md
+++ b/docs/en/sql-reference/functions/time-series-functions.md
@@ -6,7 +6,7 @@ sidebar_label: Time Series
 
 # Time Series Functions
 
-Below functions are used for time series analysis.
+Below functions are used for series data analysis.
 
 ## seriesOutliersDetectTukey
 
@@ -16,16 +16,27 @@ Detects a possible anomaly in series using [Tukey Fences](https://en.wikipedia.o
 
 ``` sql
 seriesOutliersDetectTukey(series);
+seriesOutliersDetectTukey(series, kind, min_percentile, max_percentile, K);
 ```
 
 **Arguments**
 
-- `series` - An array of numeric values
+- `series` - An array of numeric values.
+- `kind` - Kind of algorithm to use. Supported values are 'tukey' for standard tukey and 'ctukey' for custom tukey algorithm. The default is 'ctukey'.
+- `min_percentile` - The minimum percentile to be used to calculate inter-quantile range(IQR). The value must be in range [2,98]. The default is 10. This value is only supported for 'ctukey'.
+- `max_percentile` - The maximum percentile to be used to calculate inter-quantile range(IQR). The value must be in range [2,98]. The default is 90. This value is only supported for 'ctukey'.
+- `K` - Non-negative constant value to detect mild or stronger outliers. The default value is 1.5
+
+At least four data points are required in `series` to detect outliers.
+
+Default quantile range:
+- `tukey` - 25%/75%
+- `ctukey` - 10%/90%
 
 **Returned value**
 
-- Returns an array of the same length where each value represents a modified Z-score of possible anomaly of corresponding element in the series.
-- A value greater than 3 or lesser than -3 indicates a possible anomaly.
+- Returns an array of the same length where each value represents score of possible anomaly of corresponding element in the series.
+- A non-zero score indicates a possible anomaly.
 
 Type: [Array](../../sql-reference/data-types/array.md).
 
@@ -34,34 +45,34 @@ Type: [Array](../../sql-reference/data-types/array.md).
 Query:
 
 ``` sql
-seriesOutliersDetectTukey([-3,2.4,15,3.9,5,6,4.5,5.2,3,4,5,16,7,5,5,4]) AS print_0;
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4, 5, 12, 45, 12, 3, 3, 4, 5, 6]) AS print_0;
 ```
 
 Result:
 
 ``` text
-┌───────────print_0──────────────────────────────────────────────────────────────────┐
-│[-2.7121212121212137,0,4.196969696969699,0,0,0,0,0,0,0,0,4.803030303030305,0,0,0,0] │
-└────────────────────────────────────────────────────────────────────────────────────┘
+┌───────────print_0───────────────────┐
+│[0,0,0,0,0,0,0,0,0,10.5,0,0,0,0,0,0] │
+└─────────────────────────────────────┘
 ```
 
 Query:
 
 ``` sql
-seriesOutliersDetectTukey(arrayMap(x -> sin(x / 10), range(30))) AS print_0;
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.50, 5, 12, 45, 12, 3.40, 3, 4, 5, 6], 'ctukey', 20, 80, 1.5) AS print_0;
 ```
 
 Result:
 
 ``` text
-┌───────────print_0────────────────────────────────────────────┐
-│[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0] │
-└──────────────────────────────────────────────────────────────┘
+┌─print_0────────────────────────────┐
+│ [0,0,0,0,0,0,0,0,0,12,0,0,0,0,0,0] │
+└────────────────────────────────────┘
 ```
 
 ## seriesPeriodDetectFFT
 
-Finds the period of the given time series data using FFT
+Finds the period of the given series data data using FFT
 FFT - [Fast Fourier transform](https://en.wikipedia.org/wiki/Fast_Fourier_transform)
 
 **Syntax**
@@ -76,7 +87,7 @@ seriesPeriodDetectFFT(series);
 
 **Returned value**
 
-- A real value equal to the period of time series
+- A real value equal to the period of series data
 - Returns NAN when number of data points are less than four.
 
 Type: [Float64](../../sql-reference/data-types/float.md).
@@ -111,7 +122,7 @@ Result:
 
 ## seriesDecomposeSTL
 
-Decomposes a time series using STL [(Seasonal-Trend Decomposition Procedure Based on Loess)](https://www.wessa.net/download/stl.pdf) into a season, a trend and a residual component. 
+Decomposes a series data using STL [(Seasonal-Trend Decomposition Procedure Based on Loess)](https://www.wessa.net/download/stl.pdf) into a season, a trend and a residual component. 
 
 **Syntax**
 
diff --git a/src/Functions/seriesOutliersTukey.cpp b/src/Functions/seriesOutliersTukey.cpp
index 72a16949605..4c2c1ccd882 100644
--- a/src/Functions/seriesOutliersTukey.cpp
+++ b/src/Functions/seriesOutliersTukey.cpp
@@ -1,4 +1,3 @@
-#include <cmath>
 #include <Columns/ColumnArray.h>
 #include <Columns/ColumnString.h>
 #include <Columns/ColumnsNumber.h>
@@ -7,6 +6,7 @@
 #include <Functions/FunctionFactory.h>
 #include <Functions/FunctionHelpers.h>
 #include <Functions/IFunction.h>
+#include <cmath>
 
 namespace DB
 {
@@ -16,7 +16,7 @@ extern const int BAD_ARGUMENTS;
 extern const int ILLEGAL_COLUMN;
 }
 
-//Detects a possible anomaly in series using [Tukey Fences](https://en.wikipedia.org/wiki/Outlier#Tukey%27s_fences)
+///Detects a possible anomaly in series using [Tukey Fences](https://en.wikipedia.org/wiki/Outlier#Tukey%27s_fences)
 class FunctionSeriesOutliersDetectTukey : public IFunction
 {
 public:
@@ -40,28 +40,28 @@ public:
         FunctionArgumentDescriptors optional_args{
             {"kind", &isString<IDataType>, isColumnConst, "const String"},
             {"min_percentile", &isNativeNumber<IDataType>, isColumnConst, "Number"},
-            {"max_percentile", &isNativeNumber<IDataType>, isColumnConst, "Number"}
-        };
+            {"max_percentile", &isNativeNumber<IDataType>, isColumnConst, "Number"},
+            {"k", &isNativeNumber<IDataType>, isColumnConst, "Number"}};
 
         validateFunctionArgumentTypes(*this, arguments, mandatory_args, optional_args);
 
         return std::make_shared<DataTypeArray>(std::make_shared<DataTypeFloat64>());
     }
 
-    ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {1,2,3}; }
+    ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {1, 2, 3, 4}; }
 
     ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t) const override
     {
-        ColumnPtr array_ptr = arguments[0].column;
-        const ColumnArray * array = checkAndGetColumn<ColumnArray>(array_ptr.get());
+        ColumnPtr col = arguments[0].column;
+        const ColumnArray * col_arr = checkAndGetColumn<ColumnArray>(col.get());
 
-        const IColumn & src_data = array->getData();
-        const ColumnArray::Offsets & src_offsets = array->getOffsets();
+        const IColumn & arr_data = col_arr->getData();
+        const ColumnArray::Offsets & arr_offsets = col_arr->getOffsets();
 
-        Float64 min_percentile = 0.25;
-        Float64 max_percentile = 0.75;
+        Float64 min_percentile = 0.10; //default 10th percentile
+        Float64 max_percentile = 0.90; //default 90th percentile
 
-        if(arguments.size() > 1)
+        if (arguments.size() > 1)
         {
             //const IColumn * arg_column = arguments[1].column.get();
             const ColumnConst * arg_string = checkAndGetColumnConstStringOrFixedString(arguments[1].column.get());
@@ -70,44 +70,62 @@ public:
                 throw Exception(ErrorCodes::ILLEGAL_COLUMN, "The second argument of function {} must be constant String", getName());
 
             String kind = arg_string->getValue<String>();
-            if(kind == "ctukey"){
-                min_percentile = 0.10; //default 10th percentile
-                max_percentile = 0.90; //default 90th percentile
-
-                if(arguments.size() > 2)
+            if (kind == "ctukey")
+            {
+                if (arguments.size() > 2)
                 {
                     Float64 p_min = arguments[2].column->getFloat64(0);
-                    if(p_min >= 2.0 && p_min <= 98.0)
-                        min_percentile = p_min/100;
+                    if (p_min >= 2.0 && p_min <= 98.0)
+                        min_percentile = p_min / 100;
                     else
-                      throw Exception(ErrorCodes::BAD_ARGUMENTS, "The third argumet of function {} must be in range [2.0, 98.0]", getName());  
+                        throw Exception(
+                            ErrorCodes::BAD_ARGUMENTS, "The third argumet of function {} must be in range [2.0, 98.0]", getName());
                 }
 
-                if(arguments.size() == 4)
+                if (arguments.size() > 3)
                 {
                     Float64 p_max = arguments[3].column->getFloat64(0);
-                    if(p_max >= 2.0 && p_max <= 98.0 && p_max > min_percentile*100)
-                        max_percentile = p_max/100;
+                    if (p_max >= 2.0 && p_max <= 98.0 && p_max > min_percentile * 100)
+                        max_percentile = p_max / 100;
                     else
-                      throw Exception(ErrorCodes::BAD_ARGUMENTS, "The fourth argumet of function {} must be in range [2.0, 98.0]", getName());  
+                        throw Exception(
+                            ErrorCodes::BAD_ARGUMENTS, "The fourth argumet of function {} must be in range [2.0, 98.0]", getName());
                 }
-            } 
-            else
-            {
-                if(kind != "tukey")
-                    throw Exception(ErrorCodes::BAD_ARGUMENTS, "The second argument of function {} can only be 'tukey' or 'ctukey'.", getName());
             }
+            else if (kind == "tukey")
+            {
+                min_percentile = 0.25;
+                max_percentile = 0.75;
+            }
+            else
+                throw Exception(
+                    ErrorCodes::BAD_ARGUMENTS, "The second argument of function {} can only be 'tukey' or 'ctukey'.", getName());
         }
 
-        ColumnPtr res;
-
-        if (executeNumber<UInt8>(src_data, src_offsets, min_percentile, max_percentile, res) || executeNumber<UInt16>(src_data, src_offsets,min_percentile, max_percentile, res)
-            || executeNumber<UInt32>(src_data, src_offsets, min_percentile, max_percentile,res) || executeNumber<UInt64>(src_data, src_offsets,min_percentile, max_percentile, res)
-            || executeNumber<Int8>(src_data, src_offsets,min_percentile, max_percentile, res) || executeNumber<Int16>(src_data, src_offsets,min_percentile, max_percentile, res)
-            || executeNumber<Int32>(src_data, src_offsets,min_percentile, max_percentile, res) || executeNumber<Int64>(src_data, src_offsets, min_percentile, max_percentile,res)
-            || executeNumber<Float32>(src_data, src_offsets,min_percentile, max_percentile, res) || executeNumber<Float64>(src_data, src_offsets,min_percentile, max_percentile, res))
+        Float64 K = 1.50;
+        if (arguments.size() == 5)
         {
-            return res;
+            auto k_val = arguments[4].column->getFloat64(0);
+            if (k_val >= 0.0)
+                K = k_val;
+            else
+                throw Exception(ErrorCodes::BAD_ARGUMENTS, "The fifth argumet of function {} must be a positive number", getName());
+        }
+
+        ColumnPtr col_res;
+
+        if (executeNumber<UInt8>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res)
+            || executeNumber<UInt16>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res)
+            || executeNumber<UInt32>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res)
+            || executeNumber<UInt64>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res)
+            || executeNumber<Int8>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res)
+            || executeNumber<Int16>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res)
+            || executeNumber<Int32>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res)
+            || executeNumber<Int64>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res)
+            || executeNumber<Float32>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res)
+            || executeNumber<Float64>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res))
+        {
+            return col_res;
         }
         else
             throw Exception(
@@ -117,14 +135,17 @@ public:
                 getName());
     }
 
+private:
     template <typename T>
-    bool executeNumber(const IColumn & src_data, 
-    const ColumnArray::Offsets & src_offsets, 
-    Float64 min_percentile,
-    Float64 max_percentile,
-    ColumnPtr & res_ptr) const
+    bool executeNumber(
+        const IColumn & arr_data,
+        const ColumnArray::Offsets & arr_offsets,
+        Float64 min_percentile,
+        Float64 max_percentile,
+        Float64 K,
+        ColumnPtr & res_ptr) const
     {
-        const ColumnVector<T> * src_data_concrete = checkAndGetColumn<ColumnVector<T>>(&src_data);
+        const ColumnVector<T> * src_data_concrete = checkAndGetColumn<ColumnVector<T>>(&arr_data);
         if (!src_data_concrete)
             return false;
 
@@ -136,53 +157,57 @@ public:
         ColumnArray::ColumnOffsets::MutablePtr res_offsets = ColumnArray::ColumnOffsets::create();
         auto & res_offsets_data = res_offsets->getData();
 
+        std::vector<Float64> src_sorted;
+
         ColumnArray::Offset prev_src_offset = 0;
-        for (auto curr_src_offset : src_offsets)
+        for (auto src_offset : arr_offsets)
         {
-            chassert(prev_src_offset <= curr_src_offset);
-            size_t len = curr_src_offset - prev_src_offset;
+            chassert(prev_src_offset <= src_offset);
+            size_t len = src_offset - prev_src_offset;
             if (len < 4)
                 throw Exception(ErrorCodes::BAD_ARGUMENTS, "At least four data points are needed for function {}", getName());
 
-            std::vector<Float64> src_sorted(src_vec.begin() + prev_src_offset, src_vec.begin() + curr_src_offset);
+            src_sorted.assign(src_vec.begin() + prev_src_offset, src_vec.begin() + src_offset);
             std::sort(src_sorted.begin(), src_sorted.end());
 
             Float64 q1, q2;
 
             auto p1 = len * min_percentile;
-            if(p1 == static_cast<Int64>(p1)){
-                size_t index = static_cast<size_t>(p1)-1;
-                q1 = (src_sorted[index] + src_sorted[index+1])/2;
+            if (p1 == static_cast<Int64>(p1))
+            {
+                size_t index = static_cast<size_t>(p1) - 1;
+                q1 = (src_sorted[index] + src_sorted[index + 1]) / 2;
             }
             else
-            {   
-                size_t index = static_cast<size_t>(std::ceil(p1))-1;
+            {
+                size_t index = static_cast<size_t>(std::ceil(p1)) - 1;
                 q1 = src_sorted[index];
             }
 
             auto p2 = len * max_percentile;
-            if(p2 == static_cast<Int64>(p2)){
-                size_t index = static_cast<size_t>(p2)-1;
-                q2 = (src_sorted[index] + src_sorted[index+1])/2;
+            if (p2 == static_cast<Int64>(p2))
+            {
+                size_t index = static_cast<size_t>(p2) - 1;
+                q2 = (src_sorted[index] + src_sorted[index + 1]) / 2;
             }
             else
-            {   
-                size_t index = static_cast<size_t>(std::ceil(p2))-1;
+            {
+                size_t index = static_cast<size_t>(std::ceil(p2)) - 1;
                 q2 = src_sorted[index];
             }
 
-            Float64 iqr = q2 - q1;
+            Float64 iqr = q2 - q1; /// interquantile range
 
-            Float64 lower_fence = q1 - 1.5 * iqr;
-            Float64 upper_fence = q2 + 1.5 * iqr;
+            Float64 lower_fence = q1 - K * iqr;
+            Float64 upper_fence = q2 + K * iqr;
 
-            for (ColumnArray::Offset j = prev_src_offset; j < curr_src_offset; ++j)
+            for (ColumnArray::Offset j = prev_src_offset; j < src_offset; ++j)
             {
-                auto score = std::min((src_vec[j] - lower_fence) / iqr, 0.0) + std::max((src_vec[j] - upper_fence) / iqr, 0.0);
+                auto score = std::min((src_vec[j] - lower_fence), 0.0) + std::max((src_vec[j] - upper_fence), 0.0);
                 outlier_data.push_back(score);
             }
             res_offsets_data.push_back(outlier_data.size());
-            prev_src_offset = curr_src_offset;
+            prev_src_offset = src_offset;
         }
 
         res_ptr = ColumnArray::create(std::move(outliers), std::move(res_offsets));
@@ -196,20 +221,31 @@ REGISTER_FUNCTION(SeriesOutliersDetectTukey)
         .description = R"(
 Detects a possible anomaly in series using [Tukey Fences](https://en.wikipedia.org/wiki/Outlier#Tukey%27s_fences).
 
+Detects a possible anomaly in series using [Tukey Fences](https://en.wikipedia.org/wiki/Outlier#Tukey%27s_fences).
+
 **Syntax**
 
 ``` sql
 seriesOutliersDetectTukey(series);
+seriesOutliersDetectTukey(series, kind, min_percentile, max_percentile, K);
 ```
 
 **Arguments**
 
-- `series` - An array of numeric values
+- `series` - An array of numeric values.
+- `kind` - Kind of algorithm to use. Supported values are 'tukey' for standard tukey and 'ctukey' for custom tukey algorithm. The default is 'ctukey'.
+- `min_percentile` - The minimum percentile to be used to calculate inter-quantile range(IQR). The value must be in range [2,98]. The default is 10. This value is only supported for 'ctukey'.
+- `max_percentile` - The maximum percentile to be used to calculate inter-quantile range(IQR). The value must be in range [2,98]. The default is 90. This value is only supported for 'ctukey'.
+- `K` - Non-negative constant value to detect mild or stronger outliers. The default value is 1.5
+
+Default quantile range:
+- `tukey` - 25%/75%
+- `ctukey` - 10%/90%
 
 **Returned value**
 
-- Returns an array of the same length where each value represents a modified Z-score of possible anomaly of corresponding element in the series.
-- A value greater than 3 or lesser than -3 indicates a possible anomaly.
+- Returns an array of the same length where each value represents score of possible anomaly of corresponding element in the series.
+- A non-zero score indicates a possible anomaly.
 
 Type: [Array](../../sql-reference/data-types/array.md).
 
@@ -218,29 +254,29 @@ Type: [Array](../../sql-reference/data-types/array.md).
 Query:
 
 ``` sql
-seriesOutliersDetectTukey([-3,2.4,15,3.9,5,6,4.5,5.2,3,4,5,16,7,5,5,4]) AS print_0;
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4, 5, 12, 45, 12, 3, 3, 4, 5, 6]) AS print_0;
 ```
 
 Result:
 
 ``` text
-┌───────────print_0──────────────────────────────────────────────────────────────────┐
-│[-2.7121212121212137,0,4.196969696969699,0,0,0,0,0,0,0,0,4.803030303030305,0,0,0,0] │
-└────────────────────────────────────────────────────────────────────────────────────┘
+┌───────────print_0───────────────────┐
+│[0,0,0,0,0,0,0,0,0,10.5,0,0,0,0,0,0] │
+└─────────────────────────────────────┘
 ```
 
 Query:
 
 ``` sql
-seriesOutliersDetectTukey(arrayMap(x -> sin(x / 10), range(30))) AS print_0;
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.50, 5, 12, 45, 12, 3.40, 3, 4, 5, 6], 'ctukey', 20, 80, 1.5) AS print_0;
 ```
 
 Result:
 
 ``` text
-┌───────────print_0────────────────────────────────────────────┐
-│[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0] │
-└──────────────────────────────────────────────────────────────┘
+┌─print_0────────────────────────────┐
+│ [0,0,0,0,0,0,0,0,0,12,0,0,0,0,0,0] │
+└────────────────────────────────────┘
 ```)",
         .categories{"Time series analysis"}});
 }
diff --git a/tests/queries/0_stateless/02813_seriesOutliersTukey.reference b/tests/queries/0_stateless/02813_seriesOutliersTukey.reference
index 990c8c11e9e..bdcde0419a4 100644
--- a/tests/queries/0_stateless/02813_seriesOutliersTukey.reference
+++ b/tests/queries/0_stateless/02813_seriesOutliersTukey.reference
@@ -1,12 +1,14 @@
-[-2.7121212121212137,0,4.196969696969699,0,0,0,0,0,0,0,0,4.803030303030305,0,0,0,0]
-[0,0,0,0,0,0,0,0,0,5.228971962616823,0,0,0,0,0,0]
-[-2.7121212121212137,0,4.196969696969699,0,0,0,0,0,0,0,0,4.803030303030305,0,0,0,0]
-[0,0,0,0,0,0,0,0,0,5.228971962616823,0,0,0,0,0,0]
-[0,0,0,0,0,0,0,0,0,4.706896551724138,0,0,0,0,0,0]
-[-0.9615384615384613,0,1.9615384615384612,0,0,0,0,0,0.8076923076923075,19.26923076923077,0.8076923076923075,0,0,0,0,0,0,0]
-[0,0,0,0,0,0,0,0,0,4.706896551724138,0,0,0,0,0,0]
-[0,0,0,0,0,0,0,0,0,0.8076923076923077,0,0,0,0,0,0]
-[0,0,0,0,0,0,0,0,0,4.706896551724138,0,0,0,0,0,0]
+[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]
+[0,0,0,0,0,0,0,0,0,11.100000000000001,0,0,0,0,0,0]
+[-4.475000000000001,0,6.925000000000001,0,0,0,0,0,0,0,0,7.925000000000001,0,0,0,0]
+[0,0,0,0,0,0,0,0,0,27.975,0,0,0,0,0,0]
+[0,0,0,0,0,0,0,0,0,10.5,0,0,0,0,0,0]
+[0,0,0,0,0,0,0,0,0,26.1,0,0,0,0,0,0,0,0]
+[0,0,0,0,0,0,0,0,0,27.3,0,0,0,0,0,0]
+[0,0,0,0,0,0,0,0,0,10.5,0,0,0,0,0,0]
+[0,0,0,0,0,0,0,0,0,27.3,0,0,0,0,0,0]
 [0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]
 [0,0,0,0]
 [0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]
+[0,0,0,0,0,0,0,0,0,27,0,0,0,0,0,0]
+[0,0,0,0,0,0,0,0,0,18,0,0,0,0,0,0]
diff --git a/tests/queries/0_stateless/02813_seriesOutliersTukey.sql b/tests/queries/0_stateless/02813_seriesOutliersTukey.sql
index f8debc7b6db..7efe4903249 100644
--- a/tests/queries/0_stateless/02813_seriesOutliersTukey.sql
+++ b/tests/queries/0_stateless/02813_seriesOutliersTukey.sql
@@ -1,19 +1,29 @@
 DROP TABLE IF EXISTS tb1;
 
 CREATE TABLE tb1 (n UInt32, a Array(Float64)) engine=Memory;
-INSERT INTO tb1 VALUES (1, [-3,2.4,15,3.9,5,6,4.5,5.2,3,4,5,16,7,5,5,4]), (2, [-3,2.4,15,3.9,5,6,4.5,5.2,12,45,12,3.4,3,4,5,6]);
+INSERT INTO tb1 VALUES (1, [-3,2.40,15,3.90,5,6,4.50,5.20,3,4,5,16,7,5,5,4]), (2, [-3,2.40,15,3.90,5,6,4.50,5.20,12,45,12,3.40,3,4,5,6]);
 
+-- non-const inputs
 SELECT seriesOutliersDetectTukey(a) FROM tb1 ORDER BY n;
 SELECT seriesOutliersDetectTukey(a,'ctukey', 25,75) FROM tb1 ORDER BY n;
 DROP TABLE IF EXISTS tb1;
-SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.5, 5, 12, 45, 12, 3.4, 3, 4, 5, 6]);
-SELECT seriesOutliersDetectTukey([-3, 2.4, 15, 3.9, 5, 6, 4.5, 5.2, 12, 60, 12, 3.4, 3, 4, 5, 6, 3.4, 2.7]);
-SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.5, 5, 12, 45, 12, 3.4, 3, 4, 5, 6], 'ctukey', 25, 75);
-SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.5, 5, 12, 45, 12, 3.4, 3, 4, 5, 6], 'ctukey', 10, 90);
-SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.5, 5, 12, 45, 12, 3.4, 3, 4, 5, 6], 'tukey', 10, 90);
-SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.5, 5, 12, 45, 12, 3.4, 3, 4, 5, 6], 'ctukey', 2, 98)
+
+-- const inputs
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.50, 5, 12, 45, 12, 3.40, 3, 4, 5, 6]);
+SELECT seriesOutliersDetectTukey([-3, 2.40, 15, 3.90, 5, 6, 4.50, 5.20, 12, 60, 12, 3.40, 3, 4, 5, 6, 3.40, 2.7]);
+
+-- const inputs with optional arguments
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.50, 5, 12, 45, 12, 3.40, 3, 4, 5, 6], 'ctukey', 25, 75);
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.50, 5, 12, 45, 12, 3.40, 3, 4, 5, 6], 'ctukey', 10, 90);
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.50, 5, 12, 45, 12, 3.40, 3, 4, 5, 6], 'tukey', 10, 90);
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.50, 5, 12, 45, 12, 3.40, 3, 4, 5, 6], 'ctukey', 2, 98);
 SELECT seriesOutliersDetectTukey([-3, 2, 15, 3], 'ctukey', 2, 98);
-SELECT seriesOutliersDetectTukey(arrayMap(x -> sin(x / 10), range(30)));
+SELECT seriesOutliersDetectTukey(arrayMap(x -> sin(x / 10), range(30)), 'tukey');
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4, 5, 12, 45, 12, 3, 3, 4, 5, 6], 'tukey', 25, 75, 1.5);
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4, 5, 12, 45, 12, 3, 3, 4, 5, 6], 'tukey', 25, 75, 3);
+
+-- negative tests
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4, 5, 12, 45, 12, 3, 3, 4, 5, 6], 'tukey', 25, 75, -1); -- { serverError BAD_ARGUMENTS}
 SELECT seriesOutliersDetectTukey([-3, 2, 15, 3], 'xyz', 33, 53); -- { serverError BAD_ARGUMENTS}
 SELECT seriesOutliersDetectTukey([-3, 2.4, 15, NULL]); -- { serverError ILLEGAL_COLUMN}
 SELECT seriesOutliersDetectTukey([]); -- { serverError ILLEGAL_COLUMN}
diff --git a/utils/check-style/aspell-ignore/en/aspell-dict.txt b/utils/check-style/aspell-ignore/en/aspell-dict.txt
index 64327aba2d1..3c8ec34de8d 100644
--- a/utils/check-style/aspell-ignore/en/aspell-dict.txt
+++ b/utils/check-style/aspell-ignore/en/aspell-dict.txt
@@ -1332,6 +1332,7 @@ cryptographic
 csv
 csvwithnames
 csvwithnamesandtypes
+ctukey
 curdate
 currentDatabase
 currentProfiles
@@ -2271,6 +2272,7 @@ seektable
 sequenceCount
 sequenceMatch
 sequenceNextNode
+seriesOutliersDetectTukey
 seriesDecomposeSTL
 seriesPeriodDetectFFT
 serverTimeZone
@@ -2564,6 +2566,7 @@ tryPunycodeDecode
 tskv
 tsv
 tui
+tukey
 tumbleEnd
 tumbleStart
 tupleConcat

From 6d24ffc9761d05422f0d98d3a4797d0c348e3303 Mon Sep 17 00:00:00 2001
From: Bhavna Jindal <bhavna.jindal@ibm.com>
Date: Wed, 31 Jan 2024 10:57:43 -0800
Subject: [PATCH 10/73] Fix filename and spelling errors

---
 ...eriesOutliersTukey.cpp => seriesOutliersDetectTukey.cpp} | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
 rename src/Functions/{seriesOutliersTukey.cpp => seriesOutliersDetectTukey.cpp} (98%)

diff --git a/src/Functions/seriesOutliersTukey.cpp b/src/Functions/seriesOutliersDetectTukey.cpp
similarity index 98%
rename from src/Functions/seriesOutliersTukey.cpp
rename to src/Functions/seriesOutliersDetectTukey.cpp
index 4c2c1ccd882..ce5ed391fa0 100644
--- a/src/Functions/seriesOutliersTukey.cpp
+++ b/src/Functions/seriesOutliersDetectTukey.cpp
@@ -79,7 +79,7 @@ public:
                         min_percentile = p_min / 100;
                     else
                         throw Exception(
-                            ErrorCodes::BAD_ARGUMENTS, "The third argumet of function {} must be in range [2.0, 98.0]", getName());
+                            ErrorCodes::BAD_ARGUMENTS, "The third argument of function {} must be in range [2.0, 98.0]", getName());
                 }
 
                 if (arguments.size() > 3)
@@ -89,7 +89,7 @@ public:
                         max_percentile = p_max / 100;
                     else
                         throw Exception(
-                            ErrorCodes::BAD_ARGUMENTS, "The fourth argumet of function {} must be in range [2.0, 98.0]", getName());
+                            ErrorCodes::BAD_ARGUMENTS, "The fourth argument of function {} must be in range [2.0, 98.0]", getName());
                 }
             }
             else if (kind == "tukey")
@@ -109,7 +109,7 @@ public:
             if (k_val >= 0.0)
                 K = k_val;
             else
-                throw Exception(ErrorCodes::BAD_ARGUMENTS, "The fifth argumet of function {} must be a positive number", getName());
+                throw Exception(ErrorCodes::BAD_ARGUMENTS, "The fifth argument of function {} must be a positive number", getName());
         }
 
         ColumnPtr col_res;

From 2ca634702b3318441e8629985eee7d859d5e3d3f Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 31 Jan 2024 23:15:51 +0100
Subject: [PATCH 11/73] Fix test

---
 tests/queries/0_stateless/02184_default_table_engine.sql | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tests/queries/0_stateless/02184_default_table_engine.sql b/tests/queries/0_stateless/02184_default_table_engine.sql
index a984ec1b6c9..aff30eeea98 100644
--- a/tests/queries/0_stateless/02184_default_table_engine.sql
+++ b/tests/queries/0_stateless/02184_default_table_engine.sql
@@ -1,3 +1,5 @@
+SET default_table_engine = 'None';
+
 CREATE TABLE table_02184 (x UInt8); --{serverError 119}
 SET default_table_engine = 'Log';
 CREATE TABLE table_02184 (x UInt8);

From a043227685a254159a0b1469d11e5f9756f943b3 Mon Sep 17 00:00:00 2001
From: vdimir <vdimir@clickhouse.com>
Date: Thu, 1 Feb 2024 10:36:31 +0000
Subject: [PATCH 12/73] Fix ARRAY JOIN with subcolumns in analyzer

---
 src/Analyzer/Passes/QueryAnalysisPass.cpp     | 96 ++++++++++++++++++-
 tests/analyzer_integration_broken_tests.txt   |  1 -
 ...74_analyzer_array_join_subcolumn.reference | 16 ++++
 .../02974_analyzer_array_join_subcolumn.sql   | 24 +++++
 4 files changed, 133 insertions(+), 4 deletions(-)
 create mode 100644 tests/queries/0_stateless/02974_analyzer_array_join_subcolumn.reference
 create mode 100644 tests/queries/0_stateless/02974_analyzer_array_join_subcolumn.sql

diff --git a/src/Analyzer/Passes/QueryAnalysisPass.cpp b/src/Analyzer/Passes/QueryAnalysisPass.cpp
index 881d5cd00b2..d92acc2b948 100644
--- a/src/Analyzer/Passes/QueryAnalysisPass.cpp
+++ b/src/Analyzer/Passes/QueryAnalysisPass.cpp
@@ -424,6 +424,7 @@ struct TableExpressionData
     bool should_qualify_columns = true;
     NamesAndTypes column_names_and_types;
     ColumnNameToColumnNodeMap column_name_to_column_node;
+    std::unordered_set<std::string> subcolumn_names; /// Subset columns that are subcolumns of other columns
     std::unordered_set<std::string, StringTransparentHash, std::equal_to<>> column_identifier_first_parts;
 
     bool hasFullIdentifierName(IdentifierView identifier_view) const
@@ -1306,6 +1307,12 @@ private:
         const QueryTreeNodePtr & table_expression_node,
         IdentifierResolveScope & scope);
 
+    QueryTreeNodePtr matchArrayJoinSubcolumns(
+        const QueryTreeNodePtr & array_join_column_inner_expression,
+        const ColumnNode & array_join_column_expression_typed,
+        const QueryTreeNodePtr & resolved_expression,
+        IdentifierResolveScope & scope);
+
     QueryTreeNodePtr tryResolveExpressionFromArrayJoinExpressions(const QueryTreeNodePtr & resolved_expression,
         const QueryTreeNodePtr & table_expression_node,
         IdentifierResolveScope & scope);
@@ -2910,8 +2917,23 @@ QueryTreeNodePtr QueryAnalyzer::tryResolveIdentifierFromStorage(
     QueryTreeNodePtr result_expression;
     bool match_full_identifier = false;
 
-    auto it = table_expression_data.column_name_to_column_node.find(identifier_without_column_qualifier.getFullName());
-    if (it != table_expression_data.column_name_to_column_node.end())
+    const auto & identifier_full_name = identifier_without_column_qualifier.getFullName();
+    auto it = table_expression_data.column_name_to_column_node.find(identifier_full_name);
+    bool can_resolve_directly_from_storage = it != table_expression_data.column_name_to_column_node.end();
+    if (can_resolve_directly_from_storage && table_expression_data.subcolumn_names.contains(identifier_full_name))
+    {
+        /** In the case when we have an ARRAY JOIN, we should not resolve subcolumns directly from storage.
+          * For example, consider the following SQL query:
+          * SELECT ProfileEvents.Values FROM system.query_log ARRAY JOIN ProfileEvents
+          * In this case, ProfileEvents.Values should also be array joined, not directly resolved from storage.
+          */
+        auto * nearest_query_scope = scope.getNearestQueryScope();
+        auto * nearest_query_scope_query_node = nearest_query_scope ? nearest_query_scope->scope_node->as<QueryNode>() : nullptr;
+        if (nearest_query_scope_query_node && nearest_query_scope_query_node->getJoinTree()->getNodeType() == QueryTreeNodeType::ARRAY_JOIN)
+            can_resolve_directly_from_storage = false;
+    }
+
+    if (can_resolve_directly_from_storage)
     {
         match_full_identifier = true;
         result_expression = it->second;
@@ -3394,6 +3416,68 @@ QueryTreeNodePtr QueryAnalyzer::tryResolveIdentifierFromJoin(const IdentifierLoo
     return resolved_identifier;
 }
 
+QueryTreeNodePtr QueryAnalyzer::matchArrayJoinSubcolumns(
+    const QueryTreeNodePtr & array_join_column_inner_expression,
+    const ColumnNode & array_join_column_expression_typed,
+    const QueryTreeNodePtr & resolved_expression,
+    IdentifierResolveScope & scope)
+{
+    const auto * resolved_function = resolved_expression->as<FunctionNode>();
+    if (!resolved_function || resolved_function->getFunctionName() != "getSubcolumn")
+        return {};
+
+    const auto * array_join_parent_column = array_join_column_inner_expression.get();
+
+    /** If both resolved and array-joined expressions are subcolumns, try to match them:
+      * For example, in `SELECT t.map.values FROM (SELECT * FROM tbl) ARRAY JOIN t.map`
+      * Identifier `t.map.values` is resolved into `getSubcolumn(t, 'map.values')` and t.map is resolved into `getSubcolumn(t, 'map')`
+      * Since we need to perform array join on `getSubcolumn(t, 'map')`, `t.map.values` should become `getSubcolumn(getSubcolumn(t, 'map'), 'values')`
+      *
+      * Note: It doesn't work when subcolumn in ARRAY JOIN is transformed by another expression, for example
+      * SELECT c.map, c.map.values FROM (SELECT * FROM tbl) ARRAY JOIN mapApply(x -> x, t.map);
+      */
+    String array_join_subcolumn_prefix;
+    auto * array_join_column_inner_expression_function = array_join_column_inner_expression->as<FunctionNode>();
+    if (array_join_column_inner_expression_function &&
+        array_join_column_inner_expression_function->getFunctionName() == "getSubcolumn")
+    {
+        const auto & argument_nodes = array_join_column_inner_expression_function->getArguments().getNodes();
+        if (argument_nodes.size() == 2 && argument_nodes.at(1)->getNodeType() == QueryTreeNodeType::CONSTANT)
+        {
+            const auto & constant_node = argument_nodes.at(1)->as<ConstantNode &>();
+            const auto & constant_node_value = constant_node.getValue();
+            if (constant_node_value.getType() == Field::Types::String)
+            {
+                array_join_subcolumn_prefix = constant_node_value.get<String>() + ".";
+                array_join_parent_column = argument_nodes.at(0).get();
+            }
+        }
+    }
+
+    const auto & argument_nodes = resolved_function->getArguments().getNodes();
+    if (argument_nodes.size() != 2 && !array_join_parent_column->isEqual(*argument_nodes.at(0)))
+        return {};
+
+    const auto * second_argument = argument_nodes.at(1)->as<ConstantNode>();
+    if (!second_argument || second_argument->getValue().getType() != Field::Types::String)
+        throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected constant string as second argument of getSubcolumn function {}", resolved_function->dumpTree());
+
+    const auto & resolved_subcolumn_path = second_argument->getValue().get<String &>();
+    if (!startsWith(resolved_subcolumn_path, array_join_subcolumn_prefix))
+        return {};
+
+    auto get_subcolumn_function = std::make_shared<FunctionNode>("getSubcolumn");
+    get_subcolumn_function->getArguments().getNodes().push_back(
+        std::make_shared<ColumnNode>(array_join_column_expression_typed.getColumn(), array_join_column_expression_typed.getColumnSource()));
+    get_subcolumn_function->getArguments().getNodes().push_back(
+        std::make_shared<ConstantNode>(resolved_subcolumn_path.substr(array_join_subcolumn_prefix.size())));
+
+    QueryTreeNodePtr function_query_node = get_subcolumn_function;
+    resolveFunction(function_query_node, scope);
+
+    return function_query_node;
+}
+
 QueryTreeNodePtr QueryAnalyzer::tryResolveExpressionFromArrayJoinExpressions(const QueryTreeNodePtr & resolved_expression,
     const QueryTreeNodePtr & table_expression_node,
     IdentifierResolveScope & scope)
@@ -3462,8 +3546,12 @@ QueryTreeNodePtr QueryAnalyzer::tryResolveExpressionFromArrayJoinExpressions(con
                 array_join_column_expression_typed.getColumnSource());
             break;
         }
-    }
 
+        /// When we select subcolumn of array joined column it also should be array joined
+        array_join_resolved_expression = matchArrayJoinSubcolumns(array_join_column_inner_expression, array_join_column_expression_typed, resolved_expression, scope);
+        if (array_join_resolved_expression)
+            break;
+    }
     return array_join_resolved_expression;
 }
 
@@ -6424,6 +6512,8 @@ void QueryAnalyzer::initializeTableExpressionData(const QueryTreeNodePtr & table
           */
         for (const auto & column_name_and_type : table_expression_data.column_names_and_types)
         {
+            for (const auto & subcolumn : columns_description.getSubcolumns(column_name_and_type.name))
+                table_expression_data.subcolumn_names.insert(subcolumn.name);
             const auto & column_default = columns_description.getDefault(column_name_and_type.name);
 
             if (column_default && column_default->kind == ColumnDefaultKind::Alias)
diff --git a/tests/analyzer_integration_broken_tests.txt b/tests/analyzer_integration_broken_tests.txt
index e1d4de59a23..06142f98da1 100644
--- a/tests/analyzer_integration_broken_tests.txt
+++ b/tests/analyzer_integration_broken_tests.txt
@@ -10,7 +10,6 @@ test_merge_table_over_distributed/test.py::test_select_table_name_from_merge_ove
 test_mutations_with_merge_tree/test.py::test_mutations_with_merge_background_task
 test_mysql_database_engine/test.py::test_mysql_ddl_for_mysql_database
 test_passing_max_partitions_to_read_remotely/test.py::test_default_database_on_cluster
-test_profile_events_s3/test.py::test_profile_events
 test_replicating_constants/test.py::test_different_versions
 test_select_access_rights/test_main.py::test_alias_columns
 test_select_access_rights/test_main.py::test_select_count
diff --git a/tests/queries/0_stateless/02974_analyzer_array_join_subcolumn.reference b/tests/queries/0_stateless/02974_analyzer_array_join_subcolumn.reference
new file mode 100644
index 00000000000..827c710ef1a
--- /dev/null
+++ b/tests/queries/0_stateless/02974_analyzer_array_join_subcolumn.reference
@@ -0,0 +1,16 @@
+('a',(1,2))	1
+('b',(2,3))	2
+('a',(1,2))	1
+('b',(2,3))	2
+('a',(1,2))	1
+('b',(2,3))	2
+('a',(1,2))	1
+('b',(2,3))	2
+('a',(1,2))	1
+('b',(2,3))	2
+('a',(1,2))	1
+('b',(2,3))	2
+('a',(1,2))	1
+('b',(2,3))	2
+('a',(1,2))	1
+('b',(2,3))	2
diff --git a/tests/queries/0_stateless/02974_analyzer_array_join_subcolumn.sql b/tests/queries/0_stateless/02974_analyzer_array_join_subcolumn.sql
new file mode 100644
index 00000000000..14823644b96
--- /dev/null
+++ b/tests/queries/0_stateless/02974_analyzer_array_join_subcolumn.sql
@@ -0,0 +1,24 @@
+DROP TABLE IF EXISTS t2;
+DROP TABLE IF EXISTS t3;
+
+CREATE TABLE t2 (id Int32, pe Map(String, Tuple(a UInt64, b UInt64))) ENGINE = MergeTree ORDER BY id;
+INSERT INTO t2 VALUES (1, {'a': (1, 2), 'b': (2, 3)}),
+
+CREATE TABLE t3 (id Int32, c Tuple(v String, pe Map(String, Tuple(a UInt64, b UInt64)))) ENGINE = MergeTree ORDER BY id;
+INSERT INTO t3 VALUES (1, ('A', {'a':(1, 2),'b':(2, 3)}));
+
+SELECT pe, pe.values.a FROM (SELECT * FROM t2) ARRAY JOIN pe SETTINGS allow_experimental_analyzer = 1;
+SELECT p, p.values.a FROM (SELECT * FROM t2) ARRAY JOIN pe AS p SETTINGS allow_experimental_analyzer = 1;
+
+SELECT pe, pe.values.a FROM t2 ARRAY JOIN pe;
+SELECT p, p.values.a FROM t2 ARRAY JOIN pe AS p;
+
+SELECT c.pe, c.pe.values.a FROM (SELECT * FROM t3) ARRAY JOIN c.pe SETTINGS allow_experimental_analyzer = 1;
+SELECT p, p.values.a FROM (SELECT * FROM t3) ARRAY JOIN c.pe as p SETTINGS allow_experimental_analyzer = 1;
+
+SELECT c.pe, c.pe.values.a FROM t3 ARRAY JOIN c.pe SETTINGS allow_experimental_analyzer = 1;
+SELECT p, p.values.a FROM t3 ARRAY JOIN c.pe as p;
+
+
+DROP TABLE IF EXISTS t2;
+DROP TABLE IF EXISTS t3;

