Merge remote-tracking branch 'origin/master' into morton-utils

contrib/croaring

@@ -1 +1 @@
-Subproject commit e4a7ad5542746103e71ca8b5e56225baf0014c87
+Subproject commit 9b7cc0ff1c41e9457efb6228cfd2c538d0155303

docker/test/base/setup_export_logs.sh

@@ -16,7 +16,7 @@ CLICKHOUSE_CI_LOGS_USER=${CLICKHOUSE_CI_LOGS_USER:-ci}
 CLICKHOUSE_CI_LOGS_CLUSTER=${CLICKHOUSE_CI_LOGS_CLUSTER:-system_logs_export}
 
 EXTRA_COLUMNS=${EXTRA_COLUMNS:-"pull_request_number UInt32, commit_sha String, check_start_time DateTime('UTC'), check_name String, instance_type String, instance_id String, "}
-EXTRA_COLUMNS_EXPRESSION=${EXTRA_COLUMNS_EXPRESSION:-"CAST(0 AS Int32) AS pull_request_number, '' AS commit_sha, now() AS check_start_time, '' AS check_name, '' AS instance_type, '' AS instance_id"}
+EXTRA_COLUMNS_EXPRESSION=${EXTRA_COLUMNS_EXPRESSION:-"CAST(0 AS UInt32) AS pull_request_number, '' AS commit_sha, now() AS check_start_time, '' AS check_name, '' AS instance_type, '' AS instance_id"}
 EXTRA_ORDER_BY_COLUMNS=${EXTRA_ORDER_BY_COLUMNS:-"check_name, "}
 
 function __set_connection_args

docs/en/sql-reference/functions/arithmetic-functions.md

@@ -444,7 +444,7 @@ DB::Exception: Decimal result's scale is less than argument's one: While process
 
 ## byteSwap
 
-Reverses the bytes of an integer, i.e. changes its [endianness](https://en.wikipedia.org/wiki/Endianness). Currently, integers of up to 64 bit are supported.
+Reverses the bytes of an integer, i.e. changes its [endianness](https://en.wikipedia.org/wiki/Endianness).
 
 **Syntax**
 

programs/server/Server.cpp

@@ -1474,7 +1474,7 @@ try
 
     {
         std::lock_guard lock(servers_lock);
-        /// We should start interserver communications before (and more imporant shutdown after) tables.
+        /// We should start interserver communications before (and more important shutdown after) tables.
         /// Because server can wait for a long-running queries (for example in tcp_handler) after interserver handler was already shut down.
         /// In this case we will have replicated tables which are unable to send any parts to other replicas, but still can
         /// communicate with zookeeper, execute merges, etc.

src/Common/CaresPTRResolver.cpp

@@ -223,7 +223,7 @@ namespace DB
 
     void CaresPTRResolver::process_possible_timeout(ares_channel channel)
     {
-        /* Call ares_process() unconditonally here, even if we simply timed out
+        /* Call ares_process() unconditionally here, even if we simply timed out
         above, as otherwise the ares name resolve won't timeout! */
         ares_process_fd(channel, ARES_SOCKET_BAD, ARES_SOCKET_BAD);
     }

src/Common/OptimizedRegularExpression.cpp

@@ -418,7 +418,7 @@ finish:
         /// this two vals are useless, xxx|xxx cannot be trivial nor prefix.
         bool next_is_trivial = true;
         pos = analyzeImpl(regexp, pos, required_substring, next_is_trivial, next_alternatives);
-        /// For xxx|xxx|xxx, we only conbine the alternatives and return a empty required_substring.
+        /// For xxx|xxx|xxx, we only combine the alternatives and return a empty required_substring.
         if (next_alternatives.empty() || shortest_literal_length(next_alternatives) < required_substring.literal.size())
         {
             global_alternatives.push_back(required_substring);

src/Common/SpaceSaving.h

@@ -321,7 +321,7 @@ protected:
         percolate(ptr);
     }
 
-    // This is equivallent to one step of bubble sort
+    // This is equivalent to one step of bubble sort
     void percolate(Counter * counter)
     {
         while (counter->slot > 0)

src/Coordination/ZooKeeperDataReader.cpp

@@ -284,7 +284,7 @@ void deserializeLogMagic(ReadBuffer & in)
 ///     strange, that this 550 bytes obviously was a part of Create transaction,
 ///     but the operation code was -1. We have added debug prints to original
 ///     zookeeper (3.6.3) and found that it just reads 550 bytes of this "Error"
-///     transaction, tooks the first 4 bytes as an error code (it was 79, non
+///     transaction, took the first 4 bytes as an error code (it was 79, non
 ///     existing code) and skip all remaining 546 bytes. NOTE: it looks like a bug
 ///     in ZooKeeper.
 ///

src/DataTypes/ObjectUtils.cpp

@@ -339,7 +339,7 @@ static DataTypePtr getLeastCommonTypeForObject(const DataTypes & types, bool che
         for (size_t i = 1; i < subtypes.size(); ++i)
             if (first_dim != getNumberOfDimensions(*subtypes[i]))
                 throw Exception(ErrorCodes::TYPE_MISMATCH,
-                    "Uncompatible types of subcolumn '{}': {} and {}",
+                    "Incompatible types of subcolumn '{}': {} and {}",
                     key.getPath(), subtypes[0]->getName(), subtypes[i]->getName());
 
         tuple_paths.emplace_back(key);

src/Disks/ObjectStorages/DiskObjectStorageMetadata.h

@@ -33,7 +33,7 @@ private:
 
