Merge branch 'master' into metric-async-insert-bytes

src/Interpreters/ExpressionAnalyzer.cpp

@@ -100,20 +100,9 @@ bool checkPositionalArguments(ASTPtr & argument, const ASTSelectQuery * select_q
 {
     auto columns = select_query->select()->children;
 
-    const auto * group_by_expr_with_alias = dynamic_cast<const ASTWithAlias *>(argument.get());
+    const auto * expr_with_alias = dynamic_cast<const ASTWithAlias *>(argument.get());
-    if (group_by_expr_with_alias && !group_by_expr_with_alias->alias.empty())
+    if (expr_with_alias && !expr_with_alias->alias.empty())
-    {
-        for (const auto & column : columns)
-        {
-            const auto * col_with_alias = dynamic_cast<const ASTWithAlias *>(column.get());
-            if (col_with_alias)
-            {
-                const auto & alias = col_with_alias->alias;
-                if (!alias.empty() && alias == group_by_expr_with_alias->alias)
         return false;
-            }
-        }
-    }
 
     const auto * ast_literal = typeid_cast<const ASTLiteral *>(argument.get());
     if (!ast_literal)
@@ -130,7 +119,7 @@ bool checkPositionalArguments(ASTPtr & argument, const ASTSelectQuery * select_q
                         pos, columns.size());
 
     const auto & column = columns[--pos];
-    if (typeid_cast<const ASTIdentifier *>(column.get()))
+    if (typeid_cast<const ASTIdentifier *>(column.get()) || typeid_cast<const ASTLiteral *>(column.get()))
     {
         argument = column->clone();
     }
@@ -1324,8 +1313,10 @@ ActionsDAGPtr SelectQueryExpressionAnalyzer::appendOrderBy(ExpressionActionsChai
             throw Exception("Bad ORDER BY expression AST", ErrorCodes::UNKNOWN_TYPE_OF_AST_NODE);
 
         if (getContext()->getSettingsRef().enable_positional_arguments)
+        {
             replaceForPositionalArguments(ast->children.at(0), select_query, ASTSelectQuery::Expression::ORDER_BY);
         }
+    }
 
     getRootActions(select_query->orderBy(), only_types, step.actions());
 

src/Interpreters/HashJoin.cpp

@@ -962,18 +962,29 @@ public:
             /// If it's joinGetOrNull, we need to wrap not-nullable columns in StorageJoin.
             for (size_t j = 0, size = right_indexes.size(); j < size; ++j)
             {
-                const auto & column = *block.getByPosition(right_indexes[j]).column;
+                auto column_from_block = block.getByPosition(right_indexes[j]);
-                if (auto * nullable_col = typeid_cast<ColumnNullable *>(columns[j].get()); nullable_col && !column.isNullable())
+                if (type_name[j].type->lowCardinality() != column_from_block.type->lowCardinality())
-                    nullable_col->insertFromNotNullable(column, row_num);
+                {
+                    JoinCommon::changeLowCardinalityInplace(column_from_block);
+                }
+
+                if (auto * nullable_col = typeid_cast<ColumnNullable *>(columns[j].get());
+                    nullable_col && !column_from_block.column->isNullable())
+                    nullable_col->insertFromNotNullable(*column_from_block.column, row_num);
                 else
-                    columns[j]->insertFrom(column, row_num);
+                    columns[j]->insertFrom(*column_from_block.column, row_num);
             }
         }
         else
         {
             for (size_t j = 0, size = right_indexes.size(); j < size; ++j)
             {
-                columns[j]->insertFrom(*block.getByPosition(right_indexes[j]).column, row_num);
+                auto column_from_block = block.getByPosition(right_indexes[j]);
+                if (type_name[j].type->lowCardinality() != column_from_block.type->lowCardinality())
+                {
+                    JoinCommon::changeLowCardinalityInplace(column_from_block);
+                }
+                columns[j]->insertFrom(*column_from_block.column, row_num);
             }
         }
     }
@@ -1013,6 +1024,7 @@ private:
 