From 92a00d317cddec0e65d398db01ef7ee71fc818a5 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Wed, 31 Jan 2024 11:38:22 +0000
Subject: [PATCH 13/73] Cosmetics

---
 base/base/getMemoryAmount.cpp              | 17 +++++----
 src/Common/getNumberOfPhysicalCPUCores.cpp | 40 +++++++++++-----------
 2 files changed, 30 insertions(+), 27 deletions(-)

diff --git a/base/base/getMemoryAmount.cpp b/base/base/getMemoryAmount.cpp
index a46e964c5a3..2e89ac52650 100644
--- a/base/base/getMemoryAmount.cpp
+++ b/base/base/getMemoryAmount.cpp
@@ -1,8 +1,10 @@
-#include <stdexcept>
-#include <fstream>
 #include <base/getMemoryAmount.h>
+
 #include <base/getPageSize.h>
 
+#include <fstream>
+#include <stdexcept>
+
 #include <unistd.h>
 #include <sys/types.h>
 #include <sys/param.h>
@@ -27,10 +29,11 @@ uint64_t getMemoryAmountOrZero()
     uint64_t memory_amount = num_pages * page_size;
 
 #if defined(OS_LINUX)
-    // Try to lookup at the Cgroup limit
+    /// Limit the memory amount by limits set by cgroups
+    std::filesystem::path default_cgroups_mount = "/sys/fs/cgroup";
 
-    // CGroups v2
-    std::ifstream cgroupv2_limit("/sys/fs/cgroup/memory.max");
+    /// cgroups v2
+    std::ifstream cgroupv2_limit(default_cgroups_mount / "memory.max");
     if (cgroupv2_limit.is_open())
     {
         uint64_t memory_limit = 0;
@@ -40,8 +43,8 @@ uint64_t getMemoryAmountOrZero()
     }
     else
     {
-        // CGroups v1
-        std::ifstream cgroup_limit("/sys/fs/cgroup/memory/memory.limit_in_bytes");
+        /// cgroups v1
+        std::ifstream cgroup_limit(default_cgroups_mount / "memory/memory.limit_in_bytes");
         if (cgroup_limit.is_open())
         {
             uint64_t memory_limit = 0; // in case of read error
diff --git a/src/Common/getNumberOfPhysicalCPUCores.cpp b/src/Common/getNumberOfPhysicalCPUCores.cpp
index 8fbb32e911f..21060bbfaf1 100644
--- a/src/Common/getNumberOfPhysicalCPUCores.cpp
+++ b/src/Common/getNumberOfPhysicalCPUCores.cpp
@@ -1,7 +1,6 @@
 #include "getNumberOfPhysicalCPUCores.h"
 #include <filesystem>
 
-#include "config.h"
 #if defined(OS_LINUX)
 #    include <cmath>
 #    include <fstream>
@@ -34,9 +33,9 @@ int32_t readFrom(const std::filesystem::path & filename, int default_value)
 uint32_t getCGroupLimitedCPUCores(unsigned default_cpu_count)
 {
     uint32_t quota_count = default_cpu_count;
-    std::filesystem::path prefix = "/sys/fs/cgroup";
+    std::filesystem::path default_cgroups_mount = "/sys/fs/cgroup";
     /// cgroupsv2
-    std::ifstream contr_file(prefix / "cgroup.controllers");
+    std::ifstream contr_file(default_cgroups_mount / "cgroup.controllers");
     if (contr_file.is_open())
     {
         /// First, we identify the cgroup the process belongs
@@ -51,16 +50,15 @@ uint32_t getCGroupLimitedCPUCores(unsigned default_cpu_count)
 
         std::filesystem::path current_cgroup;
         if (cgroup_name.empty())
-            current_cgroup = prefix;
+            current_cgroup = default_cgroups_mount;
         else
-            current_cgroup = prefix / cgroup_name;
+            current_cgroup = default_cgroups_mount / cgroup_name;
 
         // Looking for cpu.max in directories from the current cgroup to the top level
         // It does not stop on the first time since the child could have a greater value than parent
-        while (current_cgroup != prefix.parent_path())
+        while (current_cgroup != default_cgroups_mount.parent_path())
         {
             std::ifstream cpu_max_file(current_cgroup / "cpu.max");
-            current_cgroup = current_cgroup.parent_path();
             if (cpu_max_file.is_open())
             {
                 std::string cpu_limit_str;
@@ -72,10 +70,11 @@ uint32_t getCGroupLimitedCPUCores(unsigned default_cpu_count)
                     quota_count = std::min(static_cast<uint32_t>(ceil(cpu_limit / cpu_period)), quota_count);
                 }
             }
+            current_cgroup = current_cgroup.parent_path();
         }
-        current_cgroup = prefix / cgroup_name;
+        current_cgroup = default_cgroups_mount / cgroup_name;
         // Looking for cpuset.cpus.effective in directories from the current cgroup to the top level
-        while (current_cgroup != prefix.parent_path())
+        while (current_cgroup != default_cgroups_mount.parent_path())
         {
             std::ifstream cpuset_cpus_file(current_cgroup / "cpuset.cpus.effective");
             current_cgroup = current_cgroup.parent_path();
@@ -113,8 +112,8 @@ uint32_t getCGroupLimitedCPUCores(unsigned default_cpu_count)
     /// cgroupsv1
     /// Return the number of milliseconds per period process is guaranteed to run.
     /// -1 for no quota
-    int cgroup_quota = readFrom(prefix / "cpu/cpu.cfs_quota_us", -1);
-    int cgroup_period = readFrom(prefix / "cpu/cpu.cfs_period_us", -1);
+    int cgroup_quota = readFrom(default_cgroups_mount / "cpu/cpu.cfs_quota_us", -1);
+    int cgroup_period = readFrom(default_cgroups_mount / "cpu/cpu.cfs_period_us", -1);
     if (cgroup_quota > -1 && cgroup_period > 0)
         quota_count = static_cast<uint32_t>(ceil(static_cast<float>(cgroup_quota) / static_cast<float>(cgroup_period)));
 
@@ -178,24 +177,25 @@ catch (...)
 
 unsigned getNumberOfPhysicalCPUCoresImpl()
 {
-    unsigned cpu_count = std::thread::hardware_concurrency(); /// logical cores (with SMT/HyperThreading)
+    unsigned cores = std::thread::hardware_concurrency(); /// logical cores (with SMT/HyperThreading)
 
+
+#if defined(__x86_64__) && defined(OS_LINUX)
     /// Most x86_64 CPUs have 2-way SMT (Hyper-Threading).
     /// Aarch64 and RISC-V don't have SMT so far.
     /// POWER has SMT and it can be multi-way (e.g. 8-way), but we don't know how ClickHouse really behaves, so use all of them.
-
-#if defined(__x86_64__) && defined(OS_LINUX)
+    ///
     /// On really big machines, SMT is detrimental to performance (+ ~5% overhead in ClickBench). On such machines, we limit ourself to the physical cores.
     /// Few cores indicate it is a small machine, runs in a VM or is a limited cloud instance --> it is reasonable to use all the cores.
-    if (cpu_count >= 32)
-        cpu_count = physical_concurrency();
+    if (cores >= 32)
+        cores = physical_concurrency();
 #endif
 
 #if defined(OS_LINUX)
-    cpu_count = getCGroupLimitedCPUCores(cpu_count);
+    cores = getCGroupLimitedCPUCores(cores);
 #endif
 
-    return cpu_count;
+    return cores;
 }
 
 }
@@ -203,6 +203,6 @@ unsigned getNumberOfPhysicalCPUCoresImpl()
 unsigned getNumberOfPhysicalCPUCores()
 {
     /// Calculate once.
-    static auto res = getNumberOfPhysicalCPUCoresImpl();
-    return res;
+    static auto cores = getNumberOfPhysicalCPUCoresImpl();
+    return cores;
 }

From ad0c0d87276e7a2a9eedee07a41a55895f7d88a5 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Wed, 31 Jan 2024 11:47:38 +0000
Subject: [PATCH 14/73] Factorize cgroup memory access code

This fixes the edge case that cgroup memory limit "0" is no longer
interpreted as "no limit" but as actual zero available memory, see (*).
The cgroup default setting "max" continues to be interpreted as "no
limit".

(*) https://docs.kernel.org/admin-guide/cgroup-v2.html
---
 base/base/getMemoryAmount.cpp | 51 ++++++++++++++++++++---------------
 1 file changed, 29 insertions(+), 22 deletions(-)

diff --git a/base/base/getMemoryAmount.cpp b/base/base/getMemoryAmount.cpp
index 2e89ac52650..167cd807774 100644
--- a/base/base/getMemoryAmount.cpp
+++ b/base/base/getMemoryAmount.cpp
@@ -13,6 +13,28 @@
 #endif
 
 
+namespace
+{
+
+std::optional<uint64_t> getCgroupsMemoryLimit(const std::string & setting)
+{
+#if defined(OS_LINUX)
+    std::filesystem::path default_cgroups_mount = "/sys/fs/cgroup";
+    std::ifstream file(default_cgroups_mount / setting);
+    if (!file.is_open())
+        return {};
+    uint64_t value;
+    if (file >> value)
+        return {value};
+    else
+        return {}; /// e.g. the cgroups default "max"
+#else
+    return {};
+#endif
+}
+
+}
+
 /** Returns the size of physical memory (RAM) in bytes.
   * Returns 0 on unsupported platform
   */
@@ -28,35 +50,20 @@ uint64_t getMemoryAmountOrZero()
 
     uint64_t memory_amount = num_pages * page_size;
 
-#if defined(OS_LINUX)
-    /// Limit the memory amount by limits set by cgroups
-    std::filesystem::path default_cgroups_mount = "/sys/fs/cgroup";
+    /// Respect the memory limit set by cgroups
 
     /// cgroups v2
-    std::ifstream cgroupv2_limit(default_cgroups_mount / "memory.max");
-    if (cgroupv2_limit.is_open())
-    {
-        uint64_t memory_limit = 0;
-        cgroupv2_limit >> memory_limit;
-        if (memory_limit > 0 && memory_limit < memory_amount)
-            memory_amount = memory_limit;
-    }
+    auto limit_v2 = getCgroupsMemoryLimit("memory.max");
+    if (limit_v2.has_value() && *limit_v2 < memory_amount)
+         memory_amount = *limit_v2;
     else
     {
-        /// cgroups v1
-        std::ifstream cgroup_limit(default_cgroups_mount / "memory/memory.limit_in_bytes");
-        if (cgroup_limit.is_open())
-        {
-            uint64_t memory_limit = 0; // in case of read error
-            cgroup_limit >> memory_limit;
-            if (memory_limit > 0 && memory_limit < memory_amount)
-                memory_amount = memory_limit;
-        }
+        auto limit_v1 = getCgroupsMemoryLimit("memory/memory.limit_in_bytes");
+        if (limit_v1.has_value() && *limit_v1 < memory_amount)
+             memory_amount = *limit_v1;
     }
-#endif
 
     return memory_amount;
-
 }
 
 

From 3243ea0b66ed9b435011aa6570b70ed080cd5410 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Wed, 31 Jan 2024 12:49:03 +0000
Subject: [PATCH 15/73] Remove cgroups v1 support to simplify the code going
 forward

---
 base/base/getMemoryAmount.cpp | 18 ++++++------------
 1 file changed, 6 insertions(+), 12 deletions(-)

diff --git a/base/base/getMemoryAmount.cpp b/base/base/getMemoryAmount.cpp
index 167cd807774..060de685778 100644
--- a/base/base/getMemoryAmount.cpp
+++ b/base/base/getMemoryAmount.cpp
@@ -16,7 +16,7 @@
 namespace
 {
 
-std::optional<uint64_t> getCgroupsMemoryLimit(const std::string & setting)
+std::optional<uint64_t> getCgroupsV2MemoryLimit(const std::string & setting)
 {
 #if defined(OS_LINUX)
     std::filesystem::path default_cgroups_mount = "/sys/fs/cgroup";
@@ -50,18 +50,12 @@ uint64_t getMemoryAmountOrZero()
 
     uint64_t memory_amount = num_pages * page_size;
 
-    /// Respect the memory limit set by cgroups
+    /// Respect the memory limit set by cgroups v2.
+    /// Cgroups v1 is dead since many years and its limits are not considered for simplicity.
 
-    /// cgroups v2
-    auto limit_v2 = getCgroupsMemoryLimit("memory.max");
-    if (limit_v2.has_value() && *limit_v2 < memory_amount)
-         memory_amount = *limit_v2;
-    else
-    {
-        auto limit_v1 = getCgroupsMemoryLimit("memory/memory.limit_in_bytes");
-        if (limit_v1.has_value() && *limit_v1 < memory_amount)
-             memory_amount = *limit_v1;
-    }
+    auto limit = getCgroupsV2MemoryLimit("memory.max");
+    if (limit.has_value() && *limit < memory_amount)
+         memory_amount = *limit;
 
     return memory_amount;
 }

From 1a4f56a61ac8dc5dab3245f6574769991bffc927 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <benjamin.antal@clickhouse.com>
Date: Wed, 31 Jan 2024 19:33:04 +0000
Subject: [PATCH 16/73] Fix `ASTAlterCommand::formatImpl` in case of column
 specific settings modifications

---
 src/Parsers/ASTAlterQuery.cpp | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/src/Parsers/ASTAlterQuery.cpp b/src/Parsers/ASTAlterQuery.cpp
index e229095df1b..dd7f40cff95 100644
--- a/src/Parsers/ASTAlterQuery.cpp
+++ b/src/Parsers/ASTAlterQuery.cpp
@@ -104,6 +104,14 @@ void ASTAlterCommand::formatImpl(const FormatSettings & settings, FormatState &
         {
             settings.ostr << (settings.hilite ? hilite_keyword : "") << " REMOVE " << remove_property;
         }
+        else if (settings_changes) {
+            settings.ostr << (settings.hilite ? hilite_keyword : "") << "MODIFY SETTING " << (settings.hilite ? hilite_none : "");
+            settings_changes->formatImpl(settings, state, frame);
+        }
+        else if (settings_resets) {
+            settings.ostr << (settings.hilite ? hilite_keyword : "") << "RESET SETTING " << (settings.hilite ? hilite_none : "");
+            settings_resets->formatImpl(settings, state, frame);
+        }
         else
         {
             if (first)

From 7c04a1a972f3e639f6f4e6bd3efc4e1c8ba8557d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <benjamin.antal@clickhouse.com>
Date: Wed, 31 Jan 2024 20:32:07 +0000
Subject: [PATCH 17/73] Test query formatting

---
 .../0_stateless/02870_per_column_settings.reference       | 8 ++++++--
 tests/queries/0_stateless/02870_per_column_settings.sql   | 8 ++++++--
 2 files changed, 12 insertions(+), 4 deletions(-)

diff --git a/tests/queries/0_stateless/02870_per_column_settings.reference b/tests/queries/0_stateless/02870_per_column_settings.reference
index 144c8c5ee2e..be08e7a682b 100644
--- a/tests/queries/0_stateless/02870_per_column_settings.reference
+++ b/tests/queries/0_stateless/02870_per_column_settings.reference
@@ -1,10 +1,14 @@
 CREATE TABLE default.tab\n(\n    `id` UInt64,\n    `long_string` String SETTINGS (min_compress_block_size = 163840, max_compress_block_size = 163840),\n    `v1` String,\n    `v2` UInt64,\n    `v3` Float32,\n    `v4` Float64\n)\nENGINE = ReplicatedMergeTree(\'/clickhouse/tables/default/tab/2870\', \'r1\')\nORDER BY id\nSETTINGS min_bytes_for_wide_part = 1, index_granularity = 8192
 1000
+ALTER TABLE tab\n    MODIFY COLUMN `long_string`MODIFY SETTING min_compress_block_size = 8192
 CREATE TABLE default.tab\n(\n    `id` UInt64,\n    `long_string` String SETTINGS (min_compress_block_size = 8192, max_compress_block_size = 163840),\n    `v1` String,\n    `v2` UInt64,\n    `v3` Float32,\n    `v4` Float64\n)\nENGINE = ReplicatedMergeTree(\'/clickhouse/tables/default/tab/2870\', \'r1\')\nORDER BY id\nSETTINGS min_bytes_for_wide_part = 1, index_granularity = 8192
+ALTER TABLE tab\n    MODIFY COLUMN `long_string`RESET SETTING min_compress_block_size
 CREATE TABLE default.tab\n(\n    `id` UInt64,\n    `long_string` String SETTINGS (max_compress_block_size = 163840),\n    `v1` String,\n    `v2` UInt64,\n    `v3` Float32,\n    `v4` Float64\n)\nENGINE = ReplicatedMergeTree(\'/clickhouse/tables/default/tab/2870\', \'r1\')\nORDER BY id\nSETTINGS min_bytes_for_wide_part = 1, index_granularity = 8192
+ALTER TABLE tab\n    MODIFY COLUMN `long_string` REMOVE SETTINGS
 CREATE TABLE default.tab\n(\n    `id` UInt64,\n    `long_string` String,\n    `v1` String,\n    `v2` UInt64,\n    `v3` Float32,\n    `v4` Float64\n)\nENGINE = ReplicatedMergeTree(\'/clickhouse/tables/default/tab/2870\', \'r1\')\nORDER BY id\nSETTINGS min_bytes_for_wide_part = 1, index_granularity = 8192
+ALTER TABLE tab\n    MODIFY COLUMN `long_string` String SETTINGS (min_compress_block_size = 163840, max_compress_block_size = 163840)
 CREATE TABLE default.tab\n(\n    `id` UInt64,\n    `long_string` String SETTINGS (min_compress_block_size = 163840, max_compress_block_size = 163840),\n    `v1` String,\n    `v2` UInt64,\n    `v3` Float32,\n    `v4` Float64\n)\nENGINE = ReplicatedMergeTree(\'/clickhouse/tables/default/tab/2870\', \'r1\')\nORDER BY id\nSETTINGS min_bytes_for_wide_part = 1, index_granularity = 8192
---- 
+---
 (0,0)	0
 (1,1)	1
 (2,2)	2
@@ -15,4 +19,4 @@ CREATE TABLE default.tab\n(\n    `id` UInt64,\n    `long_string` String SETTINGS
 (7,7)	7
 (8,8)	8
 (9,9)	9
---- 
+---
diff --git a/tests/queries/0_stateless/02870_per_column_settings.sql b/tests/queries/0_stateless/02870_per_column_settings.sql
index 345cf5cc744..d242ebe6c61 100644
--- a/tests/queries/0_stateless/02870_per_column_settings.sql
+++ b/tests/queries/0_stateless/02870_per_column_settings.sql
@@ -23,21 +23,25 @@ SHOW CREATE tab;
 INSERT INTO TABLE tab SELECT number, randomPrintableASCII(1000), randomPrintableASCII(10), rand(number), rand(number+1), rand(number+2) FROM numbers(1000);
 SELECT count() FROM tab;
 
+SELECT formatQuery('ALTER TABLE tab MODIFY COLUMN long_string MODIFY SETTING min_compress_block_size = 8192;');
 ALTER TABLE tab MODIFY COLUMN long_string MODIFY SETTING min_compress_block_size = 8192;
 SHOW CREATE tab;
 
+SELECT formatQuery('ALTER TABLE tab MODIFY COLUMN long_string RESET SETTING min_compress_block_size;');
 ALTER TABLE tab MODIFY COLUMN long_string RESET SETTING min_compress_block_size;
 SHOW CREATE tab;
 
+SELECT formatQuery('ALTER TABLE tab MODIFY COLUMN long_string REMOVE SETTINGS;');
 ALTER TABLE tab MODIFY COLUMN long_string REMOVE SETTINGS;
 SHOW CREATE tab;
 
+SELECT formatQuery('ALTER TABLE tab MODIFY COLUMN long_string String SETTINGS (min_compress_block_size = 163840, max_compress_block_size = 163840);');
 ALTER TABLE tab MODIFY COLUMN long_string String SETTINGS (min_compress_block_size = 163840, max_compress_block_size = 163840);
 SHOW CREATE tab;
 
 DROP TABLE tab;
 
-SELECT '--- ';
+SELECT '---';
 
 SET allow_experimental_object_type = 1;
 
@@ -56,7 +60,7 @@ SELECT tup, json.key AS key FROM tab ORDER BY key LIMIT 10;
 
 DROP TABLE tab;
 
-SELECT '--- ';
+SELECT '---';
 
 -- Unsupported column-level settings are rejected
 CREATE TABLE tab

From e8c9500981e9ebda0d4de5ff8db693fbe92a31ff Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <antaljanosbenjamin@users.noreply.github.com>
Date: Wed, 31 Jan 2024 21:56:15 +0100
Subject: [PATCH 18/73] Update src/Parsers/ASTAlterQuery.cpp

Co-authored-by: Nikolay Degterinsky <43110995+evillique@users.noreply.github.com>
---
 src/Parsers/ASTAlterQuery.cpp | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/src/Parsers/ASTAlterQuery.cpp b/src/Parsers/ASTAlterQuery.cpp
index dd7f40cff95..a6543190904 100644
--- a/src/Parsers/ASTAlterQuery.cpp
+++ b/src/Parsers/ASTAlterQuery.cpp
@@ -104,12 +104,14 @@ void ASTAlterCommand::formatImpl(const FormatSettings & settings, FormatState &
         {
             settings.ostr << (settings.hilite ? hilite_keyword : "") << " REMOVE " << remove_property;
         }
-        else if (settings_changes) {
-            settings.ostr << (settings.hilite ? hilite_keyword : "") << "MODIFY SETTING " << (settings.hilite ? hilite_none : "");
+        else if (settings_changes)
+        {
+            settings.ostr << (settings.hilite ? hilite_keyword : "") << " MODIFY SETTING " << (settings.hilite ? hilite_none : "");
             settings_changes->formatImpl(settings, state, frame);
         }
-        else if (settings_resets) {
-            settings.ostr << (settings.hilite ? hilite_keyword : "") << "RESET SETTING " << (settings.hilite ? hilite_none : "");
+        else if (settings_resets)
+        {
+            settings.ostr << (settings.hilite ? hilite_keyword : "") << " RESET SETTING " << (settings.hilite ? hilite_none : "");
             settings_resets->formatImpl(settings, state, frame);
         }
         else

From 20311791e0cb5c15273cf58de75ef70cb16bbc8c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <benjamin.antal@clickhouse.com>
Date: Thu, 1 Feb 2024 10:34:03 +0000
Subject: [PATCH 19/73] Fix tests

---
 tests/queries/0_stateless/02870_per_column_settings.reference | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/queries/0_stateless/02870_per_column_settings.reference b/tests/queries/0_stateless/02870_per_column_settings.reference
index be08e7a682b..c2ae34928bd 100644
--- a/tests/queries/0_stateless/02870_per_column_settings.reference
+++ b/tests/queries/0_stateless/02870_per_column_settings.reference
@@ -1,8 +1,8 @@
 CREATE TABLE default.tab\n(\n    `id` UInt64,\n    `long_string` String SETTINGS (min_compress_block_size = 163840, max_compress_block_size = 163840),\n    `v1` String,\n    `v2` UInt64,\n    `v3` Float32,\n    `v4` Float64\n)\nENGINE = ReplicatedMergeTree(\'/clickhouse/tables/default/tab/2870\', \'r1\')\nORDER BY id\nSETTINGS min_bytes_for_wide_part = 1, index_granularity = 8192
 1000
-ALTER TABLE tab\n    MODIFY COLUMN `long_string`MODIFY SETTING min_compress_block_size = 8192
+ALTER TABLE tab\n    MODIFY COLUMN `long_string` MODIFY SETTING min_compress_block_size = 8192
 CREATE TABLE default.tab\n(\n    `id` UInt64,\n    `long_string` String SETTINGS (min_compress_block_size = 8192, max_compress_block_size = 163840),\n    `v1` String,\n    `v2` UInt64,\n    `v3` Float32,\n    `v4` Float64\n)\nENGINE = ReplicatedMergeTree(\'/clickhouse/tables/default/tab/2870\', \'r1\')\nORDER BY id\nSETTINGS min_bytes_for_wide_part = 1, index_granularity = 8192
-ALTER TABLE tab\n    MODIFY COLUMN `long_string`RESET SETTING min_compress_block_size
+ALTER TABLE tab\n    MODIFY COLUMN `long_string` RESET SETTING min_compress_block_size
 CREATE TABLE default.tab\n(\n    `id` UInt64,\n    `long_string` String SETTINGS (max_compress_block_size = 163840),\n    `v1` String,\n    `v2` UInt64,\n    `v3` Float32,\n    `v4` Float64\n)\nENGINE = ReplicatedMergeTree(\'/clickhouse/tables/default/tab/2870\', \'r1\')\nORDER BY id\nSETTINGS min_bytes_for_wide_part = 1, index_granularity = 8192
 ALTER TABLE tab\n    MODIFY COLUMN `long_string` REMOVE SETTINGS
 CREATE TABLE default.tab\n(\n    `id` UInt64,\n    `long_string` String,\n    `v1` String,\n    `v2` UInt64,\n    `v3` Float32,\n    `v4` Float64\n)\nENGINE = ReplicatedMergeTree(\'/clickhouse/tables/default/tab/2870\', \'r1\')\nORDER BY id\nSETTINGS min_bytes_for_wide_part = 1, index_granularity = 8192

From d322accd89a79ab1c8c6817426c637b4a99b466d Mon Sep 17 00:00:00 2001
From: Max Kainov <max.kainov@clickhouse.com>
Date: Thu, 1 Feb 2024 12:46:18 +0000
Subject: [PATCH 20/73] CI: fix workflow run_command in bugfix validation

---
 .github/workflows/pull_request.yml | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/.github/workflows/pull_request.yml b/.github/workflows/pull_request.yml
index 56617294fb6..9c08363f674 100644
--- a/.github/workflows/pull_request.yml
+++ b/.github/workflows/pull_request.yml
@@ -491,11 +491,11 @@ jobs:
       run_command: |
         TEMP_PATH="${TEMP_PATH}/integration" \
           python3 integration_test_check.py "Integration $CHECK_NAME" \
-            --validate-bugfix --post-commit-status=file || echo 'ignore exit code'
+            --validate-bugfix --post-commit-status=file || echo "ignore exit code"
 
         TEMP_PATH="${TEMP_PATH}/stateless" \
           python3 functional_test_check.py "Stateless $CHECK_NAME" "$KILL_TIMEOUT" \
-            --validate-bugfix --post-commit-status=file || echo 'ignore exit code'
+            --validate-bugfix --post-commit-status=file || echo "ignore exit code"
 
         python3 bugfix_validate_check.py "${TEMP_PATH}/stateless/functional_commit_status.tsv" "${TEMP_PATH}/integration/integration_commit_status.tsv"
 ##############################################################################################

From a0935770bba84796f3f611f2022110a283114013 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <benjamin.antal@clickhouse.com>
Date: Thu, 1 Feb 2024 14:50:29 +0000
Subject: [PATCH 21/73] Fix docs

---
 .../sql-reference/statements/alter/column.md  | 22 +++++++++++--------
 1 file changed, 13 insertions(+), 9 deletions(-)

diff --git a/docs/en/sql-reference/statements/alter/column.md b/docs/en/sql-reference/statements/alter/column.md
index 676d30f5e44..193136cf8ba 100644
--- a/docs/en/sql-reference/statements/alter/column.md
+++ b/docs/en/sql-reference/statements/alter/column.md
@@ -139,8 +139,8 @@ ALTER TABLE visits COMMENT COLUMN browser 'This column shows the browser used fo
 ## MODIFY COLUMN
 
 ``` sql
-MODIFY COLUMN [IF EXISTS] name [type] [default_expr] [codec] [TTL] [AFTER name_after | FIRST]
-ALTER COLUMN [IF EXISTS] name TYPE [type] [default_expr] [codec] [TTL] [AFTER name_after | FIRST]
+MODIFY COLUMN [IF EXISTS] name [type] [default_expr] [codec] [TTL] [settings] [AFTER name_after | FIRST]
+ALTER COLUMN [IF EXISTS] name TYPE [type] [default_expr] [codec] [TTL] [settings] [AFTER name_after | FIRST]
 ```
 
 This query changes the `name` column properties:
@@ -153,10 +153,14 @@ This query changes the `name` column properties:
 
 - TTL
 
+- Column-level Settings
+
 For examples of columns compression CODECS modifying, see [Column Compression Codecs](../create/table.md/#codecs).
 
 For examples of columns TTL modifying, see [Column TTL](/docs/en/engines/table-engines/mergetree-family/mergetree.md/#mergetree-column-ttl).
 
+For examples of colum-level settings modifying, see [Column-level Settings](/docs/en/engines/table-engines/mergetree-family/mergetree.md/#column-level-settings).
+
 If the `IF EXISTS` clause is specified, the query won’t return an error if the column does not exist.
 
 When changing the type, values are converted as if the [toType](/docs/en/sql-reference/functions/type-conversion-functions.md) functions were applied to them. If only the default expression is changed, the query does not do anything complex, and is completed almost instantly.
@@ -209,7 +213,7 @@ The `ALTER` query for changing columns is replicated. The instructions are saved
 
 ## MODIFY COLUMN REMOVE
 
-Removes one of the column properties: `DEFAULT`, `ALIAS`, `MATERIALIZED`, `CODEC`, `COMMENT`, `TTL`, `SETTING`.
+Removes one of the column properties: `DEFAULT`, `ALIAS`, `MATERIALIZED`, `CODEC`, `COMMENT`, `TTL`, `SETTINGS`.
 
 Syntax:
 
@@ -237,7 +241,7 @@ Modify a column setting.
 Syntax:
 
 ```sql
-ALTER TABLE table_name MODIFY COLUMN MODIFY SETTING name=value,...;
+ALTER TABLE table_name MODIFY COLUMN column_name MODIFY SETTING name=value,...;
 ```
 
 **Example**
@@ -245,7 +249,7 @@ ALTER TABLE table_name MODIFY COLUMN MODIFY SETTING name=value,...;
 Modify column's `max_compress_block_size` to `1MB`:
 
 ```sql
-ALTER TABLE table_name MODIFY COLUMN MODIFY SETTING max_compress_block_size = 1048576;
+ALTER TABLE table_name MODIFY COLUMN column_name MODIFY SETTING max_compress_block_size = 1048576;
 ```
 
 ## MODIFY COLUMN RESET SETTING
@@ -255,21 +259,21 @@ Reset a column setting, also removes the setting declaration in the column expre
 Syntax:
 
 ```sql
-ALTER TABLE table_name MODIFY COLUMN RESET SETTING name,...;
+ALTER TABLE table_name MODIFY COLUMN column_name RESET SETTING name,...;
 ```
 
 **Example**
 
-Remove column setting `max_compress_block_size` to `1MB`:
+Reset column setting `max_compress_block_size` to it's default value:
 
 ```sql
-ALTER TABLE table_name MODIFY COLUMN REMOVE SETTING max_compress_block_size;
+ALTER TABLE table_name MODIFY COLUMN column_name RESET SETTING max_compress_block_size;
 ```
 
 ## MATERIALIZE COLUMN
 
 Materializes or updates a column with an expression for a default value (`DEFAULT` or `MATERIALIZED`).
-It is used if it is necessary to add or update a column with a complicated expression, because evaluating such an expression directly on `SELECT` executing turns out to be expensive. 
+It is used if it is necessary to add or update a column with a complicated expression, because evaluating such an expression directly on `SELECT` executing turns out to be expensive.
 Implemented as a [mutation](/docs/en/sql-reference/statements/alter/index.md#mutations).
 
 Syntax:

From 479514986378ac3bc0acdf99ca9a140b6efce7a6 Mon Sep 17 00:00:00 2001
From: Nikita Taranov <nikita.taranov@clickhouse.com>
Date: Thu, 1 Feb 2024 17:16:42 +0100
Subject: [PATCH 22/73] print stats once

---
 .../MergeTree/MergeTreeDataPartWriterOnDisk.cpp    | 14 +++++++++-----
 .../MergeTree/MergeTreeDataPartWriterOnDisk.h      | 12 ++++++++++++
 2 files changed, 21 insertions(+), 5 deletions(-)

diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp
index 153bcaa6320..6a265d79019 100644
--- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp
+++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp
@@ -6,7 +6,7 @@
 
 namespace ProfileEvents
 {
-extern const Event MergeTreeDataWriterSecondaryIndicesCalculationMicroseconds;
+extern const Event MergeTreeDataWriterSkipIndicesCalculationMicroseconds;
 }
 
 namespace DB
@@ -155,6 +155,7 @@ MergeTreeDataPartWriterOnDisk::MergeTreeDataPartWriterOnDisk(
     , default_codec(default_codec_)
     , compute_granularity(index_granularity.empty())
     , compress_primary_key(settings.compress_primary_key)
+    , execution_stats(skip_indices.size(), stats.size())
     , log(getLogger(storage.getLogName() + " (DataPartWriter)"))
 {
     if (settings.blocks_are_granules_size && !index_granularity.empty())
@@ -362,7 +363,6 @@ void MergeTreeDataPartWriterOnDisk::calculateAndSerializeSkipIndices(const Block
             store = it->second;
         }
 
-        size_t index_build_us = 0;
         for (const auto & granule : granules_to_write)
         {
             if (skip_index_accumulated_marks[i] == index_helper->index.granularity)
@@ -387,16 +387,15 @@ void MergeTreeDataPartWriterOnDisk::calculateAndSerializeSkipIndices(const Block
                     writeBinaryLittleEndian(1UL, marks_out);
             }
 
-            ProfileEventTimeIncrement<Microseconds> watch(ProfileEvents::MergeTreeDataWriterSecondaryIndicesCalculationMicroseconds);
+            ProfileEventTimeIncrement<Microseconds> watch(ProfileEvents::MergeTreeDataWriterSkipIndicesCalculationMicroseconds);
 
             size_t pos = granule.start_row;
             skip_indices_aggregators[i]->update(skip_indexes_block, &pos, granule.rows_to_write);
             if (granule.is_complete)
                 ++skip_index_accumulated_marks[i];
 
-            index_build_us += watch.elapsed<Microseconds>();
+            execution_stats.skip_indices_build_us[i] += watch.elapsed();
         }
-        LOG_DEBUG(log, "Spent {} ms calculating index {} for the part {}", index_build_us / 1000, skip_indices[i]->index.name, data_part->name);
     }
 }
 
@@ -518,6 +517,11 @@ void MergeTreeDataPartWriterOnDisk::finishSkipIndicesSerialization(bool sync)
     }
     for (auto & store: gin_index_stores)
         store.second->finalize();
+
+    chassert(execution_stats.skip_indices_build_us.size() == skip_indices.size());
+    for (size_t i = 0; i < skip_indices.size(); ++i)
+        LOG_DEBUG(log, "Spent {} ms calculating index {}", execution_stats.skip_indices_build_us[i] / 1000, skip_indices[i]->index.name);
+
     gin_index_stores.clear();
     skip_indices_streams.clear();
     skip_indices_aggregators.clear();
diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h
index 5292c0d5590..9f2cc3970fa 100644
--- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h
+++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h
@@ -191,6 +191,18 @@ private:
 
     virtual void fillIndexGranularity(size_t index_granularity_for_block, size_t rows_in_block) = 0;
 
+    struct ExecutionStatistics
+    {
+        ExecutionStatistics(size_t skip_indices_cnt, size_t stats_cnt)
+            : skip_indices_build_us(skip_indices_cnt, 0), statistics_build_us(stats_cnt, 0)
+        {
+        }
+
+        std::vector<size_t> skip_indices_build_us; // [i] corresponds to the i-th index
+        std::vector<size_t> statistics_build_us; // [i] corresponds to the i-th stat
+    };
+    ExecutionStatistics execution_stats;
+
     LoggerPtr log;
 };
 

From 76f2ae08b503bfffd732187203dbb6ea0c4bde1a Mon Sep 17 00:00:00 2001
From: Nikita Taranov <nikita.taranov@clickhouse.com>
Date: Thu, 1 Feb 2024 18:16:09 +0100
Subject: [PATCH 23/73] collect stats build time

---
 src/Common/ProfileEvents.cpp                           |  3 ++-
 .../MergeTree/MergeTreeDataPartWriterOnDisk.cpp        | 10 ++++++++--
 2 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/src/Common/ProfileEvents.cpp b/src/Common/ProfileEvents.cpp
index ef0606162a2..2e5001ce413 100644
--- a/src/Common/ProfileEvents.cpp
+++ b/src/Common/ProfileEvents.cpp
@@ -197,7 +197,8 @@
     M(MergeTreeDataWriterBlocks, "Number of blocks INSERTed to MergeTree tables. Each block forms a data part of level zero.") \
     M(MergeTreeDataWriterBlocksAlreadySorted, "Number of blocks INSERTed to MergeTree tables that appeared to be already sorted.") \
     \
-    M(MergeTreeDataWriterSecondaryIndicesCalculationMicroseconds, "Time spent calculating secondary indices") \
+    M(MergeTreeDataWriterSkipIndicesCalculationMicroseconds, "Time spent calculating skip indices") \
+    M(MergeTreeDataWriterStatisticsCalculationMicroseconds, "Time spent calculating statistics") \
     M(MergeTreeDataWriterPrimaryKeyCalculationMicroseconds, "Time spent calculating primary key") \
     M(MergeTreeDataWriterMergingBlocksMicroseconds, "Time spent merging input blocks (for special MergeTree engines)") \
     M(MergeTreeDataWriterProjectionsCalculationMicroseconds, "Time spent calculating projections") \
diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp
index 6a265d79019..b8cfdf19f54 100644
--- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp
+++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp
@@ -7,6 +7,7 @@
 namespace ProfileEvents
 {
 extern const Event MergeTreeDataWriterSkipIndicesCalculationMicroseconds;
+extern const Event MergeTreeDataWriterStatisticsCalculationMicroseconds;
 }
 
 namespace DB
@@ -338,9 +339,12 @@ void MergeTreeDataPartWriterOnDisk::calculateAndSerializePrimaryIndex(const Bloc
 
 void MergeTreeDataPartWriterOnDisk::calculateAndSerializeStatistics(const Block & block)
 {
-    for (const auto & stat_ptr : stats)
+    for (size_t i = 0; i < stats.size(); ++i)
     {
+        const auto & stat_ptr = stats[i];
+        ProfileEventTimeIncrement<Microseconds> watch(ProfileEvents::MergeTreeDataWriterStatisticsCalculationMicroseconds);
         stat_ptr->update(block.getByName(stat_ptr->columnName()).column);
+        execution_stats.statistics_build_us[i] += watch.elapsed();
     }
 }
 
@@ -494,6 +498,9 @@ void MergeTreeDataPartWriterOnDisk::finishStatisticsSerialization(bool sync)
         if (sync)
             stream->sync();
     }
+
+    for (size_t i = 0; i < stats.size(); ++i)
+        LOG_DEBUG(log, "Spent {} ms calculating statistics {}", execution_stats.statistics_build_us[i] / 1000, stats[i]->columnName());
 }
 
 void MergeTreeDataPartWriterOnDisk::fillStatisticsChecksums(MergeTreeData::DataPart::Checksums & checksums)
@@ -518,7 +525,6 @@ void MergeTreeDataPartWriterOnDisk::finishSkipIndicesSerialization(bool sync)
     for (auto & store: gin_index_stores)
         store.second->finalize();
 
-    chassert(execution_stats.skip_indices_build_us.size() == skip_indices.size());
     for (size_t i = 0; i < skip_indices.size(); ++i)
         LOG_DEBUG(log, "Spent {} ms calculating index {}", execution_stats.skip_indices_build_us[i] / 1000, skip_indices[i]->index.name);
 

From ce31fa912b2f67411350d2de85a21090726101fa Mon Sep 17 00:00:00 2001
From: Bhavna Jindal <bhavna.jindal@ibm.com>
Date: Thu, 1 Feb 2024 09:24:12 -0800
Subject: [PATCH 24/73] removed unnecessary method overloading and fixed
 documentation

---
 .../functions/time-series-functions.md        |  30 ++---
 src/Functions/seriesOutliersDetectTukey.cpp   | 121 +++++++-----------
 .../02813_seriesOutliersTukey.reference       |  10 +-
 .../0_stateless/02813_seriesOutliersTukey.sql |  23 ++--
 4 files changed, 75 insertions(+), 109 deletions(-)

diff --git a/docs/en/sql-reference/functions/time-series-functions.md b/docs/en/sql-reference/functions/time-series-functions.md
index bd50ef556f7..bb6f3da25fb 100644
--- a/docs/en/sql-reference/functions/time-series-functions.md
+++ b/docs/en/sql-reference/functions/time-series-functions.md
@@ -10,33 +10,27 @@ Below functions are used for series data analysis.
 