     /// Number of references (hardlinks) to this metadata file.
     ///
-    /// FIXME: Why we are tracking it explicetly, without
+    /// FIXME: Why we are tracking it explicitly, without
     /// info from filesystem????
     uint32_t ref_count = 0;
 

src/Functions/byteSwap.cpp

@@ -17,6 +17,15 @@ T byteSwap(T x)
     return std::byteswap(x);
 }
 
+template <typename T>
+requires std::is_same_v<T, UInt128> || std::is_same_v<T, Int128> || std::is_same_v<T, UInt256> || std::is_same_v<T, Int256>
+T byteSwap(T x)
+{
+    T dest;
+    reverseMemcpy(&dest, &x, sizeof(T));
+    return dest;
+}
+
 template <typename T>
 T byteSwap(T)
 {
@@ -55,7 +64,7 @@ REGISTER_FUNCTION(ByteSwap)
     factory.registerFunction<FunctionByteSwap>(
         FunctionDocumentation{
             .description = R"(
-Reverses the bytes of an integer, i.e. changes its [endianness](https://en.wikipedia.org/wiki/Endianness). Currently, integers of up to 64 bit are supported.
+Reverses the bytes of an integer, i.e. changes its [endianness](https://en.wikipedia.org/wiki/Endianness).
 
 **Example**
 

src/IO/tests/gtest_DateTime64_parsing_and_writing.cpp

@@ -182,7 +182,7 @@ INSTANTIATE_TEST_SUITE_P(Basic,
             DateLUT::instance("Europe/Minsk")
         },
         {
-            "When scale is 0, subsecond part (and separtor) is missing from string",
+            "When scale is 0, subsecond part (and separator) is missing from string",
             "2019-09-16 19:20:17",
             1568650817ULL,
             0,
@@ -197,4 +197,3 @@ INSTANTIATE_TEST_SUITE_P(Basic,
         }
     })
 );
-

src/Interpreters/Access/InterpreterGrantQuery.cpp

@@ -166,7 +166,7 @@ namespace
         access_to_revoke.grant(elements_to_revoke);
         access_to_revoke.makeIntersection(all_granted_access);
 
-        /// Build more accurate list of elements to revoke, now we use an intesection of the initial list of elements to revoke
+        /// Build more accurate list of elements to revoke, now we use an intersection of the initial list of elements to revoke
         /// and all the granted access rights to these grantees.
         bool grant_option = !elements_to_revoke.empty() && elements_to_revoke[0].grant_option;
         elements_to_revoke.clear();

src/Interpreters/ActionsVisitor.cpp

@@ -3,6 +3,7 @@
 #include <Common/quoteString.h>
 #include <Common/typeid_cast.h>
 #include <Common/FieldVisitorsAccurateComparison.h>
+#include <Common/checkStackSize.h>
 
 #include <Core/ColumnNumbers.h>
 #include <Core/ColumnWithTypeAndName.h>
@@ -691,6 +692,8 @@ bool ActionsMatcher::needChildVisit(const ASTPtr & node, const ASTPtr & child)
 
 void ActionsMatcher::visit(const ASTPtr & ast, Data & data)
 {
+    checkStackSize();
+
     if (const auto * identifier = ast->as<ASTIdentifier>())
         visit(*identifier, ast, data);
     else if (const auto * table = ast->as<ASTTableIdentifier>())

src/Interpreters/ConvertStringsToEnumVisitor.cpp

@@ -11,7 +11,7 @@ namespace DB
 namespace
 {
 
-/// @note We place strings in ascending order here under the assumption it colud speed up String to Enum conversion.
+/// @note We place strings in ascending order here under the assumption it could speed up String to Enum conversion.
 String makeStringsEnum(const std::set<String> & values)
 {
     String enum_string = "Enum8(";

src/Interpreters/QueryAliasesVisitor.cpp

@@ -140,7 +140,7 @@ void QueryAliasesMatcher<T>::visitOther(const ASTPtr & ast, Data & data)
         information for our ast node with query string. And this alias will be dropped because prefer_alias_to_column_name for ASTWIthAlias
         by default is false.
 