     void addColumn(const ColumnWithTypeAndName & src_column, const std::string & qualified_name)
     {
+
         columns.push_back(src_column.column->cloneEmpty());
         columns.back()->reserve(src_column.column->size());
         type_name.emplace_back(src_column.type, src_column.name, qualified_name);

src/Interpreters/join_common.cpp

@@ -326,9 +326,10 @@ ColumnRawPtrMap materializeColumnsInplaceMap(Block & block, const Names & names)
 
     for (const auto & column_name : names)
     {
-        auto & column = block.getByName(column_name).column;
+        auto & column = block.getByName(column_name);
-        column = recursiveRemoveLowCardinality(column->convertToFullColumnIfConst());
+        column.column = recursiveRemoveLowCardinality(column.column->convertToFullColumnIfConst());
-        ptrs[column_name] = column.get();
+        column.type = recursiveRemoveLowCardinality(column.type);
+        ptrs[column_name] = column.column.get();
     }
 
     return ptrs;

src/Storages/MergeTree/MergeFromLogEntryTask.cpp

@@ -279,14 +279,17 @@ bool MergeFromLogEntryTask::finalize(ReplicatedMergeMutateTaskBase::PartLogWrite
             ProfileEvents::increment(ProfileEvents::DataAfterMergeDiffersFromReplica);
 
             LOG_ERROR(log,
-                "{}. Data after merge is not byte-identical to data on another replicas. There could be several"
+                "{}. Data after merge is not byte-identical to data on another replicas. There could be several reasons:"
-                " reasons: 1. Using newer version of compression library after server update. 2. Using another"
+                " 1. Using newer version of compression library after server update."
-                " compression method. 3. Non-deterministic compression algorithm (highly unlikely). 4."
+                " 2. Using another compression method."
-                " Non-deterministic merge algorithm due to logical error in code. 5. Data corruption in memory due"
+                " 3. Non-deterministic compression algorithm (highly unlikely)."
-                " to bug in code. 6. Data corruption in memory due to hardware issue. 7. Manual modification of"
+                " 4. Non-deterministic merge algorithm due to logical error in code."
-                " source data after server startup. 8. Manual modification of checksums stored in ZooKeeper. 9."
+                " 5. Data corruption in memory due to bug in code."
-                " Part format related settings like 'enable_mixed_granularity_parts' are different on different"
+                " 6. Data corruption in memory due to hardware issue."
-                " replicas. We will download merged part from replica to force byte-identical result.",
+                " 7. Manual modification of source data after server startup."
+                " 8. Manual modification of checksums stored in ZooKeeper."
+                " 9. Part format related settings like 'enable_mixed_granularity_parts' are different on different replicas."
+                " We will download merged part from replica to force byte-identical result.",
                 getCurrentExceptionMessage(false));
 
             write_part_log(ExecutionStatus::fromCurrentException());

src/Storages/MergeTree/MutateFromLogEntryTask.cpp

@@ -185,7 +185,8 @@ bool MutateFromLogEntryTask::finalize(ReplicatedMergeMutateTaskBase::PartLogWrit
 
             ProfileEvents::increment(ProfileEvents::DataAfterMutationDiffersFromReplica);
 
-            LOG_ERROR(log, "{}. Data after mutation is not byte-identical to data on another replicas. We will download merged part from replica to force byte-identical result.", getCurrentExceptionMessage(false));
+            LOG_ERROR(log, "{}. Data after mutation is not byte-identical to data on another replicas. "
+                           "We will download merged part from replica to force byte-identical result.", getCurrentExceptionMessage(false));
 
             write_part_log(ExecutionStatus::fromCurrentException());
 

src/Storages/StorageReplicatedMergeTree.cpp

@@ -1312,10 +1312,14 @@ void StorageReplicatedMergeTree::checkPartChecksumsAndAddCommitOps(const zkutil:
 