 ## seriesOutliersDetectTukey
 
-Detects a possible anomaly in series using [Tukey Fences](https://en.wikipedia.org/wiki/Outlier#Tukey%27s_fences).
+Detects outliers in series data using [Tukey Fences](https://en.wikipedia.org/wiki/Outlier#Tukey%27s_fences).
 
 **Syntax**
 
 ``` sql
 seriesOutliersDetectTukey(series);
-seriesOutliersDetectTukey(series, kind, min_percentile, max_percentile, K);
+seriesOutliersDetectTukey(series, min_percentile, max_percentile, K);
 ```
 
 **Arguments**
 
 - `series` - An array of numeric values.
-- `kind` - Kind of algorithm to use. Supported values are 'tukey' for standard tukey and 'ctukey' for custom tukey algorithm. The default is 'ctukey'.
-- `min_percentile` - The minimum percentile to be used to calculate inter-quantile range(IQR). The value must be in range [2,98]. The default is 10. This value is only supported for 'ctukey'.
-- `max_percentile` - The maximum percentile to be used to calculate inter-quantile range(IQR). The value must be in range [2,98]. The default is 90. This value is only supported for 'ctukey'.
+- `min_percentile` - The minimum percentile to be used to calculate inter-quantile range [(IQR)](https://en.wikipedia.org/wiki/Interquartile_range). The value must be in range [2,98]. The default is 25.
+- `max_percentile` - The maximum percentile to be used to calculate inter-quantile range (IQR). The value must be in range [2,98]. The default is 75.
 - `K` - Non-negative constant value to detect mild or stronger outliers. The default value is 1.5
 
 At least four data points are required in `series` to detect outliers.
 
-Default quantile range:
-- `tukey` - 25%/75%
-- `ctukey` - 10%/90%
-
 **Returned value**
 
-- Returns an array of the same length where each value represents score of possible anomaly of corresponding element in the series.
-- A non-zero score indicates a possible anomaly.
+- Returns an array of the same length as the input array where each value represents score of possible anomaly of corresponding element in the series. A non-zero score indicates a possible anomaly.
 
 Type: [Array](../../sql-reference/data-types/array.md).
 
@@ -51,23 +45,23 @@ SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4, 5, 12, 45, 12, 3, 3, 4,
 Result:
 
 ``` text
-┌───────────print_0───────────────────┐
-│[0,0,0,0,0,0,0,0,0,10.5,0,0,0,0,0,0] │
-└─────────────────────────────────────┘
+┌───────────print_0─────────────────┐
+│[0,0,0,0,0,0,0,0,0,27,0,0,0,0,0,0] │
+└───────────────────────────────────┘
 ```
 
 Query:
 
 ``` sql
-SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.50, 5, 12, 45, 12, 3.40, 3, 4, 5, 6], 'ctukey', 20, 80, 1.5) AS print_0;
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.50, 5, 12, 45, 12, 3.40, 3, 4, 5, 6], 20, 80, 1.5) AS print_0;
 ```
 
 Result:
 
 ``` text
-┌─print_0────────────────────────────┐
-│ [0,0,0,0,0,0,0,0,0,12,0,0,0,0,0,0] │
-└────────────────────────────────────┘
+┌─print_0──────────────────────────────┐
+│ [0,0,0,0,0,0,0,0,0,19.5,0,0,0,0,0,0] │
+└──────────────────────────────────────┘
 ```
 
 ## seriesPeriodDetectFFT
diff --git a/src/Functions/seriesOutliersDetectTukey.cpp b/src/Functions/seriesOutliersDetectTukey.cpp
index ce5ed391fa0..66fda8ce976 100644
--- a/src/Functions/seriesOutliersDetectTukey.cpp
+++ b/src/Functions/seriesOutliersDetectTukey.cpp
@@ -14,9 +14,10 @@ namespace ErrorCodes
 {
 extern const int BAD_ARGUMENTS;
 extern const int ILLEGAL_COLUMN;
+extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH;
 }
 
-///Detects a possible anomaly in series using [Tukey Fences](https://en.wikipedia.org/wiki/Outlier#Tukey%27s_fences)
+/// Detects a possible anomaly in series using [Tukey Fences](https://en.wikipedia.org/wiki/Outlier#Tukey%27s_fences)
 class FunctionSeriesOutliersDetectTukey : public IFunction
 {
 public:
@@ -36,9 +37,15 @@ public:
 
     DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override
     {
+        if (arguments.size() != 1 && arguments.size() != 4)
+            throw Exception(
+                ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH,
+                "Function {} needs either 1 or 4 arguments; passed {}.",
+                getName(),
+                arguments.size());
+
         FunctionArgumentDescriptors mandatory_args{{"time_series", &isArray<IDataType>, nullptr, "Array"}};
         FunctionArgumentDescriptors optional_args{
-            {"kind", &isString<IDataType>, isColumnConst, "const String"},
             {"min_percentile", &isNativeNumber<IDataType>, isColumnConst, "Number"},
             {"max_percentile", &isNativeNumber<IDataType>, isColumnConst, "Number"},
             {"k", &isNativeNumber<IDataType>, isColumnConst, "Number"}};
@@ -48,9 +55,9 @@ public:
         return std::make_shared<DataTypeArray>(std::make_shared<DataTypeFloat64>());
     }
 
-    ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {1, 2, 3, 4}; }
+    ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {1, 2, 3}; }
 
-    ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t) const override
+    ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override
     {
         ColumnPtr col = arguments[0].column;
         const ColumnArray * col_arr = checkAndGetColumn<ColumnArray>(col.get());
@@ -58,62 +65,36 @@ public:
         const IColumn & arr_data = col_arr->getData();
         const ColumnArray::Offsets & arr_offsets = col_arr->getOffsets();
 
-        Float64 min_percentile = 0.10; //default 10th percentile
-        Float64 max_percentile = 0.90; //default 90th percentile
+        ColumnPtr col_res;
+        if (input_rows_count == 0)
+            return ColumnArray::create(ColumnFloat64::create());
+
+
+        Float64 min_percentile = 0.25; /// default 25th percentile
+        Float64 max_percentile = 0.75; /// default 75th percentile
+        Float64 K = 1.50;
 
         if (arguments.size() > 1)
         {
-            //const IColumn * arg_column = arguments[1].column.get();
-            const ColumnConst * arg_string = checkAndGetColumnConstStringOrFixedString(arguments[1].column.get());
+            Float64 p_min = arguments[1].column->getFloat64(0);
+            if (p_min < 2.0 || p_min > 98.0)
+                throw Exception(ErrorCodes::BAD_ARGUMENTS, "The second argument of function {} must be in range [2.0, 98.0]", getName());
 
-            if (!arg_string)
-                throw Exception(ErrorCodes::ILLEGAL_COLUMN, "The second argument of function {} must be constant String", getName());
+            min_percentile = p_min / 100;
 
-            String kind = arg_string->getValue<String>();
-            if (kind == "ctukey")
-            {
-                if (arguments.size() > 2)
-                {
-                    Float64 p_min = arguments[2].column->getFloat64(0);
-                    if (p_min >= 2.0 && p_min <= 98.0)
-                        min_percentile = p_min / 100;
-                    else
-                        throw Exception(
-                            ErrorCodes::BAD_ARGUMENTS, "The third argument of function {} must be in range [2.0, 98.0]", getName());
-                }
+            Float64 p_max = arguments[2].column->getFloat64(0);
+            if (p_max < 2.0 || p_max > 98.0 || p_max < min_percentile * 100)
+                throw Exception(ErrorCodes::BAD_ARGUMENTS, "The third argument of function {} must be in range [2.0, 98.0]", getName());
 
-                if (arguments.size() > 3)
-                {
-                    Float64 p_max = arguments[3].column->getFloat64(0);
-                    if (p_max >= 2.0 && p_max <= 98.0 && p_max > min_percentile * 100)
-                        max_percentile = p_max / 100;
-                    else
-                        throw Exception(
-                            ErrorCodes::BAD_ARGUMENTS, "The fourth argument of function {} must be in range [2.0, 98.0]", getName());
-                }
-            }
-            else if (kind == "tukey")
-            {
-                min_percentile = 0.25;
-                max_percentile = 0.75;
-            }
-            else
-                throw Exception(
-                    ErrorCodes::BAD_ARGUMENTS, "The second argument of function {} can only be 'tukey' or 'ctukey'.", getName());
+            max_percentile = p_max / 100;
+
+            auto k_val = arguments[3].column->getFloat64(0);
+            if (k_val < 0.0)
+                throw Exception(ErrorCodes::BAD_ARGUMENTS, "The fourth argument of function {} must be a positive number", getName());
+
+            K = k_val;
         }
 
-        Float64 K = 1.50;
-        if (arguments.size() == 5)
-        {
-            auto k_val = arguments[4].column->getFloat64(0);
-            if (k_val >= 0.0)
-                K = k_val;
-            else
-                throw Exception(ErrorCodes::BAD_ARGUMENTS, "The fifth argument of function {} must be a positive number", getName());
-        }
-
-        ColumnPtr col_res;
-
         if (executeNumber<UInt8>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res)
             || executeNumber<UInt16>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res)
             || executeNumber<UInt32>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res)
@@ -172,7 +153,7 @@ private:
 
             Float64 q1, q2;
 
-            auto p1 = len * min_percentile;
+            Float64 p1 = len * min_percentile;
             if (p1 == static_cast<Int64>(p1))
             {
                 size_t index = static_cast<size_t>(p1) - 1;
@@ -184,7 +165,7 @@ private:
                 q1 = src_sorted[index];
             }
 
-            auto p2 = len * max_percentile;
+            Float64 p2 = len * max_percentile;
             if (p2 == static_cast<Int64>(p2))
             {
                 size_t index = static_cast<size_t>(p2) - 1;
@@ -219,33 +200,27 @@ REGISTER_FUNCTION(SeriesOutliersDetectTukey)
 {
     factory.registerFunction<FunctionSeriesOutliersDetectTukey>(FunctionDocumentation{
         .description = R"(
-Detects a possible anomaly in series using [Tukey Fences](https://en.wikipedia.org/wiki/Outlier#Tukey%27s_fences).
-
-Detects a possible anomaly in series using [Tukey Fences](https://en.wikipedia.org/wiki/Outlier#Tukey%27s_fences).
+Detects outliers in series data using [Tukey Fences](https://en.wikipedia.org/wiki/Outlier#Tukey%27s_fences).
 
 **Syntax**
 
 ``` sql
 seriesOutliersDetectTukey(series);
-seriesOutliersDetectTukey(series, kind, min_percentile, max_percentile, K);
+seriesOutliersDetectTukey(series, min_percentile, max_percentile, K);
 ```
 
 **Arguments**
 
 - `series` - An array of numeric values.
-- `kind` - Kind of algorithm to use. Supported values are 'tukey' for standard tukey and 'ctukey' for custom tukey algorithm. The default is 'ctukey'.
-- `min_percentile` - The minimum percentile to be used to calculate inter-quantile range(IQR). The value must be in range [2,98]. The default is 10. This value is only supported for 'ctukey'.
-- `max_percentile` - The maximum percentile to be used to calculate inter-quantile range(IQR). The value must be in range [2,98]. The default is 90. This value is only supported for 'ctukey'.
+- `min_percentile` - The minimum percentile to be used to calculate inter-quantile range [(IQR)](https://en.wikipedia.org/wiki/Interquartile_range). The value must be in range [2,98]. The default is 25.
+- `max_percentile` - The maximum percentile to be used to calculate inter-quantile range (IQR). The value must be in range [2,98]. The default is 75.
 - `K` - Non-negative constant value to detect mild or stronger outliers. The default value is 1.5
 
-Default quantile range:
-- `tukey` - 25%/75%
-- `ctukey` - 10%/90%
+At least four data points are required in `series` to detect outliers.
 
 **Returned value**
 
-- Returns an array of the same length where each value represents score of possible anomaly of corresponding element in the series.
-- A non-zero score indicates a possible anomaly.
+- Returns an array of the same length as the input array where each value represents score of possible anomaly of corresponding element in the series. A non-zero score indicates a possible anomaly.
 
 Type: [Array](../../sql-reference/data-types/array.md).
 
@@ -260,23 +235,23 @@ SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4, 5, 12, 45, 12, 3, 3, 4,
 Result:
 
 ``` text
-┌───────────print_0───────────────────┐
-│[0,0,0,0,0,0,0,0,0,10.5,0,0,0,0,0,0] │
-└─────────────────────────────────────┘
+┌───────────print_0─────────────────┐
+│[0,0,0,0,0,0,0,0,0,27,0,0,0,0,0,0] │
+└───────────────────────────────────┘
 ```
 
 Query:
 
 ``` sql
-SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.50, 5, 12, 45, 12, 3.40, 3, 4, 5, 6], 'ctukey', 20, 80, 1.5) AS print_0;
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.50, 5, 12, 45, 12, 3.40, 3, 4, 5, 6], 20, 80, 1.5) AS print_0;
 ```
 
 Result:
 
 ``` text
-┌─print_0────────────────────────────┐
-│ [0,0,0,0,0,0,0,0,0,12,0,0,0,0,0,0] │
-└────────────────────────────────────┘
+┌─print_0──────────────────────────────┐
+│ [0,0,0,0,0,0,0,0,0,19.5,0,0,0,0,0,0] │
+└──────────────────────────────────────┘
 ```)",
         .categories{"Time series analysis"}});
 }
diff --git a/tests/queries/0_stateless/02813_seriesOutliersTukey.reference b/tests/queries/0_stateless/02813_seriesOutliersTukey.reference
index bdcde0419a4..85c65ab10ba 100644
--- a/tests/queries/0_stateless/02813_seriesOutliersTukey.reference
+++ b/tests/queries/0_stateless/02813_seriesOutliersTukey.reference
@@ -1,14 +1,12 @@
-[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]
-[0,0,0,0,0,0,0,0,0,11.100000000000001,0,0,0,0,0,0]
 [-4.475000000000001,0,6.925000000000001,0,0,0,0,0,0,0,0,7.925000000000001,0,0,0,0]
 [0,0,0,0,0,0,0,0,0,27.975,0,0,0,0,0,0]
-[0,0,0,0,0,0,0,0,0,10.5,0,0,0,0,0,0]
-[0,0,0,0,0,0,0,0,0,26.1,0,0,0,0,0,0,0,0]
+[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]
+[0,0,0,0,0,0,0,0,0,11.100000000000001,0,0,0,0,0,0]
+[0,0,0,0,0,0,0,0,0,27.3,0,0,0,0,0,0]
+[-2.4999999999999996,0,5.1,0,0,0,0,0,2.0999999999999996,50.1,2.0999999999999996,0,0,0,0,0,0,0]
 [0,0,0,0,0,0,0,0,0,27.3,0,0,0,0,0,0]
 [0,0,0,0,0,0,0,0,0,10.5,0,0,0,0,0,0]
-[0,0,0,0,0,0,0,0,0,27.3,0,0,0,0,0,0]
 [0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]
 [0,0,0,0]
 [0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]
-[0,0,0,0,0,0,0,0,0,27,0,0,0,0,0,0]
 [0,0,0,0,0,0,0,0,0,18,0,0,0,0,0,0]
diff --git a/tests/queries/0_stateless/02813_seriesOutliersTukey.sql b/tests/queries/0_stateless/02813_seriesOutliersTukey.sql
index 7efe4903249..b43fa40e82b 100644
--- a/tests/queries/0_stateless/02813_seriesOutliersTukey.sql
+++ b/tests/queries/0_stateless/02813_seriesOutliersTukey.sql
@@ -1,11 +1,11 @@
 DROP TABLE IF EXISTS tb1;
 
 CREATE TABLE tb1 (n UInt32, a Array(Float64)) engine=Memory;
-INSERT INTO tb1 VALUES (1, [-3,2.40,15,3.90,5,6,4.50,5.20,3,4,5,16,7,5,5,4]), (2, [-3,2.40,15,3.90,5,6,4.50,5.20,12,45,12,3.40,3,4,5,6]);
+INSERT INTO tb1 VALUES (1, [-3, 2.40, 15, 3.90, 5, 6, 4.50, 5.20, 3, 4, 5, 16, 7, 5, 5, 4]), (2, [-3, 2.40, 15, 3.90, 5, 6, 4.50, 5.20, 12, 45, 12, 3.40, 3, 4, 5, 6]);
 
 -- non-const inputs
 SELECT seriesOutliersDetectTukey(a) FROM tb1 ORDER BY n;
-SELECT seriesOutliersDetectTukey(a,'ctukey', 25,75) FROM tb1 ORDER BY n;
+SELECT seriesOutliersDetectTukey(a,10,90,1.5) FROM tb1 ORDER BY n;
 DROP TABLE IF EXISTS tb1;
 
 -- const inputs
@@ -13,18 +13,17 @@ SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.50, 5, 12, 45, 12, 3.40,
 SELECT seriesOutliersDetectTukey([-3, 2.40, 15, 3.90, 5, 6, 4.50, 5.20, 12, 60, 12, 3.40, 3, 4, 5, 6, 3.40, 2.7]);
 
 -- const inputs with optional arguments
-SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.50, 5, 12, 45, 12, 3.40, 3, 4, 5, 6], 'ctukey', 25, 75);
-SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.50, 5, 12, 45, 12, 3.40, 3, 4, 5, 6], 'ctukey', 10, 90);
-SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.50, 5, 12, 45, 12, 3.40, 3, 4, 5, 6], 'tukey', 10, 90);
-SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.50, 5, 12, 45, 12, 3.40, 3, 4, 5, 6], 'ctukey', 2, 98);
-SELECT seriesOutliersDetectTukey([-3, 2, 15, 3], 'ctukey', 2, 98);
-SELECT seriesOutliersDetectTukey(arrayMap(x -> sin(x / 10), range(30)), 'tukey');
-SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4, 5, 12, 45, 12, 3, 3, 4, 5, 6], 'tukey', 25, 75, 1.5);
-SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4, 5, 12, 45, 12, 3, 3, 4, 5, 6], 'tukey', 25, 75, 3);
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.50, 5, 12, 45, 12, 3.40, 3, 4, 5, 6], 25, 75, 1.5);
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.50, 5, 12, 45, 12, 3.40, 3, 4, 5, 6], 10, 90, 1.5);
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4.50, 5, 12, 45, 12, 3.40, 3, 4, 5, 6], 2, 98, 1.5);
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3], 2, 98, 1.5);
+SELECT seriesOutliersDetectTukey(arrayMap(x -> sin(x / 10), range(30)));
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4, 5, 12, 45, 12, 3, 3, 4, 5, 6], 25, 75, 3);
 
 -- negative tests
-SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4, 5, 12, 45, 12, 3, 3, 4, 5, 6], 'tukey', 25, 75, -1); -- { serverError BAD_ARGUMENTS}
-SELECT seriesOutliersDetectTukey([-3, 2, 15, 3], 'xyz', 33, 53); -- { serverError BAD_ARGUMENTS}
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3, 5, 6, 4, 5, 12, 45, 12, 3, 3, 4, 5, 6], 25, 75, -1); -- { serverError BAD_ARGUMENTS}
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3], 33, 53); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH}
+SELECT seriesOutliersDetectTukey([-3, 2, 15, 3], 33); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH}
 SELECT seriesOutliersDetectTukey([-3, 2.4, 15, NULL]); -- { serverError ILLEGAL_COLUMN}
 SELECT seriesOutliersDetectTukey([]); -- { serverError ILLEGAL_COLUMN}
 SELECT seriesOutliersDetectTukey([-3, 2.4, 15]); -- { serverError BAD_ARGUMENTS}
\ No newline at end of file

From 132c7362cf585f255971b094f62a9120fce1dc31 Mon Sep 17 00:00:00 2001
From: Nikita Taranov <nikita.taranov@clickhouse.com>
Date: Thu, 1 Feb 2024 18:51:05 +0100
Subject: [PATCH 25/73] Fix `00191_aggregating_merge_tree_and_final`

It became flaky after we merged #39663. When external aggregation happens it seems like we get more than one part, and because we used `OPTIMIZE` without `FINAL` data wasn't properly merged.
`order by` on the line 12 is not needed by now, but let's add it just in case.
---
 .../0_stateless/00191_aggregating_merge_tree_and_final.sql    | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/queries/0_stateless/00191_aggregating_merge_tree_and_final.sql b/tests/queries/0_stateless/00191_aggregating_merge_tree_and_final.sql
index 8160d4dee9e..4f73a9e9a54 100644
--- a/tests/queries/0_stateless/00191_aggregating_merge_tree_and_final.sql
+++ b/tests/queries/0_stateless/00191_aggregating_merge_tree_and_final.sql
@@ -7,9 +7,9 @@ INSERT INTO aggregating_00191 (k, u) SELECT intDiv(number, 100) AS k, uniqState(
 
 SELECT k, finalizeAggregation(u) FROM aggregating_00191 FINAL order by k;
 
-OPTIMIZE TABLE aggregating_00191;
+OPTIMIZE TABLE aggregating_00191 FINAL;
 
-SELECT k, finalizeAggregation(u) FROM aggregating_00191;
+SELECT k, finalizeAggregation(u) FROM aggregating_00191 order by k;
 SELECT k, finalizeAggregation(u) FROM aggregating_00191 FINAL order by k;
 
 DROP TABLE aggregating_00191;

From 698d00dbfee5f03badbfa182865ca4f2c9cb627d Mon Sep 17 00:00:00 2001
From: Nikita Taranov <nikita.taranov@clickhouse.com>
Date: Thu, 1 Feb 2024 20:43:43 +0100
Subject: [PATCH 26/73] add to MutateTask

---
 src/Common/ProfileEvents.cpp                        |  1 +
 src/Storages/MergeTree/MutateTask.cpp               | 13 ++++++++++++-
 .../02982_perf_introspection_for_inserts.sh         |  2 +-
 3 files changed, 14 insertions(+), 2 deletions(-)

diff --git a/src/Common/ProfileEvents.cpp b/src/Common/ProfileEvents.cpp
index 2e5001ce413..4d1a30d7676 100644
--- a/src/Common/ProfileEvents.cpp
+++ b/src/Common/ProfileEvents.cpp
@@ -204,6 +204,7 @@
     M(MergeTreeDataWriterProjectionsCalculationMicroseconds, "Time spent calculating projections") \
     M(MergeTreeDataProjectionWriterPrimaryKeyCalculationMicroseconds, "Time spent calculating primary key") \
     M(MergeTreeDataProjectionWriterMergingBlocksMicroseconds, "Time spent merging input blocks") \
+    M(MutateTaskProjectionsCalculationMicroseconds, "Time spent calculating projections") \
     \
     M(InsertedWideParts, "Number of parts inserted in Wide format.") \
     M(InsertedCompactParts, "Number of parts inserted in Compact format.") \
diff --git a/src/Storages/MergeTree/MutateTask.cpp b/src/Storages/MergeTree/MutateTask.cpp
index e325bb5d720..2affb474104 100644
--- a/src/Storages/MergeTree/MutateTask.cpp
+++ b/src/Storages/MergeTree/MutateTask.cpp
@@ -28,6 +28,11 @@
 #include <Common/ProfileEventsScope.h>
 
 
+namespace ProfileEvents
+{
+extern const Event MutateTaskProjectionsCalculationMicroseconds;
+}
+
 namespace CurrentMetrics
 {
     extern const Metric PartMutation;
@@ -1239,7 +1244,13 @@ bool PartMergerWriter::mutateOriginalPartAndPrepareProjections()
         for (size_t i = 0, size = ctx->projections_to_build.size(); i < size; ++i)
         {
             const auto & projection = *ctx->projections_to_build[i];
-            auto projection_block = projection_squashes[i].add(projection.calculate(cur_block, ctx->context));
+
+            Block projection_block;
+            {
+                ProfileEventTimeIncrement<Microseconds> watch(ProfileEvents::MutateTaskProjectionsCalculationMicroseconds);
+                projection_block = projection_squashes[i].add(projection.calculate(cur_block, ctx->context));
+            }
+
             if (projection_block)
             {
                 auto tmp_part = MergeTreeDataWriter::writeTempProjectionPart(
diff --git a/tests/queries/0_stateless/02982_perf_introspection_for_inserts.sh b/tests/queries/0_stateless/02982_perf_introspection_for_inserts.sh
index a45f88fc10f..a249cbef6a2 100755
--- a/tests/queries/0_stateless/02982_perf_introspection_for_inserts.sh
+++ b/tests/queries/0_stateless/02982_perf_introspection_for_inserts.sh
@@ -40,7 +40,7 @@ SELECT
     ProfileEvents['MergeTreeDataProjectionWriterPrimaryKeyCalculationMicroseconds'] > 0,
     ProfileEvents['MergeTreeDataWriterPrimaryKeyCalculationMicroseconds'] > 0,
     ProfileEvents['MergeTreeDataWriterProjectionsCalculationMicroseconds'] > 0,
-    ProfileEvents['MergeTreeDataWriterSecondaryIndicesCalculationMicroseconds'] > 0
+    ProfileEvents['MergeTreeDataWriterSkipIndicesCalculationMicroseconds'] > 0
 FROM system.query_log
 WHERE current_database = currentDatabase() AND query_id='$query_id' AND type = 'QueryFinish';
 """

From d73abc17f53282a39ee085a0ae3dfcc51f92a86e Mon Sep 17 00:00:00 2001
From: Nikita Taranov <nikita.taranov@clickhouse.com>
Date: Thu, 1 Feb 2024 21:06:44 +0100
Subject: [PATCH 27/73] better

---
 src/Common/ProfileEvents.cpp                              | 6 +++---
 src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp  | 4 ++--
 src/Storages/MergeTree/MergeTreeDataWriter.cpp            | 8 ++++----
 .../0_stateless/02982_perf_introspection_for_inserts.sh   | 4 ++--
 4 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/src/Common/ProfileEvents.cpp b/src/Common/ProfileEvents.cpp
index 4d1a30d7676..d831402fd37 100644
--- a/src/Common/ProfileEvents.cpp
+++ b/src/Common/ProfileEvents.cpp
@@ -199,11 +199,11 @@
     \
     M(MergeTreeDataWriterSkipIndicesCalculationMicroseconds, "Time spent calculating skip indices") \
     M(MergeTreeDataWriterStatisticsCalculationMicroseconds, "Time spent calculating statistics") \
-    M(MergeTreeDataWriterPrimaryKeyCalculationMicroseconds, "Time spent calculating primary key") \
+    M(MergeTreeDataWriterSortingBlocksMicroseconds, "Time spent sorting blocks") \
     M(MergeTreeDataWriterMergingBlocksMicroseconds, "Time spent merging input blocks (for special MergeTree engines)") \
     M(MergeTreeDataWriterProjectionsCalculationMicroseconds, "Time spent calculating projections") \
-    M(MergeTreeDataProjectionWriterPrimaryKeyCalculationMicroseconds, "Time spent calculating primary key") \
-    M(MergeTreeDataProjectionWriterMergingBlocksMicroseconds, "Time spent merging input blocks") \
+    M(MergeTreeDataProjectionWriterSortingBlocksMicroseconds, "Time spent sorting blocks (for projection it might be a key different from table's sorting key)") \
+    M(MergeTreeDataProjectionWriterMergingBlocksMicroseconds, "Time spent merging blocks") \
     M(MutateTaskProjectionsCalculationMicroseconds, "Time spent calculating projections") \
     \
     M(InsertedWideParts, "Number of parts inserted in Wide format.") \
diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp
index b8cfdf19f54..fd83d2ebfe9 100644
--- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp
+++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp
@@ -500,7 +500,7 @@ void MergeTreeDataPartWriterOnDisk::finishStatisticsSerialization(bool sync)
     }
 
     for (size_t i = 0; i < stats.size(); ++i)
-        LOG_DEBUG(log, "Spent {} ms calculating statistics {}", execution_stats.statistics_build_us[i] / 1000, stats[i]->columnName());
+        LOG_DEBUG(log, "Spent {} ms calculating statistics {} for the part {}", execution_stats.statistics_build_us[i] / 1000, stats[i]->columnName(), data_part->name);
 }
 
 void MergeTreeDataPartWriterOnDisk::fillStatisticsChecksums(MergeTreeData::DataPart::Checksums & checksums)
@@ -526,7 +526,7 @@ void MergeTreeDataPartWriterOnDisk::finishSkipIndicesSerialization(bool sync)
         store.second->finalize();
 
     for (size_t i = 0; i < skip_indices.size(); ++i)
-        LOG_DEBUG(log, "Spent {} ms calculating index {}", execution_stats.skip_indices_build_us[i] / 1000, skip_indices[i]->index.name);
+        LOG_DEBUG(log, "Spent {} ms calculating index {} for the part {}", execution_stats.skip_indices_build_us[i] / 1000, skip_indices[i]->index.name, data_part->name);
 
     gin_index_stores.clear();
     skip_indices_streams.clear();
diff --git a/src/Storages/MergeTree/MergeTreeDataWriter.cpp b/src/Storages/MergeTree/MergeTreeDataWriter.cpp
index 8f522832bce..0418bf6fc7d 100644
--- a/src/Storages/MergeTree/MergeTreeDataWriter.cpp
+++ b/src/Storages/MergeTree/MergeTreeDataWriter.cpp
@@ -36,7 +36,7 @@ namespace ProfileEvents
     extern const Event MergeTreeDataWriterRows;
     extern const Event MergeTreeDataWriterUncompressedBytes;
     extern const Event MergeTreeDataWriterCompressedBytes;
-    extern const Event MergeTreeDataWriterPrimaryKeyCalculationMicroseconds;
+    extern const Event MergeTreeDataWriterSortingBlocksMicroseconds;
     extern const Event MergeTreeDataWriterMergingBlocksMicroseconds;
     extern const Event MergeTreeDataWriterProjectionsCalculationMicroseconds;
     extern const Event MergeTreeDataProjectionWriterBlocks;
@@ -44,7 +44,7 @@ namespace ProfileEvents
     extern const Event MergeTreeDataProjectionWriterRows;
     extern const Event MergeTreeDataProjectionWriterUncompressedBytes;
     extern const Event MergeTreeDataProjectionWriterCompressedBytes;
-    extern const Event MergeTreeDataProjectionWriterPrimaryKeyCalculationMicroseconds;
+    extern const Event MergeTreeDataProjectionWriterSortingBlocksMicroseconds;
     extern const Event MergeTreeDataProjectionWriterMergingBlocksMicroseconds;
     extern const Event RejectedInserts;
 }
@@ -478,7 +478,7 @@ MergeTreeDataWriter::TemporaryPart MergeTreeDataWriter::writeTempPartImpl(
     IColumn::Permutation perm;
     if (!sort_description.empty())
     {
-        ProfileEventTimeIncrement<Microseconds> watch(ProfileEvents::MergeTreeDataWriterPrimaryKeyCalculationMicroseconds);
+        ProfileEventTimeIncrement<Microseconds> watch(ProfileEvents::MergeTreeDataWriterSortingBlocksMicroseconds);
 
         if (!isAlreadySorted(block, sort_description))
         {
@@ -702,7 +702,7 @@ MergeTreeDataWriter::TemporaryPart MergeTreeDataWriter::writeProjectionPartImpl(
     IColumn::Permutation perm;
     if (!sort_description.empty())
     {
-        ProfileEventTimeIncrement<Microseconds> watch(ProfileEvents::MergeTreeDataProjectionWriterPrimaryKeyCalculationMicroseconds);
+        ProfileEventTimeIncrement<Microseconds> watch(ProfileEvents::MergeTreeDataProjectionWriterSortingBlocksMicroseconds);
 
         if (!isAlreadySorted(block, sort_description))
         {
diff --git a/tests/queries/0_stateless/02982_perf_introspection_for_inserts.sh b/tests/queries/0_stateless/02982_perf_introspection_for_inserts.sh
index a249cbef6a2..f5fb54b54d3 100755
--- a/tests/queries/0_stateless/02982_perf_introspection_for_inserts.sh
+++ b/tests/queries/0_stateless/02982_perf_introspection_for_inserts.sh
@@ -37,8 +37,8 @@ $CLICKHOUSE_CLIENT -q "SYSTEM FLUSH LOGS"
 $CLICKHOUSE_CLIENT -q """
 SELECT
     ProfileEvents['MergeTreeDataProjectionWriterMergingBlocksMicroseconds'] > 0,
-    ProfileEvents['MergeTreeDataProjectionWriterPrimaryKeyCalculationMicroseconds'] > 0,
-    ProfileEvents['MergeTreeDataWriterPrimaryKeyCalculationMicroseconds'] > 0,
+    ProfileEvents['MergeTreeDataProjectionWriterSortingBlocksMicroseconds'] > 0,
+    ProfileEvents['MergeTreeDataWriterSortingBlocksMicroseconds'] > 0,
     ProfileEvents['MergeTreeDataWriterProjectionsCalculationMicroseconds'] > 0,
     ProfileEvents['MergeTreeDataWriterSkipIndicesCalculationMicroseconds'] > 0
 FROM system.query_log

From 1faa04de48eb7114620a74ce4f5d4e1ad83e490a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?=
 <benjamin.antal@clickhouse.com>
Date: Fri, 2 Feb 2024 08:58:43 +0000
Subject: [PATCH 28/73] Fix typo

---
 docs/en/sql-reference/statements/alter/column.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/en/sql-reference/statements/alter/column.md b/docs/en/sql-reference/statements/alter/column.md
index 193136cf8ba..f6d9668e628 100644
--- a/docs/en/sql-reference/statements/alter/column.md
+++ b/docs/en/sql-reference/statements/alter/column.md
@@ -159,7 +159,7 @@ For examples of columns compression CODECS modifying, see [Column Compression Co
 
 For examples of columns TTL modifying, see [Column TTL](/docs/en/engines/table-engines/mergetree-family/mergetree.md/#mergetree-column-ttl).
 
-For examples of colum-level settings modifying, see [Column-level Settings](/docs/en/engines/table-engines/mergetree-family/mergetree.md/#column-level-settings).
+For examples of column-level settings modifying, see [Column-level Settings](/docs/en/engines/table-engines/mergetree-family/mergetree.md/#column-level-settings).
 
 If the `IF EXISTS` clause is specified, the query won’t return an error if the column does not exist.
 

From e997168b50a7d2f471d1d8f552e0b7f6a2df495a Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Wed, 31 Jan 2024 12:46:35 +0000
Subject: [PATCH 29/73] Support memory limits in nested cgroups

---
 base/base/getMemoryAmount.cpp | 63 +++++++++++++++++++++++++++++------
 1 file changed, 52 insertions(+), 11 deletions(-)

diff --git a/base/base/getMemoryAmount.cpp b/base/base/getMemoryAmount.cpp
index 060de685778..a55dabdcf46 100644
--- a/base/base/getMemoryAmount.cpp
+++ b/base/base/getMemoryAmount.cpp
@@ -3,6 +3,7 @@
 #include <base/getPageSize.h>
 
 #include <fstream>
+#include <sstream>
 #include <stdexcept>
 
 #include <unistd.h>
@@ -16,18 +17,59 @@
 namespace
 {
 
-std::optional<uint64_t> getCgroupsV2MemoryLimit(const std::string & setting)
+std::optional<uint64_t> getCgroupsV2MemoryLimit()
 {
 #if defined(OS_LINUX)
-    std::filesystem::path default_cgroups_mount = "/sys/fs/cgroup";
-    std::ifstream file(default_cgroups_mount / setting);
-    if (!file.is_open())
+    const std::filesystem::path default_cgroups_mount = "/sys/fs/cgroup";
+
+    /// This file exists iff the host has cgroups v2 enabled.
+    std::ifstream controllers_file(default_cgroups_mount / "cgroup.controllers");
+    if (!controllers_file.is_open())
         return {};
-    uint64_t value;
-    if (file >> value)
-        return {value};
-    else
-        return {}; /// e.g. the cgroups default "max"
+
+    /// We also need the memory controller enabled
+    std::stringstream controllers_buf;
+    controllers_buf << controllers_file.rdbuf();
+    std::string controllers = controllers_buf.str();
+    if (controllers.find("memory") == std::string::npos)
+        return {};
+
+    /// Identify the cgroup the process belongs to
+    std::ifstream cgroup_name_file("/proc/self/cgroup");
+    if (!cgroup_name_file.is_open())
+        return {};
+
+    std::stringstream cgroup_name_buf;
+    cgroup_name_buf << cgroup_name_file.rdbuf();
+    std::string cgroup_name = cgroup_name_buf.str();
+    if (!cgroup_name.empty() && cgroup_name.back() == '\n')
+        cgroup_name.pop_back(); /// remove trailing newline, if any
+    /// cgroups v2 will show a single line with prefix "0::/"
+    /// - https://book.hacktricks.xyz/linux-hardening/privilege-escalation/docker-security/cgroups
+    const std::string v2_prefix = "0::/";
+    if (cgroup_name.find('\n') != std::string::npos || !cgroup_name.starts_with(v2_prefix))
+        return {};
+    cgroup_name = cgroup_name.substr(v2_prefix.length());
+
+    std::filesystem::path current_cgroup = cgroup_name.empty() ? default_cgroups_mount : (default_cgroups_mount / cgroup_name);
+
+    /// Open the bottom-most nested memory limit setting file. If there is no such file at the current
+    /// level, try again at the parent level as memory settings are inherited.
+    while (current_cgroup != default_cgroups_mount.parent_path())
+    {
+        std::ifstream setting_file(current_cgroup / "memory.max");
+        if (setting_file.is_open())
+        {
+            uint64_t value;
+            if (setting_file >> value)
+                return {value};
+            else
+                return {}; /// e.g. the cgroups default "max"
+        }
+        current_cgroup = current_cgroup.parent_path();
+    }
+
+    return {};
 #else
     return {};
 #endif
@@ -52,8 +94,7 @@ uint64_t getMemoryAmountOrZero()
 
     /// Respect the memory limit set by cgroups v2.
     /// Cgroups v1 is dead since many years and its limits are not considered for simplicity.
-
-    auto limit = getCgroupsV2MemoryLimit("memory.max");
+    auto limit = getCgroupsV2MemoryLimit();
     if (limit.has_value() && *limit < memory_amount)
          memory_amount = *limit;
 

From 80a0e05ac4e3fc6d1837877bad2be7a1016ee4e6 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Wed, 31 Jan 2024 15:04:24 +0000
Subject: [PATCH 30/73] Resurrect v1

---
 base/base/getMemoryAmount.cpp | 22 ++++++++++++++++++----
 1 file changed, 18 insertions(+), 4 deletions(-)

diff --git a/base/base/getMemoryAmount.cpp b/base/base/getMemoryAmount.cpp
index a55dabdcf46..7f50683944e 100644
--- a/base/base/getMemoryAmount.cpp
+++ b/base/base/getMemoryAmount.cpp
@@ -93,10 +93,24 @@ uint64_t getMemoryAmountOrZero()
     uint64_t memory_amount = num_pages * page_size;
 
     /// Respect the memory limit set by cgroups v2.
-    /// Cgroups v1 is dead since many years and its limits are not considered for simplicity.
-    auto limit = getCgroupsV2MemoryLimit();
-    if (limit.has_value() && *limit < memory_amount)
-         memory_amount = *limit;
+    auto limit_v2 = getCgroupsV2MemoryLimit();
+    if (limit_v2.has_value() && *limit_v2 < memory_amount)
+         memory_amount = *limit_v2;
+    else
+    {
+        /// Cgroups v1 were replaced by v2 in 2015. The only reason we keep supporting v1 is that the transition to v2
+        /// has been slow. Caveat : Hierarchical groups as in v2 are not supported for v1, the location of the memory
+        /// limit (virtual) file is hard-coded.
+        /// TODO: check at the end of 2024 if we can get rid of v1.
+        std::ifstream limit_file_v1("/sys/fs/cgroup/memory/memory.limit_in_bytes");
+        if (limit_file_v1.is_open())
+        {
+            uint64_t limit_v1;
+            if (limit_file_v1 >> limit_v1)
+                if (limit_v1 < memory_amount)
+                    memory_amount = limit_v1;
+        }
+    }
 
     return memory_amount;
 }

From 28e6e290045c54dffedcbdb8879e785368f7bf24 Mon Sep 17 00:00:00 2001
From: Bhavna Jindal <bhavna.jindal@ibm.com>
Date: Fri, 2 Feb 2024 07:46:36 -0800
Subject: [PATCH 31/73] Minor fixes

---
 .../functions/time-series-functions.md        |  2 +-
 src/Functions/seriesOutliersDetectTukey.cpp   | 41 ++++++++++---------
 ...02813_seriesOutliersDetectTukey.reference} |  0
 ...ql => 02813_seriesOutliersDetectTukey.sql} |  3 ++
 4 files changed, 26 insertions(+), 20 deletions(-)
 rename tests/queries/0_stateless/{02813_seriesOutliersTukey.reference => 02813_seriesOutliersDetectTukey.reference} (100%)
 rename tests/queries/0_stateless/{02813_seriesOutliersTukey.sql => 02813_seriesOutliersDetectTukey.sql} (94%)

diff --git a/docs/en/sql-reference/functions/time-series-functions.md b/docs/en/sql-reference/functions/time-series-functions.md
index bb6f3da25fb..ce36c89f473 100644
--- a/docs/en/sql-reference/functions/time-series-functions.md
+++ b/docs/en/sql-reference/functions/time-series-functions.md
@@ -24,7 +24,7 @@ seriesOutliersDetectTukey(series, min_percentile, max_percentile, K);
 - `series` - An array of numeric values.
 - `min_percentile` - The minimum percentile to be used to calculate inter-quantile range [(IQR)](https://en.wikipedia.org/wiki/Interquartile_range). The value must be in range [2,98]. The default is 25.
 - `max_percentile` - The maximum percentile to be used to calculate inter-quantile range (IQR). The value must be in range [2,98]. The default is 75.
-- `K` - Non-negative constant value to detect mild or stronger outliers. The default value is 1.5
+- `K` - Non-negative constant value to detect mild or stronger outliers. The default value is 1.5.
 