-        It is imporant that subquery can be converted to literal during ExecuteScalarSubqueriesVisitor.
+        It is important that subquery can be converted to literal during ExecuteScalarSubqueriesVisitor.
         And code below check if we previously set for subquery alias as _subquery, and if it is true
         then set prefer_alias_to_column_name = true for node that was optimized during ExecuteScalarSubqueriesVisitor.
      */

src/Parsers/ExpressionListParsers.cpp

@@ -2435,6 +2435,7 @@ const std::vector<std::pair<std::string_view, Operator>> ParserExpressionImpl::o
     {"||",            Operator("concat",          10, 2, OperatorType::Mergeable)},
     {"+",             Operator("plus",            11, 2)},
     {"-",             Operator("minus",           11, 2)},
+    {"−",             Operator("minus",           11, 2)},
     {"*",             Operator("multiply",        12, 2)},
     {"/",             Operator("divide",          12, 2)},
     {"%",             Operator("modulo",          12, 2)},
@@ -2448,7 +2449,8 @@ const std::vector<std::pair<std::string_view, Operator>> ParserExpressionImpl::o
 const std::vector<std::pair<std::string_view, Operator>> ParserExpressionImpl::unary_operators_table
 {
     {"NOT",           Operator("not",             5,  1)},
-    {"-",             Operator("negate",          13, 1)}
+    {"-",             Operator("negate",          13, 1)},
+    {"−",             Operator("negate",          13, 1)}
 };
 
 const Operator ParserExpressionImpl::finish_between_operator("", 8, 0, OperatorType::FinishBetween);

src/Parsers/Kusto/KustoFunctions/IParserKQLFunction.cpp

@@ -110,7 +110,7 @@ String IParserKQLFunction::generateUniqueIdentifier()
     // This particular random generator hits each number exactly once before looping over.
     // Because of this, it's sufficient for queries consisting of up to 2^16 (= 65536) distinct function calls.
     // Reference: https://www.pcg-random.org/using-pcg-cpp.html#insecure-generators
-    static pcg32_once_insecure random_generator;
+    static thread_local pcg32_once_insecure random_generator;
     return std::to_string(random_generator());
 }
 

src/Parsers/Lexer.cpp

@@ -426,7 +426,17 @@ Token Lexer::nextTokenImpl()
                 return Token(TokenType::VerticalDelimiter, token_begin, ++pos);
             return Token(TokenType::Error, token_begin, pos);
         }
-
+        case '\xE2':
+        {
+            /// Mathematical minus symbol, UTF-8
+            if (pos + 3 <= end && pos[1] == '\x88' && pos[2] == '\x92')
+            {
+                pos += 3;
+                return Token(TokenType::Minus, token_begin, pos);
+            }
+            /// Other characters starting at E2 can be parsed, see skipWhitespacesUTF8
+            [[fallthrough]];
+        }
         default:
             if (*pos == '$')
             {

src/Parsers/parseQuery.cpp

@@ -263,6 +263,10 @@ ASTPtr tryParseQuery(
     const auto last_token = token_iterator.max();
     _out_query_end = last_token.end;
 
+    /// Also check on the AST level, because the generated AST depth can be greater than the recursion depth of the parser.
+    if (res && max_parser_depth)
+        res->checkDepth(max_parser_depth);
+
     ASTInsertQuery * insert = nullptr;
     if (parse_res)
     {

src/Processors/Formats/Impl/Parquet/Write.cpp

@@ -226,7 +226,7 @@ struct StatisticsStringRef
 /// or [element of ColumnString] -> std::string_view.
 /// We do this conversion in small batches rather than all at once, just before encoding the batch,
 /// in hopes of getting better performance through cache locality.
-/// The Coverter* structs below are responsible for that.
+/// The Converter* structs below are responsible for that.
 /// When conversion is not needed, getBatch() will just return pointer into original data.
 
 template <typename Col, typename To, typename MinMaxType = typename std::conditional_t<

src/Storages/MergeTree/MergeTreeRangeReader.cpp

@@ -1174,7 +1174,7 @@ void MergeTreeRangeReader::fillPartOffsetColumn(ReadResult & result, UInt64 lead
     UInt64 * pos = vec.data();
     UInt64 * end = &vec[num_rows];
 
-    /// Fill the reamining part of the previous range (it was started in the previous read request).
+    /// Fill the remaining part of the previous range (it was started in the previous read request).
     while (pos < end && leading_begin_part_offset < leading_end_part_offset)
         *pos++ = leading_begin_part_offset++;
 

src/Storages/MergeTree/MergeTreeReaderCompact.cpp

@@ -152,7 +152,7 @@ void MergeTreeReaderCompact::fillColumnPositions()
             auto [name_in_storage, subcolumn_name] = Nested::splitName(column_to_read.name);
 
             /// If it is a part of Nested, we need to get the column from
-            /// storage metatadata which is converted to Nested type with subcolumns.
+            /// storage metadata which is converted to Nested type with subcolumns.
             /// It is required for proper counting of shared streams.
             if (!subcolumn_name.empty())
             {

src/Storages/MergeTree/ReplicatedMergeTreeQueue.cpp

@@ -749,7 +749,7 @@ QueueRepresentation getQueueRepresentation(const std::list<ReplicatedMergeTreeLo
         const auto & key = entry->znode_name;
         switch (entry->type)
         {
-            /// explicetely specify all types of entries without default, so if
+            /// explicitly specify all types of entries without default, so if
             /// someone decide to add new type it will produce a compiler warning (error in our case)
             case LogEntryType::GET_PART:
             case LogEntryType::ATTACH_PART:

src/Storages/MergeTree/ReplicatedMergeTreeQueue.h

@@ -240,7 +240,7 @@ private:
     /// by first argument. If remove_part == true, than also remove part itself.
     /// Both negative flags will throw exception.
     ///
-    /// Part removed from mutations which satisfy contitions:
+    /// Part removed from mutations which satisfy conditions:
     /// block_number > part.getDataVersion()
     /// or block_number == part.getDataVersion()
     ///    ^ (this may happen if we downloaded mutated part from other replica)