         if (replica_part_header.getColumnsHash() != local_part_header.getColumnsHash())
         {
-            /// Either it's a bug or ZooKeeper contains broken data.
+            /// Currently there are two (known) cases when it may happen:
-            /// TODO Fix KILL MUTATION and replace CHECKSUM_DOESNT_MATCH with LOGICAL_ERROR
+            ///  - KILL MUTATION query had removed mutation before all replicas have executed assigned MUTATE_PART entries.
-            /// (some replicas may skip killed mutation even if it was executed on other replicas)
+            ///    Some replicas may skip this mutation and update part version without actually applying any changes.
-            throw Exception(ErrorCodes::CHECKSUM_DOESNT_MATCH, "Part {} from {} has different columns hash", part_name, replica);
+            ///    It leads to mismatching checksum if changes were applied on other replicas.
+            ///  - ALTER_METADATA and MERGE_PARTS were reordered on some replicas.
+            ///    It may lead to different number of columns in merged parts on these replicas.
+            throw Exception(ErrorCodes::CHECKSUM_DOESNT_MATCH, "Part {} from {} has different columns hash "
+                            "(it may rarely happen on race condition with KILL MUTATION or ALTER COLUMN).", part_name, replica);
         }
 
         replica_part_header.getChecksums().checkEqual(local_part_header.getChecksums(), true);

tests/queries/0_stateless/02006_test_positional_arguments.reference

@@ -111,3 +111,9 @@ select substr('aaaaaaaaaaaaaa', 8) as a  group by a;
 aaaaaaa
 select substr('aaaaaaaaaaaaaa', 8) as a  group by substr('aaaaaaaaaaaaaa', 8);
 aaaaaaa
+select b from (select 5 as a, 'Hello' as b order by a);
+Hello
+select b from (select 5 as a, 'Hello' as b group by a);
+Hello
+select b from (select 5 as a, 'Hello' as b order by 1);
+Hello

tests/queries/0_stateless/02006_test_positional_arguments.sql

@@ -45,3 +45,6 @@ explain syntax select plus(1, 1) as a group by a;
 select substr('aaaaaaaaaaaaaa', 8) as a  group by a;
 select substr('aaaaaaaaaaaaaa', 8) as a  group by substr('aaaaaaaaaaaaaa', 8);
 
+select b from (select 5 as a, 'Hello' as b order by a);
+select b from (select 5 as a, 'Hello' as b group by a);
+select b from (select 5 as a, 'Hello' as b order by 1);

tests/queries/0_stateless/02244_lowcardinality_hash_join.reference

@@ -0,0 +1,4 @@
+x	x
+x	x
+x	x
+x	x

tests/queries/0_stateless/02244_lowcardinality_hash_join.sql

@@ -0,0 +1,27 @@
+-- Tags: no-parallel
+DROP TABLE IF EXISTS lc_table;
+
+CREATE TABLE lc_table
+(
+    col LowCardinality(String)
+) ENGINE=TinyLog;
+
+INSERT INTO lc_table VALUES('x');
+
+SELECT *
+FROM lc_table
+INNER JOIN lc_table AS lc_table2 ON lc_table.col = lc_table2.col;
+
+SELECT *
+FROM lc_table
+INNER JOIN lc_table AS lc_table2 ON CAST(lc_table.col AS String) = CAST(lc_table2.col AS String);
+
+SELECT *
+FROM lc_table
+INNER JOIN lc_table AS lc_table2 ON (lc_table.col = lc_table2.col) OR (lc_table.col = lc_table2.col);
+
+SELECT *
+FROM lc_table
+INNER JOIN lc_table AS lc_table2 ON (CAST(lc_table.col AS String) = CAST(lc_table2.col AS String)) OR (CAST(lc_table.col AS String) = CAST(lc_table2.col AS String));
+
+DROP TABLE IF EXISTS lc_table;

tests/queries/0_stateless/02245_weird_partitions_pruning.reference

@@ -0,0 +1,14 @@
+202112-0	(202112,0)
+202201-0	(202201,0)
+202301-0	(202301,0)
+202112-0	2021-12-31 22:30:00	2021-12-31 22:30:00	2021-12-31 14:30:00	2021-12-31 14:30:00	1000
+202201-0	2022-01-01 00:30:00	2022-01-31 22:30:00	2021-12-31 16:30:00	2022-01-31 14:30:00	2000
+202301-0	2023-01-31 22:30:00	2023-01-31 22:30:00	2023-01-31 14:30:00	2023-01-31 14:30:00	1000
+202112-0
+default	weird_partitions_02245	1	1000	1
+202201-0
+default	weird_partitions_02245	1	2000	1
+202112-0
+202201-0
+default	weird_partitions_02245	2	3000	2
+default	weird_partitions_02245	0	0	0