 At least four data points are required in `series` to detect outliers.
 
diff --git a/src/Functions/seriesOutliersDetectTukey.cpp b/src/Functions/seriesOutliersDetectTukey.cpp
index 66fda8ce976..a2fc9cf8eb6 100644
--- a/src/Functions/seriesOutliersDetectTukey.cpp
+++ b/src/Functions/seriesOutliersDetectTukey.cpp
@@ -23,6 +23,9 @@ class FunctionSeriesOutliersDetectTukey : public IFunction
 public:
     static constexpr auto name = "seriesOutliersDetectTukey";
 
+    static constexpr Float64 min_quartile = 2.0;
+    static constexpr Float64 max_quartile = 98.0;
+
     static FunctionPtr create(ContextPtr) { return std::make_shared<FunctionSeriesOutliersDetectTukey>(); }
 
     std::string getName() const override { return name; }
@@ -72,18 +75,18 @@ public:
 
         Float64 min_percentile = 0.25; /// default 25th percentile
         Float64 max_percentile = 0.75; /// default 75th percentile
-        Float64 K = 1.50;
+        Float64 k = 1.50;
 
         if (arguments.size() > 1)
         {
             Float64 p_min = arguments[1].column->getFloat64(0);
-            if (p_min < 2.0 || p_min > 98.0)
+            if (p_min < min_quartile|| p_min > max_quartile)
                 throw Exception(ErrorCodes::BAD_ARGUMENTS, "The second argument of function {} must be in range [2.0, 98.0]", getName());
 
             min_percentile = p_min / 100;
 
             Float64 p_max = arguments[2].column->getFloat64(0);
-            if (p_max < 2.0 || p_max > 98.0 || p_max < min_percentile * 100)
+            if (p_max < min_quartile || p_max > max_quartile || p_max < min_percentile * 100)
                 throw Exception(ErrorCodes::BAD_ARGUMENTS, "The third argument of function {} must be in range [2.0, 98.0]", getName());
 
             max_percentile = p_max / 100;
@@ -92,19 +95,19 @@ public:
             if (k_val < 0.0)
                 throw Exception(ErrorCodes::BAD_ARGUMENTS, "The fourth argument of function {} must be a positive number", getName());
 
-            K = k_val;
+            k = k_val;
         }
 
-        if (executeNumber<UInt8>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res)
-            || executeNumber<UInt16>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res)
-            || executeNumber<UInt32>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res)
-            || executeNumber<UInt64>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res)
-            || executeNumber<Int8>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res)
-            || executeNumber<Int16>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res)
-            || executeNumber<Int32>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res)
-            || executeNumber<Int64>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res)
-            || executeNumber<Float32>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res)
-            || executeNumber<Float64>(arr_data, arr_offsets, min_percentile, max_percentile, K, col_res))
+        if (executeNumber<UInt8>(arr_data, arr_offsets, min_percentile, max_percentile, k, col_res)
+            || executeNumber<UInt16>(arr_data, arr_offsets, min_percentile, max_percentile, k, col_res)
+            || executeNumber<UInt32>(arr_data, arr_offsets, min_percentile, max_percentile, k, col_res)
+            || executeNumber<UInt64>(arr_data, arr_offsets, min_percentile, max_percentile, k, col_res)
+            || executeNumber<Int8>(arr_data, arr_offsets, min_percentile, max_percentile, k, col_res)
+            || executeNumber<Int16>(arr_data, arr_offsets, min_percentile, max_percentile, k, col_res)
+            || executeNumber<Int32>(arr_data, arr_offsets, min_percentile, max_percentile, k, col_res)
+            || executeNumber<Int64>(arr_data, arr_offsets, min_percentile, max_percentile, k, col_res)
+            || executeNumber<Float32>(arr_data, arr_offsets, min_percentile, max_percentile, k, col_res)
+            || executeNumber<Float64>(arr_data, arr_offsets, min_percentile, max_percentile, k, col_res))
         {
             return col_res;
         }
@@ -123,7 +126,7 @@ private:
         const ColumnArray::Offsets & arr_offsets,
         Float64 min_percentile,
         Float64 max_percentile,
-        Float64 K,
+        Float64 k,
         ColumnPtr & res_ptr) const
     {
         const ColumnVector<T> * src_data_concrete = checkAndGetColumn<ColumnVector<T>>(&arr_data);
@@ -179,8 +182,8 @@ private:
 
             Float64 iqr = q2 - q1; /// interquantile range
 
-            Float64 lower_fence = q1 - K * iqr;
-            Float64 upper_fence = q2 + K * iqr;
+            Float64 lower_fence = q1 - k * iqr;
+            Float64 upper_fence = q2 + k * iqr;
 
             for (ColumnArray::Offset j = prev_src_offset; j < src_offset; ++j)
             {
@@ -206,7 +209,7 @@ Detects outliers in series data using [Tukey Fences](https://en.wikipedia.org/wi
 
 ``` sql
 seriesOutliersDetectTukey(series);
-seriesOutliersDetectTukey(series, min_percentile, max_percentile, K);
+seriesOutliersDetectTukey(series, min_percentile, max_percentile, k);
 ```
 
 **Arguments**
@@ -214,7 +217,7 @@ seriesOutliersDetectTukey(series, min_percentile, max_percentile, K);
 - `series` - An array of numeric values.
 - `min_percentile` - The minimum percentile to be used to calculate inter-quantile range [(IQR)](https://en.wikipedia.org/wiki/Interquartile_range). The value must be in range [2,98]. The default is 25.
 - `max_percentile` - The maximum percentile to be used to calculate inter-quantile range (IQR). The value must be in range [2,98]. The default is 75.
-- `K` - Non-negative constant value to detect mild or stronger outliers. The default value is 1.5
+- `k` - Non-negative constant value to detect mild or stronger outliers. The default value is 1.5
 
 At least four data points are required in `series` to detect outliers.
 
diff --git a/tests/queries/0_stateless/02813_seriesOutliersTukey.reference b/tests/queries/0_stateless/02813_seriesOutliersDetectTukey.reference
similarity index 100%
rename from tests/queries/0_stateless/02813_seriesOutliersTukey.reference
rename to tests/queries/0_stateless/02813_seriesOutliersDetectTukey.reference
diff --git a/tests/queries/0_stateless/02813_seriesOutliersTukey.sql b/tests/queries/0_stateless/02813_seriesOutliersDetectTukey.sql
similarity index 94%
rename from tests/queries/0_stateless/02813_seriesOutliersTukey.sql
rename to tests/queries/0_stateless/02813_seriesOutliersDetectTukey.sql
index b43fa40e82b..ca116e8b7ed 100644
--- a/tests/queries/0_stateless/02813_seriesOutliersTukey.sql
+++ b/tests/queries/0_stateless/02813_seriesOutliersDetectTukey.sql
@@ -1,3 +1,6 @@
+-- Tags: no-cpu-aarch64
+-- Tag no-cpu-aarch64: values generated are slighly different on aarch64
+
 DROP TABLE IF EXISTS tb1;
 
 CREATE TABLE tb1 (n UInt32, a Array(Float64)) engine=Memory;

From 82caaa744a51ba11e0b3be803956c875b43a6aad Mon Sep 17 00:00:00 2001
From: Alexander Gololobov <davenger@clickhouse.com>
Date: Sat, 3 Feb 2024 16:08:41 +0100
Subject: [PATCH 32/73] Produce stream of chunks instead of accumulating the
 whole result

---
 .../System/StorageSystemZooKeeper.cpp         | 67 ++++++++++++++-----
 1 file changed, 50 insertions(+), 17 deletions(-)

diff --git a/src/Storages/System/StorageSystemZooKeeper.cpp b/src/Storages/System/StorageSystemZooKeeper.cpp
index 7a2b830b088..cacd7cc2133 100644
--- a/src/Storages/System/StorageSystemZooKeeper.cpp
+++ b/src/Storages/System/StorageSystemZooKeeper.cpp
@@ -180,7 +180,7 @@ using Paths = std::deque<std::pair<String, ZkPathType>>;
 class ReadFromSystemZooKeeper final : public SourceStepWithFilter
 {
 public:
-    ReadFromSystemZooKeeper(const Block & header, SelectQueryInfo & query_info_, ContextPtr context_);
+    ReadFromSystemZooKeeper(const Block & header, SelectQueryInfo & query_info_, UInt64 max_block_size_, ContextPtr context_);
 
     String getName() const override { return "ReadFromSystemZooKeeper"; }
 
@@ -189,13 +189,40 @@ public:
     void applyFilters() override;
 
 private:
-    void fillData(MutableColumns & res_columns);
-
     std::shared_ptr<const StorageLimitsList> storage_limits;
+    const UInt64 max_block_size;
     ContextPtr context;
     Paths paths;
 };
 
+
+class SystemZooKeeperSource : public ISource
+{
+public:
+    SystemZooKeeperSource(
+        Paths && paths_,
+        Block header_,
+        UInt64 max_block_size_,
+        ContextPtr context_)
+        : ISource(header_)
+        , max_block_size(max_block_size_)
+        , paths(std::move(paths_))
+        , context(std::move(context_))
+    {
+    }
+
+    String getName() const override { return "SystemZooKeeper"; }
+
+protected:
+    Chunk generate() override;
+
+private:
+    const UInt64 max_block_size;
+    Paths paths;
+    ContextPtr context;
+};
+
+
 StorageSystemZooKeeper::StorageSystemZooKeeper(const StorageID & table_id_)
         : IStorage(table_id_)
 {
@@ -211,11 +238,11 @@ void StorageSystemZooKeeper::read(
     SelectQueryInfo & query_info,
     ContextPtr context,
     QueryProcessingStage::Enum /*processed_stage*/,
-    size_t /*max_block_size*/,
+    size_t max_block_size,
     size_t /*num_streams*/)
 {
     auto header = storage_snapshot->metadata->getSampleBlockWithVirtuals(getVirtuals());
-    auto read_step = std::make_unique<ReadFromSystemZooKeeper>(header, query_info, context);
+    auto read_step = std::make_unique<ReadFromSystemZooKeeper>(header, query_info, max_block_size, context);
     query_plan.addStep(std::move(read_step));
 }
 
@@ -426,8 +453,15 @@ void ReadFromSystemZooKeeper::applyFilters()
     paths = extractPath(getFilterNodes().nodes, context, context->getSettingsRef().allow_unrestricted_reads_from_keeper);
 }
 
-void ReadFromSystemZooKeeper::fillData(MutableColumns & res_columns)
+
+Chunk SystemZooKeeperSource::generate()
 {
+    if (paths.empty())
+        return {};
+
+    MutableColumns res_columns = getPort().getHeader().cloneEmptyColumns();
+    size_t row_count = 0;
+
     QueryStatusPtr query_status = context->getProcessListElement();
 
     const auto & settings = context->getSettingsRef();
@@ -471,7 +505,7 @@ void ReadFromSystemZooKeeper::fillData(MutableColumns & res_columns)
     };
     std::vector<ListTask> list_tasks;
     std::unordered_set<String> added;
-    while (!paths.empty())
+    while (!paths.empty() && (max_block_size == 0 || row_count < max_block_size))
     {
         if (query_status)
             query_status->checkTimeLimit();
@@ -519,8 +553,8 @@ void ReadFromSystemZooKeeper::fillData(MutableColumns & res_columns)
                 continue;
 
             auto & task = list_tasks[list_task_idx];
-            if (auto elem = context->getProcessListElement())
-                elem->checkTimeLimit();
+            if (query_status)
+                query_status->checkTimeLimit();
 
             Strings nodes = std::move(list_result.names);
 
@@ -584,17 +618,22 @@ void ReadFromSystemZooKeeper::fillData(MutableColumns & res_columns)
             res_columns[col_num++]->insert(
                 list_task.path); /// This is the original path. In order to process the request, condition in WHERE should be triggered.
 
+            ++row_count;
+
             if (list_task.path_type != ZkPathType::Exact && res.stat.numChildren > 0)
             {
                 paths.emplace_back(key, ZkPathType::Recurse);
             }
         }
     }
+
+    return Chunk(std::move(res_columns), row_count);
 }
 
-ReadFromSystemZooKeeper::ReadFromSystemZooKeeper(const Block & header, SelectQueryInfo & query_info, ContextPtr context_)
+ReadFromSystemZooKeeper::ReadFromSystemZooKeeper(const Block & header, SelectQueryInfo & query_info, UInt64 max_block_size_, ContextPtr context_)
     : SourceStepWithFilter({.header = header})
     , storage_limits(query_info.storage_limits)
+    , max_block_size(max_block_size_)
     , context(std::move(context_))
 {
 }
@@ -602,13 +641,7 @@ ReadFromSystemZooKeeper::ReadFromSystemZooKeeper(const Block & header, SelectQue
 void ReadFromSystemZooKeeper::initializePipeline(QueryPipelineBuilder & pipeline, const BuildQueryPipelineSettings &)
 {
     const auto & header = getOutputStream().header;
-    MutableColumns res_columns = header.cloneEmptyColumns();
-    fillData(res_columns);
-
-    UInt64 num_rows = res_columns.at(0)->size();
-    Chunk chunk(std::move(res_columns), num_rows);
-
-    auto source = std::make_shared<SourceFromSingleChunk>(header, std::move(chunk));
+    auto source = std::make_shared<SystemZooKeeperSource>(std::move(paths), header, max_block_size, context);
     source->setStorageLimits(storage_limits);
     processors.emplace_back(source);
     pipeline.init(Pipe(std::move(source)));

From 6a2867819ce091c313b1a14c3c39d42244553dab Mon Sep 17 00:00:00 2001
From: Alexander Gololobov <davenger@clickhouse.com>
Date: Sat, 3 Feb 2024 18:00:50 +0100
Subject: [PATCH 33/73] Test for filters w/ and w/o path column

---
 .../02976_system_zookeeper_filters.reference    |  6 ++++++
 .../02976_system_zookeeper_filters.sql          | 17 +++++++++++++++++
 2 files changed, 23 insertions(+)
 create mode 100644 tests/queries/0_stateless/02976_system_zookeeper_filters.reference
 create mode 100644 tests/queries/0_stateless/02976_system_zookeeper_filters.sql

diff --git a/tests/queries/0_stateless/02976_system_zookeeper_filters.reference b/tests/queries/0_stateless/02976_system_zookeeper_filters.reference
new file mode 100644
index 00000000000..a9e2f17562a
--- /dev/null
+++ b/tests/queries/0_stateless/02976_system_zookeeper_filters.reference
@@ -0,0 +1,6 @@
+1
+1
+1
+1
+1
+1
diff --git a/tests/queries/0_stateless/02976_system_zookeeper_filters.sql b/tests/queries/0_stateless/02976_system_zookeeper_filters.sql
new file mode 100644
index 00000000000..3664c817da7
--- /dev/null
+++ b/tests/queries/0_stateless/02976_system_zookeeper_filters.sql
@@ -0,0 +1,17 @@
+-- Tags: zookeeper, no-parallel, no-fasttest, long
+
+SET allow_unrestricted_reads_from_keeper = 'false';
+
+SELECT count() > 0 FROM system.zookeeper; -- { serverError BAD_ARGUMENTS }
+SELECT count() > 0 FROM system.zookeeper WHERE name LIKE '%_%'; -- { serverError BAD_ARGUMENTS }
+SELECT count() > 0 FROM system.zookeeper WHERE value LIKE '%'; -- { serverError BAD_ARGUMENTS }
+SELECT count() > 0 FROM system.zookeeper WHERE path LIKE '/%'; -- { serverError BAD_ARGUMENTS }
+SELECT count() > 0 FROM system.zookeeper WHERE path = '/';
+
+SET allow_unrestricted_reads_from_keeper = 'true';
+
+SELECT count() > 0 FROM system.zookeeper;
+SELECT count() > 0 FROM system.zookeeper WHERE name LIKE '%_%';
+SELECT count() > 0 FROM system.zookeeper WHERE value LIKE '%';
+SELECT count() > 0 FROM system.zookeeper WHERE path LIKE '/%';
+SELECT count() > 0 FROM system.zookeeper WHERE path = '/';

From 454b03270988494e4291cacef2361af5586c1260 Mon Sep 17 00:00:00 2001
From: Maksim Kita <kitaetoya@gmail.com>
Date: Mon, 5 Feb 2024 12:52:18 +0300
Subject: [PATCH 34/73] MultiVersion use mutex

---
 src/Common/MultiVersion.h | 28 ++++++++++++++++++++++------
 1 file changed, 22 insertions(+), 6 deletions(-)

diff --git a/src/Common/MultiVersion.h b/src/Common/MultiVersion.h
index 6ea337e6a7c..8f488f9fcbc 100644
--- a/src/Common/MultiVersion.h
+++ b/src/Common/MultiVersion.h
@@ -2,6 +2,7 @@
 
 #include <atomic>
 #include <memory>
+#include <mutex>
 #include <base/defines.h>
 
 
@@ -20,6 +21,9 @@
   * }   // now we finish own current version; if the version is outdated and no one else is using it - it will be destroyed.
   *
   * All methods are thread-safe.
+  *
+  * Standard library does not have atomic_shared_ptr, and we do not use std::atomic* operations on shared_ptr,
+  * because standard library implementation uses fixed table of mutexes, and it is better to avoid contention here.
   */
 template <typename T>
 class MultiVersion
@@ -42,25 +46,37 @@ public:
     MultiVersion & operator=(MultiVersion && src)
     {
         if (this != &src)
-            std::atomic_store(&current_version, std::atomic_exchange(&src.current_version, Version{}));
+        {
+            Version version;
+
+            {
+                std::lock_guard<std::mutex> lock(src.mutex);
+                src.current_version.swap(version);
+            }
+
+            std::lock_guard<std::mutex> lock(mutex);
+            current_version = std::move(version);
+        }
+
         return *this;
     }
 
     /// Obtain current version for read-only usage. Returns shared_ptr, that manages lifetime of version.
     Version get() const
     {
-        return std::atomic_load(&current_version);
+        std::lock_guard<std::mutex> lock(mutex);
+        return current_version;
     }
 
-    /// TODO: replace atomic_load/store() on shared_ptr (which is deprecated as of C++20) by C++20 std::atomic<std::shared_ptr>.
-    /// Clang 15 currently does not support it.
-
     /// Update an object with new version.
     void set(std::unique_ptr<const T> && value)
     {
-        std::atomic_store(&current_version, Version{std::move(value)});
+        Version version{std::move(value)};
+        std::lock_guard<std::mutex> lock(mutex);
+        current_version = std::move(version);
     }
 
 private:
+    mutable std::mutex mutex;
     Version current_version;
 };

From a2fc3facca0f62405addeb4119c745faaa191c11 Mon Sep 17 00:00:00 2001
From: mikhnenko <mikhnenko@yandex-team.com>
Date: Mon, 5 Feb 2024 13:21:06 +0300
Subject: [PATCH 35/73] Add missed #include <bit> and <exception>

---
 base/base/bit_cast.h          | 1 +
 base/base/wide_integer_impl.h | 1 +
 src/Common/Exception.h        | 1 +
 3 files changed, 3 insertions(+)

diff --git a/base/base/bit_cast.h b/base/base/bit_cast.h
index 4783a84586b..9a92b7660f1 100644
--- a/base/base/bit_cast.h
+++ b/base/base/bit_cast.h
@@ -1,5 +1,6 @@
 #pragma once
 
+#include <bit>
 #include <cstring>
 #include <algorithm>
 #include <type_traits>
diff --git a/base/base/wide_integer_impl.h b/base/base/wide_integer_impl.h
index c1fd7b69b7f..17b1fa7cd6a 100644
--- a/base/base/wide_integer_impl.h
+++ b/base/base/wide_integer_impl.h
@@ -6,6 +6,7 @@
 
 #include "throwError.h"
 
+#include <bit>
 #include <cmath>
 #include <cfloat>
 #include <cassert>
diff --git a/src/Common/Exception.h b/src/Common/Exception.h
index 817a4400080..94e4a986feb 100644
--- a/src/Common/Exception.h
+++ b/src/Common/Exception.h
@@ -1,6 +1,7 @@
 #pragma once
 
 #include <cerrno>
+#include <exception>
 #include <vector>
 #include <memory>
 

From 8581bdbe58bdeea57c33f1e499624111ad26d09b Mon Sep 17 00:00:00 2001
From: Sema Checherinda <Sema.Checherinda@clickhouse.com>
Date: Fri, 2 Feb 2024 18:44:01 +0100
Subject: [PATCH 36/73] do not call value() on empty optional query_info

---
 src/Server/GRPCServer.cpp | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/src/Server/GRPCServer.cpp b/src/Server/GRPCServer.cpp
index f31a8d6feb5..7c532312612 100644
--- a/src/Server/GRPCServer.cpp
+++ b/src/Server/GRPCServer.cpp
@@ -419,7 +419,11 @@ namespace
         void read(GRPCQueryInfo & query_info_, const CompletionCallback & callback) override
         {
             if (!query_info.has_value())
+            {
                 callback(false);
+                return;
+            }
+
             query_info_ = std::move(query_info).value();
             query_info.reset();
             callback(true);
@@ -486,7 +490,11 @@ namespace
         void read(GRPCQueryInfo & query_info_, const CompletionCallback & callback) override
         {
             if (!query_info.has_value())
+            {
                 callback(false);
+                return;
+            }
+
             query_info_ = std::move(query_info).value();
             query_info.reset();
             callback(true);

From ad13a57b7f462fa7c0ed0a50a2572969d334e845 Mon Sep 17 00:00:00 2001
From: Robert Schulze <robert@clickhouse.com>
Date: Mon, 5 Feb 2024 13:43:55 +0000
Subject: [PATCH 37/73] Improve detection of v2 capabilities

---
 base/base/getMemoryAmount.cpp | 27 +++++++++++++++++++--------
 1 file changed, 19 insertions(+), 8 deletions(-)

diff --git a/base/base/getMemoryAmount.cpp b/base/base/getMemoryAmount.cpp
index 7f50683944e..ccdc0f0f976 100644
--- a/base/base/getMemoryAmount.cpp
+++ b/base/base/getMemoryAmount.cpp
@@ -27,14 +27,26 @@ std::optional<uint64_t> getCgroupsV2MemoryLimit()
     if (!controllers_file.is_open())
         return {};
 
-    /// We also need the memory controller enabled
-    std::stringstream controllers_buf;
-    controllers_buf << controllers_file.rdbuf();
-    std::string controllers = controllers_buf.str();
-    if (controllers.find("memory") == std::string::npos)
+    /// Make sure that the memory controller is enabled.
+    /// - cgroup.controllers defines which controllers *can* be enabled.
+    /// - cgroup.subtree_control defines which controllers *are* enabled.
+    /// (see https://docs.kernel.org/admin-guide/cgroup-v2.html)
+    /// Caveat: nested groups may disable controllers. For simplicity, check only the top-level group.
+    /// ReadBufferFromFile subtree_control_file(default_cgroups_mount / "cgroup.subtree_control");
+    /// std::string subtree_control;
+    /// readString(subtree_control, subtree_control_file);
+    /// if (subtree_control.find("memory") == std::string::npos)
+    ///     return {};
+    std::ifstream subtree_control_file(default_cgroups_mount / "cgroup.subtree_control");
+    std::stringstream subtree_control_buf;
+    subtree_control_buf << subtree_control_file.rdbuf();
+    std::string subtree_control = subtree_control_buf.str();
+    if (subtree_control.find("memory") == std::string::npos)
         return {};
 
     /// Identify the cgroup the process belongs to
+    /// All PIDs assigned to a cgroup are in /sys/fs/cgroups/{cgroup_name}/cgroup.procs
+    /// A simpler way to get the membership is:
     std::ifstream cgroup_name_file("/proc/self/cgroup");
     if (!cgroup_name_file.is_open())
         return {};
@@ -44,10 +56,9 @@ std::optional<uint64_t> getCgroupsV2MemoryLimit()
     std::string cgroup_name = cgroup_name_buf.str();
     if (!cgroup_name.empty() && cgroup_name.back() == '\n')
         cgroup_name.pop_back(); /// remove trailing newline, if any
-    /// cgroups v2 will show a single line with prefix "0::/"
-    /// - https://book.hacktricks.xyz/linux-hardening/privilege-escalation/docker-security/cgroups
+    /// With cgroups v2, there will be a *single* line with prefix "0::/"
     const std::string v2_prefix = "0::/";
-    if (cgroup_name.find('\n') != std::string::npos || !cgroup_name.starts_with(v2_prefix))
+    if (!cgroup_name.starts_with(v2_prefix))
         return {};
     cgroup_name = cgroup_name.substr(v2_prefix.length());
 

From 9b5ea7bb8a966da3681b8887ab6c87debc4e3e39 Mon Sep 17 00:00:00 2001
From: vdimir <vdimir@clickhouse.com>
Date: Thu, 1 Feb 2024 14:29:25 +0000
Subject: [PATCH 38/73] Analyzer: Fix 00917_multiple_joins_denny_crane

---
 src/Analyzer/Passes/QueryAnalysisPass.cpp | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/src/Analyzer/Passes/QueryAnalysisPass.cpp b/src/Analyzer/Passes/QueryAnalysisPass.cpp
index cb1e94305fb..5e1e42b25ac 100644
--- a/src/Analyzer/Passes/QueryAnalysisPass.cpp
+++ b/src/Analyzer/Passes/QueryAnalysisPass.cpp
@@ -2759,7 +2759,13 @@ QueryTreeNodePtr QueryAnalyzer::tryResolveIdentifierFromAliases(const Identifier
     {
         if (identifier_lookup.isExpressionLookup())
         {
-            return tryResolveIdentifierFromCompoundExpression(identifier_lookup.identifier, 1 /*identifier_bind_size*/, it->second, {}, scope);
+            return tryResolveIdentifierFromCompoundExpression(
+                identifier_lookup.identifier,
+                1 /*identifier_bind_size*/,
+                it->second,
+                {} /* compound_expression_source */,
+                scope,
+                identifier_resolve_settings.allow_to_check_join_tree /* can_be_not_found */);
         }
         else if (identifier_lookup.isFunctionLookup() || identifier_lookup.isTableExpressionLookup())
         {

From cbb63e333c1e263e89accc5262e4076753453fa3 Mon Sep 17 00:00:00 2001
From: Dmitry Novik <mrnovikd@gmail.com>
Date: Fri, 2 Feb 2024 14:48:38 +0100
Subject: [PATCH 39/73] Update analyzer_tech_debt.txt

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

diff --git a/tests/analyzer_tech_debt.txt b/tests/analyzer_tech_debt.txt
index d944dd96e25..b84b46114b4 100644
--- a/tests/analyzer_tech_debt.txt
+++ b/tests/analyzer_tech_debt.txt
@@ -23,7 +23,6 @@
 02428_parameterized_view
 02493_inconsistent_hex_and_binary_number
 02575_merge_prewhere_different_default_kind
-00917_multiple_joins_denny_crane
 02725_agg_projection_resprect_PK
 02763_row_policy_storage_merge_alias
 02818_parameterized_view_with_cte_multiple_usage

From c4e1d09c66914b95bb8e8d82bff6ee7d24ecc986 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Mon, 5 Feb 2024 17:33:14 +0100
Subject: [PATCH 40/73] Reserve enough memory for
 memcpySmallAllowReadWriteOverflow15

---
 src/Functions/array/arrayElement.cpp | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/src/Functions/array/arrayElement.cpp b/src/Functions/array/arrayElement.cpp
index cea407aee02..8669fd1f3a7 100644
--- a/src/Functions/array/arrayElement.cpp
+++ b/src/Functions/array/arrayElement.cpp
@@ -670,8 +670,7 @@ struct ArrayElementStringImpl
         ColumnArray::Offset current_offset = 0;
         /// get the total result bytes at first, and reduce the cost of result_data.resize.
         size_t total_result_bytes = 0;
-        ColumnString::Chars zero_buf(1);
-        zero_buf.push_back(0);
+        ColumnString::Chars zero_buf(16, '\0'); /// Needs 15 extra bytes for memcpySmallAllowReadWriteOverflow15
         std::vector<std::pair<const ColumnString::Char *, UInt64>> selected_bufs;
         selected_bufs.reserve(size);
         for (size_t i = 0; i < size; ++i)
@@ -737,8 +736,7 @@ struct ArrayElementStringImpl
         size_t size = offsets.size();
         result_offsets.resize(size);
 
-        ColumnString::Chars zero_buf(1);
-        zero_buf.push_back(0);
+        ColumnString::Chars zero_buf(16, '\0'); /// Needs 15 extra bytes for memcpySmallAllowReadWriteOverflow15
         ColumnArray::Offset current_offset = 0;
         /// get the total result bytes at first, and reduce the cost of result_data.resize.
         size_t total_result_bytes = 0;

From ca9e2cac8a6c6fa368a38f8b81360072343aef7b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Mon, 5 Feb 2024 17:37:06 +0100
Subject: [PATCH 41/73] Add test

---
 .../0_stateless/02983_empty_map.reference     |  7 +++++++
 tests/queries/0_stateless/02983_empty_map.sql | 21 +++++++++++++++++++
 2 files changed, 28 insertions(+)
 create mode 100644 tests/queries/0_stateless/02983_empty_map.reference
 create mode 100644 tests/queries/0_stateless/02983_empty_map.sql

diff --git a/tests/queries/0_stateless/02983_empty_map.reference b/tests/queries/0_stateless/02983_empty_map.reference
new file mode 100644
index 00000000000..fadedaf23ae
--- /dev/null
+++ b/tests/queries/0_stateless/02983_empty_map.reference
@@ -0,0 +1,7 @@
+-- { echoOn }
+SELECT f1, f2['2'], count() FROM t1 GROUP BY 1,2 order by 1,2;
+1		1000111
+SELECT f1, f3['2'], count() FROM t1 GROUP BY 1,2 order by 1,2;
+1		1000111
+SELECT f1, f4[2], count() FROM t1 GROUP BY 1,2 order by 1,2;
+1	0	1000111
diff --git a/tests/queries/0_stateless/02983_empty_map.sql b/tests/queries/0_stateless/02983_empty_map.sql
new file mode 100644
index 00000000000..78bc5d8736f
--- /dev/null
+++ b/tests/queries/0_stateless/02983_empty_map.sql
@@ -0,0 +1,21 @@
+--https://github.com/ClickHouse/ClickHouse/issues/59402
+CREATE TABLE t1
+(
+    f1 Int32,
+    f2 Map(LowCardinality(String),LowCardinality(String)),
+    f3 Map(String,String),
+    f4 Map(Int32,Int32)
+)
+ENGINE=Memory AS
+SELECT 1 as f1,
+       map(number%2,number%10) as f2,
+       f2 as f3,
+       f2 as f4
+from numbers(1000111);
+
+SET max_block_size=10;
+
+-- { echoOn }
+SELECT f1, f2['2'], count() FROM t1 GROUP BY 1,2 order by 1,2;
+SELECT f1, f3['2'], count() FROM t1 GROUP BY 1,2 order by 1,2;
+SELECT f1, f4[2], count() FROM t1 GROUP BY 1,2 order by 1,2;

From 3ddaa1b5f19251974d16431d6c35f0965f639610 Mon Sep 17 00:00:00 2001
From: Max K <max.kainov@clickhouse.com>
Date: Mon, 5 Feb 2024 18:09:43 +0100
Subject: [PATCH 42/73] CI: nightly job to update latest docker tag only
 (#59586)

* CI: nightly job to update latest docker tag only

---------

Co-authored-by: robot-clickhouse <robot-clickhouse@users.noreply.github.com>
---
 .github/workflows/master.yml          | 10 ---
 .github/workflows/nightly.yml         |  2 +-
 .github/workflows/reusable_docker.yml | 16 ++---
 tests/ci/ci.py                        | 99 ++++++++++-----------------
 tests/ci/ci_config.py                 |  9 ++-
 tests/ci/docker_manifests_merge.py    | 14 ++--
 6 files changed, 58 insertions(+), 92 deletions(-)

diff --git a/.github/workflows/master.yml b/.github/workflows/master.yml
index 5d57e6fc1d8..2471e4f9194 100644
--- a/.github/workflows/master.yml
+++ b/.github/workflows/master.yml
@@ -55,7 +55,6 @@ jobs:
     uses: ./.github/workflows/reusable_docker.yml
     with:
       data: ${{ needs.RunConfig.outputs.data }}
-      set_latest: true
   StyleCheck:
     needs: [RunConfig, BuildDockers]
     if: ${{ !failure() && !cancelled() }}
@@ -362,14 +361,6 @@ jobs:
       test_name: Stateless tests (release)
       runner_type: func-tester
       data: ${{ needs.RunConfig.outputs.data }}
-  FunctionalStatelessTestReleaseDatabaseOrdinary:
-    needs: [RunConfig, BuilderDebRelease]
-    if: ${{ !failure() && !cancelled() }}
-    uses: ./.github/workflows/reusable_test.yml
-    with:
-      test_name: Stateless tests (release, DatabaseOrdinary)
-      runner_type: func-tester
-      data: ${{ needs.RunConfig.outputs.data }}
   FunctionalStatelessTestReleaseDatabaseReplicated:
     needs: [RunConfig, BuilderDebRelease]
     if: ${{ !failure() && !cancelled() }}
@@ -733,7 +724,6 @@ jobs:
       - MarkReleaseReady
       - FunctionalStatelessTestDebug
       - FunctionalStatelessTestRelease
-      - FunctionalStatelessTestReleaseDatabaseOrdinary
       - FunctionalStatelessTestReleaseDatabaseReplicated
       - FunctionalStatelessTestReleaseAnalyzer
       - FunctionalStatelessTestReleaseS3
diff --git a/.github/workflows/nightly.yml b/.github/workflows/nightly.yml
index 2774eae24cc..770e1ec3789 100644
--- a/.github/workflows/nightly.yml
+++ b/.github/workflows/nightly.yml
@@ -28,7 +28,7 @@ jobs:
         id: runconfig
         run: |
             echo "::group::configure CI run"
-            python3 "$GITHUB_WORKSPACE/tests/ci/ci.py" --configure --skip-jobs --rebuild-all-docker --outfile ${{ runner.temp }}/ci_run_data.json
+            python3 "$GITHUB_WORKSPACE/tests/ci/ci.py" --configure --skip-jobs --outfile ${{ runner.temp }}/ci_run_data.json
             echo "::endgroup::"
 
             echo "::group::CI run configure results"
diff --git a/.github/workflows/reusable_docker.yml b/.github/workflows/reusable_docker.yml
index 08a5740e7e0..1b186350580 100644
--- a/.github/workflows/reusable_docker.yml
+++ b/.github/workflows/reusable_docker.yml
@@ -46,7 +46,7 @@ jobs:
     needs: [DockerBuildAmd64, DockerBuildAarch64]
     runs-on: [self-hosted, style-checker]
     if: |
-      !failure() && !cancelled() && toJson(fromJson(inputs.data).docker_data.missing_multi) != '[]'
+      !failure() && !cancelled() && (toJson(fromJson(inputs.data).docker_data.missing_multi) != '[]' || inputs.set_latest)
     steps:
       - name: Check out repository code
         uses: ClickHouse/checkout@v1
@@ -55,14 +55,12 @@ jobs:
       - name: Build images
         run: |
           cd "$GITHUB_WORKSPACE/tests/ci"
+          FLAG_LATEST=''
           if [ "${{ inputs.set_latest }}" == "true" ]; then
+            FLAG_LATEST='--set-latest'
             echo "latest tag will be set for resulting manifests"
-            python3 docker_manifests_merge.py --suffix amd64 --suffix aarch64 \
-              --image-tags '${{ toJson(fromJson(inputs.data).docker_data.images) }}' \
-              --missing-images '${{ toJson(fromJson(inputs.data).docker_data.missing_multi) }}' \
-              --set-latest
-          else
-            python3 docker_manifests_merge.py --suffix amd64 --suffix aarch64 \
-              --image-tags '${{ toJson(fromJson(inputs.data).docker_data.images) }}' \
-              --missing-images '${{ toJson(fromJson(inputs.data).docker_data.missing_multi) }}'
           fi
+          python3 docker_manifests_merge.py --suffix amd64 --suffix aarch64 \
+            --image-tags '${{ toJson(fromJson(inputs.data).docker_data.images) }}' \
+            --missing-images '${{ toJson(fromJson(inputs.data).docker_data.missing_multi) }}' \
+            "$FLAG_LATEST"
diff --git a/tests/ci/ci.py b/tests/ci/ci.py
index 12e27b532db..d2d05b1ed2d 100644
--- a/tests/ci/ci.py
+++ b/tests/ci/ci.py
@@ -734,12 +734,6 @@ def parse_args(parser: argparse.ArgumentParser) -> argparse.Namespace:
         default=False,
         help="skip fetching data about job runs, used in --configure action (for debugging and nigthly ci)",
     )
-    parser.add_argument(
-        "--rebuild-all-docker",
-        action="store_true",
-        default=False,
-        help="will create run config for rebuilding all dockers, used in --configure action (for nightly docker job)",
-    )
     # FIXME: remove, not used
     parser.add_argument(
         "--rebuild-all-binaries",
@@ -939,9 +933,7 @@ def _update_config_for_docs_only(jobs_data: dict) -> None:
     }
 
 
-def _configure_docker_jobs(
-    rebuild_all_dockers: bool, docker_digest_or_latest: bool = False
-) -> Dict:
+def _configure_docker_jobs(docker_digest_or_latest: bool) -> Dict:
     print("::group::Docker images check")
     # generate docker jobs data
     docker_digester = DockerDigester()
@@ -950,50 +942,33 @@ def _configure_docker_jobs(
     )  # 'image name - digest' mapping
     images_info = docker_images_helper.get_images_info()
 