src/Storages/VirtualColumnUtils.cpp

@@ -1,3 +1,4 @@
+#include <algorithm>
 #include <memory>
 #include <Core/NamesAndTypes.h>
 #include <Core/TypeId.h>
@@ -81,14 +82,33 @@ bool extractFunctions(const ASTPtr & expression, const std::function<bool(const
         }
         else if (function->name == "or")
         {
-            bool ret = true;
+            bool ret = false;
             ASTs or_args;
             for (const auto & child : function->arguments->children)
-                ret &= extractFunctions(child, is_constant, or_args);
-            /// We can keep condition only if it still OR condition (i.e. we
-            /// have dependent conditions for columns at both sides)
-            if (or_args.size() == 2)
+                ret |= extractFunctions(child, is_constant, or_args);
+
+            if (!or_args.empty())
+            {
+                /// In case of there are less number of arguments for which
+                /// is_constant() == true, we need to add always-true
+                /// implicitly to avoid breaking AND invariant.
+                ///
+                /// Consider the following:
+                ///
+                ///     ((value = 10) OR (_table = 'v2')) AND ((_table = 'v1') OR (value = 20))
+                ///
+                /// Without implicit always-true:
+                ///
+                ///     (_table = 'v2') AND (_table = 'v1')
+                ///
+                /// With:
+                ///
+                ///     (_table = 'v2' OR 1) AND (_table = 'v1' OR 1) -> (_table = 'v2') OR (_table = 'v1')
+                ///
+                if (or_args.size() != function->arguments->children.size())
+                    or_args.push_back(std::make_shared<ASTLiteral>(Field(1)));
                 result.push_back(makeASTForLogicalOr(std::move(or_args)));
+            }
             return ret;
         }
     }
@@ -165,8 +185,10 @@ bool prepareFilterBlockWithQuery(const ASTPtr & query, ContextPtr context, Block
     if (!select.where() && !select.prewhere())
         return unmodified;
 
-    // Provide input columns as constant columns to check if an expression is constant.
-    std::function<bool(const ASTPtr &)> is_constant = [&block, &context](const ASTPtr & node)
+    // Provide input columns as constant columns to check if an expression is
+    // constant and depends on the columns from provided block (the last is
+    // required to allow skipping some conditions for handling OR).
+    std::function<bool(const ASTPtr &)> is_constant = [&block, &context](const ASTPtr & expr)
     {
         auto actions = std::make_shared<ActionsDAG>(block.getColumnsWithTypeAndName());
         PreparedSetsPtr prepared_sets = std::make_shared<PreparedSets>();
@@ -178,13 +200,26 @@ bool prepareFilterBlockWithQuery(const ASTPtr & query, ContextPtr context, Block
             context, SizeLimits{}, 1, source_columns, std::move(actions), prepared_sets, true, true, true,
             { aggregation_keys, grouping_set_keys, GroupByKind::NONE });
 
-        ActionsVisitor(visitor_data).visit(node);
+        ActionsVisitor(visitor_data).visit(expr);
         actions = visitor_data.getActions();
+        auto expr_column_name = expr->getColumnName();
+
+        const auto * expr_const_node = actions->tryFindInOutputs(expr_column_name);
+        if (!expr_const_node)
+            return false;
+        auto filter_actions = ActionsDAG::buildFilterActionsDAG({expr_const_node}, {}, context);
+        const auto & nodes = filter_actions->getNodes();
+        bool has_dependent_columns = std::any_of(nodes.begin(), nodes.end(), [&](const auto & node)
+        {
+            return block.has(node.result_name);
+        });
+        if (!has_dependent_columns)
+            return false;
+
         auto expression_actions = std::make_shared<ExpressionActions>(actions);
         auto block_with_constants = block;
         expression_actions->execute(block_with_constants);
-        auto column_name = node->getColumnName();
-        return block_with_constants.has(column_name) && isColumnConst(*block_with_constants.getByName(column_name).column);
+        return block_with_constants.has(expr_column_name) && isColumnConst(*block_with_constants.getByName(expr_column_name).column);
     };
 
     /// Create an expression that evaluates the expressions in WHERE and PREWHERE, depending only on the existing columns.

