Merge pull request #11818 from ClickHouse/fix-bad-code-optimizations

Fix another 10% of bad code.

src/Core/Settings.h

@@ -158,7 +158,7 @@ struct Settings : public SettingsCollection<Settings>
     M(SettingInt64, os_thread_priority, 0, "If non zero - set corresponding 'nice' value for query processing threads. Can be used to adjust query priority for OS scheduler.", 0) \
     \
     M(SettingBool, log_queries, 1, "Log requests and write the log to the system table.", 0) \
-    M(SettingLogQueriesType, log_queries_min_type, QueryLogElementType::QUERY_START, "query_log minimal type to log, possible values (from low to high): QUERY_START, QUERY_FINISH, EXCEPTION_BEFORE_START, EXCEPTION_WHILE_PROCESSING.", 0) \
+    M(SettingLogQueriesType, log_queries_min_type, QueryLogElementType::QUERY_START, "Minimal type in query_log to log, possible values (from low to high): QUERY_START, QUERY_FINISH, EXCEPTION_BEFORE_START, EXCEPTION_WHILE_PROCESSING.", 0) \
     M(SettingUInt64, log_queries_cut_to_length, 100000, "If query length is greater than specified threshold (in bytes), then cut query when writing to query log. Also limit length of printed query in ordinary text log.", 0) \
     \
     M(SettingDistributedProductMode, distributed_product_mode, DistributedProductMode::DENY, "How are distributed subqueries performed inside IN or JOIN sections?", IMPORTANT) \
@@ -360,7 +360,7 @@ struct Settings : public SettingsCollection<Settings>
     M(SettingBool, 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) \
     M(SettingBool, optimize_trivial_count_query, true, "Process trivial 'SELECT count() FROM table' query from metadata.", 0) \
     M(SettingUInt64, mutations_sync, 0, "Wait for synchronous execution of ALTER TABLE UPDATE/DELETE queries (mutations). 0 - execute asynchronously. 1 - wait current server. 2 - wait all replicas if they exist.", 0) \
-    M(SettingBool, optimize_any_input, true, "removal of any operations from Any", 0) \
+    M(SettingBool, optimize_move_functions_out_of_any, true, "Move functions out of aggregate functions 'any', 'anyLast'.", 0) \
     M(SettingBool, optimize_arithmetic_operations_in_aggregate_functions, true, "Move arithmetic operations out of aggregation functions", 0) \
     M(SettingBool, optimize_duplicate_order_by_and_distinct, true, "Remove duplicate ORDER BY and DISTINCT if it's possible", 0) \
     M(SettingBool, optimize_if_chain_to_miltiif, false, "Replace if(cond1, then1, if(cond2, ...)) chains to multiIf. Currently it's not beneficial for numeric types.", 0) \

src/Interpreters/SyntaxAnalyzer.cpp

@@ -394,11 +394,8 @@ GroupByKeysInfo getGroupByKeysInfo(ASTs & group_keys)
 }
 
 ///eliminate functions of other GROUP BY keys
-void optimizeGroupByFunctionKeys(ASTSelectQuery * select_query, bool optimize_group_by_function_keys)
+void optimizeGroupByFunctionKeys(ASTSelectQuery * select_query)
 {
-    if (!optimize_group_by_function_keys)
-        return;
-
     if (!select_query->groupBy())
         return;
 
@@ -446,11 +443,8 @@ void optimizeGroupByFunctionKeys(ASTSelectQuery * select_query, bool optimize_gr
 }
 
 /// Eliminates min/max/any-aggregators of functions of GROUP BY keys
-void optimizeAggregateFunctionsOfGroupByKeys(ASTSelectQuery * select_query, bool optimize_aggregators_of_group_by_keys)
+void optimizeAggregateFunctionsOfGroupByKeys(ASTSelectQuery * select_query)
 {
-    if (!optimize_aggregators_of_group_by_keys)
-        return;
-
     if (!select_query->groupBy())
         return;
 
@@ -493,15 +487,12 @@ void optimizeOrderBy(const ASTSelectQuery * select_query)
 }
 