-    # a. check missing images
-    if not rebuild_all_dockers:
-        # FIXME: we need login as docker manifest inspect goes directly to one of the *.docker.com hosts instead of "registry-mirrors" : ["http://dockerhub-proxy.dockerhub-proxy-zone:5000"]
-        #         find if it's possible to use the setting of /etc/docker/daemon.json
-        docker_images_helper.docker_login()
-        missing_multi_dict = check_missing_images_on_dockerhub(imagename_digest_dict)
-        missing_multi = list(missing_multi_dict)
-        missing_amd64 = []
-        missing_aarch64 = []
-        if not docker_digest_or_latest:
-            # look for missing arm and amd images only among missing multiarch manifests @missing_multi_dict
-            # to avoid extra dockerhub api calls
-            missing_amd64 = list(
-                check_missing_images_on_dockerhub(missing_multi_dict, "amd64")
-            )
-            # FIXME: WA until full arm support: skip not supported arm images
-            missing_aarch64 = list(
-                check_missing_images_on_dockerhub(
-                    {
-                        im: digest
-                        for im, digest in missing_multi_dict.items()
-                        if not images_info[im]["only_amd64"]
-                    },
-                    "aarch64",
-                )
-            )
-        # FIXME: temporary hack, remove after transition to docker digest as tag
-        else:
-            if missing_multi:
-                print(
-                    f"WARNING: Missing images {list(missing_multi)} - fallback to latest tag"
-                )
-                for image in missing_multi:
-                    imagename_digest_dict[image] = "latest"
-    else:
-        # add all images to missing
-        missing_multi = list(imagename_digest_dict)
-        missing_amd64 = missing_multi
+    # FIXME: we need login as docker manifest inspect goes directly to one of the *.docker.com hosts instead of "registry-mirrors" : ["http://dockerhub-proxy.dockerhub-proxy-zone:5000"]
+    #   find if it's possible to use the setting of /etc/docker/daemon.json (https://github.com/docker/cli/issues/4484#issuecomment-1688095463)
+    docker_images_helper.docker_login()
+    missing_multi_dict = check_missing_images_on_dockerhub(imagename_digest_dict)
+    missing_multi = list(missing_multi_dict)
+    missing_amd64 = []
+    missing_aarch64 = []
+    if not docker_digest_or_latest:
+        # look for missing arm and amd images only among missing multiarch manifests @missing_multi_dict
+        # to avoid extra dockerhub api calls
+        missing_amd64 = list(
+            check_missing_images_on_dockerhub(missing_multi_dict, "amd64")
+        )
         # FIXME: WA until full arm support: skip not supported arm images
-        missing_aarch64 = [
-            name
-            for name in imagename_digest_dict
-            if not images_info[name]["only_amd64"]
-        ]
+        missing_aarch64 = list(
+            check_missing_images_on_dockerhub(
+                {
+                    im: digest
+                    for im, digest in missing_multi_dict.items()
+                    if not images_info[im]["only_amd64"]
+                },
+                "aarch64",
+            )
+        )
+    else:
+        if missing_multi:
+            assert False, f"Missing images [{missing_multi}], cannot proceed"
     print("::endgroup::")
 
     return {
@@ -1502,9 +1477,7 @@ def main() -> int:
         print(f"Got CH version for this commit: [{version}]")
 
         docker_data = (
-            _configure_docker_jobs(
-                args.rebuild_all_docker, args.docker_digest_or_latest
-            )
+            _configure_docker_jobs(args.docker_digest_or_latest)
             if not args.skip_docker
             else {}
         )
@@ -1528,17 +1501,16 @@ def main() -> int:
             else {}
         )
 
-        # FIXME: Early style check manipulates with job names might be not robust with await feature
-        if pr_info.number != 0 and not args.docker_digest_or_latest:
-            # FIXME: it runs style check before docker build if possible (style-check images is not changed)
-            #    find a way to do style check always before docker build and others
-            _check_and_update_for_early_style_check(jobs_data, docker_data)
-        if args.skip_jobs and pr_info.has_changes_in_documentation_only():
+        # # FIXME: Early style check manipulates with job names might be not robust with await feature
+        # if pr_info.number != 0:
+        #     # FIXME: it runs style check before docker build if possible (style-check images is not changed)
+        #     #    find a way to do style check always before docker build and others
+        #     _check_and_update_for_early_style_check(jobs_data, docker_data)
+        if not args.skip_jobs and pr_info.has_changes_in_documentation_only():
             _update_config_for_docs_only(jobs_data)
 
         # TODO: await pending jobs
         # wait for pending jobs to be finished, await_jobs is a long blocking call if any job has to be awaited
-        ci_cache = CiCache(s3, jobs_data["digests"])
         # awaited_jobs = ci_cache.await_jobs(jobs_data.get("jobs_to_wait", {}))
         # for job in awaited_jobs:
         #     jobs_to_do = jobs_data["jobs_to_do"]
@@ -1548,7 +1520,8 @@ def main() -> int:
         #         assert False, "BUG"
 
         # set planned jobs as pending in the CI cache if on the master
-        if pr_info.is_master():
+        if pr_info.is_master() and not args.skip_jobs:
+            ci_cache = CiCache(s3, jobs_data["digests"])
             for job in jobs_data["jobs_to_do"]:
                 config = CI_CONFIG.get_job_config(job)
                 if config.run_always or config.run_by_label:
diff --git a/tests/ci/ci_config.py b/tests/ci/ci_config.py
index 23270aae8ec..6036a04080c 100644
--- a/tests/ci/ci_config.py
+++ b/tests/ci/ci_config.py
@@ -296,6 +296,7 @@ stateless_check_digest = DigestConfig(
     include_paths=[
         "./tests/queries/0_stateless/",
         "./tests/clickhouse-test",
+        "./tests/config",
         "./tests/*.txt",
     ],
     exclude_files=[".md"],
@@ -305,6 +306,7 @@ stateful_check_digest = DigestConfig(
     include_paths=[
         "./tests/queries/1_stateful/",
         "./tests/clickhouse-test",
+        "./tests/config",
         "./tests/*.txt",
     ],
     exclude_files=[".md"],
@@ -316,6 +318,7 @@ stress_check_digest = DigestConfig(
         "./tests/queries/0_stateless/",
         "./tests/queries/1_stateful/",
         "./tests/clickhouse-test",
+        "./tests/config",
         "./tests/*.txt",
     ],
     exclude_files=[".md"],
@@ -962,10 +965,6 @@ CI_CONFIG = CiConfig(
         JobNames.STATELESS_TEST_ANALYZER_RELEASE: TestConfig(
             Build.PACKAGE_RELEASE, job_config=JobConfig(**statless_test_common_params)  # type: ignore
         ),
-        # delete?
-        # "Stateless tests (release, DatabaseOrdinary)": TestConfig(
-        #     Build.PACKAGE_RELEASE, job_config=JobConfig(**statless_test_common_params)  # type: ignore
-        # ),
         JobNames.STATELESS_TEST_DB_REPL_RELEASE: TestConfig(
             Build.PACKAGE_RELEASE,
             job_config=JobConfig(num_batches=4, **statless_test_common_params),  # type: ignore
@@ -1112,7 +1111,7 @@ CI_CONFIG = CiConfig(
         JobNames.SQL_LOGIC_TEST: TestConfig(
             Build.PACKAGE_RELEASE, job_config=JobConfig(**sqllogic_test_params)  # type: ignore
         ),
-        JobNames.SQL_LOGIC_TEST: TestConfig(
+        JobNames.SQLTEST: TestConfig(
             Build.PACKAGE_RELEASE, job_config=JobConfig(**sql_test_params)  # type: ignore
         ),
         JobNames.CLCIKBENCH_TEST: TestConfig(Build.PACKAGE_RELEASE),
diff --git a/tests/ci/docker_manifests_merge.py b/tests/ci/docker_manifests_merge.py
index f87246be24b..1ae92807c16 100644
--- a/tests/ci/docker_manifests_merge.py
+++ b/tests/ci/docker_manifests_merge.py
@@ -154,22 +154,28 @@ def main():
     ok_cnt, fail_cnt = 0, 0
     images = get_images_oredered_list()
     for image_obj in images:
-        if image_obj.repo not in missing_images:
-            continue
         tag = image_tags[image_obj.repo]
         if image_obj.only_amd64:
             # FIXME: WA until full arm support
             tags = [f"{tag}-{arch}" for arch in archs if arch != "aarch64"]
         else:
             tags = [f"{tag}-{arch}" for arch in archs]
-        manifest, test_result = create_manifest(image_obj.repo, tag, tags, args.push)
-        test_results.append(TestResult(manifest, test_result))
+
+        # 1. update multiarch latest manifest for every image
         if args.set_latest:
             manifest, test_result = create_manifest(
                 image_obj.repo, "latest", tags, args.push
             )
             test_results.append(TestResult(manifest, test_result))
 
+        # 2. skip manifest create if not missing
+        if image_obj.repo not in missing_images:
+            continue
+
+        # 3. created image:digest multiarch manifest for changed images only
+        manifest, test_result = create_manifest(image_obj.repo, tag, tags, args.push)
+        test_results.append(TestResult(manifest, test_result))
+
         if test_result != "OK":
             status = "failure"
             fail_cnt += 1

From c9954a7eb8e75c130435b84fd14e0be17765adb2 Mon Sep 17 00:00:00 2001
From: Bhavna Jindal <bhavna.jindal@ibm.com>
Date: Mon, 5 Feb 2024 09:23:11 -0800
Subject: [PATCH 43/73] added check for NaN and infinite values

---
 src/Functions/seriesOutliersDetectTukey.cpp | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/src/Functions/seriesOutliersDetectTukey.cpp b/src/Functions/seriesOutliersDetectTukey.cpp
index a2fc9cf8eb6..8a2e276c74a 100644
--- a/src/Functions/seriesOutliersDetectTukey.cpp
+++ b/src/Functions/seriesOutliersDetectTukey.cpp
@@ -6,6 +6,7 @@
 #include <Functions/FunctionFactory.h>
 #include <Functions/FunctionHelpers.h>
 #include <Functions/IFunction.h>
+#include <Common/NaNUtils.h>
 #include <cmath>
 
 namespace DB
@@ -80,19 +81,19 @@ public:
         if (arguments.size() > 1)
         {
             Float64 p_min = arguments[1].column->getFloat64(0);
-            if (p_min < min_quartile|| p_min > max_quartile)
+            if (isnan(p_min) || !isFinite(p_min) || p_min < min_quartile|| p_min > max_quartile)
                 throw Exception(ErrorCodes::BAD_ARGUMENTS, "The second argument of function {} must be in range [2.0, 98.0]", getName());
 
             min_percentile = p_min / 100;
 
             Float64 p_max = arguments[2].column->getFloat64(0);
-            if (p_max < min_quartile || p_max > max_quartile || p_max < min_percentile * 100)
+            if (isnan(p_max) || !isFinite(p_max) || p_max < min_quartile || p_max > max_quartile || p_max < min_percentile * 100)
                 throw Exception(ErrorCodes::BAD_ARGUMENTS, "The third argument of function {} must be in range [2.0, 98.0]", getName());
 
             max_percentile = p_max / 100;
 
             auto k_val = arguments[3].column->getFloat64(0);
-            if (k_val < 0.0)
+            if (k_val < 0.0 || isnan(k_val) || !isFinite(k_val))
                 throw Exception(ErrorCodes::BAD_ARGUMENTS, "The fourth argument of function {} must be a positive number", getName());
 
             k = k_val;

From d782c9bde622f5fbe10cebf6bd76df363a13a5c8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Mon, 5 Feb 2024 20:52:14 +0100
Subject: [PATCH 44/73] Fix crash in topK when merging empty states

---
 src/AggregateFunctions/AggregateFunctionTopK.cpp           | 3 +++
 src/Common/SpaceSaving.h                                   | 3 +++
 tests/queries/0_stateless/02984_topk_empty_merge.reference | 1 +
 tests/queries/0_stateless/02984_topk_empty_merge.sql       | 2 ++
 4 files changed, 9 insertions(+)
 create mode 100644 tests/queries/0_stateless/02984_topk_empty_merge.reference
 create mode 100644 tests/queries/0_stateless/02984_topk_empty_merge.sql

diff --git a/src/AggregateFunctions/AggregateFunctionTopK.cpp b/src/AggregateFunctions/AggregateFunctionTopK.cpp
index 660f136e4dc..dcbd5586406 100644
--- a/src/AggregateFunctions/AggregateFunctionTopK.cpp
+++ b/src/AggregateFunctions/AggregateFunctionTopK.cpp
@@ -234,6 +234,9 @@ public:
 
     void merge(AggregateDataPtr __restrict place, ConstAggregateDataPtr rhs, Arena *) const override
     {
+        if (!this->data(rhs).value.size())
+            return;
+
         auto & set = this->data(place).value;
         if (set.capacity() != reserved)
             set.resize(reserved);
diff --git a/src/Common/SpaceSaving.h b/src/Common/SpaceSaving.h
index 8dc0ee074c0..7a740ae6c9b 100644
--- a/src/Common/SpaceSaving.h
+++ b/src/Common/SpaceSaving.h
@@ -206,6 +206,9 @@ public:
      */
     void merge(const Self & rhs)
     {
+        if (!rhs.size())
+            return;
+
         UInt64 m1 = 0;
         UInt64 m2 = 0;
 
diff --git a/tests/queries/0_stateless/02984_topk_empty_merge.reference b/tests/queries/0_stateless/02984_topk_empty_merge.reference
new file mode 100644
index 00000000000..fe51488c706
--- /dev/null
+++ b/tests/queries/0_stateless/02984_topk_empty_merge.reference
@@ -0,0 +1 @@
+[]
diff --git a/tests/queries/0_stateless/02984_topk_empty_merge.sql b/tests/queries/0_stateless/02984_topk_empty_merge.sql
new file mode 100644
index 00000000000..754b0cb26a2
--- /dev/null
+++ b/tests/queries/0_stateless/02984_topk_empty_merge.sql
@@ -0,0 +1,2 @@
+-- https://github.com/ClickHouse/ClickHouse/issues/59107
+SELECT topK('102.4') FROM remote('127.0.0.{1,2}', view(SELECT NULL FROM system.one WHERE dummy = 1));

From 40885d7f16d2a5e72d500a799c8b718c36626e63 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Mon, 5 Feb 2024 21:51:26 +0100
Subject: [PATCH 45/73] Add extra sanity check

---
 src/Analyzer/Passes/QueryAnalysisPass.cpp | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/src/Analyzer/Passes/QueryAnalysisPass.cpp b/src/Analyzer/Passes/QueryAnalysisPass.cpp
index cb1e94305fb..de7788997be 100644
--- a/src/Analyzer/Passes/QueryAnalysisPass.cpp
+++ b/src/Analyzer/Passes/QueryAnalysisPass.cpp
@@ -5570,6 +5570,14 @@ ProjectionNames QueryAnalyzer::resolveFunction(QueryTreeNodePtr & node, Identifi
                 column = function_base->getConstantResultForNonConstArguments(argument_columns, result_type);
             }
 
+            if (column->getDataType() != result_type->getColumnType())
+                throw Exception(
+                    ErrorCodes::LOGICAL_ERROR,
+                    "Unexpected return type from {}. Expected {}. Got {}",
+                    function->getName(),
+                    result_type->getColumnType(),
+                    column->getDataType());
+
             /** Do not perform constant folding if there are aggregate or arrayJoin functions inside function.
               * Example: SELECT toTypeName(sum(number)) FROM numbers(10);
               */

From 86ac7b14d0bc1ea1796285755a87c68ba59e46b2 Mon Sep 17 00:00:00 2001
From: Vitaly Baranov <vitlibar@clickhouse.com>
Date: Mon, 5 Feb 2024 21:55:32 +0100
Subject: [PATCH 46/73] Fix Distributed table engine with a constant sharding
 key.

---
 src/Storages/StorageDistributed.cpp | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/src/Storages/StorageDistributed.cpp b/src/Storages/StorageDistributed.cpp
index f711ed5ac92..8b0c1b63306 100644
--- a/src/Storages/StorageDistributed.cpp
+++ b/src/Storages/StorageDistributed.cpp
@@ -1533,10 +1533,7 @@ ClusterPtr StorageDistributed::getOptimizedCluster(
 IColumn::Selector StorageDistributed::createSelector(const ClusterPtr cluster, const ColumnWithTypeAndName & result)
 {
     const auto & slot_to_shard = cluster->getSlotToShard();
-
     const IColumn * column = result.column.get();
-    if (const auto * col_const = typeid_cast<const ColumnConst *>(column))
-        column = &col_const->getDataColumn();
 
 // If result.type is DataTypeLowCardinality, do shard according to its dictionaryType
 #define CREATE_FOR_TYPE(TYPE)                                                                                       \

From 15b6dd143175e1540139cf4a9076a3a170d463b5 Mon Sep 17 00:00:00 2001
From: Tyler Hannan <tyler@clickhouse.com>
Date: Mon, 5 Feb 2024 22:27:13 +0100
Subject: [PATCH 47/73] Update README.md

Fixing dated Recordings link
---
 README.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/README.md b/README.md
index d356e429892..9ada350d173 100644
--- a/README.md
+++ b/README.md
@@ -37,7 +37,7 @@ Keep an eye out for upcoming meetups around the world. Somewhere else you want u
 
 ## Recent Recordings
 * **Recent Meetup Videos**: [Meetup Playlist](https://www.youtube.com/playlist?list=PL0Z2YDlm0b3iNDUzpY1S3L_iV4nARda_U) Whenever possible recordings of the ClickHouse Community Meetups are edited and presented as individual talks. Current featuring "Modern SQL in 2023", "Fast, Concurrent, and Consistent Asynchronous INSERTS in ClickHouse", and "Full-Text Indices: Design and Experiments"
-* **Recording available**: [**v23.10 Release Webinar**](https://www.youtube.com/watch?v=PGQS6uPb970) All the features of 23.10, one convenient video! Watch it now!
+* **Recording available**: [**v24.1 Release Webinar**](https://www.youtube.com/watch?v=pBF9g0wGAGs) All the features of 24.1, one convenient video! Watch it now!
 * **All release webinar recordings**: [YouTube playlist](https://www.youtube.com/playlist?list=PL0Z2YDlm0b3jAlSy1JxyP8zluvXaN3nxU)
 
   

From d8555af44836d6ecf4ce2d9bae04b1e7d7ae8f6b Mon Sep 17 00:00:00 2001
From: Nikita Mikhaylov <mikhaylovnikitka@gmail.com>
Date: Mon, 5 Feb 2024 22:41:28 +0100
Subject: [PATCH 48/73] Change the default value for
 `async_insert_max_data_size` to 10 MiB (#59536)

---
 src/Core/Settings.h               | 2 +-
 src/Core/SettingsChangesHistory.h | 3 ++-
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/src/Core/Settings.h b/src/Core/Settings.h
index 2b36b83edd2..dc863576a85 100644
--- a/src/Core/Settings.h
+++ b/src/Core/Settings.h
@@ -748,7 +748,7 @@ class IColumn;
     M(Bool, async_insert, false, "If true, data from INSERT query is stored in queue and later flushed to table in background. If wait_for_async_insert is false, INSERT query is processed almost instantly, otherwise client will wait until data will be flushed to table", 0) \
     M(Bool, wait_for_async_insert, true, "If true wait for processing of asynchronous insertion", 0) \
     M(Seconds, wait_for_async_insert_timeout, DBMS_DEFAULT_LOCK_ACQUIRE_TIMEOUT_SEC, "Timeout for waiting for processing asynchronous insertion", 0) \
-    M(UInt64, async_insert_max_data_size, 1000000, "Maximum size in bytes of unparsed data collected per query before being inserted", 0) \
+    M(UInt64, async_insert_max_data_size, 10485760, "Maximum size in bytes of unparsed data collected per query before being inserted", 0) \
     M(UInt64, async_insert_max_query_number, 450, "Maximum number of insert queries before being inserted", 0) \
     M(Milliseconds, async_insert_poll_timeout_ms, 10, "Timeout for polling data from asynchronous insert queue", 0) \
     M(Bool, async_insert_use_adaptive_busy_timeout, true, "If it is set to true, use adaptive busy timeout for asynchronous inserts", 0) \
diff --git a/src/Core/SettingsChangesHistory.h b/src/Core/SettingsChangesHistory.h
index c3b0cee00a4..db3a76e29cd 100644
--- a/src/Core/SettingsChangesHistory.h
+++ b/src/Core/SettingsChangesHistory.h
@@ -84,7 +84,8 @@ namespace SettingsChangesHistory
 /// It's used to implement `compatibility` setting (see https://github.com/ClickHouse/ClickHouse/issues/35972)
 static std::map<ClickHouseVersion, SettingsChangesHistory::SettingsChanges> settings_changes_history =
 {
-    {"24.2", {{"async_insert_poll_timeout_ms", 10, 10, "Timeout in milliseconds for polling data from asynchronous insert queue"},
+    {"24.2", {{"async_insert_max_data_size", 1000000, 10485760, "The previous value appeared to be too small."},
+              {"async_insert_poll_timeout_ms", 10, 10, "Timeout in milliseconds for polling data from asynchronous insert queue"},
               {"async_insert_use_adaptive_busy_timeout", true, true, "Use adaptive asynchronous insert timeout"},
               {"async_insert_busy_timeout_min_ms", 50, 50, "The minimum value of the asynchronous insert timeout in milliseconds; it also serves as the initial value, which may be increased later by the adaptive algorithm"},
               {"async_insert_busy_timeout_max_ms", 200, 200, "The minimum value of the asynchronous insert timeout in milliseconds; async_insert_busy_timeout_ms is aliased to async_insert_busy_timeout_max_ms"},

From a26e8b3a9e4594d648c7a7693b27114038aee89f Mon Sep 17 00:00:00 2001
From: Alexander Gololobov <davenger@clickhouse.com>
Date: Sat, 3 Feb 2024 18:03:38 +0100
Subject: [PATCH 49/73] Fix for filters w/o path column

---
 .../System/StorageSystemZooKeeper.cpp         | 24 ++++++++++++-------
 1 file changed, 15 insertions(+), 9 deletions(-)

diff --git a/src/Storages/System/StorageSystemZooKeeper.cpp b/src/Storages/System/StorageSystemZooKeeper.cpp
index cacd7cc2133..6d2166ff5f9 100644
--- a/src/Storages/System/StorageSystemZooKeeper.cpp
+++ b/src/Storages/System/StorageSystemZooKeeper.cpp
@@ -220,6 +220,7 @@ private:
     const UInt64 max_block_size;
     Paths paths;
     ContextPtr context;
+    bool started = false;
 };
 
 
@@ -441,7 +442,7 @@ static Paths extractPath(const ActionsDAG::NodeRawConstPtrs & filter_nodes, Cont
     for (const auto * node : filter_nodes)
         extractPathImpl(*node, res, context, allow_unrestricted);
 
-    if (filter_nodes.empty() && allow_unrestricted)
+    if (res.empty() && allow_unrestricted)
         res.emplace_back("/", ZkPathType::Recurse);
 
     return res;
@@ -457,7 +458,18 @@ void ReadFromSystemZooKeeper::applyFilters()
 Chunk SystemZooKeeperSource::generate()
 {
     if (paths.empty())
+    {
+        if (!started)
+            throw Exception(ErrorCodes::BAD_ARGUMENTS,
+                        "SELECT from system.zookeeper table must contain condition like path = 'path' "
+                        "or path IN ('path1','path2'...) or path IN (subquery) "
+                        "in WHERE clause unless `set allow_unrestricted_reads_from_keeper = 'true'`.");
+
+        /// No more work
         return {};
+    }
+
+    started = true;
 
     MutableColumns res_columns = getPort().getHeader().cloneEmptyColumns();
     size_t row_count = 0;
@@ -487,12 +499,6 @@ Chunk SystemZooKeeperSource::generate()
         return zookeeper;
     };
 
-    if (paths.empty())
-        throw Exception(ErrorCodes::BAD_ARGUMENTS,
-                        "SELECT from system.zookeeper table must contain condition like path = 'path' "
-                        "or path IN ('path1','path2'...) or path IN (subquery) "
-                        "in WHERE clause unless `set allow_unrestricted_reads_from_keeper = 'true'`.");
-
     const Int64 max_inflight_requests = std::max<Int64>(1, context->getSettingsRef().max_download_threads.value);
 
     struct ListTask
@@ -591,8 +597,8 @@ Chunk SystemZooKeeperSource::generate()
 
             auto & get_task = get_tasks[i];
             auto & list_task = list_tasks[get_task.list_task_idx];
-            if (auto elem = context->getProcessListElement())
-                elem->checkTimeLimit();
+            if (query_status)
+                query_status->checkTimeLimit();
 
             // Deduplication
             String key = list_task.path_part + '/' + get_task.node;

From e673d8bd0cc08d3a53587425b5ddd4f48c5bf943 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Tue, 6 Feb 2024 01:50:59 +0300
Subject: [PATCH 50/73] Update 85_bug-report.md

---
 .github/ISSUE_TEMPLATE/85_bug-report.md | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/.github/ISSUE_TEMPLATE/85_bug-report.md b/.github/ISSUE_TEMPLATE/85_bug-report.md
index 93b2342af70..6bf265260ac 100644
--- a/.github/ISSUE_TEMPLATE/85_bug-report.md
+++ b/.github/ISSUE_TEMPLATE/85_bug-report.md
@@ -17,7 +17,7 @@ assignees: ''
 
 > A link to reproducer in [https://fiddle.clickhouse.com/](https://fiddle.clickhouse.com/).
 
-**Does it reproduce on recent release?**
+**Does it reproduce on the most recent release?**
 
 [The list of releases](https://github.com/ClickHouse/ClickHouse/blob/master/utils/list-versions/version_date.tsv)
 
@@ -34,11 +34,11 @@ assignees: ''
 **How to reproduce**
 
 * Which ClickHouse server version to use
-* Which interface to use, if matters
+* Which interface to use, if it matters
 * Non-default settings, if any
 * `CREATE TABLE` statements for all tables involved
 * Sample data for all these tables, use [clickhouse-obfuscator](https://github.com/ClickHouse/ClickHouse/blob/master/programs/obfuscator/Obfuscator.cpp#L42-L80) if necessary
-* Queries to run that lead to unexpected result
+* Queries to run that lead to an unexpected result
 
 **Expected behavior**
 

From 1c0fa345ac341030d76a687b0900f8e66739d384 Mon Sep 17 00:00:00 2001
From: Justin de Guzman <justin@justindeguzman.net>
Date: Mon, 5 Feb 2024 15:51:28 -0800
Subject: [PATCH 51/73] Fix broken link for Variant data type (#59539)

---
 docs/en/sql-reference/data-types/variant.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/en/sql-reference/data-types/variant.md b/docs/en/sql-reference/data-types/variant.md
index 17d51878420..f027e3fe343 100644
--- a/docs/en/sql-reference/data-types/variant.md
+++ b/docs/en/sql-reference/data-types/variant.md
@@ -1,5 +1,5 @@
 ---
-slug: /en/sql-reference/data-types/json
+slug: /en/sql-reference/data-types/variant
 sidebar_position: 55
 sidebar_label: Variant
 ---

From 2c882f012ef23efa95d577cb00d08dbeb34e2783 Mon Sep 17 00:00:00 2001
From: taiyang-li <654010905@qq.com>
Date: Tue, 6 Feb 2024 11:35:14 +0800
Subject: [PATCH 52/73] opt isNotnull

---
 src/Functions/isNotNull.cpp | 93 ++++++++++++++++++++++++++++++-------
 1 file changed, 76 insertions(+), 17 deletions(-)

diff --git a/src/Functions/isNotNull.cpp b/src/Functions/isNotNull.cpp
index 360c2fc7f9f..a15a0e82e3c 100644
--- a/src/Functions/isNotNull.cpp
+++ b/src/Functions/isNotNull.cpp
@@ -1,19 +1,25 @@
-#include <Functions/IFunction.h>
-#include <Functions/FunctionHelpers.h>
-#include <Functions/FunctionFactory.h>
-#include <DataTypes/DataTypesNumber.h>
-#include <Core/ColumnNumbers.h>
-#include <Columns/ColumnNullable.h>
 #include <Columns/ColumnLowCardinality.h>
+#include <Columns/ColumnNullable.h>
 #include <Columns/ColumnVariant.h>
+#include <Core/ColumnNumbers.h>
+#include <DataTypes/DataTypesNumber.h>
+#include <Functions/FunctionFactory.h>
+#include <Functions/FunctionHelpers.h>
+#include <Functions/IFunction.h>
+#include <Functions/PerformanceAdaptors.h>
 #include <Common/assert_cast.h>
 
-
 namespace DB
 {
 namespace
 {
 
+
+#define DECLARE_SEVERAL_IMPLEMENTATIONS(...) \
+DECLARE_DEFAULT_CODE      (__VA_ARGS__) \
+DECLARE_AVX2_SPECIFIC_CODE(__VA_ARGS__)
+
+DECLARE_SEVERAL_IMPLEMENTATIONS(
 /// Implements the function isNotNull which returns true if a value
 /// is not null, false otherwise.
 class FunctionIsNotNull : public IFunction
@@ -21,11 +27,6 @@ class FunctionIsNotNull : public IFunction
 public:
     static constexpr auto name = "isNotNull";
 
-    static FunctionPtr create(ContextPtr)
-    {
-        return std::make_shared<FunctionIsNotNull>();
-    }
-
     std::string getName() const override
     {
         return name;
@@ -52,9 +53,9 @@ public:
             const auto & discriminators = checkAndGetColumn<ColumnVariant>(*elem.column)->getLocalDiscriminators();
             auto res = DataTypeUInt8().createColumn();
             auto & data = typeid_cast<ColumnUInt8 &>(*res).getData();
-            data.reserve(discriminators.size());
-            for (auto discr : discriminators)
-                data.push_back(discr != ColumnVariant::NULL_DISCRIMINATOR);
+            data.resize(discriminators.size());
+            for (size_t i = 0; i < discriminators.size(); ++i)
+                data[i] = discriminators[i] != ColumnVariant::NULL_DISCRIMINATOR;
             return res;
         }
 
@@ -64,9 +65,9 @@ public:
             const size_t null_index = low_cardinality_column->getDictionary().getNullValueIndex();
             auto res = DataTypeUInt8().createColumn();
             auto & data = typeid_cast<ColumnUInt8 &>(*res).getData();
-            data.reserve(low_cardinality_column->size());
+            data.resize(low_cardinality_column->size());
             for (size_t i = 0; i != low_cardinality_column->size(); ++i)
-                data.push_back(low_cardinality_column->getIndexAt(i) != null_index);
+                data[i] = (low_cardinality_column->getIndexAt(i) != null_index);
             return res;
         }
 
@@ -88,6 +89,64 @@ public:
             return DataTypeUInt8().createColumnConst(elem.column->size(), 1u);
         }
     }
+
+private:
+    MULTITARGET_FUNCTION_AVX2_SSE42(
+    MULTITARGET_FUNCTION_HEADER(static void NO_INLINE), vectorImpl, MULTITARGET_FUNCTION_BODY((const PaddedPODArray<UInt8> & null_map, PaddedPODArray<UInt8> & res) /// NOLINT
+    {
+        size_t size = null_map.size();
+        for (size_t i = 0; i < size; ++i)
+            res[i] = !null_map[i];
+    }))
+
+    static void NO_INLINE vector(const PaddedPODArray<UInt8> & null_map, PaddedPODArray<UInt8> & res)
+    {
+#if USE_MULTITARGET_CODE
+        if (isArchSupported(TargetArch::AVX2))
+        {
+            vectorImplAVX2(null_map, res);
+            return;
+        }
+
+        if (isArchSupported(TargetArch::SSE42))
+        {
+            vectorImplSSE42(null_map, res);
+            return;
+        }
+#endif
+        vectorImpl(null_map, res);
+    }
+};
+
+) // DECLARE_SEVERAL_IMPLEMENTATIONS
+#undef DECLARE_SEVERAL_IMPLEMENTATIONS
+
+class FunctionIsNotNull : public TargetSpecific::Default::FunctionIsNotNull
+{
+public:
+    explicit FunctionIsNotNull(ContextPtr context) : selector(context)
+    {
+        selector.registerImplementation<TargetArch::Default,
+            TargetSpecific::Default::FunctionIsNotNull>();
+
+    #if USE_MULTITARGET_CODE
+        selector.registerImplementation<TargetArch::AVX2,
+            TargetSpecific::AVX2::FunctionIsNotNull>();
+    #endif
+    }
+
+    ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const override
+    {
+        return selector.selectAndExecute(arguments, result_type, input_rows_count);
+    }
+
+    static FunctionPtr create(ContextPtr context)
+    {
+        return std::make_shared<FunctionIsNotNull>(context);
+    }
+
+private:
+    ImplementationSelector<IFunction> selector;
 };
 
 }

From bc9c1bf34583ebb7e5974a96725bbcb68da54008 Mon Sep 17 00:00:00 2001
From: taiyang-li <654010905@qq.com>
Date: Tue, 6 Feb 2024 11:48:13 +0800
Subject: [PATCH 53/73] improve ifnotnull

---
 src/Functions/isNotNull.cpp | 45 +++----------------------------------
 1 file changed, 3 insertions(+), 42 deletions(-)

diff --git a/src/Functions/isNotNull.cpp b/src/Functions/isNotNull.cpp
index a15a0e82e3c..dd5182aeade 100644
--- a/src/Functions/isNotNull.cpp
+++ b/src/Functions/isNotNull.cpp
@@ -14,12 +14,6 @@ namespace DB
 namespace
 {
 
-
-#define DECLARE_SEVERAL_IMPLEMENTATIONS(...) \
-DECLARE_DEFAULT_CODE      (__VA_ARGS__) \
-DECLARE_AVX2_SPECIFIC_CODE(__VA_ARGS__)
-
-DECLARE_SEVERAL_IMPLEMENTATIONS(
 /// Implements the function isNotNull which returns true if a value
 /// is not null, false otherwise.
 class FunctionIsNotNull : public IFunction
@@ -27,6 +21,8 @@ class FunctionIsNotNull : public IFunction
 public:
     static constexpr auto name = "isNotNull";
 
+    static FunctionPtr create(ContextPtr) { return std::make_shared<FunctionIsNotNull>(); }
+
     std::string getName() const override
     {
         return name;
@@ -77,10 +73,7 @@ public:
             auto res_column = ColumnUInt8::create(input_rows_count);
             const auto & src_data = nullable->getNullMapData();
             auto & res_data = assert_cast<ColumnUInt8 &>(*res_column).getData();
-
-            for (size_t i = 0; i < input_rows_count; ++i)
-                res_data[i] = !src_data[i];
-
+            vector(src_data, res_data);
             return res_column;
         }
         else
@@ -117,38 +110,6 @@ private:
         vectorImpl(null_map, res);
     }
 };
-
-) // DECLARE_SEVERAL_IMPLEMENTATIONS
-#undef DECLARE_SEVERAL_IMPLEMENTATIONS
-
-class FunctionIsNotNull : public TargetSpecific::Default::FunctionIsNotNull
-{
-public:
-    explicit FunctionIsNotNull(ContextPtr context) : selector(context)
-    {
-        selector.registerImplementation<TargetArch::Default,
-            TargetSpecific::Default::FunctionIsNotNull>();
-
-    #if USE_MULTITARGET_CODE
-        selector.registerImplementation<TargetArch::AVX2,
-            TargetSpecific::AVX2::FunctionIsNotNull>();
-    #endif
-    }
-
-    ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const override
-    {
-        return selector.selectAndExecute(arguments, result_type, input_rows_count);
-    }
-
-    static FunctionPtr create(ContextPtr context)
-    {
-        return std::make_shared<FunctionIsNotNull>(context);
-    }
-
-private:
-    ImplementationSelector<IFunction> selector;
-};
-
 }
 
 REGISTER_FUNCTION(IsNotNull)

From 5962ed0bd3ee89ba283c541982f60d13b1f010fa Mon Sep 17 00:00:00 2001
From: Vitaly Baranov <vitlibar@clickhouse.com>
Date: Mon, 5 Feb 2024 21:54:27 +0100
Subject: [PATCH 54/73] Add test.