tests/analyzer_integration_broken_tests.txt

@@ -1,94 +1,48 @@
 test_access_for_functions/test.py::test_access_rights_for_function
+test_build_sets_from_multiple_threads/test.py::test_set
 test_concurrent_backups_s3/test.py::test_concurrent_backups
-test_distributed_ddl/test.py::test_default_database[configs]
-test_distributed_ddl/test.py::test_default_database[configs_secure]
-test_distributed_ddl/test.py::test_on_server_fail[configs]
-test_distributed_ddl/test.py::test_on_server_fail[configs_secure]
-test_distributed_insert_backward_compatibility/test.py::test_distributed_in_tuple
+test_dictionaries_update_and_reload/test.py::test_reload_after_fail_in_cache_dictionary
+test_distributed_backward_compatability/test.py::test_distributed_in_tuple
+test_distributed_type_object/test.py::test_distributed_type_object
+test_drop_is_lock_free/test.py::test_query_is_lock_free[detach table]
+test_executable_table_function/test.py::test_executable_function_input_python
 test_mask_sensitive_info/test.py::test_encryption_functions
 test_merge_table_over_distributed/test.py::test_global_in
 test_merge_table_over_distributed/test.py::test_select_table_name_from_merge_over_distributed
+test_merge_tree_s3/test.py::test_heavy_insert_select_check_memory[node]
 test_mutations_with_merge_tree/test.py::test_mutations_with_merge_background_task
-test_passing_max_partitions_to_read_remotely/test.py::test_default_database_on_cluster
-test_row_policy/test.py::test_change_of_users_xml_changes_row_policies
-test_row_policy/test.py::test_change_of_users_xml_changes_row_policies
-test_row_policy/test.py::test_dcl_introspection
-test_row_policy/test.py::test_dcl_introspection
-test_row_policy/test.py::test_dcl_management
-test_row_policy/test.py::test_dcl_management
-test_row_policy/test.py::test_dcl_users_with_policies_from_users_xml
-test_row_policy/test.py::test_dcl_users_with_policies_from_users_xml
-test_row_policy/test.py::test_grant_create_row_policy
-test_row_policy/test.py::test_grant_create_row_policy
-test_row_policy/test.py::test_introspection
-test_row_policy/test.py::test_introspection
-test_row_policy/test.py::test_join
-test_row_policy/test.py::test_join
-test_row_policy/test.py::test_miscellaneous_engines
-test_row_policy/test.py::test_miscellaneous_engines
-test_row_policy/test.py::test_policy_from_users_xml_affects_only_user_assigned
-test_row_policy/test.py::test_policy_from_users_xml_affects_only_user_assigned
-test_row_policy/test.py::test_policy_on_distributed_table_via_role
-test_row_policy/test.py::test_policy_on_distributed_table_via_role
-test_row_policy/test.py::test_reload_users_xml_by_timer
-test_row_policy/test.py::test_reload_users_xml_by_timer
-test_row_policy/test.py::test_row_policy_filter_with_subquery
-test_row_policy/test.py::test_row_policy_filter_with_subquery
-test_row_policy/test.py::test_smoke
-test_row_policy/test.py::test_smoke
-test_row_policy/test.py::test_some_users_without_policies
-test_row_policy/test.py::test_some_users_without_policies
-test_row_policy/test.py::test_tags_with_db_and_table_names
-test_row_policy/test.py::test_tags_with_db_and_table_names
-test_row_policy/test.py::test_throwif_error_in_prewhere_with_same_condition_as_filter
-test_row_policy/test.py::test_throwif_error_in_prewhere_with_same_condition_as_filter
-test_row_policy/test.py::test_throwif_error_in_where_with_same_condition_as_filter
-test_row_policy/test.py::test_throwif_error_in_where_with_same_condition_as_filter
-test_row_policy/test.py::test_throwif_in_prewhere_doesnt_expose_restricted_data
-test_row_policy/test.py::test_throwif_in_prewhere_doesnt_expose_restricted_data
-test_row_policy/test.py::test_throwif_in_where_doesnt_expose_restricted_data
-test_row_policy/test.py::test_throwif_in_where_doesnt_expose_restricted_data
-test_row_policy/test.py::test_users_xml_is_readonly
-test_row_policy/test.py::test_users_xml_is_readonly
-test_row_policy/test.py::test_with_prewhere
-test_row_policy/test.py::test_with_prewhere
-test_settings_constraints_distributed/test.py::test_select_clamps_settings
-test_compression_codec_read/test.py::test_default_codec_read