 /// Optimize duplicate ORDER BY and DISTINCT
-void optimizeDuplicateOrderByAndDistinct(ASTPtr & query, bool optimize_duplicate_order_by_and_distinct, const Context & context)
+void optimizeDuplicateOrderByAndDistinct(ASTPtr & query, const Context & context)
 {
-    if (optimize_duplicate_order_by_and_distinct)
-    {
-        DuplicateOrderByVisitor::Data order_by_data{context, false};
-        DuplicateOrderByVisitor(order_by_data).visit(query);
-        DuplicateDistinctVisitor::Data distinct_data{};
-        DuplicateDistinctVisitor(distinct_data).visit(query);
-    }
+    DuplicateOrderByVisitor::Data order_by_data{context, false};
+    DuplicateOrderByVisitor(order_by_data).visit(query);
+    DuplicateDistinctVisitor::Data distinct_data{};
+    DuplicateDistinctVisitor(distinct_data).visit(query);
 }
 
 /// Remove duplicate items from LIMIT BY.
@@ -574,14 +565,11 @@ void optimizeArithmeticOperationsInAgr(ASTPtr & query, bool optimize_arithmetic_
     }
 }
 
-void optimizeAnyInput(ASTPtr & query, bool optimize_any_input)
+void optimizeAnyInput(ASTPtr & query)
 {
-    if (optimize_any_input)
-    {
-        /// Removing arithmetic operations from functions
-        AnyInputVisitor::Data data = {};
-        AnyInputVisitor(data).visit(query);
-    }
+    /// Removing arithmetic operations from functions
+    AnyInputVisitor::Data data = {};
+    AnyInputVisitor(data).visit(query);
 }
 
 void getArrayJoinedColumns(ASTPtr & query, SyntaxAnalyzerResult & result, const ASTSelectQuery * select_query,
@@ -973,19 +961,23 @@ SyntaxAnalyzerResultPtr SyntaxAnalyzer::analyzeSelect(
         optimizeGroupBy(select_query, source_columns_set, context);
 
         /// GROUP BY functions of other keys elimination.
-        optimizeGroupByFunctionKeys(select_query, settings.optimize_group_by_function_keys);
+        if (settings.optimize_group_by_function_keys)
+            optimizeGroupByFunctionKeys(select_query);
 
         ///Move all operations out of any function
-        optimizeAnyInput(query, settings.optimize_any_input);
+        if (settings.optimize_move_functions_out_of_any)
+            optimizeAnyInput(query);
 
         /// Eliminate min/max/any aggregators of functions of GROUP BY keys
-        optimizeAggregateFunctionsOfGroupByKeys(select_query, settings.optimize_aggregators_of_group_by_keys);
+        if (settings.optimize_aggregators_of_group_by_keys)
+            optimizeAggregateFunctionsOfGroupByKeys(select_query);
 
         /// Remove duplicate items from ORDER BY.
         optimizeOrderBy(select_query);
 
         /// Remove duplicate ORDER BY and DISTINCT from subqueries.
-        optimizeDuplicateOrderByAndDistinct(query, settings.optimize_duplicate_order_by_and_distinct, context);
+        if (settings.optimize_duplicate_order_by_and_distinct)
+            optimizeDuplicateOrderByAndDistinct(query, context);
 
         /// Remove duplicated elements from LIMIT BY clause.
         optimizeLimitBy(select_query);

tests/queries/0_stateless/01321_aggregate_functions_of_group_by_keys.sql

@@ -1,6 +1,6 @@
 set optimize_aggregators_of_group_by_keys = 1;
 set enable_debug_queries = 1;
-set optimize_any_input = 0;
+set optimize_move_functions_out_of_any = 0;
 
 SELECT min(number % 2) AS a, max(number % 3) AS b FROM numbers(10000000) GROUP BY number % 2, number % 3 ORDER BY a, b;
 SELECT any(number % 2) AS a, anyLast(number % 3) AS b FROM numbers(10000000) GROUP BY number % 2, number % 3 ORDER BY a, b;

tests/queries/0_stateless/01322_any_input_optimize.sql

@@ -1,4 +1,4 @@
-SET optimize_any_input=1;
+SET optimize_move_functions_out_of_any=1;
 SET enable_debug_queries=1;
 SELECT any(number + number * 2) FROM  numbers(3, 10);
 ANALYZE SELECT any(number + number * 2) FROM  numbers(3, 10);

tests/queries/0_stateless/01324_settings_documentation.reference

@@ -0,0 +1 @@
+Settings description should start with capital letter

tests/queries/0_stateless/01324_settings_documentation.sql

@@ -0,0 +1,2 @@
+SELECT 'Settings description should start with capital letter';
+SELECT name, description FROM system.settings WHERE substring(description, 1, 1) != upper(substring(description, 1, 1));