---
 .../02983_const_sharding_key.reference        |  7 +++++
 .../0_stateless/02983_const_sharding_key.sql  | 26 +++++++++++++++++++
 2 files changed, 33 insertions(+)
 create mode 100644 tests/queries/0_stateless/02983_const_sharding_key.reference
 create mode 100644 tests/queries/0_stateless/02983_const_sharding_key.sql

diff --git a/tests/queries/0_stateless/02983_const_sharding_key.reference b/tests/queries/0_stateless/02983_const_sharding_key.reference
new file mode 100644
index 00000000000..06e567b11df
--- /dev/null
+++ b/tests/queries/0_stateless/02983_const_sharding_key.reference
@@ -0,0 +1,7 @@
+1
+2
+3
+4
+5
+6
+7
diff --git a/tests/queries/0_stateless/02983_const_sharding_key.sql b/tests/queries/0_stateless/02983_const_sharding_key.sql
new file mode 100644
index 00000000000..339293b8b81
--- /dev/null
+++ b/tests/queries/0_stateless/02983_const_sharding_key.sql
@@ -0,0 +1,26 @@
+-- Tags: distributed, no-parallel
+
+DROP DATABASE IF EXISTS shard_0;
+DROP DATABASE IF EXISTS shard_1;
+DROP TABLE IF EXISTS t_distr;
+
+CREATE DATABASE IF NOT EXISTS shard_0;
+CREATE DATABASE IF NOT EXISTS shard_1;
+
+CREATE TABLE shard_0.t_local (a Int) ENGINE = Memory;
+CREATE TABLE shard_1.t_local (a Int) ENGINE = Memory;
+CREATE TABLE t_distr (a Int) ENGINE = Distributed(test_cluster_two_shards_different_databases, '', 't_local', 1000);
+
+SET distributed_foreground_insert=0;
+INSERT INTO t_distr VALUES (1), (2), (3);
+
+SET distributed_foreground_insert=1;
+INSERT INTO t_distr VALUES (4), (5), (6), (7);
+
+SYSTEM FLUSH DISTRIBUTED t_distr;
+
+SELECT * FROM t_distr ORDER BY a;
+
+DROP TABLE t_distr;
+DROP DATABASE shard_0;
+DROP DATABASE shard_1;

From cbd81b506fb9acddd59087eddc2d431db91c26af Mon Sep 17 00:00:00 2001
From: Max Kainov <max.kainov@clickhouse.com>
Date: Tue, 6 Feb 2024 09:56:15 +0000
Subject: [PATCH 55/73] CI: fix ast fuzzer job report (slack bot issue)

 #job_ast_fuzzer_debug #job_style_check
---
 tests/ci/ast_fuzzer_check.py | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/tests/ci/ast_fuzzer_check.py b/tests/ci/ast_fuzzer_check.py
index 26ce7f5140b..5a0eca51570 100644
--- a/tests/ci/ast_fuzzer_check.py
+++ b/tests/ci/ast_fuzzer_check.py
@@ -13,7 +13,7 @@ from clickhouse_helper import (
 from docker_images_helper import DockerImage, get_docker_image, pull_image
 from env_helper import REPORT_PATH, TEMP_PATH
 from pr_info import PRInfo
-from report import JobReport
+from report import FAIL, FAILURE, OK, SUCCESS, JobReport, TestResult
 from stopwatch import Stopwatch
 from tee_popen import TeePopen
 
@@ -144,12 +144,16 @@ def main():
         with open(workspace_path / "description.txt", "r", encoding="utf-8") as desc_f:
             description = desc_f.readline().rstrip("\n")
     except:
-        status = "failure"
+        status = FAILURE
         description = "Task failed: $?=" + str(retcode)
 
+    test_result = TestResult(description, OK)
+    if "fail" in status:
+        test_result.status = FAIL
+
     JobReport(
         description=description,
-        test_results=[],
+        test_results=[test_result],
         status=status,
         start_time=stopwatch.start_time_str,
         duration=stopwatch.duration_seconds,
@@ -158,7 +162,7 @@ def main():
     ).dump()
 
     logging.info("Result: '%s', '%s'", status, description)
-    if status == "failure":
+    if status != SUCCESS:
         sys.exit(1)
 
 

From 306e739daadaf0ae52f9cb33e8fd13c852e9b4a1 Mon Sep 17 00:00:00 2001
From: Nikita Mikhaylov <mikhaylovnikitka@gmail.com>
Date: Tue, 6 Feb 2024 12:33:11 +0100
Subject: [PATCH 56/73] Updated a list of trusted contributors (#59616)

* Updated a list of trusted contributors

* Automatic style fix

---------

Co-authored-by: robot-clickhouse <robot-clickhouse@users.noreply.github.com>
---
 tests/ci/lambda_shared_package/lambda_shared/pr.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tests/ci/lambda_shared_package/lambda_shared/pr.py b/tests/ci/lambda_shared_package/lambda_shared/pr.py
index 4872ecb4d59..ce38475b3ee 100644
--- a/tests/ci/lambda_shared_package/lambda_shared/pr.py
+++ b/tests/ci/lambda_shared_package/lambda_shared/pr.py
@@ -43,6 +43,7 @@ TRUSTED_CONTRIBUTORS = {
         "tsolodov",  # ClickHouse, Inc
         "kitaisreal",
         "k-morozov",  # Konstantin Morozov, Yandex Cloud
+        "justindeguzman",  # ClickHouse, Inc
     ]
 }
 

From df93062472c32402ca96c9ff7d92fbc481df9008 Mon Sep 17 00:00:00 2001
From: "Mikhail f. Shiryaev" <felixoid@clickhouse.com>
Date: Fri, 2 Feb 2024 15:33:08 +0100
Subject: [PATCH 57/73] Apply and fix isort in scripts

---
 tests/ci/clickbench.py             | 14 +++++---------
 tests/ci/commit_status_helper.py   | 17 ++++++++++-------
 tests/ci/docker_images_check.py    | 11 +++++++----
 tests/ci/docker_manifests_merge.py | 11 +++++------
 tests/ci/functional_test_check.py  |  9 ++++++---
 tests/ci/report.py                 | 15 +++++++--------
 tests/ci/sqllogic_test.py          | 11 +++++------
 7 files changed, 45 insertions(+), 43 deletions(-)

diff --git a/tests/ci/clickbench.py b/tests/ci/clickbench.py
index 72827929ff9..f81a946e994 100644
--- a/tests/ci/clickbench.py
+++ b/tests/ci/clickbench.py
@@ -10,18 +10,14 @@ from pathlib import Path
 from typing import List, Tuple
 
 from build_download_helper import download_all_deb_packages
-from clickhouse_helper import (
-    CiLogsCredentials,
-)
-from commit_status_helper import (
-    override_status,
-)
-from docker_images_helper import get_docker_image, pull_image, DockerImage
-from env_helper import TEMP_PATH, REPORT_PATH
+from clickhouse_helper import CiLogsCredentials
+from commit_status_helper import override_status
+from docker_images_helper import DockerImage, get_docker_image, pull_image
+from env_helper import REPORT_PATH, TEMP_PATH
 from pr_info import FORCE_TESTS_LABEL, PRInfo
+from report import JobReport, TestResults
 from stopwatch import Stopwatch
 from tee_popen import TeePopen
-from report import JobReport, TestResults
 
 
 def get_image_name() -> str:
diff --git a/tests/ci/commit_status_helper.py b/tests/ci/commit_status_helper.py
index 851a4cc5298..c8dc6f713b3 100644
--- a/tests/ci/commit_status_helper.py
+++ b/tests/ci/commit_status_helper.py
@@ -1,14 +1,15 @@
 #!/usr/bin/env python3
 
-from collections import defaultdict
-import json
-from pathlib import Path
-from typing import Dict, List, Optional, Union
 import csv
+import json
 import logging
 import time
+from collections import defaultdict
 from dataclasses import asdict, dataclass
+from pathlib import Path
+from typing import Dict, List, Optional, Union
 
+# isort: off
 from github import Github
 from github.Commit import Commit
 from github.CommitStatus import CommitStatus
@@ -17,15 +18,17 @@ from github.GithubObject import NotSet
 from github.IssueComment import IssueComment
 from github.Repository import Repository
 
-from ci_config import CI_CONFIG, REQUIRED_CHECKS, CHECK_DESCRIPTIONS, CheckDescription
+# isort: on
+
+from ci_config import CHECK_DESCRIPTIONS, CI_CONFIG, REQUIRED_CHECKS, CheckDescription
 from env_helper import GITHUB_JOB_URL, GITHUB_REPOSITORY, TEMP_PATH
-from pr_info import PRInfo, SKIP_MERGEABLE_CHECK_LABEL
+from pr_info import SKIP_MERGEABLE_CHECK_LABEL, PRInfo
 from report import (
     ERROR,
     FAILURE,
     PENDING,
-    StatusType,
     SUCCESS,
+    StatusType,
     TestResult,
     TestResults,
     get_worst_status,
diff --git a/tests/ci/docker_images_check.py b/tests/ci/docker_images_check.py
index a25669d85d0..e88deae9a38 100644
--- a/tests/ci/docker_images_check.py
+++ b/tests/ci/docker_images_check.py
@@ -3,24 +3,27 @@ import argparse
 import json
 import logging
 import os
-import time
 import sys
+import time
 from pathlib import Path
 from typing import List, Optional, Tuple
 
+# isort: off
 from github import Github
 
+# isort: on
+
 from clickhouse_helper import ClickHouseHelper, prepare_tests_results_for_clickhouse
 from commit_status_helper import format_description, get_commit, post_commit_status
-from env_helper import RUNNER_TEMP, GITHUB_RUN_URL
+from docker_images_helper import DockerImageData, docker_login, get_images_oredered_list
+from env_helper import GITHUB_RUN_URL, RUNNER_TEMP
 from get_robot_token import get_best_robot_token
 from pr_info import PRInfo
-from report import TestResults, TestResult
+from report import TestResult, TestResults
 from s3_helper import S3Helper
 from stopwatch import Stopwatch
 from tee_popen import TeePopen
 from upload_result_helper import upload_results
-from docker_images_helper import DockerImageData, docker_login, get_images_oredered_list
 
 NAME = "Push to Dockerhub"
 TEMP_PATH = Path(RUNNER_TEMP) / "docker_images_check"
diff --git a/tests/ci/docker_manifests_merge.py b/tests/ci/docker_manifests_merge.py
index 1ae92807c16..f99bfc247cc 100644
--- a/tests/ci/docker_manifests_merge.py
+++ b/tests/ci/docker_manifests_merge.py
@@ -5,24 +5,23 @@ import json
 import logging
 import os
 import subprocess
-
 import sys
 from typing import List, Tuple
 
+# isort: off
 from github import Github
 
-from clickhouse_helper import (
-    ClickHouseHelper,
-    prepare_tests_results_for_clickhouse,
-)
+# isort: on
+
+from clickhouse_helper import ClickHouseHelper, prepare_tests_results_for_clickhouse
 from commit_status_helper import format_description, get_commit, post_commit_status
+from docker_images_helper import docker_login, get_images_oredered_list
 from get_robot_token import get_best_robot_token
 from pr_info import PRInfo
 from report import TestResult
 from s3_helper import S3Helper
 from stopwatch import Stopwatch
 from upload_result_helper import upload_results
-from docker_images_helper import docker_login, get_images_oredered_list
 
 NAME = "Push multi-arch images to Dockerhub"
 
diff --git a/tests/ci/functional_test_check.py b/tests/ci/functional_test_check.py
index b7e6c656b1f..f5dabe0c26d 100644
--- a/tests/ci/functional_test_check.py
+++ b/tests/ci/functional_test_check.py
@@ -1,18 +1,21 @@
 #!/usr/bin/env python3
 
 import argparse
+import atexit
 import csv
 import logging
 import os
 import re
 import subprocess
 import sys
-import atexit
 from pathlib import Path
 from typing import List, Tuple
 
+# isort: off
 from github import Github
 
+# isort: on
+
 from build_download_helper import download_all_deb_packages
 from clickhouse_helper import (
     CiLogsCredentials,
@@ -26,9 +29,9 @@ from commit_status_helper import (
     post_commit_status_to_file,
     update_mergeable_check,
 )
-from docker_images_helper import DockerImage, pull_image, get_docker_image
+from docker_images_helper import DockerImage, get_docker_image, pull_image
 from download_release_packages import download_last_release
-from env_helper import REPORT_PATH, TEMP_PATH, REPO_COPY
+from env_helper import REPO_COPY, REPORT_PATH, TEMP_PATH
 from get_robot_token import get_best_robot_token
 from pr_info import FORCE_TESTS_LABEL, PRInfo
 from report import TestResults, read_test_results
diff --git a/tests/ci/report.py b/tests/ci/report.py
index b478f737963..a1213300151 100644
--- a/tests/ci/report.py
+++ b/tests/ci/report.py
@@ -1,6 +1,12 @@
 # -*- coding: utf-8 -*-
+import csv
+import datetime
+import json
+import logging
+import os
 from ast import literal_eval
 from dataclasses import asdict, dataclass
+from html import escape
 from pathlib import Path
 from typing import (
     Dict,
@@ -13,18 +19,11 @@ from typing import (
     Tuple,
     Union,
 )
-from html import escape
-import csv
-import datetime
-import json
-import logging
-import os
 
 from build_download_helper import get_gh_api
-from ci_config import BuildConfig, CI_CONFIG
+from ci_config import CI_CONFIG, BuildConfig
 from env_helper import REPORT_PATH, TEMP_PATH
 
-
 logger = logging.getLogger(__name__)
 
 ERROR: Final = "error"
diff --git a/tests/ci/sqllogic_test.py b/tests/ci/sqllogic_test.py
index a7b3e3cf69e..14018eaa7a7 100755
--- a/tests/ci/sqllogic_test.py
+++ b/tests/ci/sqllogic_test.py
@@ -10,22 +10,21 @@ from typing import Tuple
 
 from build_download_helper import download_all_deb_packages
 from commit_status_helper import override_status
-from docker_images_helper import DockerImage, pull_image, get_docker_image
-from env_helper import REPORT_PATH, TEMP_PATH, REPO_COPY
+from docker_images_helper import DockerImage, get_docker_image, pull_image
+from env_helper import REPO_COPY, REPORT_PATH, TEMP_PATH
 from report import (
-    OK,
-    FAIL,
     ERROR,
+    FAIL,
+    OK,
     SUCCESS,
     JobReport,
-    TestResults,
     TestResult,
+    TestResults,
     read_test_results,
 )
 from stopwatch import Stopwatch
 from tee_popen import TeePopen
 
-
 NO_CHANGES_MSG = "Nothing to run"
 IMAGE_NAME = "clickhouse/sqllogic-test"
 

From 5b9bf80f0960f79c59dd0f0ec07a1999090d9db3 Mon Sep 17 00:00:00 2001
From: "Mikhail f. Shiryaev" <felixoid@clickhouse.com>
Date: Fri, 2 Feb 2024 15:35:13 +0100
Subject: [PATCH 58/73] Fix usage of status types to prevent API exceptions

---
 tests/ci/clickbench.py             |  6 +++---
 tests/ci/commit_status_helper.py   |  8 +++++---
 tests/ci/docker_images_check.py    |  6 +++---
 tests/ci/docker_manifests_merge.py |  6 +++---
 tests/ci/functional_test_check.py  |  6 +++---
 tests/ci/integration_test_check.py |  6 +++---
 tests/ci/report.py                 | 26 ++++++++++++++------------
 tests/ci/sqllogic_test.py          |  7 ++++---
 8 files changed, 38 insertions(+), 33 deletions(-)

diff --git a/tests/ci/clickbench.py b/tests/ci/clickbench.py
index f81a946e994..61d15373a78 100644
--- a/tests/ci/clickbench.py
+++ b/tests/ci/clickbench.py
@@ -15,7 +15,7 @@ from commit_status_helper import override_status
 from docker_images_helper import DockerImage, get_docker_image, pull_image
 from env_helper import REPORT_PATH, TEMP_PATH
 from pr_info import FORCE_TESTS_LABEL, PRInfo
-from report import JobReport, TestResults
+from report import JobReport, StatusType, TestResults
 from stopwatch import Stopwatch
 from tee_popen import TeePopen
 
@@ -48,7 +48,7 @@ def get_run_command(
 def process_results(
     result_directory: Path,
     server_log_path: Path,
-) -> Tuple[str, str, TestResults, List[Path]]:
+) -> Tuple[StatusType, str, TestResults, List[Path]]:
     test_results = []  # type: TestResults
     additional_files = []  # type: List[Path]
     # Just upload all files from result_directory.
@@ -90,7 +90,7 @@ def process_results(
             additional_files,
         )
 
-    return state, description, test_results, additional_files
+    return state, description, test_results, additional_files  # type: ignore
 
 
 def parse_args():
diff --git a/tests/ci/commit_status_helper.py b/tests/ci/commit_status_helper.py
index c8dc6f713b3..9f1a33d7db6 100644
--- a/tests/ci/commit_status_helper.py
+++ b/tests/ci/commit_status_helper.py
@@ -67,7 +67,9 @@ class RerunHelper:
         return None
 
 
-def override_status(status: str, check_name: str, invert: bool = False) -> str:
+def override_status(
+    status: StatusType, check_name: str, invert: bool = False
+) -> StatusType:
     test_config = CI_CONFIG.test_configs.get(check_name)
     if test_config and test_config.force_tests:
         return SUCCESS
@@ -96,7 +98,7 @@ def get_commit(gh: Github, commit_sha: str, retry_count: int = RETRY) -> Commit:
 
 def post_commit_status(
     commit: Commit,
-    state: str,
+    state: StatusType,
     report_url: Optional[str] = None,
     description: Optional[str] = None,
     check_name: Optional[str] = None,
@@ -291,7 +293,7 @@ def generate_status_comment(pr_info: PRInfo, statuses: CommitStatuses) -> str:
     return "".join(result)
 
 
-def get_worst_state(statuses: CommitStatuses) -> str:
+def get_worst_state(statuses: CommitStatuses) -> StatusType:
     return get_worst_status(status.state for status in statuses)
 
 
diff --git a/tests/ci/docker_images_check.py b/tests/ci/docker_images_check.py
index e88deae9a38..4fdaabac633 100644
--- a/tests/ci/docker_images_check.py
+++ b/tests/ci/docker_images_check.py
@@ -19,7 +19,7 @@ from docker_images_helper import DockerImageData, docker_login, get_images_orede
 from env_helper import GITHUB_RUN_URL, RUNNER_TEMP
 from get_robot_token import get_best_robot_token
 from pr_info import PRInfo
-from report import TestResult, TestResults
+from report import FAILURE, SUCCESS, StatusType, TestResult, TestResults
 from s3_helper import S3Helper
 from stopwatch import Stopwatch
 from tee_popen import TeePopen
@@ -192,7 +192,7 @@ def main():
     #     additional_cache.append(str(pr_info.merged_pr))
 
     ok_cnt = 0
-    status = "success"
+    status = SUCCESS  # type: StatusType
     image_tags = (
         json.loads(args.image_tags)
         if not os.path.isfile(args.image_tags)
@@ -236,7 +236,7 @@ def main():
         if all(x.status == "OK" for x in res):
             ok_cnt += 1
         else:
-            status = "failure"
+            status = FAILURE
             break  # No need to continue with next images
 
     description = format_description(
diff --git a/tests/ci/docker_manifests_merge.py b/tests/ci/docker_manifests_merge.py
index f99bfc247cc..9d25ad65ce6 100644
--- a/tests/ci/docker_manifests_merge.py
+++ b/tests/ci/docker_manifests_merge.py
@@ -18,7 +18,7 @@ from commit_status_helper import format_description, get_commit, post_commit_sta
 from docker_images_helper import docker_login, get_images_oredered_list
 from get_robot_token import get_best_robot_token
 from pr_info import PRInfo
-from report import TestResult
+from report import FAILURE, SUCCESS, StatusType, TestResult
 from s3_helper import S3Helper
 from stopwatch import Stopwatch
 from upload_result_helper import upload_results
@@ -148,7 +148,7 @@ def main():
         else json.load(open(args.missing_images))
     )
     test_results = []
-    status = "success"
+    status = SUCCESS  # type: StatusType
 
     ok_cnt, fail_cnt = 0, 0
     images = get_images_oredered_list()
@@ -176,7 +176,7 @@ def main():
         test_results.append(TestResult(manifest, test_result))
 
         if test_result != "OK":
-            status = "failure"
+            status = FAILURE
             fail_cnt += 1
         else:
             ok_cnt += 1
diff --git a/tests/ci/functional_test_check.py b/tests/ci/functional_test_check.py
index f5dabe0c26d..a05e2501504 100644
--- a/tests/ci/functional_test_check.py
+++ b/tests/ci/functional_test_check.py
@@ -34,7 +34,7 @@ from download_release_packages import download_last_release
 from env_helper import REPO_COPY, REPORT_PATH, TEMP_PATH
 from get_robot_token import get_best_robot_token
 from pr_info import FORCE_TESTS_LABEL, PRInfo
-from report import TestResults, read_test_results
+from report import StatusType, TestResults, read_test_results
 from s3_helper import S3Helper
 from stopwatch import Stopwatch
 from tee_popen import TeePopen
@@ -155,7 +155,7 @@ def get_tests_to_run(pr_info: PRInfo) -> List[str]:
 def process_results(
     result_directory: Path,
     server_log_path: Path,
-) -> Tuple[str, str, TestResults, List[Path]]:
+) -> Tuple[StatusType, str, TestResults, List[Path]]:
     test_results = []  # type: TestResults
     additional_files = []
     # Just upload all files from result_directory.
@@ -200,7 +200,7 @@ def process_results(
             additional_files,
         )
 
-    return state, description, test_results, additional_files
+    return state, description, test_results, additional_files  # type: ignore
 
 
 def parse_args():
diff --git a/tests/ci/integration_test_check.py b/tests/ci/integration_test_check.py
index fa2529d1b89..bab89ea0941 100644
--- a/tests/ci/integration_test_check.py
+++ b/tests/ci/integration_test_check.py
@@ -25,7 +25,7 @@ from get_robot_token import get_best_robot_token
 from github_helper import GitHub
 from integration_test_images import IMAGES
 from pr_info import PRInfo
-from report import ERROR, TestResult, TestResults, read_test_results
+from report import ERROR, StatusType, TestResult, TestResults, read_test_results
 from s3_helper import S3Helper
 from stopwatch import Stopwatch
 from tee_popen import TeePopen
@@ -84,7 +84,7 @@ def get_env_for_runner(
 
 def process_results(
     result_directory: Path,
-) -> Tuple[str, str, TestResults, List[Path]]:
+) -> Tuple[StatusType, str, TestResults, List[Path]]:
     test_results = []  # type: TestResults
     additional_files = []
     # Just upload all files from result_directory.
@@ -118,7 +118,7 @@ def process_results(
             additional_files,
         )
 
-    return state, description, test_results, additional_files
+    return state, description, test_results, additional_files  # type: ignore
 
 
 def parse_args():
diff --git a/tests/ci/report.py b/tests/ci/report.py
index a1213300151..ce20c7293f9 100644
--- a/tests/ci/report.py
+++ b/tests/ci/report.py
@@ -35,26 +35,28 @@ OK: Final = "OK"
 FAIL: Final = "FAIL"
 
 StatusType = Literal["error", "failure", "pending", "success"]
+STATUSES = [ERROR, FAILURE, PENDING, SUCCESS]  # type: List[StatusType]
+
+
 # The order of statuses from the worst to the best
-_STATES = {ERROR: 0, FAILURE: 1, PENDING: 2, SUCCESS: 3}
+def _state_rank(status: str) -> int:
+    "return the index of status or index of SUCCESS in case of wrong status"
+    try:
+        return STATUSES.index(status)  # type: ignore
+    except ValueError:
+        return 3
 
 
-def get_worst_status(statuses: Iterable[str]) -> str:
-    worst_status = None
+def get_worst_status(statuses: Iterable[str]) -> StatusType:
+    worst_status = SUCCESS  # type: StatusType
     for status in statuses:
-        if _STATES.get(status) is None:
-            continue
-        if worst_status is None:
-            worst_status = status
-            continue
-        if _STATES.get(status) < _STATES.get(worst_status):
-            worst_status = status
+        ind = _state_rank(status)
+        if ind < _state_rank(worst_status):
+            worst_status = STATUSES[ind]
 
         if worst_status == ERROR:
             break
 
-    if worst_status is None:
-        return ""
     return worst_status
 
 
diff --git a/tests/ci/sqllogic_test.py b/tests/ci/sqllogic_test.py
index 14018eaa7a7..344eaea8277 100755
--- a/tests/ci/sqllogic_test.py
+++ b/tests/ci/sqllogic_test.py
@@ -18,6 +18,7 @@ from report import (
     OK,
     SUCCESS,
     JobReport,
+    StatusType,
     TestResult,
     TestResults,
     read_test_results,
@@ -46,7 +47,7 @@ def get_run_command(
     )
 
 
-def read_check_status(result_folder: Path) -> Tuple[str, str]:
+def read_check_status(result_folder: Path) -> Tuple[StatusType, str]:
     status_path = result_folder / "check_status.tsv"
     if not status_path.exists():
         return ERROR, "Not found check_status.tsv"
@@ -59,9 +60,9 @@ def read_check_status(result_folder: Path) -> Tuple[str, str]:
         if len(row) != 2:
             return ERROR, "Invalid check_status.tsv"
         if row[0] != SUCCESS:
-            return row[0], row[1]
+            return row[0], row[1]  # type: ignore
 
-    return status_rows[-1][0], status_rows[-1][1]
+    return status_rows[-1][0], status_rows[-1][1]  # type: ignore
 
 
 def parse_args() -> argparse.Namespace:

From c4846b661373dd59609583f2f5de79313d65a39d Mon Sep 17 00:00:00 2001
From: Maksim Kita <kitaetoya@gmail.com>
Date: Tue, 6 Feb 2024 15:55:16 +0300
Subject: [PATCH 59/73] MergeTree FINAL optimization diagnostics and settings

---
 src/Core/Settings.h                           |  2 +
 src/Core/SettingsChangesHistory.h             |  4 +-
 src/Processors/QueryPlan/PartsSplitter.cpp    | 38 +++++++++++++++----
 src/Processors/QueryPlan/PartsSplitter.h      |  3 +-
 .../QueryPlan/ReadFromMergeTree.cpp           |  6 ++-
 5 files changed, 42 insertions(+), 11 deletions(-)

diff --git a/src/Core/Settings.h b/src/Core/Settings.h
index dc863576a85..67bd721d3d1 100644
--- a/src/Core/Settings.h
+++ b/src/Core/Settings.h
@@ -215,6 +215,8 @@ class IColumn;
     M(UInt64, merge_tree_max_rows_to_use_cache, (128 * 8192), "The maximum number of rows per request, to use the cache of uncompressed data. If the request is large, the cache is not used. (For large queries not to flush out the cache.)", 0) \
     M(UInt64, merge_tree_max_bytes_to_use_cache, (192 * 10 * 1024 * 1024), "The maximum number of bytes per request, to use the cache of uncompressed data. If the request is large, the cache is not used. (For large queries not to flush out the cache.)", 0) \
     M(Bool, do_not_merge_across_partitions_select_final, false, "Merge parts only in one partition in select final", 0) \
+    M(Bool, split_parts_ranges_into_intersecting_and_non_intersecting_final, true, "Split parts ranges into intersecting and non intersecting during FINAL optimization", 0) \
+    M(Bool, split_intersecting_parts_ranges_into_layers_final, true, "Split intersecting parts ranges into layers during FINAL optimization", 0) \
     M(Bool, allow_experimental_inverted_index, false, "If it is set to true, allow to use experimental inverted index.", 0) \
     \
     M(UInt64, mysql_max_rows_to_insert, 65536, "The maximum number of rows in MySQL batch insertion of the MySQL storage engine", 0) \
diff --git a/src/Core/SettingsChangesHistory.h b/src/Core/SettingsChangesHistory.h
index db3a76e29cd..a2612cab850 100644
--- a/src/Core/SettingsChangesHistory.h
+++ b/src/Core/SettingsChangesHistory.h
@@ -115,7 +115,9 @@ static std::map<ClickHouseVersion, SettingsChangesHistory::SettingsChanges> sett
     {"23.12", {{"allow_suspicious_ttl_expressions", true, false, "It is a new setting, and in previous versions the behavior was equivalent to allowing."},
               {"input_format_parquet_allow_missing_columns", false, true, "Allow missing columns in Parquet files by default"},
               {"input_format_orc_allow_missing_columns", false, true, "Allow missing columns in ORC files by default"},
-              {"input_format_arrow_allow_missing_columns", false, true, "Allow missing columns in Arrow files by default"}}},
+              {"input_format_arrow_allow_missing_columns", false, true, "Allow missing columns in Arrow files by default"},
+              {"split_parts_ranges_into_intersecting_and_non_intersecting_final", false, true, "Allow to split parts ranges into intersecting and non intersecting during FINAL optimization"},
+              {"split_intersecting_parts_ranges_into_layers_final", true, true, "Allow to split intersecting parts ranges into layers during FINAL optimization"}}},
     {"23.9", {{"optimize_group_by_constant_keys", false, true, "Optimize group by constant keys by default"},
               {"input_format_json_try_infer_named_tuples_from_objects", false, true, "Try to infer named Tuples from JSON objects by default"},
               {"input_format_json_read_numbers_as_strings", false, true, "Allow to read numbers as strings in JSON formats by default"},
diff --git a/src/Processors/QueryPlan/PartsSplitter.cpp b/src/Processors/QueryPlan/PartsSplitter.cpp
index 7c66c0cc8df..363fdca22c5 100644
--- a/src/Processors/QueryPlan/PartsSplitter.cpp
+++ b/src/Processors/QueryPlan/PartsSplitter.cpp
@@ -228,7 +228,7 @@ struct SplitPartsRangesResult
     RangesInDataParts intersecting_parts_ranges;
 };
 
-SplitPartsRangesResult splitPartsRanges(RangesInDataParts ranges_in_data_parts)
+SplitPartsRangesResult splitPartsRanges(RangesInDataParts ranges_in_data_parts, const LoggerPtr & logger)
 {
     /** Split ranges in data parts into intersecting ranges in data parts and non intersecting ranges in data parts.
       *
@@ -483,10 +483,15 @@ SplitPartsRangesResult splitPartsRanges(RangesInDataParts ranges_in_data_parts)
         intersecting_ranges_in_data_parts.end(),
         [](const auto & lhs, const auto & rhs) { return lhs.part_index_in_query < rhs.part_index_in_query; });
 
+    LOG_TEST(logger, "Non intersecting ranges in data parts {}", non_intersecting_ranges_in_data_parts.getDescriptions().describe());
+    LOG_TEST(logger, "Intersecting ranges in data parts {}", intersecting_ranges_in_data_parts.getDescriptions().describe());
+
     return {std::move(non_intersecting_ranges_in_data_parts), std::move(intersecting_ranges_in_data_parts)};
 }
 
-std::pair<std::vector<RangesInDataParts>, std::vector<Values>> splitIntersectingPartsRangesIntoLayers(RangesInDataParts intersecting_ranges_in_data_parts, size_t max_layers)
+std::pair<std::vector<RangesInDataParts>, std::vector<Values>> splitIntersectingPartsRangesIntoLayers(RangesInDataParts intersecting_ranges_in_data_parts,
+    size_t max_layers,
+    const LoggerPtr & logger)
 {
     // We will advance the iterator pointing to the mark with the smallest PK value until
     // there will be not less than rows_per_layer rows in the current layer (roughly speaking).
@@ -591,8 +596,18 @@ std::pair<std::vector<RangesInDataParts>, std::vector<Values>> splitIntersecting
         result_layers.back() = std::move(current_layer_builder.getCurrentRangesInDataParts());
     }
 
-    for (auto & layer : result_layers)
+    size_t result_layers_size = result_layers.size();
+    LOG_TEST(logger, "Split intersecting ranges into {} layers", result_layers_size);
+
+    for (size_t i = 0; i < result_layers_size; ++i)
     {
+        auto & layer = result_layers[i];
+
+        LOG_TEST(logger, "Layer {} {} filter values in ({}, {}])",
+            i,
+            layer.getDescriptions().describe(),
+            i ? ::toString(borders[i - 1]) : "-inf", i < borders.size() ? ::toString(borders[i]) : "+inf");
+
         std::stable_sort(
             layer.begin(),
             layer.end(),
@@ -712,23 +727,32 @@ SplitPartsWithRangesByPrimaryKeyResult splitPartsWithRangesByPrimaryKey(
     size_t max_layers,
     ContextPtr context,
     ReadingInOrderStepGetter && in_order_reading_step_getter,
-    bool force_process_all_ranges)
+    bool split_parts_ranges_into_intersecting_and_non_intersecting_final,
+    bool split_intersecting_parts_ranges_into_layers)
 {
     if (max_layers <= 1)
         throw Exception(ErrorCodes::LOGICAL_ERROR, "max_layer should be greater than 1");
 
+    auto logger = getLogger("PartsSplitter");
+
     SplitPartsWithRangesByPrimaryKeyResult result;
 
     RangesInDataParts intersecting_parts_ranges = std::move(parts);
 
-    if (!force_process_all_ranges)
+    if (split_parts_ranges_into_intersecting_and_non_intersecting_final)
     {
-        SplitPartsRangesResult split_result = splitPartsRanges(intersecting_parts_ranges);
+        SplitPartsRangesResult split_result = splitPartsRanges(intersecting_parts_ranges, logger);
         result.non_intersecting_parts_ranges = std::move(split_result.non_intersecting_parts_ranges);
         intersecting_parts_ranges = std::move(split_result.intersecting_parts_ranges);
     }
 
-    auto && [layers, borders] = splitIntersectingPartsRangesIntoLayers(intersecting_parts_ranges, max_layers);
+    if (!split_intersecting_parts_ranges_into_layers)
+    {
+        result.merging_pipes.emplace_back(in_order_reading_step_getter(intersecting_parts_ranges));
+        return result;
+    }
+
+    auto && [layers, borders] = splitIntersectingPartsRangesIntoLayers(intersecting_parts_ranges, max_layers, logger);
     auto filters = buildFilters(primary_key, borders);
     result.merging_pipes.resize(layers.size());
 
diff --git a/src/Processors/QueryPlan/PartsSplitter.h b/src/Processors/QueryPlan/PartsSplitter.h
index f1ed1cb0b9c..9bceb344589 100644
--- a/src/Processors/QueryPlan/PartsSplitter.h
+++ b/src/Processors/QueryPlan/PartsSplitter.h
@@ -34,5 +34,6 @@ SplitPartsWithRangesByPrimaryKeyResult splitPartsWithRangesByPrimaryKey(
     size_t max_layers,
     ContextPtr context,
     ReadingInOrderStepGetter && in_order_reading_step_getter,
-    bool force_process_all_ranges);
+    bool split_parts_ranges_into_intersecting_and_non_intersecting,
+    bool split_intersecting_parts_ranges_into_layers);
 }
diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp
index 5ed56f59fc1..8a04caede80 100644
--- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp
+++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp
@@ -1175,7 +1175,8 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsFinal(
 
                 /// Parts of non-zero level still may contain duplicate PK values to merge on FINAL if there's is_deleted column,
                 /// so we have to process all ranges. It would be more optimal to remove this flag and add an extra filtering step.
-                bool force_process_all_ranges = !data.merging_params.is_deleted_column.empty();
+                bool split_parts_ranges_into_intersecting_and_non_intersecting_final = settings.split_parts_ranges_into_intersecting_and_non_intersecting_final &&
+                    data.merging_params.is_deleted_column.empty();
 
                 SplitPartsWithRangesByPrimaryKeyResult split_ranges_result = splitPartsWithRangesByPrimaryKey(
                     metadata_for_reading->getPrimaryKey(),
@@ -1184,7 +1185,8 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsFinal(
                     num_streams,
                     context,
                     std::move(in_order_reading_step_getter),
-                    force_process_all_ranges);
+                    split_parts_ranges_into_intersecting_and_non_intersecting_final,
+                    settings.split_intersecting_parts_ranges_into_layers_final);
 
                 for (auto && non_intersecting_parts_range : split_ranges_result.non_intersecting_parts_ranges)
                     non_intersecting_parts_by_primary_key.push_back(std::move(non_intersecting_parts_range));

From ca4f46ab418afab56deea3ae51c2378e42489db7 Mon Sep 17 00:00:00 2001
From: Alexander Gololobov <davenger@clickhouse.com>
Date: Tue, 6 Feb 2024 14:27:15 +0100
Subject: [PATCH 60/73] Check block size in bytes instead of rows

---
 src/Storages/System/StorageSystemZooKeeper.cpp | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/src/Storages/System/StorageSystemZooKeeper.cpp b/src/Storages/System/StorageSystemZooKeeper.cpp
index 6d2166ff5f9..a4920d49b6d 100644
--- a/src/Storages/System/StorageSystemZooKeeper.cpp
+++ b/src/Storages/System/StorageSystemZooKeeper.cpp
@@ -511,11 +511,21 @@ Chunk SystemZooKeeperSource::generate()
     };
     std::vector<ListTask> list_tasks;
     std::unordered_set<String> added;
-    while (!paths.empty() && (max_block_size == 0 || row_count < max_block_size))
+    while (!paths.empty())
     {
         if (query_status)
             query_status->checkTimeLimit();
 
+        /// Check if the block is big enough already
+        if (max_block_size > 0 && row_count > 0)
+        {
+            size_t total_size = 0;
+            for (const auto & column : res_columns)
+                total_size += column->byteSize();
+            if (total_size > max_block_size)
+                break;
+        }
+
         list_tasks.clear();
         std::vector<String> paths_to_list;
         while (!paths.empty() && static_cast<Int64>(list_tasks.size()) < max_inflight_requests)

From 13948cbb201cb76a335bcccfe86cca49313dac05 Mon Sep 17 00:00:00 2001
From: "Mikhail f. Shiryaev" <felixoid@clickhouse.com>
Date: Tue, 6 Feb 2024 13:39:34 +0100
Subject: [PATCH 61/73] Use constants from report instead of string literals

---
 tests/ci/artifacts_helper.py             |  9 +++---
 tests/ci/ast_fuzzer_check.py             |  4 +--
 tests/ci/bugfix_validate_check.py        | 13 +++++----
 tests/ci/ci.py                           | 12 ++++----
 tests/ci/clickbench.py                   | 10 +++----
 tests/ci/commit_status_helper.py         | 12 ++++----
 tests/ci/compatibility_check.py          | 24 ++++++++--------
 tests/ci/docker_images_check.py          |  2 +-
 tests/ci/docker_manifests_merge.py       |  2 +-
 tests/ci/docker_server.py                | 15 +++++-----
 tests/ci/docs_check.py                   | 16 +++++------
 tests/ci/fast_test_check.py              | 32 ++++++++++++++--------
 tests/ci/finish_check.py                 |  9 ++++--
 tests/ci/functional_test_check.py        | 16 +++++------
 tests/ci/integration_test_check.py       | 19 +++++++++----
 tests/ci/jepsen_check.py                 | 13 ++++-----
 tests/ci/mark_release_ready.py           |  5 ++--
 tests/ci/merge_pr.py                     |  8 ++++--
 tests/ci/performance_comparison_check.py | 35 +++++++++++++-----------
 tests/ci/release.py                      |  3 +-
 tests/ci/run_check.py                    |  4 +--
 tests/ci/sqlancer_check.py               | 13 ++++-----
 tests/ci/sqllogic_test.py                |  2 +-
 tests/ci/sqltest.py                      | 15 ++++------
 tests/ci/stress_check.py                 | 11 ++++----
 tests/ci/style_check.py                  | 10 +++----
 tests/ci/unit_tests_check.py             | 10 +++----
 27 files changed, 171 insertions(+), 153 deletions(-)

diff --git a/tests/ci/artifacts_helper.py b/tests/ci/artifacts_helper.py
index a9f3385585b..5feca927a96 100644
--- a/tests/ci/artifacts_helper.py
+++ b/tests/ci/artifacts_helper.py
@@ -10,14 +10,17 @@ from pathlib import Path
 from shutil import copy2
 from typing import List, Optional, Union
 
+# isort: off
 from github.Commit import Commit
 
+# isort: on
+
 from build_download_helper import download_build_with_progress
 from commit_status_helper import post_commit_status
 from compress_files import SUFFIX, compress_fast, decompress_fast
 from env_helper import CI, RUNNER_TEMP, S3_BUILDS_BUCKET
 from git_helper import SHA_REGEXP
-from report import HEAD_HTML_TEMPLATE, FOOTER_HTML_TEMPLATE
+from report import FOOTER_HTML_TEMPLATE, HEAD_HTML_TEMPLATE, SUCCESS
 from s3_helper import S3Helper
 
 ARTIFACTS_PATH = Path(RUNNER_TEMP) / "artifacts"
@@ -128,9 +131,7 @@ class ArtifactsHelper:
 
     @staticmethod
     def post_commit_status(commit: Commit, url: str) -> None:
-        post_commit_status(
-            commit, "success", url, "Artifacts for workflow", "Artifacts"
-        )
+        post_commit_status(commit, SUCCESS, url, "Artifacts for workflow", "Artifacts")
 
     def _regenerate_index(self) -> None:
         if CI:
diff --git a/tests/ci/ast_fuzzer_check.py b/tests/ci/ast_fuzzer_check.py
index 5a0eca51570..0a69d8aab49 100644
--- a/tests/ci/ast_fuzzer_check.py
+++ b/tests/ci/ast_fuzzer_check.py
@@ -7,9 +7,7 @@ import sys
 from pathlib import Path
 
 from build_download_helper import get_build_name_for_check, read_build_urls
-from clickhouse_helper import (
-    CiLogsCredentials,
-)
+from clickhouse_helper import CiLogsCredentials
 from docker_images_helper import DockerImage, get_docker_image, pull_image
 from env_helper import REPORT_PATH, TEMP_PATH
 from pr_info import PRInfo
diff --git a/tests/ci/bugfix_validate_check.py b/tests/ci/bugfix_validate_check.py
index 107c02a0f56..7fda81f11b2 100644
--- a/tests/ci/bugfix_validate_check.py
+++ b/tests/ci/bugfix_validate_check.py
@@ -1,17 +1,20 @@
 #!/usr/bin/env python3
 
-from pathlib import Path
-from typing import List, Tuple, Optional
 import argparse
 import csv
 import logging
+from pathlib import Path
+from typing import List, Optional, Tuple
 
+# isort: off
 from github import Github
 
+# isort: on
+
 from commit_status_helper import get_commit, post_commit_status
 from get_robot_token import get_best_robot_token
 from pr_info import PRInfo
-from report import TestResults, TestResult
+from report import ERROR, SUCCESS, TestResult, TestResults
 from s3_helper import S3Helper
 from upload_result_helper import upload_results
 
@@ -49,7 +52,7 @@ def process_result(file_path: Path) -> Tuple[bool, TestResults, Optional[str]]:
         )
         return False, [TestResult(f"{prefix}: {description}", status)], "Check failed"
 
-    is_ok = state == "success"
+    is_ok = state == SUCCESS
     if is_ok and report_url == "null":
         return is_ok, test_results, None
 
@@ -111,7 +114,7 @@ def main():
     commit = get_commit(gh, pr_info.sha)
     post_commit_status(
         commit,
-        "success" if is_ok else "error",
+        SUCCESS if is_ok else ERROR,
         report_url,
         description,
         check_name_with_group,
diff --git a/tests/ci/ci.py b/tests/ci/ci.py
index d2d05b1ed2d..622b7bb005a 100644
--- a/tests/ci/ci.py
+++ b/tests/ci/ci.py
@@ -1,21 +1,21 @@
 import argparse
 import concurrent.futures
-from dataclasses import asdict, dataclass
-from enum import Enum
 import json
 import logging
 import os
 import re
 import subprocess
 import sys
-from pathlib import Path
 import time
+from dataclasses import asdict, dataclass
+from enum import Enum
+from pathlib import Path
 from typing import Any, Dict, List, Optional, Sequence, Union
 
 import docker_images_helper
 import upload_result_helper
 from build_check import get_release_or_pr
-from ci_config import CI_CONFIG, Build, Labels, JobNames
+from ci_config import CI_CONFIG, Build, JobNames, Labels
 from ci_utils import GHActions, is_hex
 from clickhouse_helper import (
     CiLogsCredentials,
@@ -859,7 +859,7 @@ def _mark_success_action(
         # there is no status for build jobs
         # create dummy success to mark it as done
         # FIXME: consider creating commit status for build jobs too, to treat everything the same way
-        CommitStatusData("success", "dummy description", "dummy_url").dump_status()
+        CommitStatusData(SUCCESS, "dummy description", "dummy_url").dump_status()
 
     job_status = None
     if CommitStatusData.exist():
@@ -1142,7 +1142,7 @@ def _update_gh_statuses_action(indata: Dict, s3: S3Helper) -> None:
         if not job_status:
             return
         print(f"Going to re-create GH status for job [{job}] sha [{pr_info.sha}]")
-        assert job_status.status == "success", "BUG!"
+        assert job_status.status == SUCCESS, "BUG!"
         commit.create_status(
             state=job_status.status,
             target_url=job_status.report_url,
diff --git a/tests/ci/clickbench.py b/tests/ci/clickbench.py
index 61d15373a78..f8707cbcff7 100644
--- a/tests/ci/clickbench.py
+++ b/tests/ci/clickbench.py
@@ -15,7 +15,7 @@ from commit_status_helper import override_status
 from docker_images_helper import DockerImage, get_docker_image, pull_image
 from env_helper import REPORT_PATH, TEMP_PATH
 from pr_info import FORCE_TESTS_LABEL, PRInfo
-from report import JobReport, StatusType, TestResults
+from report import ERROR, SUCCESS, JobReport, StatusType, TestResults
 from stopwatch import Stopwatch
 from tee_popen import TeePopen
 