-test_dictionaries_update_and_reload/test.py::test_reload_after_fail_in_cache_dictionary
-test_distributed_type_object/test.py::test_distributed_type_object
-test_materialized_mysql_database/test.py::test_select_without_columns_5_7
-test_materialized_mysql_database/test.py::test_select_without_columns_8_0
-test_shard_level_const_function/test.py::test_remote
-test_storage_postgresql/test.py::test_postgres_select_insert
-test_storage_rabbitmq/test.py::test_rabbitmq_materialized_view
-test_system_merges/test.py::test_mutation_simple[]
-test_system_merges/test.py::test_mutation_simple[replicated]
-test_disk_over_web_server/test.py::test_cache[node2]
-test_disk_over_web_server/test.py::test_incorrect_usage
-test_disk_over_web_server/test.py::test_replicated_database
-test_disk_over_web_server/test.py::test_unavailable_server
-test_disk_over_web_server/test.py::test_usage[node2]
-test_distributed_backward_compatability/test.py::test_distributed_in_tuple
-test_executable_table_function/test.py::test_executable_function_input_python
-test_settings_profile/test.py::test_show_profiles
-test_sql_user_defined_functions_on_cluster/test.py::test_sql_user_defined_functions_on_cluster
-test_postgresql_protocol/test.py::test_python_client
 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_user_defined_object_persistence/test.py::test_persistence
-test_settings_profile/test.py::test_show_profiles
-test_sql_user_defined_functions_on_cluster/test.py::test_sql_user_defined_functions_on_cluster
+test_replicating_constants/test.py::test_different_versions
+test_row_policy/test.py::test_change_of_users_xml_changes_row_policies
+test_row_policy/test.py::test_dcl_introspection
+test_row_policy/test.py::test_dcl_management
+test_row_policy/test.py::test_dcl_users_with_policies_from_users_xml
+test_row_policy/test.py::test_grant_create_row_policy
+test_row_policy/test.py::test_policy_from_users_xml_affects_only_user_assigned
+test_row_policy/test.py::test_policy_on_distributed_table_via_role
+test_row_policy/test.py::test_reload_users_xml_by_timer
+test_row_policy/test.py::test_row_policy_filter_with_subquery
+test_row_policy/test.py::test_smoke
+test_row_policy/test.py::test_some_users_without_policies
+test_row_policy/test.py::test_tags_with_db_and_table_names
+test_row_policy/test.py::test_throwif_error_in_prewhere_with_same_condition_as_filter
+test_row_policy/test.py::test_throwif_error_in_where_with_same_condition_as_filter
+test_row_policy/test.py::test_throwif_in_prewhere_doesnt_expose_restricted_data
+test_row_policy/test.py::test_throwif_in_where_doesnt_expose_restricted_data
+test_row_policy/test.py::test_users_xml_is_readonly
+test_row_policy/test.py::test_with_prewhere
 test_select_access_rights/test_main.py::test_alias_columns
 test_select_access_rights/test_main.py::test_select_count
 test_select_access_rights/test_main.py::test_select_join
-test_postgresql_protocol/test.py::test_python_client
-test_replicating_constants/test.py::test_different_versions
-test_merge_tree_s3/test.py::test_heavy_insert_select_check_memory[node]
+test_settings_profile/test.py::test_show_profiles
+test_shard_level_const_function/test.py::test_remote
+test_sql_user_defined_functions_on_cluster/test.py::test_sql_user_defined_functions_on_cluster
+test_storage_rabbitmq/test.py::test_rabbitmq_materialized_view
+test_system_merges/test.py::test_mutation_simple[]
+test_system_merges/test.py::test_mutation_simple[replicated]
+test_user_defined_object_persistence/test.py::test_persistence
 test_wrong_db_or_table_name/test.py::test_wrong_table_name
-test_drop_is_lock_free/test.py::test_query_is_lock_free[detach table]
-test_odbc_interaction/test.py::test_postgres_insert
-test_zookeeper_config/test.py::test_chroot_with_different_root
 test_zookeeper_config/test.py::test_chroot_with_same_root
-test_merge_tree_azure_blob_storage/test.py::test_table_manipulations
-test_parallel_replicas_skip_shards/test.py::test_skip_unavailable_shards
-test_build_sets_from_multiple_threads/test.py::test_set
+test_zookeeper_config/test.py::test_chroot_with_different_root