tests/queries/0_stateless/02245_weird_partitions_pruning.sql

@@ -0,0 +1,61 @@
+-- We use a hack - partition by ignore(d1). In some cases there are two columns 
+-- not fully correlated (<1) (date_begin - date_end or datetime - datetime_in_TZ_with_DST)
+-- If we partition by these columns instead of one it will be twice more partitions.
+-- Partition by (.., ignore(d1)) allows to partition by the first column but build 
+-- min_max indexes for both column, so partition pruning works for both columns.
+-- It's very similar to min_max skip index but gives bigger performance boost, 
+-- because partition pruning happens on very early query stage.
+
+
+DROP TABLE IF EXISTS weird_partitions_02245;
+
+CREATE TABLE weird_partitions_02245(d DateTime, d1 DateTime default d - toIntervalHour(8), id Int64)
+Engine=MergeTree
+PARTITION BY (toYYYYMM(toDateTime(d)), ignore(d1))
+ORDER BY id;
+
+INSERT INTO weird_partitions_02245(d, id)
+SELECT
+    toDateTime('2021-12-31 22:30:00') AS d,
+    number
+FROM numbers(1000);
+
+INSERT INTO weird_partitions_02245(d, id)
+SELECT
+    toDateTime('2022-01-01 00:30:00') AS d,
+    number
+FROM numbers(1000);
+
+INSERT INTO weird_partitions_02245(d, id)
+SELECT
+    toDateTime('2022-01-31 22:30:00') AS d,
+    number
+FROM numbers(1000);
+
+INSERT INTO weird_partitions_02245(d, id)
+SELECT
+    toDateTime('2023-01-31 22:30:00') AS d,
+    number
+FROM numbers(1000);
+
+OPTIMIZE TABLE weird_partitions_02245;
+OPTIMIZE TABLE weird_partitions_02245;
+
+SELECT DISTINCT _partition_id, _partition_value FROM weird_partitions_02245 ORDER BY _partition_id ASC;
+
+SELECT _partition_id, min(d), max(d), min(d1), max(d1), count() FROM weird_partitions_02245 GROUP BY _partition_id ORDER BY _partition_id ASC;
+
+select  DISTINCT _partition_id from weird_partitions_02245 where d >= '2021-12-31 00:00:00' and d < '2022-01-01 00:00:00' ORDER BY _partition_id;
+explain estimate select  DISTINCT _partition_id from weird_partitions_02245 where d >= '2021-12-31 00:00:00' and d < '2022-01-01 00:00:00';
+
+select  DISTINCT _partition_id from weird_partitions_02245 where d >= '2022-01-01 00:00:00' and  d1 >= '2021-12-31 00:00:00' and d1 < '2022-01-01 00:00:00' ORDER BY _partition_id;;
+explain estimate select  DISTINCT _partition_id from weird_partitions_02245 where d >= '2022-01-01 00:00:00' and d1 >= '2021-12-31 00:00:00' and d1 < '2022-01-01 00:00:00';
+
+select  DISTINCT _partition_id from weird_partitions_02245 where d1 >= '2021-12-31 00:00:00' and d1 < '2022-01-01 00:00:00' ORDER BY _partition_id;;
+explain estimate select  DISTINCT _partition_id from weird_partitions_02245 where d1 >= '2021-12-31 00:00:00' and d1 < '2022-01-01 00:00:00';
+
+select  DISTINCT _partition_id from weird_partitions_02245 where d >= '2022-01-01 00:00:00' and  d1 >= '2021-12-31 00:00:00' and d1 < '2020-01-01 00:00:00' ORDER BY _partition_id;;
+explain estimate select  DISTINCT _partition_id from weird_partitions_02245 where d >= '2022-01-01 00:00:00' and d1 >= '2021-12-31 00:00:00' and d1 < '2020-01-01 00:00:00';
+
+DROP TABLE weird_partitions_02245;
+