@@ -70,7 +70,7 @@ def process_results(
 
     if len(status) != 1 or len(status[0]) != 2:
         logging.info("Files in result folder %s", os.listdir(result_directory))
-        return "error", "Invalid check_status.tsv", test_results, additional_files
+        return ERROR, "Invalid check_status.tsv", test_results, additional_files
     state, description = status[0][0], status[0][1]
 
     try:
@@ -80,11 +80,11 @@ def process_results(
             logging.info("Found %s", results_path.name)
         else:
             logging.info("Files in result folder %s", os.listdir(result_directory))
-            return "error", "Not found test_results.tsv", test_results, additional_files
+            return ERROR, "Not found test_results.tsv", test_results, additional_files
 
     except Exception as e:
         return (
-            "error",
+            ERROR,
             f"Cannot parse test_results.tsv ({e})",
             test_results,
             additional_files,
@@ -175,7 +175,7 @@ def main():
         additional_files=[run_log_path] + additional_logs,
     ).dump()
 
-    if state != "success":
+    if state != SUCCESS:
         if FORCE_TESTS_LABEL in pr_info.labels:
             print(f"'{FORCE_TESTS_LABEL}' enabled, will report success")
         else:
diff --git a/tests/ci/commit_status_helper.py b/tests/ci/commit_status_helper.py
index 9f1a33d7db6..5dd2a33adaf 100644
--- a/tests/ci/commit_status_helper.py
+++ b/tests/ci/commit_status_helper.py
@@ -378,12 +378,12 @@ class CommitStatusData:
 def get_commit_filtered_statuses(commit: Commit) -> CommitStatuses:
     """
     Squash statuses to latest state
-    1. context="first", state="success", update_time=1
-    2. context="second", state="success", update_time=2
-    3. context="first", stat="failure", update_time=3
+    1. context="first", state=SUCCESS, update_time=1
+    2. context="second", state=SUCCESS, update_time=2
+    3. context="first", stat=FAILURE, update_time=3
     =========>
-    1. context="second", state="success"
-    2. context="first", stat="failure"
+    1. context="second", state=SUCCESS
+    2. context="first", stat=FAILURE
     """
     filtered = {}
     for status in sorted(commit.get_statuses(), key=lambda x: x.updated_at):
@@ -435,7 +435,7 @@ def format_description(description: str) -> str:
 def set_mergeable_check(
     commit: Commit,
     description: str = "",
-    state: StatusType = "success",
+    state: StatusType = SUCCESS,
 ) -> None:
     commit.create_status(
         context=MERGEABLE_NAME,
diff --git a/tests/ci/compatibility_check.py b/tests/ci/compatibility_check.py
index a0c6294d8fd..a2e6c94cf48 100644
--- a/tests/ci/compatibility_check.py
+++ b/tests/ci/compatibility_check.py
@@ -1,17 +1,17 @@
 #!/usr/bin/env python3
 
-from distutils.version import StrictVersion
-from pathlib import Path
-from typing import List, Tuple
 import argparse
 import logging
 import subprocess
 import sys
+from distutils.version import StrictVersion
+from pathlib import Path
+from typing import List, Tuple
 
 from build_download_helper import download_builds_filter
 from docker_images_helper import DockerImage, get_docker_image, pull_image
-from env_helper import TEMP_PATH, REPORT_PATH
-from report import JobReport, TestResults, TestResult
+from env_helper import REPORT_PATH, TEMP_PATH
+from report import FAILURE, SUCCESS, JobReport, TestResult, TestResults
 from stopwatch import Stopwatch
 
 IMAGE_UBUNTU = "clickhouse/test-old-ubuntu"
@@ -55,19 +55,19 @@ def process_result(
     glibc_log_path = result_directory / "glibc.log"
     test_results = process_glibc_check(glibc_log_path, max_glibc_version)
 
-    status = "success"
+    status = SUCCESS
     description = "Compatibility check passed"
 
     if check_glibc:
         if len(test_results) > 1 or test_results[0].status != "OK":
-            status = "failure"
+            status = FAILURE
             description = "glibc check failed"
 
-    if status == "success" and check_distributions:
+    if status == SUCCESS and check_distributions:
         for operating_system in ("ubuntu:12.04", "centos:5"):
             test_result = process_os_check(result_directory / operating_system)
             if test_result.status != "OK":
-                status = "failure"
+                status = FAILURE
                 description = f"Old {operating_system} failed"
                 test_results += [test_result]
                 break
@@ -178,14 +178,14 @@ def main():
         )
         run_commands.extend(check_distributions_commands)
 
-    state = "success"
+    state = SUCCESS
     for run_command in run_commands:
         try:
             logging.info("Running command %s", run_command)
             subprocess.check_call(run_command, shell=True)
         except subprocess.CalledProcessError as ex:
             logging.info("Exception calling command %s", ex)
-            state = "failure"
+            state = FAILURE
 
     subprocess.check_call(f"sudo chown -R ubuntu:ubuntu {temp_path}", shell=True)
 
@@ -215,7 +215,7 @@ def main():
         additional_files=additional_logs,
     ).dump()
 
-    if state == "failure":
+    if state == FAILURE:
         sys.exit(1)
 
 
diff --git a/tests/ci/docker_images_check.py b/tests/ci/docker_images_check.py
index 4fdaabac633..af0416d83dc 100644
--- a/tests/ci/docker_images_check.py
+++ b/tests/ci/docker_images_check.py
@@ -271,7 +271,7 @@ def main():
     ch_helper = ClickHouseHelper()
     ch_helper.insert_events_into(db="default", table="checks", events=prepared_events)
 
-    if status == "failure":
+    if status == FAILURE:
         sys.exit(1)
 
 
diff --git a/tests/ci/docker_manifests_merge.py b/tests/ci/docker_manifests_merge.py
index 9d25ad65ce6..fc00969d5d6 100644
--- a/tests/ci/docker_manifests_merge.py
+++ b/tests/ci/docker_manifests_merge.py
@@ -212,7 +212,7 @@ def main():
     )
     ch_helper = ClickHouseHelper()
     ch_helper.insert_events_into(db="default", table="checks", events=prepared_events)
-    if status == "failure":
+    if status == FAILURE:
         sys.exit(1)
 
 
diff --git a/tests/ci/docker_server.py b/tests/ci/docker_server.py
index b9e5c13ec42..7f53034fd0f 100644
--- a/tests/ci/docker_server.py
+++ b/tests/ci/docker_server.py
@@ -6,25 +6,26 @@ import json
 import logging
 import sys
 import time
+from os import makedirs
+from os import path as p
 from pathlib import Path
-from os import path as p, makedirs
 from typing import Dict, List
 
 from build_check import get_release_or_pr
+from build_download_helper import read_build_urls
 from docker_images_helper import DockerImageData, docker_login
 from env_helper import (
     GITHUB_RUN_URL,
     REPORT_PATH,
-    TEMP_PATH,
     S3_BUILDS_BUCKET,
     S3_DOWNLOAD,
+    TEMP_PATH,
 )
 from git_helper import Git
 from pr_info import PRInfo
-from report import JobReport, TestResults, TestResult
+from report import FAILURE, SUCCESS, JobReport, TestResult, TestResults
 from stopwatch import Stopwatch
 from tee_popen import TeePopen
-from build_download_helper import read_build_urls
 from version_helper import (
     ClickHouseVersion,
     get_tagged_versions,
@@ -378,7 +379,7 @@ def main():
         docker_login()
 
     logging.info("Following tags will be created: %s", ", ".join(tags))
-    status = "success"
+    status = SUCCESS
     test_results = []  # type: TestResults
     for os in args.os:
         for tag in tags:
@@ -388,7 +389,7 @@ def main():
                 )
             )
             if test_results[-1].status != "OK":
-                status = "failure"
+                status = FAILURE
     pr_info = pr_info or PRInfo()
 
     description = f"Processed tags: {', '.join(tags)}"
@@ -401,7 +402,7 @@ def main():
         additional_files=[],
     ).dump()
 
-    if status != "success":
+    if status != SUCCESS:
         sys.exit(1)
 
 
diff --git a/tests/ci/docs_check.py b/tests/ci/docs_check.py
index 400d4bc6ad5..6bd4ef49675 100644
--- a/tests/ci/docs_check.py
+++ b/tests/ci/docs_check.py
@@ -8,7 +8,7 @@ from pathlib import Path
 from docker_images_helper import get_docker_image, pull_image
 from env_helper import REPO_COPY, TEMP_PATH
 from pr_info import PRInfo
-from report import JobReport, TestResult, TestResults
+from report import FAILURE, SUCCESS, JobReport, TestResult, TestResults
 from stopwatch import Stopwatch
 from tee_popen import TeePopen
 
@@ -49,7 +49,7 @@ def main():
         JobReport(
             description="No changes in docs",
             test_results=[],
-            status="success",
+            status=SUCCESS,
             start_time=stopwatch.start_time_str,
             duration=stopwatch.duration_seconds,
             additional_files=[],
@@ -79,11 +79,11 @@ def main():
         retcode = process.wait()
         if retcode == 0:
             logging.info("Run successfully")
-            status = "success"
+            status = SUCCESS
             description = "Docs check passed"
         else:
             description = "Docs check failed (non zero exit code)"
-            status = "failure"
+            status = FAILURE
             logging.info("Run failed")
 
     subprocess.check_call(f"sudo chown -R ubuntu:ubuntu {temp_path}", shell=True)
@@ -92,7 +92,7 @@ def main():
     if not any(test_output.iterdir()):
         logging.error("No output files after docs check")
         description = "No output files after docs check"
-        status = "failure"
+        status = FAILURE
     else:
         for p in test_output.iterdir():
             additional_files.append(p)
@@ -101,9 +101,9 @@ def main():
                     if "ERROR" in line:
                         test_results.append(TestResult(line.split(":")[-1], "FAIL"))
         if test_results:
-            status = "failure"
+            status = FAILURE
             description = "Found errors in docs"
-        elif status != "failure":
+        elif status != FAILURE:
             test_results.append(TestResult("No errors found", "OK"))
         else:
             test_results.append(TestResult("Non zero exit code", "FAIL"))
@@ -117,7 +117,7 @@ def main():
         additional_files=additional_files,
     ).dump()
 
-    if status == "failure":
+    if status == FAILURE:
         sys.exit(1)
 
 
diff --git a/tests/ci/fast_test_check.py b/tests/ci/fast_test_check.py
index b558253ca95..e483e9d4ac2 100644
--- a/tests/ci/fast_test_check.py
+++ b/tests/ci/fast_test_check.py
@@ -11,7 +11,15 @@ from typing import Tuple
 from docker_images_helper import DockerImage, get_docker_image, pull_image
 from env_helper import REPO_COPY, S3_BUILDS_BUCKET, TEMP_PATH
 from pr_info import FORCE_TESTS_LABEL, PRInfo
-from report import JobReport, TestResult, TestResults, read_test_results
+from report import (
+    ERROR,
+    FAILURE,
+    SUCCESS,
+    JobReport,
+    TestResult,
+    TestResults,
+    read_test_results,
+)
 from stopwatch import Stopwatch
 from tee_popen import TeePopen
 
@@ -56,16 +64,16 @@ def process_results(result_directory: Path) -> Tuple[str, str, TestResults]:
             status = list(csv.reader(status_file, delimiter="\t"))
     if len(status) != 1 or len(status[0]) != 2:
         logging.info("Files in result folder %s", os.listdir(result_directory))
-        return "error", "Invalid check_status.tsv", test_results
+        return ERROR, "Invalid check_status.tsv", test_results
     state, description = status[0][0], status[0][1]
 
     try:
         results_path = result_directory / "test_results.tsv"
         test_results = read_test_results(results_path)
         if len(test_results) == 0:
-            return "error", "Empty test_results.tsv", test_results
+            return ERROR, "Empty test_results.tsv", test_results
     except Exception as e:
-        return ("error", f"Cannot parse test_results.tsv ({e})", test_results)
+        return (ERROR, f"Cannot parse test_results.tsv ({e})", test_results)
 
     return state, description, test_results
 
@@ -149,25 +157,25 @@ def main():
     test_results = []  # type: TestResults
     if "submodule_log.txt" not in test_output_files:
         description = "Cannot clone repository"
-        state = "failure"
+        state = FAILURE
     elif "cmake_log.txt" not in test_output_files:
         description = "Cannot fetch submodules"
-        state = "failure"
+        state = FAILURE
     elif "build_log.txt" not in test_output_files:
         description = "Cannot finish cmake"
-        state = "failure"
+        state = FAILURE
     elif "install_log.txt" not in test_output_files:
         description = "Cannot build ClickHouse"
-        state = "failure"
+        state = FAILURE
     elif not test_log_exists and not test_result_exists:
         description = "Cannot install or start ClickHouse"
-        state = "failure"
+        state = FAILURE
     else:
         state, description, test_results = process_results(output_path)
 
     if timeout_expired:
         test_results.append(TestResult.create_check_timeout_expired(args.timeout))
-        state = "failure"
+        state = FAILURE
         description = test_results[-1].name
 
     JobReport(
@@ -181,8 +189,8 @@ def main():
     ).dump()
 
     # Refuse other checks to run if fast test failed
-    if state != "success":
-        if state == "error":
+    if state != SUCCESS:
+        if state == ERROR:
             print("The status is 'error', report failure disregard the labels")
             sys.exit(1)
         elif FORCE_TESTS_LABEL in pr_info.labels:
diff --git a/tests/ci/finish_check.py b/tests/ci/finish_check.py
index 6c615817164..e5268947304 100644
--- a/tests/ci/finish_check.py
+++ b/tests/ci/finish_check.py
@@ -1,7 +1,11 @@
 #!/usr/bin/env python3
 import logging
+
+# isort: off
 from github import Github
 
+# isort: on
+
 from commit_status_helper import (
     CI_STATUS_NAME,
     get_commit,
@@ -11,6 +15,7 @@ from commit_status_helper import (
 )
 from get_robot_token import get_best_robot_token
 from pr_info import PRInfo
+from report import PENDING, SUCCESS
 
 
 def main():
@@ -31,10 +36,10 @@ def main():
         return
     # Take the latest status
     status = statuses[-1]
-    if status.state == "pending":
+    if status.state == PENDING:
         post_commit_status(
             commit,
-            "success",
+            SUCCESS,
             status.target_url,
             "All checks finished",
             CI_STATUS_NAME,
diff --git a/tests/ci/functional_test_check.py b/tests/ci/functional_test_check.py
index a05e2501504..e230aa5a679 100644
--- a/tests/ci/functional_test_check.py
+++ b/tests/ci/functional_test_check.py
@@ -34,7 +34,7 @@ from download_release_packages import download_last_release
 from env_helper import REPO_COPY, REPORT_PATH, TEMP_PATH
 from get_robot_token import get_best_robot_token
 from pr_info import FORCE_TESTS_LABEL, PRInfo
-from report import StatusType, TestResults, read_test_results
+from report import ERROR, SUCCESS, StatusType, TestResults, read_test_results
 from s3_helper import S3Helper
 from stopwatch import Stopwatch
 from tee_popen import TeePopen
@@ -177,7 +177,7 @@ def process_results(
 
     if len(status) != 1 or len(status[0]) != 2:
         logging.info("Files in result folder %s", os.listdir(result_directory))
-        return "error", "Invalid check_status.tsv", test_results, additional_files
+        return ERROR, "Invalid check_status.tsv", test_results, additional_files
     state, description = status[0][0], status[0][1]
 
     try:
@@ -187,14 +187,14 @@ def process_results(
             logging.info("Found test_results.tsv")
         else:
             logging.info("Files in result folder %s", os.listdir(result_directory))
-            return "error", "Not found test_results.tsv", test_results, additional_files
+            return ERROR, "Not found test_results.tsv", test_results, additional_files
 
         test_results = read_test_results(results_path)
         if len(test_results) == 0:
-            return "error", "Empty test_results.tsv", test_results, additional_files
+            return ERROR, "Empty test_results.tsv", test_results, additional_files
     except Exception as e:
         return (
-            "error",
+            ERROR,
             f"Cannot parse test_results.tsv ({e})",
             test_results,
             additional_files,
@@ -265,7 +265,7 @@ def main():
             post_commit_status_to_file(
                 post_commit_path,
                 f"Skipped (no pr-bugfix in {pr_info.labels})",
-                "success",
+                SUCCESS,
                 "null",
             )
         logging.info("Skipping '%s' (no pr-bugfix in %s)", check_name, pr_info.labels)
@@ -286,7 +286,7 @@ def main():
     if run_changed_tests:
         tests_to_run = get_tests_to_run(pr_info)
         if not tests_to_run:
-            state = override_status("success", check_name, validate_bugfix_check)
+            state = override_status(SUCCESS, check_name, validate_bugfix_check)
             if args.post_commit_status == "commit_status":
                 post_commit_status(
                     commit,
@@ -418,7 +418,7 @@ def main():
     )
     ch_helper.insert_events_into(db="default", table="checks", events=prepared_events)
 
-    if state != "success":
+    if state != SUCCESS:
         if FORCE_TESTS_LABEL in pr_info.labels:
             print(f"'{FORCE_TESTS_LABEL}' enabled, will report success")
         else:
diff --git a/tests/ci/integration_test_check.py b/tests/ci/integration_test_check.py
index bab89ea0941..5af4d5e625b 100644
--- a/tests/ci/integration_test_check.py
+++ b/tests/ci/integration_test_check.py
@@ -25,7 +25,14 @@ from get_robot_token import get_best_robot_token
 from github_helper import GitHub
 from integration_test_images import IMAGES
 from pr_info import PRInfo
-from report import ERROR, StatusType, TestResult, TestResults, read_test_results
+from report import (
+    ERROR,
+    SUCCESS,
+    StatusType,
+    TestResult,
+    TestResults,
+    read_test_results,
+)
 from s3_helper import S3Helper
 from stopwatch import Stopwatch
 from tee_popen import TeePopen
@@ -102,17 +109,17 @@ def process_results(
 
     if len(status) != 1 or len(status[0]) != 2:
         logging.info("Files in result folder %s", os.listdir(result_directory))
-        return "error", "Invalid check_status.tsv", test_results, additional_files
+        return ERROR, "Invalid check_status.tsv", test_results, additional_files
     state, description = status[0][0], status[0][1]
 
     try:
         results_path = result_directory / "test_results.tsv"
         test_results = read_test_results(results_path, False)
         if len(test_results) == 0:
-            return "error", "Empty test_results.tsv", test_results, additional_files
+            return ERROR, "Empty test_results.tsv", test_results, additional_files
     except Exception as e:
         return (
-            "error",
+            ERROR,
             f"Cannot parse test_results.tsv ({e})",
             test_results,
             additional_files,
@@ -182,7 +189,7 @@ def main():
             post_commit_status_to_file(
                 post_commit_path,
                 f"Skipped (no pr-bugfix in {pr_info.labels})",
-                "success",
+                SUCCESS,
                 "null",
             )
         logging.info("Skipping '%s' (no pr-bugfix in '%s')", check_name, pr_info.labels)
@@ -315,7 +322,7 @@ def main():
 
     ch_helper.insert_events_into(db="default", table="checks", events=prepared_events)
 
-    if state == "failure":
+    if state != SUCCESS:
         sys.exit(1)
 
 
diff --git a/tests/ci/jepsen_check.py b/tests/ci/jepsen_check.py
index 93e33d62293..fb7540abda3 100644
--- a/tests/ci/jepsen_check.py
+++ b/tests/ci/jepsen_check.py
@@ -5,29 +5,26 @@ import logging
 import os
 import sys
 import time
-
 from pathlib import Path
 from typing import Any, List
 
 import boto3  # type: ignore
 import requests  # type: ignore
-
 from build_download_helper import (
     download_build_with_progress,
     get_build_name_for_check,
     read_build_urls,
 )
 from compress_files import compress_fast
-from env_helper import REPO_COPY, REPORT_PATH, S3_URL, TEMP_PATH, S3_BUILDS_BUCKET
+from env_helper import REPO_COPY, REPORT_PATH, S3_BUILDS_BUCKET, S3_URL, TEMP_PATH
 from get_robot_token import get_parameter_from_ssm
 from git_helper import git_runner
 from pr_info import PRInfo
-from report import JobReport, TestResults, TestResult
+from report import FAILURE, SUCCESS, JobReport, TestResult, TestResults
 from ssh import SSHKey
 from stopwatch import Stopwatch
 from tee_popen import TeePopen
 
-
 JEPSEN_GROUP_NAME = "jepsen_group"
 
 KEEPER_DESIRED_INSTANCE_COUNT = 3
@@ -263,21 +260,21 @@ def main():
             else:
                 logging.info("Run failed")
 
-    status = "success"
+    status = SUCCESS
     description = "No invalid analysis found ヽ(‘ー`)ノ"
     jepsen_log_path = result_path / "jepsen_run_all_tests.log"
     additional_data = []
     try:
         test_result = _parse_jepsen_output(jepsen_log_path)
         if any(r.status == "FAIL" for r in test_result):
-            status = "failure"
+            status = FAILURE
             description = "Found invalid analysis (ﾉಥ益ಥ）ﾉ ┻━┻"
 
         compress_fast(result_path / "store", result_path / "jepsen_store.tar.zst")
         additional_data.append(result_path / "jepsen_store.tar.zst")
     except Exception as ex:
         print("Exception", ex)
-        status = "failure"
+        status = FAILURE
         description = "No Jepsen output log"
         test_result = [TestResult("No Jepsen output log", "FAIL")]
 
diff --git a/tests/ci/mark_release_ready.py b/tests/ci/mark_release_ready.py
index 0ad4b2bd2ed..011b3f28843 100755
--- a/tests/ci/mark_release_ready.py
+++ b/tests/ci/mark_release_ready.py
@@ -7,10 +7,11 @@ import os
 from commit_status_helper import get_commit, post_commit_status
 from env_helper import GITHUB_JOB_URL
 from get_robot_token import get_best_robot_token
+from git_helper import commit as commit_arg
 from github_helper import GitHub
 from pr_info import PRInfo
 from release import RELEASE_READY_STATUS
-from git_helper import commit as commit_arg
+from report import SUCCESS
 
 
 def main():
@@ -50,7 +51,7 @@ def main():
     gh.get_rate_limit()
     post_commit_status(
         commit,
-        "success",
+        SUCCESS,
         url,
         description,
         RELEASE_READY_STATUS,
diff --git a/tests/ci/merge_pr.py b/tests/ci/merge_pr.py
index 772821f4960..cc92fe4f42c 100644
--- a/tests/ci/merge_pr.py
+++ b/tests/ci/merge_pr.py
@@ -4,21 +4,23 @@
 
 import argparse
 import logging
-
 from datetime import datetime
 from os import getenv
 from pprint import pformat
 from typing import Dict, List
 
+# isort: off
 from github.PaginatedList import PaginatedList
 from github.PullRequestReview import PullRequestReview
 from github.WorkflowRun import WorkflowRun
 
+# isort: on
+
 from commit_status_helper import get_commit_filtered_statuses
 from get_robot_token import get_best_robot_token
 from github_helper import GitHub, NamedUser, PullRequest, Repository
 from pr_info import PRInfo
-
+from report import SUCCESS
 
 # The team name for accepted approvals
 TEAM_NAME = getenv("GITHUB_TEAM_NAME", "core")
@@ -269,7 +271,7 @@ def main():
         failed_statuses = [
             status.context
             for status in get_commit_filtered_statuses(commit)
-            if status.state != "success"
+            if status.state != SUCCESS
         ]
         if failed_statuses:
             logging.warning(
diff --git a/tests/ci/performance_comparison_check.py b/tests/ci/performance_comparison_check.py
index 5a98fa06b39..f0af15397c7 100644
--- a/tests/ci/performance_comparison_check.py
+++ b/tests/ci/performance_comparison_check.py
@@ -1,35 +1,38 @@
 #!/usr/bin/env python3
 
-import os
-import logging
-import sys
 import json
-import subprocess
-import traceback
+import logging
+import os
 import re
+import subprocess
+import sys
+import traceback
 from pathlib import Path
 
+# isort: off
 from github import Github
 
-from commit_status_helper import get_commit
+# isort: on
+
+from build_download_helper import download_builds_filter
 from ci_config import CI_CONFIG
-from docker_images_helper import pull_image, get_docker_image
+from clickhouse_helper import get_instance_id, get_instance_type
+from commit_status_helper import get_commit
+from docker_images_helper import get_docker_image, pull_image
 from env_helper import (
     GITHUB_EVENT_PATH,
     GITHUB_RUN_URL,
     REPO_COPY,
+    REPORT_PATH,
     S3_BUILDS_BUCKET,
     S3_DOWNLOAD,
     TEMP_PATH,
-    REPORT_PATH,
 )
 from get_robot_token import get_best_robot_token, get_parameter_from_ssm
 from pr_info import PRInfo
-from tee_popen import TeePopen
-from clickhouse_helper import get_instance_type, get_instance_id
+from report import FAILURE, SUCCESS, JobReport
 from stopwatch import Stopwatch
-from build_download_helper import download_builds_filter
-from report import SUCCESS, JobReport
+from tee_popen import TeePopen
 
 IMAGE_NAME = "clickhouse/performance-comparison"
 
@@ -225,18 +228,18 @@ def main():
         # TODO: Remove me, always green mode for the first time, unless errors
         status = SUCCESS
         if "errors" in message.lower() or too_many_slow(message.lower()):
-            status = "failure"
+            status = FAILURE
         # TODO: Remove until here
     except Exception:
         traceback.print_exc()
-        status = "failure"
+        status = FAILURE
         message = "Failed to parse the report."
 
     if not status:
-        status = "failure"
+        status = FAILURE
         message = "No status in report."
     elif not message:
-        status = "failure"
+        status = FAILURE
         message = "No message in report."
 
     JobReport(
diff --git a/tests/ci/release.py b/tests/ci/release.py
index f96845dad95..2b3331938e7 100755
--- a/tests/ci/release.py
+++ b/tests/ci/release.py
@@ -18,6 +18,7 @@ from contextlib import contextmanager
 from typing import Any, Final, Iterator, List, Optional, Tuple
 
 from git_helper import Git, commit, release_branch
+from report import SUCCESS
 from version_helper import (
     FILE_WITH_VERSION_PATH,
     GENERATED_CONTRIBUTORS,
@@ -142,7 +143,7 @@ class Release:
 
             for status in statuses:
                 if status["context"] == RELEASE_READY_STATUS:
-                    if not status["state"] == "success":
+                    if not status["state"] == SUCCESS:
                         raise Exception(
                             f"the status {RELEASE_READY_STATUS} is {status['state']}"
                             ", not success"
diff --git a/tests/ci/run_check.py b/tests/ci/run_check.py
index a6312872c2a..2aeac5b5740 100644
--- a/tests/ci/run_check.py
+++ b/tests/ci/run_check.py
@@ -27,7 +27,7 @@ from lambda_shared_package.lambda_shared.pr import (
     check_pr_description,
 )
 from pr_info import FORCE_TESTS_LABEL, PRInfo
-from report import FAILURE
+from report import FAILURE, PENDING
 
 TRUSTED_ORG_IDS = {
     54801242,  # clickhouse
@@ -188,7 +188,7 @@ def main():
     print("::notice ::Can run")
     post_commit_status(
         commit,
-        "pending",
+        PENDING,
         ci_report_url,
         description,
         CI_STATUS_NAME,
diff --git a/tests/ci/sqlancer_check.py b/tests/ci/sqlancer_check.py
index f85ab2be9a3..59d2a3d6275 100644
--- a/tests/ci/sqlancer_check.py
+++ b/tests/ci/sqlancer_check.py
@@ -7,12 +7,9 @@ import sys
 from pathlib import Path
 
 from build_download_helper import get_build_name_for_check, read_build_urls
-from docker_images_helper import DockerImage, pull_image, get_docker_image
-from env_helper import (
-    REPORT_PATH,
-    TEMP_PATH,
-)
-from report import JobReport, TestResults, TestResult
+from docker_images_helper import DockerImage, get_docker_image, pull_image
+from env_helper import REPORT_PATH, TEMP_PATH
+from report import FAILURE, SUCCESS, JobReport, TestResult, TestResults
 from stopwatch import Stopwatch
 from tee_popen import TeePopen
 
@@ -94,7 +91,7 @@ def main():
     paths += [workspace_path / f"{t}.err" for t in tests]
     paths += [workspace_path / f"{t}.out" for t in tests]
 
-    status = "success"
+    status = SUCCESS
     test_results = []  # type: TestResults
     # Try to get status message saved by the SQLancer
     try:
@@ -109,7 +106,7 @@ def main():
         with open(workspace_path / "description.txt", "r", encoding="utf-8") as desc_f:
             description = desc_f.readline().rstrip("\n")
     except:
-        status = "failure"
+        status = FAILURE
         description = "Task failed: $?=" + str(retcode)
 
     if not test_results:
diff --git a/tests/ci/sqllogic_test.py b/tests/ci/sqllogic_test.py
index 344eaea8277..bbd81fd76bb 100755
--- a/tests/ci/sqllogic_test.py
+++ b/tests/ci/sqllogic_test.py
@@ -172,7 +172,7 @@ def main():
         )
     )
 
-    # Until it pass all tests, do not block CI, report "success"
+    # Until it pass all tests, do not block CI, report SUCCESS
     assert description is not None
     # FIXME: force SUCCESS until all cases are fixed
     status = SUCCESS
diff --git a/tests/ci/sqltest.py b/tests/ci/sqltest.py
index b2105d4f5c0..2fe6aabd69c 100644
--- a/tests/ci/sqltest.py
+++ b/tests/ci/sqltest.py
@@ -1,21 +1,16 @@
 #!/usr/bin/env python3
 
 import logging
-import subprocess
 import os
+import subprocess
 import sys
 from pathlib import Path
-from typing import Dict
-
 
 from build_download_helper import get_build_name_for_check, read_build_urls
-from docker_images_helper import pull_image, get_docker_image
-from env_helper import (
-    REPORT_PATH,
-    TEMP_PATH,
-)
+from docker_images_helper import get_docker_image, pull_image
+from env_helper import REPORT_PATH, TEMP_PATH
 from pr_info import PRInfo
-from report import JobReport, TestResult
+from report import SUCCESS, JobReport, TestResult
 from stopwatch import Stopwatch
 
 IMAGE_NAME = "clickhouse/sqltest"
@@ -98,7 +93,7 @@ def main():
         "report.html": workspace_path / "report.html",
         "test.log": workspace_path / "test.log",
     }
-    status = "success"
+    status = SUCCESS
     description = "See the report"
     test_results = [TestResult(description, "OK")]
 
diff --git a/tests/ci/stress_check.py b/tests/ci/stress_check.py
index 46bb2261aba..49c1515c69f 100644
--- a/tests/ci/stress_check.py
+++ b/tests/ci/stress_check.py
@@ -10,11 +10,10 @@ from typing import List, Tuple
 
 from build_download_helper import download_all_deb_packages
 from clickhouse_helper import CiLogsCredentials
-
-from docker_images_helper import DockerImage, pull_image, get_docker_image
-from env_helper import REPORT_PATH, TEMP_PATH, REPO_COPY
+from docker_images_helper import DockerImage, get_docker_image, pull_image
+from env_helper import REPO_COPY, REPORT_PATH, TEMP_PATH
 from pr_info import PRInfo
-from report import JobReport, TestResult, TestResults, read_test_results
+from report import ERROR, JobReport, TestResult, TestResults, read_test_results
 from stopwatch import Stopwatch
 from tee_popen import TeePopen
 
@@ -89,7 +88,7 @@ def process_results(
         status = list(csv.reader(status_file, delimiter="\t"))
 
     if len(status) != 1 or len(status[0]) != 2:
-        return "error", "Invalid check_status.tsv", test_results, additional_files
+        return ERROR, "Invalid check_status.tsv", test_results, additional_files
     state, description = status[0][0], status[0][1]
 
     try:
@@ -99,7 +98,7 @@ def process_results(
             raise Exception("Empty results")
     except Exception as e:
         return (
-            "error",
+            ERROR,
             f"Cannot parse test_results.tsv ({e})",
             test_results,
             additional_files,
diff --git a/tests/ci/style_check.py b/tests/ci/style_check.py
index 785e29b2359..0c7160aeea4 100644
--- a/tests/ci/style_check.py
+++ b/tests/ci/style_check.py
@@ -12,7 +12,7 @@ from docker_images_helper import get_docker_image, pull_image
 from env_helper import REPO_COPY, TEMP_PATH
 from git_helper import GIT_PREFIX, git_runner
 from pr_info import PRInfo
-from report import JobReport, TestResults, read_test_results
+from report import ERROR, FAILURE, SUCCESS, JobReport, TestResults, read_test_results
 from ssh import SSHKey
 from stopwatch import Stopwatch
 
@@ -36,7 +36,7 @@ def process_result(
             status = list(csv.reader(status_file, delimiter="\t"))
     if len(status) != 1 or len(status[0]) != 2:
         logging.info("Files in result folder %s", os.listdir(result_directory))
-        return "error", "Invalid check_status.tsv", test_results, additional_files
+        return ERROR, "Invalid check_status.tsv", test_results, additional_files
     state, description = status[0][0], status[0][1]
 
     try:
@@ -47,8 +47,8 @@ def process_result(
 
         return state, description, test_results, additional_files
     except Exception:
-        if state == "success":
-            state, description = "error", "Failed to read test_results.tsv"
+        if state == SUCCESS:
+            state, description = ERROR, "Failed to read test_results.tsv"
         return state, description, test_results, additional_files
 
 
@@ -161,7 +161,7 @@ def main():
         additional_files=additional_files,
     ).dump()
 
-    if state in ["error", "failure"]:
+    if state in [ERROR, FAILURE]:
         print(f"Style check failed: [{description}]")
         sys.exit(1)
 
diff --git a/tests/ci/unit_tests_check.py b/tests/ci/unit_tests_check.py
index 495547e1dfc..41c52d53020 100644
--- a/tests/ci/unit_tests_check.py
+++ b/tests/ci/unit_tests_check.py
@@ -3,15 +3,15 @@
 import json
 import logging
 import os
-import sys
 import subprocess
+import sys
 from pathlib import Path
 from typing import Tuple
 
 from build_download_helper import download_unit_tests
-from docker_images_helper import pull_image, get_docker_image
+from docker_images_helper import get_docker_image, pull_image
 from env_helper import REPORT_PATH, TEMP_PATH
-from report import ERROR, FAILURE, FAIL, OK, SUCCESS, JobReport, TestResults, TestResult
+from report import ERROR, FAIL, FAILURE, OK, SUCCESS, JobReport, TestResult, TestResults
 from stopwatch import Stopwatch
 from tee_popen import TeePopen
 
@@ -104,7 +104,7 @@ def process_results(
             if "failures" in test_case:
                 raw_logs = ""
                 for failure in test_case["failures"]:
-                    raw_logs += failure["failure"]
+                    raw_logs += failure[FAILURE]
                 if (
                     "Segmentation fault" in raw_logs  # type: ignore
                     and SEGFAULT not in description
@@ -205,7 +205,7 @@ def main():
         additional_files=additional_files,
     ).dump()
 
-    if state == "failure":
+    if state == FAILURE:
         sys.exit(1)
 
 

From 5d042dc815a8e5ccf9136fb4191d610e8d193814 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Tue, 6 Feb 2024 17:42:24 +0100
Subject: [PATCH 62/73] Check pointer

---
 src/Analyzer/Passes/QueryAnalysisPass.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/Analyzer/Passes/QueryAnalysisPass.cpp b/src/Analyzer/Passes/QueryAnalysisPass.cpp
index de7788997be..99425518623 100644
--- a/src/Analyzer/Passes/QueryAnalysisPass.cpp
+++ b/src/Analyzer/Passes/QueryAnalysisPass.cpp
@@ -5570,7 +5570,7 @@ ProjectionNames QueryAnalyzer::resolveFunction(QueryTreeNodePtr & node, Identifi
                 column = function_base->getConstantResultForNonConstArguments(argument_columns, result_type);
             }
 
-            if (column->getDataType() != result_type->getColumnType())
+            if (column && column->getDataType() != result_type->getColumnType())
                 throw Exception(
                     ErrorCodes::LOGICAL_ERROR,
                     "Unexpected return type from {}. Expected {}. Got {}",

From 8a428eb7cfbe72acdbd197a924fad604765e217e Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Tue, 6 Feb 2024 18:37:14 +0100
Subject: [PATCH 63/73] Better usability for server versions in ClickHouse
 Cloud

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

diff --git a/programs/client/Client.cpp b/programs/client/Client.cpp
index 0988e1eb4a1..cc142470d7f 100644
--- a/programs/client/Client.cpp
+++ b/programs/client/Client.cpp
@@ -504,7 +504,7 @@ void Client::connect()
                         << "It may lack support for new features." << std::endl
                         << std::endl;
         }
-        else if (client_version_tuple > server_version_tuple)
+        else if (client_version_tuple > server_version_tuple && server_display_name != "clickhouse-cloud")
         {
             std::cout << "ClickHouse server version is older than ClickHouse client. "
                         << "It may indicate that the server is out of date and can be upgraded." << std::endl

From 5c68f9cabdd44ed147c586702793980f7c70a2e5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Tue, 6 Feb 2024 19:09:28 +0100
Subject: [PATCH 64/73] Maintain function alias in
 RewriteSumFunctionWithSumAndCountVisitor