tests/analyzer_tech_debt.txt

@@ -1,68 +1,46 @@
 00223_shard_distributed_aggregation_memory_efficient
-00562_in_subquery_merge_tree
 00593_union_all_assert_columns_removed
-00673_subquery_prepared_set_performance
 00717_merge_and_distributed
 00725_memory_tracking
 00754_distributed_optimize_skip_select_on_unused_shards
 00754_distributed_optimize_skip_select_on_unused_shards_with_prewhere
-00927_asof_joins
 00940_order_by_read_in_order_query_plan
-00945_bloom_filter_index
-00981_in_subquery_with_tuple
-01049_join_low_card_bug_long
 01062_pm_all_join_with_block_continuation
 01064_incremental_streaming_from_2_src_with_feedback
 01071_force_optimize_skip_unused_shards
 01072_optimize_skip_unused_shards_const_expr_eval
 01083_expressions_in_engine_arguments
-01086_odbc_roundtrip
 01155_rename_move_materialized_view
 01173_transaction_control_queries
-01211_optimize_skip_unused_shards_type_mismatch
 01213_optimize_skip_unused_shards_DISTINCT
 01214_test_storage_merge_aliases_with_where
-01231_distributed_aggregation_memory_efficient_mix_levels
 01244_optimize_distributed_group_by_sharding_key
 01247_optimize_distributed_group_by_sharding_key_dist_on_dist
 01268_mv_scalars
 01268_shard_avgweighted
 01270_optimize_skip_unused_shards_low_cardinality
 01319_optimize_skip_unused_shards_nesting
-01428_nullable_asof_join
 01455_shard_leaf_max_rows_bytes_to_read
 01495_subqueries_in_with_statement
 01504_rocksdb
-01526_client_start_and_exit
 01527_dist_sharding_key_dictGet_reload
 01528_allow_nondeterministic_optimize_skip_unused_shards
-01540_verbatim_partition_pruning
 01560_merge_distributed_join
 01563_distributed_query_finish
-01576_alias_column_rewrite
-01583_const_column_in_set_index
 01584_distributed_buffer_cannot_find_column
-01585_use_index_for_global_in
-01585_use_index_for_global_in_with_null
 01586_columns_pruning
 01624_soft_constraints
 01651_bugs_from_15889
 01656_test_query_log_factories_info
-01676_clickhouse_client_autocomplete
-01681_bloom_filter_nullable_column
-01700_system_zookeeper_path_in
-01710_projection_additional_filters
 01739_index_hint
 02880_indexHint__partition_id
 01747_join_view_filter_dictionary
-01748_partition_id_pruning
 01756_optimize_skip_unused_shards_rewrite_in
 01757_optimize_skip_unused_shards_limit
 01758_optimize_skip_unused_shards_once
 01759_optimize_skip_unused_shards_zero_shards
 01761_cast_to_enum_nullable
 01786_explain_merge_tree
-01889_key_condition_function_chains
 01890_materialized_distributed_join
 01901_in_literal_shard_prune
 01925_join_materialized_columns
@@ -70,8 +48,6 @@
 01930_optimize_skip_unused_shards_rewrite_in
 01947_mv_subquery
 01952_optimize_distributed_group_by_sharding_key
-02000_join_on_const
-02001_shard_num_shard_count
 02131_used_row_policies_in_query_log
 02139_MV_with_scalar_subquery
 02174_cte_scalar_cache_mv
@@ -93,35 +69,18 @@
 02554_fix_grouping_sets_predicate_push_down
 02575_merge_prewhere_different_default_kind
 02713_array_low_cardinality_string
-02707_skip_index_with_in
-02241_join_rocksdb_bs
 02003_WithMergeableStateAfterAggregationAndLimit_LIMIT_BY_LIMIT_OFFSET
-01115_join_with_dictionary
 01009_global_array_join_names
 00917_multiple_joins_denny_crane
-00725_join_on_bug_1
 00636_partition_key_parts_pruning
-00261_storage_aliases_and_array_join
 01825_type_json_multiple_files
-01281_group_by_limit_memory_tracking
-02723_zookeeper_name
 00002_log_and_exception_messages_formatting
-01646_rewrite_sum_if_bug
 02725_agg_projection_resprect_PK
-01019_alter_materialized_view_consistent
-01600_parts_states_metrics_long
-01600_parts_types_metrics_long
-01287_max_execution_speed
 02703_row_policy_for_database
 02721_url_cluster
 02534_s3_cluster_insert_select_schema_inference
 02765_parallel_replicas_final_modifier
-02784_parallel_replicas_automatic_disabling
-02581_share_big_sets_between_mutation_tasks_long
-02581_share_big_sets_between_multiple_mutations_tasks_long
-00992_system_parts_race_condition_zookeeper_long
 02818_parameterized_view_with_cte_multiple_usage
-02790_optimize_skip_unused_shards_join
 01940_custom_tld_sharding_key
 02815_range_dict_no_direct_join
 02861_join_on_nullsafe_compare

tests/ci/clickhouse_helper.py

@@ -296,7 +296,7 @@ class CiLogsCredentials:
             logging.info("Do not use external logs pushing")
             return ""
         extra_columns = (
-            f"CAST({pr_info.number} AS Int32) AS pull_request_number, '{pr_info.sha}' AS commit_sha, "
+            f"CAST({pr_info.number} AS UInt32) AS pull_request_number, '{pr_info.sha}' AS commit_sha, "
             f"toDateTime('{check_start_time}', 'UTC') AS check_start_time, '{check_name}' AS check_name, "
             f"'{get_instance_type()}' AS instance_type, '{get_instance_id()}' AS instance_id"
         )

tests/integration/test_postgresql_replica_database_engine_2/test.py

@@ -719,6 +719,36 @@ def test_too_many_parts(started_cluster):
     pg_manager2.drop_materialized_db()
 
 
+def test_toast(started_cluster):
+    table = "test_toast"
+    pg_manager.execute(
+        f"CREATE TABLE {table} (id integer PRIMARY KEY, txt text, other text)"
+    )
+    pg_manager.create_materialized_db(
+        ip=started_cluster.postgres_ip,
+        port=started_cluster.postgres_port,
+        settings=[
+            f"materialized_postgresql_tables_list = '{table}'",
+            "materialized_postgresql_backoff_min_ms = 100",
+            "materialized_postgresql_backoff_max_ms = 100",
+        ],
+    )
+
+    pg_manager.execute(
+        f"""\
+INSERT INTO {table} (id, txt)\
+VALUES (1, (SELECT array_to_string(ARRAY(SELECT chr((100 + round(random() * 25)) :: integer) FROM generate_series(1,30000) as t(i)), '')))
+    """
+    )
+
+    check_tables_are_synchronized(
+        instance,
+        table,
+        postgres_database=pg_manager.get_default_database(),
+        order_by="id",
+    )
+
+
 if __name__ == "__main__":
     cluster.start()
     input("Cluster created, press any key to destroy...")