---
 ...writeSumFunctionWithSumAndCountVisitor.cpp |  6 ++
 ..._rewrite_sum_column_and_constant.reference | 72 ++++++++++---------
 .../02931_rewrite_sum_column_and_constant.sql |  5 ++
 3 files changed, 51 insertions(+), 32 deletions(-)

diff --git a/src/Interpreters/RewriteSumFunctionWithSumAndCountVisitor.cpp b/src/Interpreters/RewriteSumFunctionWithSumAndCountVisitor.cpp
index b654d28d750..2f5e597bdab 100644
--- a/src/Interpreters/RewriteSumFunctionWithSumAndCountVisitor.cpp
+++ b/src/Interpreters/RewriteSumFunctionWithSumAndCountVisitor.cpp
@@ -100,7 +100,10 @@ void RewriteSumFunctionWithSumAndCountMatcher::visit(const ASTFunction & functio
         if (!new_ast)
             return;
         else
+        {
+            new_ast->setAlias(ast->tryGetAlias());
             ast = new_ast;
+        }
     }
     else if (column_id == 1)
     {
@@ -116,7 +119,10 @@ void RewriteSumFunctionWithSumAndCountMatcher::visit(const ASTFunction & functio
         if (!new_ast)
             return;
         else
+        {
+            new_ast->setAlias(ast->tryGetAlias());
             ast = new_ast;
+        }
     }
 }
 
diff --git a/tests/queries/0_stateless/02931_rewrite_sum_column_and_constant.reference b/tests/queries/0_stateless/02931_rewrite_sum_column_and_constant.reference
index 8f29910e9ae..3124698d218 100644
--- a/tests/queries/0_stateless/02931_rewrite_sum_column_and_constant.reference
+++ b/tests/queries/0_stateless/02931_rewrite_sum_column_and_constant.reference
@@ -47,24 +47,24 @@ SELECT sum(uint64) + (1 * count(uint64))
 FROM test_table
 WHERE ((uint64 + 1) AS i) > 0
 EXPLAIN SYNTAX (SELECT sum(uint64 + 1) AS j from test_table having j > 0);
-SELECT sum(uint64) + (1 * count(uint64))
+SELECT sum(uint64) + (1 * count(uint64)) AS j
 FROM test_table
-HAVING (sum(uint64) + (1 * count(uint64))) > 0
+HAVING j > 0
 EXPLAIN SYNTAX (SELECT sum(uint64 + 1 AS i) j from test_table where i > 0 having j > 0);
-SELECT sum(uint64) + (1 * count(uint64))
+SELECT sum(uint64) + (1 * count(uint64)) AS j
 FROM test_table
 WHERE ((uint64 + 1) AS i) > 0
-HAVING (sum(uint64) + (1 * count(uint64))) > 0
+HAVING j > 0
 EXPLAIN SYNTAX (SELECT sum((uint64 AS m) + (1 AS n)) j from test_table where m > 0 and n > 0 having j > 0);
-SELECT sum(uint64) + ((1 AS n) * count(uint64))
+SELECT sum(uint64) + ((1 AS n) * count(uint64)) AS j
 FROM test_table
 WHERE ((uint64 AS m) > 0) AND (n > 0)
-HAVING (sum(uint64) + (n * count(uint64))) > 0
+HAVING j > 0
 EXPLAIN SYNTAX (SELECT sum(((uint64 AS m) + (1 AS n)) AS i) j from test_table where m > 0 and n > 0 and i > 0 having j > 0);
-SELECT sum(uint64) + ((1 AS n) * count(uint64))
+SELECT sum(uint64) + ((1 AS n) * count(uint64)) AS j
 FROM test_table
 WHERE ((uint64 AS m) > 0) AND (n > 0) AND (((m + n) AS i) > 0)
-HAVING (sum(uint64) + (n * count(uint64))) > 0
+HAVING j > 0
 SELECT sum(1 + uint64 AS i) from test_table where i > 0;
 20
 SELECT sum(1 + uint64) AS j from test_table having j > 0;
@@ -80,24 +80,24 @@ SELECT (1 * count(uint64)) + sum(uint64)
 FROM test_table
 WHERE ((1 + uint64) AS i) > 0
 EXPLAIN SYNTAX (SELECT sum(1 + uint64) AS j from test_table having j > 0);
-SELECT (1 * count(uint64)) + sum(uint64)
+SELECT (1 * count(uint64)) + sum(uint64) AS j
 FROM test_table
-HAVING ((1 * count(uint64)) + sum(uint64)) > 0
+HAVING j > 0
 EXPLAIN SYNTAX (SELECT sum(1 + uint64 AS i) j from test_table where i > 0 having j > 0);
-SELECT (1 * count(uint64)) + sum(uint64)
+SELECT (1 * count(uint64)) + sum(uint64) AS j
 FROM test_table
 WHERE ((1 + uint64) AS i) > 0
-HAVING ((1 * count(uint64)) + sum(uint64)) > 0
+HAVING j > 0
 EXPLAIN SYNTAX (SELECT sum((1 AS m) + (uint64 AS n)) j from test_table where m > 0 and n > 0 having j > 0);
-SELECT ((1 AS m) * count(uint64)) + sum(uint64)
+SELECT ((1 AS m) * count(uint64)) + sum(uint64) AS j
 FROM test_table
 WHERE (m > 0) AND ((uint64 AS n) > 0)
-HAVING ((m * count(uint64)) + sum(uint64)) > 0
+HAVING j > 0
 EXPLAIN SYNTAX (SELECT sum(((1 AS m) + (uint64 AS n)) AS i) j from test_table where m > 0 and n > 0 and i > 0 having j > 0);
-SELECT ((1 AS m) * count(uint64)) + sum(uint64)
+SELECT ((1 AS m) * count(uint64)) + sum(uint64) AS j
 FROM test_table
 WHERE (m > 0) AND ((uint64 AS n) > 0) AND (((m + n) AS i) > 0)
-HAVING ((m * count(uint64)) + sum(uint64)) > 0
+HAVING j > 0
 SELECT sum(uint64 - 1 AS i) from test_table where i > 0;
 10
 SELECT sum(uint64 - 1) AS j from test_table having j > 0;
@@ -113,24 +113,24 @@ SELECT sum(uint64) - (1 * count(uint64))
 FROM test_table
 WHERE ((uint64 - 1) AS i) > 0
 EXPLAIN SYNTAX (SELECT sum(uint64 - 1) AS j from test_table having j > 0);
-SELECT sum(uint64) - (1 * count(uint64))
+SELECT sum(uint64) - (1 * count(uint64)) AS j
 FROM test_table
-HAVING (sum(uint64) - (1 * count(uint64))) > 0
+HAVING j > 0
 EXPLAIN SYNTAX (SELECT sum(uint64 - 1 AS i) j from test_table where i > 0 having j > 0);
-SELECT sum(uint64) - (1 * count(uint64))
+SELECT sum(uint64) - (1 * count(uint64)) AS j
 FROM test_table
 WHERE ((uint64 - 1) AS i) > 0
-HAVING (sum(uint64) - (1 * count(uint64))) > 0
+HAVING j > 0
 EXPLAIN SYNTAX (SELECT sum((uint64 AS m) - (1 AS n)) j from test_table where m > 0 and n > 0 having j > 0);
-SELECT sum(uint64) - ((1 AS n) * count(uint64))
+SELECT sum(uint64) - ((1 AS n) * count(uint64)) AS j
 FROM test_table
 WHERE ((uint64 AS m) > 0) AND (n > 0)
-HAVING (sum(uint64) - (n * count(uint64))) > 0
+HAVING j > 0
 EXPLAIN SYNTAX (SELECT sum(((uint64 AS m) - (1 AS n)) AS i) j from test_table where m > 0 and n > 0 and i > 0 having j > 0);
-SELECT sum(uint64) - ((1 AS n) * count(uint64))
+SELECT sum(uint64) - ((1 AS n) * count(uint64)) AS j
 FROM test_table
 WHERE ((uint64 AS m) > 0) AND (n > 0) AND (((m - n) AS i) > 0)
-HAVING (sum(uint64) - (n * count(uint64))) > 0
+HAVING j > 0
 SELECT sum(1 - uint64 AS i) from test_table;
 -10
 SELECT sum(1 - uint64) AS j from test_table;
@@ -146,24 +146,24 @@ SELECT (1 * count(uint64)) - sum(uint64)
 FROM test_table
 WHERE ((1 - uint64) AS i) > 0
 EXPLAIN SYNTAX (SELECT sum(1 - uint64) AS j from test_table having j < 0);
-SELECT (1 * count(uint64)) - sum(uint64)
+SELECT (1 * count(uint64)) - sum(uint64) AS j
 FROM test_table
-HAVING ((1 * count(uint64)) - sum(uint64)) < 0
+HAVING j < 0
 EXPLAIN SYNTAX (SELECT sum(1 - uint64 AS i) j from test_table where i > 0 having j < 0);
-SELECT (1 * count(uint64)) - sum(uint64)
+SELECT (1 * count(uint64)) - sum(uint64) AS j
 FROM test_table
 WHERE ((1 - uint64) AS i) > 0
-HAVING ((1 * count(uint64)) - sum(uint64)) < 0
+HAVING j < 0
 EXPLAIN SYNTAX (SELECT sum((1 AS m) - (uint64 AS n)) j from test_table where m > 0 and n > 0 having j < 0);
-SELECT ((1 AS m) * count(uint64)) - sum(uint64)
+SELECT ((1 AS m) * count(uint64)) - sum(uint64) AS j
 FROM test_table
 WHERE (m > 0) AND ((uint64 AS n) > 0)
-HAVING ((m * count(uint64)) - sum(uint64)) < 0
+HAVING j < 0
 EXPLAIN SYNTAX (SELECT sum(((1 AS m) - (uint64 AS n)) AS i) j from test_table where m > 0 and n > 0 and i < 0 having j < 0);
-SELECT ((1 AS m) * count(uint64)) - sum(uint64)
+SELECT ((1 AS m) * count(uint64)) - sum(uint64) AS j
 FROM test_table
 WHERE (m > 0) AND ((uint64 AS n) > 0) AND (((m - n) AS i) < 0)
-HAVING ((m * count(uint64)) - sum(uint64)) < 0
+HAVING j < 0
 SELECT sum(uint64 + 2.11) From test_table;
 25.549999999999997
 SELECT sum(2.11 + uint64) From test_table;
@@ -474,3 +474,11 @@ FROM test_table
 EXPLAIN SYNTAX (SELECT (2 * count(decimal32) - sum(decimal32)) + (3 * count(decimal32) - sum(decimal32)) From test_table);
 SELECT ((2 * count(decimal32)) - sum(decimal32)) + ((3 * count(decimal32)) - sum(decimal32))
 FROM test_table
+-- https://github.com/ClickHouse/ClickHouse/issues/59414
+SELECT sum(uint64 + 2) as j, j + 5 as t from test_table;
+25	30
+EXPLAIN SYNTAX SELECT sum(uint64 + 2) as j, j + 5 as t from test_table;
+SELECT
+    sum(uint64) + (2 * count(uint64)) AS j,
+    j + 5 AS t
+FROM test_table
diff --git a/tests/queries/0_stateless/02931_rewrite_sum_column_and_constant.sql b/tests/queries/0_stateless/02931_rewrite_sum_column_and_constant.sql
index b29407d7208..c7b0ff82442 100644
--- a/tests/queries/0_stateless/02931_rewrite_sum_column_and_constant.sql
+++ b/tests/queries/0_stateless/02931_rewrite_sum_column_and_constant.sql
@@ -204,6 +204,11 @@ EXPLAIN SYNTAX (SELECT (sum(decimal32) + 2 * count(decimal32)) - (sum(decimal32)
 EXPLAIN SYNTAX (SELECT (sum(decimal32) - 2 * count(decimal32)) + (sum(decimal32) - 3 * count(decimal32)) From test_table);
 EXPLAIN SYNTAX (SELECT (sum(decimal32) - 2 * count(decimal32)) - (sum(decimal32) - 3 * count(decimal32)) From test_table);
 EXPLAIN SYNTAX (SELECT (2 * count(decimal32) - sum(decimal32)) + (3 * count(decimal32) - sum(decimal32)) From test_table);
+
+-- https://github.com/ClickHouse/ClickHouse/issues/59414
+SELECT sum(uint64 + 2) as j, j + 5 as t from test_table;
+EXPLAIN SYNTAX SELECT sum(uint64 + 2) as j, j + 5 as t from test_table;
 -- { echoOff }
 
+
 DROP TABLE IF EXISTS test_table;

From 648fa525832644a89b55dba9f6957969060e1a69 Mon Sep 17 00:00:00 2001
From: Yakov Olkhovskiy <yakov@clickhouse.com>
Date: Tue, 6 Feb 2024 18:48:57 +0000
Subject: [PATCH 65/73] move instead copy

---
 utils/self-extracting-executable/decompressor.cpp | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/utils/self-extracting-executable/decompressor.cpp b/utils/self-extracting-executable/decompressor.cpp
index 1f19a349d65..6614403c0ab 100644
--- a/utils/self-extracting-executable/decompressor.cpp
+++ b/utils/self-extracting-executable/decompressor.cpp
@@ -322,6 +322,8 @@ int decompressFiles(int input_fd, char * path, char * name, bool & have_compress
             return 1;
         }
 
+        if (0 != munmap(output, le64toh(file_info.uncompressed_size)))
+            perror("munmap");
         if (0 != fsync(output_fd))
             perror("fsync");
         if (0 != close(output_fd))
@@ -528,10 +530,10 @@ int main(int/* argc*/, char* argv[])
         (void)snprintf(decompressed_name, decompressed_name_len + 1, decompressed_name_fmt, self, decompressed_suffix);
 
         std::error_code ec;
-        std::filesystem::copy_file(static_cast<char *>(decompressed_name), static_cast<char *>(self), ec);
-        if (ec)
+
+        if (link(decompressed_name, self))
         {
-            std::cerr << ec.message() << std::endl;
+            perror("link");
             return 1;
         }
 

From 0a4e73a98dc4d0471957ad65aecb8b16bae1070f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Thu, 1 Feb 2024 15:50:48 +0100
Subject: [PATCH 66/73] Strict aliasing for Int8

---
 base/base/CMakeLists.txt                      |  1 +
 base/base/int8_to_string.cpp                  |  9 +++++
 base/base/int8_to_string.h                    | 34 +++++++++++++++++++
 base/base/types.h                             | 19 +++++++++--
 src/Common/Exception.h                        |  5 +--
 src/Core/Field.h                              |  3 +-
 src/Functions/divide/divide.cpp               |  4 +--
 src/Functions/divide/divideImpl.cpp           |  8 +++--
 .../Formats/Impl/CHColumnToArrowColumn.cpp    |  5 +++
 9 files changed, 78 insertions(+), 10 deletions(-)
 create mode 100644 base/base/int8_to_string.cpp
 create mode 100644 base/base/int8_to_string.h

diff --git a/base/base/CMakeLists.txt b/base/base/CMakeLists.txt
index 3886932d198..025687d2c59 100644
--- a/base/base/CMakeLists.txt
+++ b/base/base/CMakeLists.txt
@@ -17,6 +17,7 @@ set (SRCS
     getMemoryAmount.cpp
     getPageSize.cpp
     getThreadId.cpp
+    int8_to_string.cpp
     JSON.cpp
     mremap.cpp
     phdr_cache.cpp
diff --git a/base/base/int8_to_string.cpp b/base/base/int8_to_string.cpp
new file mode 100644
index 00000000000..09da940e1a0
--- /dev/null
+++ b/base/base/int8_to_string.cpp
@@ -0,0 +1,9 @@
+#include <base/int8_to_string.h>
+
+namespace std
+{
+std::string to_string(Int8 v)
+{
+    return to_string(int8_t{v});
+}
+}
diff --git a/base/base/int8_to_string.h b/base/base/int8_to_string.h
new file mode 100644
index 00000000000..db92630ad86
--- /dev/null
+++ b/base/base/int8_to_string.h
@@ -0,0 +1,34 @@
+#pragma once
+
+#include <base/defines.h>
+#include <base/types.h>
+
+#include <fmt/format.h>
+
+template <>
+struct fmt::formatter<Int8>
+{
+    constexpr auto parse(format_parse_context & ctx)
+    {
+        const auto * it = ctx.begin();
+        const auto * end = ctx.end();
+
+        /// Only support {}.
+        if (it != end && *it != '}')
+            throw format_error("invalid format");
+
+        return it;
+    }
+
+    template <typename FormatContext>
+    auto format(const Int8 & value, FormatContext & ctx) -> decltype(ctx.out())
+    {
+        return format<FormatContext>(int8_t{value}, ctx);
+    }
+};
+
+
+namespace std
+{
+std::string to_string(Int8 v);
+}
diff --git a/base/base/types.h b/base/base/types.h
index 3a7760eae91..52e9df65f04 100644
--- a/base/base/types.h
+++ b/base/base/types.h
@@ -3,14 +3,29 @@
 #include <cstdint>
 #include <string>
 
-/// This is needed for more strict aliasing. https://godbolt.org/z/xpJBSb https://stackoverflow.com/a/57453713
+/// Using char8_t more strict aliasing (https://stackoverflow.com/a/57453713)
 using UInt8 = char8_t;
 
+/// Same for using signed _BitInt(8) (there isn't a signed char8_t, which would be more convenient)
+/// See https://godbolt.org/z/fafnWEnnf
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wbit-int-extension"
+using Int8 = signed _BitInt(8);
+#pragma clang diagnostic pop
+
+namespace std
+{
+template <>
+struct hash<Int8>
+{
+    size_t operator()(const Int8 x) const { return std::hash<int8_t>()(int8_t{x}); }
+};
+}
+
 using UInt16 = uint16_t;
 using UInt32 = uint32_t;
 using UInt64 = uint64_t;
 
-using Int8 = int8_t;
 using Int16 = int16_t;
 using Int32 = int32_t;
 using Int64 = int64_t;
diff --git a/src/Common/Exception.h b/src/Common/Exception.h
index 94e4a986feb..21f9529a49c 100644
--- a/src/Common/Exception.h
+++ b/src/Common/Exception.h
@@ -9,10 +9,11 @@
 
 #include <base/defines.h>
 #include <base/errnoToString.h>
+#include <base/int8_to_string.h>
 #include <base/scope_guard.h>
-#include <Common/LoggingFormatStringHelpers.h>
-#include <Common/Logger.h>
 #include <Common/AtomicLogger.h>
+#include <Common/Logger.h>
+#include <Common/LoggingFormatStringHelpers.h>
 #include <Common/StackTrace.h>
 
 #include <fmt/format.h>
diff --git a/src/Core/Field.h b/src/Core/Field.h
index f9ce70efbc5..414874310a6 100644
--- a/src/Core/Field.h
+++ b/src/Core/Field.h
@@ -216,9 +216,8 @@ using NearestFieldType = typename NearestFieldTypeImpl<T>::Type;
 template <> struct NearestFieldTypeImpl<char> { using Type = std::conditional_t<is_signed_v<char>, Int64, UInt64>; };
 template <> struct NearestFieldTypeImpl<signed char> { using Type = Int64; };
 template <> struct NearestFieldTypeImpl<unsigned char> { using Type = UInt64; };
-#ifdef __cpp_char8_t
 template <> struct NearestFieldTypeImpl<char8_t> { using Type = UInt64; };
-#endif
+template <> struct NearestFieldTypeImpl<Int8> { using Type = Int64; };
 
 template <> struct NearestFieldTypeImpl<UInt16> { using Type = UInt64; };
 template <> struct NearestFieldTypeImpl<UInt32> { using Type = UInt64; };
diff --git a/src/Functions/divide/divide.cpp b/src/Functions/divide/divide.cpp
index cf2cd354a7d..6262d42a666 100644
--- a/src/Functions/divide/divide.cpp
+++ b/src/Functions/divide/divide.cpp
@@ -49,9 +49,9 @@ template void divideImpl<uint32_t, char8_t, uint32_t>(const uint32_t * __restric
 template void divideImpl<int64_t, int64_t, int64_t>(const int64_t * __restrict, int64_t, int64_t * __restrict, size_t);
 template void divideImpl<int64_t, int32_t, int64_t>(const int64_t * __restrict, int32_t, int64_t * __restrict, size_t);
 template void divideImpl<int64_t, int16_t, int64_t>(const int64_t * __restrict, int16_t, int64_t * __restrict, size_t);
-template void divideImpl<int64_t, int8_t, int64_t>(const int64_t * __restrict, int8_t, int64_t * __restrict, size_t);
+template void divideImpl<int64_t, Int8, int64_t>(const int64_t * __restrict, Int8, int64_t * __restrict, size_t);
 
 template void divideImpl<int32_t, int64_t, int32_t>(const int32_t * __restrict, int64_t, int32_t * __restrict, size_t);
 template void divideImpl<int32_t, int32_t, int32_t>(const int32_t * __restrict, int32_t, int32_t * __restrict, size_t);
 template void divideImpl<int32_t, int16_t, int32_t>(const int32_t * __restrict, int16_t, int32_t * __restrict, size_t);
-template void divideImpl<int32_t, int8_t, int32_t>(const int32_t * __restrict, int8_t, int32_t * __restrict, size_t);
+template void divideImpl<int32_t, Int8, int32_t>(const int32_t * __restrict, Int8, int32_t * __restrict, size_t);
diff --git a/src/Functions/divide/divideImpl.cpp b/src/Functions/divide/divideImpl.cpp
index 966d5777c1d..6d44b427582 100644
--- a/src/Functions/divide/divideImpl.cpp
+++ b/src/Functions/divide/divideImpl.cpp
@@ -12,6 +12,10 @@
 
 #include <libdivide.h>
 
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wbit-int-extension"
+using Int8 = signed _BitInt(8);
+#pragma clang diagnostic pop
 
 namespace NAMESPACE
 {
@@ -62,11 +66,11 @@ template void divideImpl<uint32_t, char8_t, uint32_t>(const uint32_t * __restric
 template void divideImpl<int64_t, int64_t, int64_t>(const int64_t * __restrict, int64_t, int64_t * __restrict, size_t);
 template void divideImpl<int64_t, int32_t, int64_t>(const int64_t * __restrict, int32_t, int64_t * __restrict, size_t);
 template void divideImpl<int64_t, int16_t, int64_t>(const int64_t * __restrict, int16_t, int64_t * __restrict, size_t);
-template void divideImpl<int64_t, int8_t, int64_t>(const int64_t * __restrict, int8_t, int64_t * __restrict, size_t);
+template void divideImpl<int64_t, Int8, int64_t>(const int64_t * __restrict, Int8, int64_t * __restrict, size_t);
 
 template void divideImpl<int32_t, int64_t, int32_t>(const int32_t * __restrict, int64_t, int32_t * __restrict, size_t);
 template void divideImpl<int32_t, int32_t, int32_t>(const int32_t * __restrict, int32_t, int32_t * __restrict, size_t);
 template void divideImpl<int32_t, int16_t, int32_t>(const int32_t * __restrict, int16_t, int32_t * __restrict, size_t);
-template void divideImpl<int32_t, int8_t, int32_t>(const int32_t * __restrict, int8_t, int32_t * __restrict, size_t);
+template void divideImpl<int32_t, Int8, int32_t>(const int32_t * __restrict, Int8, int32_t * __restrict, size_t);
 
 }
diff --git a/src/Processors/Formats/Impl/CHColumnToArrowColumn.cpp b/src/Processors/Formats/Impl/CHColumnToArrowColumn.cpp
index 1048bdad22f..9d6c8420069 100644
--- a/src/Processors/Formats/Impl/CHColumnToArrowColumn.cpp
+++ b/src/Processors/Formats/Impl/CHColumnToArrowColumn.cpp
@@ -130,6 +130,11 @@ namespace DB
                 reinterpret_cast<const uint8_t *>(internal_data.data() + start),
                 end - start,
                 reinterpret_cast<const uint8_t *>(arrow_null_bytemap_raw_ptr));
+        else if constexpr (std::is_same_v<NumericType, Int8>)
+            status = builder.AppendValues(
+                reinterpret_cast<const int8_t *>(internal_data.data() + start),
+                end - start,
+                reinterpret_cast<const uint8_t *>(arrow_null_bytemap_raw_ptr));
         else
             status = builder.AppendValues(internal_data.data() + start, end - start, reinterpret_cast<const uint8_t *>(arrow_null_bytemap_raw_ptr));
         checkStatus(status, write_column->getName(), format_name);

From 333c45634fa8fd239d562b958146fc2178e1cd2a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Thu, 1 Feb 2024 16:53:31 +0100
Subject: [PATCH 67/73] Infinite recursion

---
 base/base/int8_to_string.h | 19 +------------------
 1 file changed, 1 insertion(+), 18 deletions(-)

diff --git a/base/base/int8_to_string.h b/base/base/int8_to_string.h
index db92630ad86..715473b1482 100644
--- a/base/base/int8_to_string.h
+++ b/base/base/int8_to_string.h
@@ -6,25 +6,8 @@
 #include <fmt/format.h>
 
 template <>
-struct fmt::formatter<Int8>
+struct fmt::formatter<Int8> : fmt::formatter<int8_t>
 {
-    constexpr auto parse(format_parse_context & ctx)
-    {
-        const auto * it = ctx.begin();
-        const auto * end = ctx.end();
-
-        /// Only support {}.
-        if (it != end && *it != '}')
-            throw format_error("invalid format");
-
-        return it;
-    }
-
-    template <typename FormatContext>
-    auto format(const Int8 & value, FormatContext & ctx) -> decltype(ctx.out())
-    {
-        return format<FormatContext>(int8_t{value}, ctx);
-    }
 };
 
 

From b609fc157fc9a0ca8a6cd0511c93599c401164ad Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Fri, 2 Feb 2024 16:28:09 +0100
Subject: [PATCH 68/73] Silence tidy

---
 base/base/int8_to_string.cpp | 2 +-
 base/base/int8_to_string.h   | 2 +-
 base/base/types.h            | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/base/base/int8_to_string.cpp b/base/base/int8_to_string.cpp
index 09da940e1a0..f74a6b8077e 100644
--- a/base/base/int8_to_string.cpp
+++ b/base/base/int8_to_string.cpp
@@ -2,7 +2,7 @@
 
 namespace std
 {
-std::string to_string(Int8 v)
+std::string to_string(Int8 v) /// NOLINT (cert-dcl58-cpp)
 {
     return to_string(int8_t{v});
 }
diff --git a/base/base/int8_to_string.h b/base/base/int8_to_string.h
index 715473b1482..af0914f4312 100644
--- a/base/base/int8_to_string.h
+++ b/base/base/int8_to_string.h
@@ -13,5 +13,5 @@ struct fmt::formatter<Int8> : fmt::formatter<int8_t>
 
 namespace std
 {
-std::string to_string(Int8 v);
+std::string to_string(Int8 v); /// NOLINT (cert-dcl58-cpp)
 }
diff --git a/base/base/types.h b/base/base/types.h
index 52e9df65f04..a4874860514 100644
--- a/base/base/types.h
+++ b/base/base/types.h
@@ -16,7 +16,7 @@ using Int8 = signed _BitInt(8);
 namespace std
 {
 template <>
-struct hash<Int8>
+struct hash<Int8> /// NOLINT (cert-dcl58-cpp)
 {
     size_t operator()(const Int8 x) const { return std::hash<int8_t>()(int8_t{x}); }
 };

From ff63a1cf9f578fa4f33433b9969d0eb844383d1e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Mon, 5 Feb 2024 13:41:07 +0100
Subject: [PATCH 69/73] Fix tidy

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

diff --git a/src/Storages/PostgreSQL/MaterializedPostgreSQLConsumer.cpp b/src/Storages/PostgreSQL/MaterializedPostgreSQLConsumer.cpp
index f99ebf51792..ba3cc6f58d0 100644
--- a/src/Storages/PostgreSQL/MaterializedPostgreSQLConsumer.cpp
+++ b/src/Storages/PostgreSQL/MaterializedPostgreSQLConsumer.cpp
@@ -313,7 +313,7 @@ void MaterializedPostgreSQLConsumer::readTupleData(
                 Int32 col_len = readInt32(message, pos, size);
                 String value;
                 for (Int32 i = 0; i < col_len; ++i)
-                    value += readInt8(message, pos, size);
+                    value += static_cast<char>(readInt8(message, pos, size));
 
                 insertValue(storage_data, value, column_idx);
                 break;

From 25738c8313bff80f3a2e3a8e3a53101041a62f2c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Mon, 5 Feb 2024 14:26:46 +0100
Subject: [PATCH 70/73] Remove no longer needed specialization

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

diff --git a/src/Functions/if.cpp b/src/Functions/if.cpp
index 5346f3a811c..7306dc4173e 100644
--- a/src/Functions/if.cpp
+++ b/src/Functions/if.cpp
@@ -230,13 +230,8 @@ inline void fillConstantConstant(const ArrayCond & cond, A a, B b, ArrayResult &
 {
     size_t size = cond.size();
 
-    /// Int8(alias type of uint8_t) has special aliasing properties that prevents compiler from auto-vectorizing for below codes, refer to https://gist.github.com/alexei-zaripov/dcc14c78819c5f1354afe8b70932007c
-    ///
-    /// for (size_t i = 0; i < size; ++i)
-    ///     res[i] = cond[i] ? static_cast<Int8>(a) : static_cast<Int8>(b);
-    ///
-    /// Therefore, we manually optimize it by avoiding branch miss when ResultType is Int8. Other types like (U)Int128|256 or Decimal128/256 also benefit from this optimization.
-    if constexpr (std::is_same_v<ResultType, Int8> || is_over_big_int<ResultType>)
+    /// We manually optimize the loop for types like (U)Int128|256 or Decimal128/256 to avoid branches
+    if constexpr (is_over_big_int<ResultType>)
     {
         alignas(64) const ResultType ab[2] = {static_cast<ResultType>(a), static_cast<ResultType>(b)};
         for (size_t i = 0; i < size; ++i)

From f7a0c38a3ddc2caf555daae22badf56c38430800 Mon Sep 17 00:00:00 2001
From: "Mikhail f. Shiryaev" <felixoid@clickhouse.com>
Date: Tue, 6 Feb 2024 21:31:44 +0100
Subject: [PATCH 71/73] Unquote FLAG_LATEST to fix issue with empty argument

---
 .github/workflows/reusable_docker.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/reusable_docker.yml b/.github/workflows/reusable_docker.yml
index 1b186350580..3fe1a8883c6 100644
--- a/.github/workflows/reusable_docker.yml
+++ b/.github/workflows/reusable_docker.yml
@@ -63,4 +63,4 @@ jobs:
           python3 docker_manifests_merge.py --suffix amd64 --suffix aarch64 \
             --image-tags '${{ toJson(fromJson(inputs.data).docker_data.images) }}' \
             --missing-images '${{ toJson(fromJson(inputs.data).docker_data.missing_multi) }}' \
-            "$FLAG_LATEST"
+            $FLAG_LATEST

From 90f2c4286d647b9888eda0a70e4f5379207c7f76 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@clickhouse.com>
Date: Wed, 7 Feb 2024 05:33:32 +0300
Subject: [PATCH 72/73] Update CHANGELOG.md

---
 CHANGELOG.md | 1 -
 1 file changed, 1 deletion(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 60618402174..b3e5dd709ab 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -22,7 +22,6 @@
 * Add `quantileDD` aggregate function as well as the corresponding `quantilesDD` and `medianDD`. It is based on the DDSketch https://www.vldb.org/pvldb/vol12/p2195-masson.pdf. ### Documentation entry for user-facing changes. [#56342](https://github.com/ClickHouse/ClickHouse/pull/56342) ([Srikanth Chekuri](https://github.com/srikanthccv)).
 * Allow to configure any kind of object storage with any kind of metadata type. [#58357](https://github.com/ClickHouse/ClickHouse/pull/58357) ([Kseniia Sumarokova](https://github.com/kssenii)).
 * Added `null_status_on_timeout_only_active` and `throw_only_active` modes for `distributed_ddl_output_mode` that allow to avoid waiting for inactive replicas. [#58350](https://github.com/ClickHouse/ClickHouse/pull/58350) ([Alexander Tokmakov](https://github.com/tavplubix)).
-* Allow partitions from tables with different partition expressions to be attached when the destination table partition expression doesn't re-partition/split the part. [#39507](https://github.com/ClickHouse/ClickHouse/pull/39507) ([Arthur Passos](https://github.com/arthurpassos)).
 * Add function `arrayShingles` to compute subarrays, e.g. `arrayShingles([1, 2, 3, 4, 5], 3)` returns `[[1,2,3],[2,3,4],[3,4,5]]`. [#58396](https://github.com/ClickHouse/ClickHouse/pull/58396) ([Zheng Miao](https://github.com/zenmiao7)).
 * Added functions `punycodeEncode`, `punycodeDecode`, `idnaEncode` and `idnaDecode` which are useful for translating international domain names to an ASCII representation according to the IDNA standard. [#58454](https://github.com/ClickHouse/ClickHouse/pull/58454) ([Robert Schulze](https://github.com/rschu1ze)).
 * Added string similarity functions `dramerauLevenshteinDistance`, `jaroSimilarity` and `jaroWinklerSimilarity`. [#58531](https://github.com/ClickHouse/ClickHouse/pull/58531) ([Robert Schulze](https://github.com/rschu1ze)).

From b5b7847036452f73f0e8d7ebab7af9fcb2497ccb Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= <git@rmr.ninja>
Date: Wed, 7 Feb 2024 10:59:32 +0100
Subject: [PATCH 73/73] Revert "MergeTree FINAL optimization diagnostics and
 settings"

---
 src/Core/Settings.h                           |  2 -
 src/Core/SettingsChangesHistory.h             |  4 +-
 src/Processors/QueryPlan/PartsSplitter.cpp    | 38 ++++---------------
 src/Processors/QueryPlan/PartsSplitter.h      |  3 +-
 .../QueryPlan/ReadFromMergeTree.cpp           |  6 +--
 5 files changed, 11 insertions(+), 42 deletions(-)

diff --git a/src/Core/Settings.h b/src/Core/Settings.h
index 221f8a01966..a433d523358 100644
--- a/src/Core/Settings.h
+++ b/src/Core/Settings.h
@@ -215,8 +215,6 @@ class IColumn;
     M(UInt64, merge_tree_max_rows_to_use_cache, (128 * 8192), "The maximum number of rows per request, to use the cache of uncompressed data. If the request is large, the cache is not used. (For large queries not to flush out the cache.)", 0) \
     M(UInt64, merge_tree_max_bytes_to_use_cache, (192 * 10 * 1024 * 1024), "The maximum number of bytes per request, to use the cache of uncompressed data. If the request is large, the cache is not used. (For large queries not to flush out the cache.)", 0) \
     M(Bool, do_not_merge_across_partitions_select_final, false, "Merge parts only in one partition in select final", 0) \
-    M(Bool, split_parts_ranges_into_intersecting_and_non_intersecting_final, true, "Split parts ranges into intersecting and non intersecting during FINAL optimization", 0) \
-    M(Bool, split_intersecting_parts_ranges_into_layers_final, true, "Split intersecting parts ranges into layers during FINAL optimization", 0) \
     M(Bool, allow_experimental_inverted_index, false, "If it is set to true, allow to use experimental inverted index.", 0) \
     \
     M(UInt64, mysql_max_rows_to_insert, 65536, "The maximum number of rows in MySQL batch insertion of the MySQL storage engine", 0) \
diff --git a/src/Core/SettingsChangesHistory.h b/src/Core/SettingsChangesHistory.h
index 27412ec20a4..a1c1523aaa9 100644
--- a/src/Core/SettingsChangesHistory.h
+++ b/src/Core/SettingsChangesHistory.h
@@ -116,9 +116,7 @@ static std::map<ClickHouseVersion, SettingsChangesHistory::SettingsChanges> sett
     {"23.12", {{"allow_suspicious_ttl_expressions", true, false, "It is a new setting, and in previous versions the behavior was equivalent to allowing."},
               {"input_format_parquet_allow_missing_columns", false, true, "Allow missing columns in Parquet files by default"},
               {"input_format_orc_allow_missing_columns", false, true, "Allow missing columns in ORC files by default"},
-              {"input_format_arrow_allow_missing_columns", false, true, "Allow missing columns in Arrow files by default"},
-              {"split_parts_ranges_into_intersecting_and_non_intersecting_final", false, true, "Allow to split parts ranges into intersecting and non intersecting during FINAL optimization"},
-              {"split_intersecting_parts_ranges_into_layers_final", true, true, "Allow to split intersecting parts ranges into layers during FINAL optimization"}}},
+              {"input_format_arrow_allow_missing_columns", false, true, "Allow missing columns in Arrow files by default"}}},
     {"23.9", {{"optimize_group_by_constant_keys", false, true, "Optimize group by constant keys by default"},
               {"input_format_json_try_infer_named_tuples_from_objects", false, true, "Try to infer named Tuples from JSON objects by default"},
               {"input_format_json_read_numbers_as_strings", false, true, "Allow to read numbers as strings in JSON formats by default"},
diff --git a/src/Processors/QueryPlan/PartsSplitter.cpp b/src/Processors/QueryPlan/PartsSplitter.cpp
index 363fdca22c5..7c66c0cc8df 100644
--- a/src/Processors/QueryPlan/PartsSplitter.cpp
+++ b/src/Processors/QueryPlan/PartsSplitter.cpp
@@ -228,7 +228,7 @@ struct SplitPartsRangesResult
     RangesInDataParts intersecting_parts_ranges;
 };
 
-SplitPartsRangesResult splitPartsRanges(RangesInDataParts ranges_in_data_parts, const LoggerPtr & logger)
+SplitPartsRangesResult splitPartsRanges(RangesInDataParts ranges_in_data_parts)
 {
     /** Split ranges in data parts into intersecting ranges in data parts and non intersecting ranges in data parts.
       *
@@ -483,15 +483,10 @@ SplitPartsRangesResult splitPartsRanges(RangesInDataParts ranges_in_data_parts,
         intersecting_ranges_in_data_parts.end(),
         [](const auto & lhs, const auto & rhs) { return lhs.part_index_in_query < rhs.part_index_in_query; });
 
-    LOG_TEST(logger, "Non intersecting ranges in data parts {}", non_intersecting_ranges_in_data_parts.getDescriptions().describe());
-    LOG_TEST(logger, "Intersecting ranges in data parts {}", intersecting_ranges_in_data_parts.getDescriptions().describe());
-
     return {std::move(non_intersecting_ranges_in_data_parts), std::move(intersecting_ranges_in_data_parts)};
 }
 
-std::pair<std::vector<RangesInDataParts>, std::vector<Values>> splitIntersectingPartsRangesIntoLayers(RangesInDataParts intersecting_ranges_in_data_parts,
-    size_t max_layers,
-    const LoggerPtr & logger)
+std::pair<std::vector<RangesInDataParts>, std::vector<Values>> splitIntersectingPartsRangesIntoLayers(RangesInDataParts intersecting_ranges_in_data_parts, size_t max_layers)
 {
     // We will advance the iterator pointing to the mark with the smallest PK value until
     // there will be not less than rows_per_layer rows in the current layer (roughly speaking).
@@ -596,18 +591,8 @@ std::pair<std::vector<RangesInDataParts>, std::vector<Values>> splitIntersecting
         result_layers.back() = std::move(current_layer_builder.getCurrentRangesInDataParts());
     }
 
-    size_t result_layers_size = result_layers.size();
-    LOG_TEST(logger, "Split intersecting ranges into {} layers", result_layers_size);
-
-    for (size_t i = 0; i < result_layers_size; ++i)
+    for (auto & layer : result_layers)
     {
-        auto & layer = result_layers[i];
-
-        LOG_TEST(logger, "Layer {} {} filter values in ({}, {}])",
-            i,
-            layer.getDescriptions().describe(),
-            i ? ::toString(borders[i - 1]) : "-inf", i < borders.size() ? ::toString(borders[i]) : "+inf");
-
         std::stable_sort(
             layer.begin(),
             layer.end(),
@@ -727,32 +712,23 @@ SplitPartsWithRangesByPrimaryKeyResult splitPartsWithRangesByPrimaryKey(
     size_t max_layers,
     ContextPtr context,
     ReadingInOrderStepGetter && in_order_reading_step_getter,
-    bool split_parts_ranges_into_intersecting_and_non_intersecting_final,
-    bool split_intersecting_parts_ranges_into_layers)
+    bool force_process_all_ranges)
 {
     if (max_layers <= 1)
         throw Exception(ErrorCodes::LOGICAL_ERROR, "max_layer should be greater than 1");
 
-    auto logger = getLogger("PartsSplitter");
-
     SplitPartsWithRangesByPrimaryKeyResult result;
 
     RangesInDataParts intersecting_parts_ranges = std::move(parts);
 
-    if (split_parts_ranges_into_intersecting_and_non_intersecting_final)
+    if (!force_process_all_ranges)
     {
-        SplitPartsRangesResult split_result = splitPartsRanges(intersecting_parts_ranges, logger);
+        SplitPartsRangesResult split_result = splitPartsRanges(intersecting_parts_ranges);
         result.non_intersecting_parts_ranges = std::move(split_result.non_intersecting_parts_ranges);
         intersecting_parts_ranges = std::move(split_result.intersecting_parts_ranges);
     }
 
-    if (!split_intersecting_parts_ranges_into_layers)
-    {
-        result.merging_pipes.emplace_back(in_order_reading_step_getter(intersecting_parts_ranges));
-        return result;
-    }
-
-    auto && [layers, borders] = splitIntersectingPartsRangesIntoLayers(intersecting_parts_ranges, max_layers, logger);
+    auto && [layers, borders] = splitIntersectingPartsRangesIntoLayers(intersecting_parts_ranges, max_layers);
     auto filters = buildFilters(primary_key, borders);
     result.merging_pipes.resize(layers.size());
 
diff --git a/src/Processors/QueryPlan/PartsSplitter.h b/src/Processors/QueryPlan/PartsSplitter.h
index 9bceb344589..f1ed1cb0b9c 100644
--- a/src/Processors/QueryPlan/PartsSplitter.h
+++ b/src/Processors/QueryPlan/PartsSplitter.h
@@ -34,6 +34,5 @@ SplitPartsWithRangesByPrimaryKeyResult splitPartsWithRangesByPrimaryKey(
     size_t max_layers,
     ContextPtr context,
     ReadingInOrderStepGetter && in_order_reading_step_getter,
-    bool split_parts_ranges_into_intersecting_and_non_intersecting,
-    bool split_intersecting_parts_ranges_into_layers);
+    bool force_process_all_ranges);
 }
diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp
index 8a04caede80..5ed56f59fc1 100644
--- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp
+++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp
@@ -1175,8 +1175,7 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsFinal(
 
                 /// Parts of non-zero level still may contain duplicate PK values to merge on FINAL if there's is_deleted column,
                 /// so we have to process all ranges. It would be more optimal to remove this flag and add an extra filtering step.
-                bool split_parts_ranges_into_intersecting_and_non_intersecting_final = settings.split_parts_ranges_into_intersecting_and_non_intersecting_final &&
-                    data.merging_params.is_deleted_column.empty();
+                bool force_process_all_ranges = !data.merging_params.is_deleted_column.empty();
 
                 SplitPartsWithRangesByPrimaryKeyResult split_ranges_result = splitPartsWithRangesByPrimaryKey(
                     metadata_for_reading->getPrimaryKey(),
@@ -1185,8 +1184,7 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsFinal(
                     num_streams,
                     context,
                     std::move(in_order_reading_step_getter),
-                    split_parts_ranges_into_intersecting_and_non_intersecting_final,
-                    settings.split_intersecting_parts_ranges_into_layers_final);
+                    force_process_all_ranges);
 
                 for (auto && non_intersecting_parts_range : split_ranges_result.non_intersecting_parts_ranges)
                     non_intersecting_parts_by_primary_key.push_back(std::move(non_intersecting_parts_range));