tests/integration/test_store_cleanup/configs/store_cleanup.xml

@@ -0,0 +1,11 @@
+<clickhouse>
+    <database_catalog_unused_dir_hide_timeout_sec>0</database_catalog_unused_dir_hide_timeout_sec>
+    <database_catalog_unused_dir_rm_timeout_sec>60</database_catalog_unused_dir_rm_timeout_sec>
+    <database_catalog_unused_dir_cleanup_period_sec>1</database_catalog_unused_dir_cleanup_period_sec>
+
+    <!-- We don't really need [Zoo]Keeper for this test.
+    And it makes sense to have at least one test with TestKeeper. -->
+    <zookeeper>
+        <implementation>testkeeper</implementation>
+    </zookeeper>
+</clickhouse>

tests/integration/test_store_cleanup/test.py

@@ -0,0 +1,212 @@
+import pytest
+
+from helpers.cluster import ClickHouseCluster
+
+cluster = ClickHouseCluster(__file__)
+
+node1 = cluster.add_instance(
+    "node1", stay_alive=True, main_configs=["configs/store_cleanup.xml"]
+)
+
+path_to_data = "/var/lib/clickhouse/"
+
+
+@pytest.fixture(scope="module")
+def started_cluster():
+    try:
+        cluster.start()
+        yield cluster
+
+    finally:
+        cluster.shutdown()
+
+
+def test_store_cleanup(started_cluster):
+    node1.query("CREATE DATABASE db UUID '10000000-1000-4000-8000-000000000001'")
+    node1.query(
+        "CREATE TABLE db.log UUID '10000000-1000-4000-8000-000000000002' ENGINE=Log AS SELECT 1"
+    )
+    node1.query(
+        "CREATE TABLE db.mt UUID '10000000-1000-4000-8000-000000000003' ENGINE=MergeTree ORDER BY tuple() AS SELECT 1"
+    )
+    node1.query(
+        "CREATE TABLE db.mem UUID '10000000-1000-4000-8000-000000000004' ENGINE=Memory AS SELECT 1"
+    )
+
+    node1.query("CREATE DATABASE db2 UUID '20000000-1000-4000-8000-000000000001'")
+    node1.query(
+        "CREATE TABLE db2.log UUID '20000000-1000-4000-8000-000000000002' ENGINE=Log AS SELECT 1"
+    )
+    node1.query("DETACH DATABASE db2")
+
+    node1.query("CREATE DATABASE db3 UUID '30000000-1000-4000-8000-000000000001'")
+    node1.query(
+        "CREATE TABLE db3.log UUID '30000000-1000-4000-8000-000000000002' ENGINE=Log AS SELECT 1"
+    )
+    node1.query(
+        "CREATE TABLE db3.log2 UUID '30000000-1000-4000-8000-000000000003' ENGINE=Log AS SELECT 1"
+    )
+    node1.query("DETACH TABLE db3.log")
+    node1.query("DETACH TABLE db3.log2 PERMANENTLY")
+
+    assert "d---------" not in node1.exec_in_container(
+        ["ls", "-l", f"{path_to_data}/store"]
+    )
+    assert "d---------" not in node1.exec_in_container(
+        ["ls", "-l", f"{path_to_data}/store/100"]
+    )
+    assert "d---------" not in node1.exec_in_container(
+        ["ls", "-l", f"{path_to_data}/store/200"]
+    )
+    assert "d---------" not in node1.exec_in_container(
+        ["ls", "-l", f"{path_to_data}/store/300"]
+    )
+
+    node1.stop_clickhouse(kill=True)
+    # All dirs related to `db` will be removed
+    node1.exec_in_container(["rm", f"{path_to_data}/metadata/db.sql"])
+
+    node1.exec_in_container(["mkdir", f"{path_to_data}/store/kek"])
+    node1.exec_in_container(["touch", f"{path_to_data}/store/12"])
+    try:
+        node1.exec_in_container(["mkdir", f"{path_to_data}/store/456"])
+    except Exception as e:
+        print("Failed to create 456/:", str(e))
+    node1.exec_in_container(["mkdir", f"{path_to_data}/store/456/testgarbage"])
+    node1.exec_in_container(
+        ["mkdir", f"{path_to_data}/store/456/30000000-1000-4000-8000-000000000003"]
+    )
+    node1.exec_in_container(
+        ["touch", f"{path_to_data}/store/456/45600000-1000-4000-8000-000000000003"]
+    )
+    node1.exec_in_container(
+        ["mkdir", f"{path_to_data}/store/456/45600000-1000-4000-8000-000000000004"]
+    )
+
+    node1.start_clickhouse()
+    node1.query("DETACH DATABASE db2")
+    node1.query("DETACH TABLE db3.log")
+
+    node1.wait_for_log_line(
+        "Removing access rights for unused directory",
+        timeout=60,
+        look_behind_lines=1000000,
+    )
+    node1.wait_for_log_line(
+        "directories from store", timeout=60, look_behind_lines=1000000
+    )
+
+    store = node1.exec_in_container(["ls", f"{path_to_data}/store"])
+    assert "100" in store
+    assert "200" in store
+    assert "300" in store
+    assert "456" in store
+    assert "kek" in store
+    assert "12" in store
+    assert "d---------" in node1.exec_in_container(
+        ["ls", "-l", f"{path_to_data}/store"]
+    )
+    assert "d---------" in node1.exec_in_container(
+        ["ls", "-l", f"{path_to_data}/store/456"]
+    )
+
+    # Metadata is removed, so store/100 contains garbage
+    store100 = node1.exec_in_container(["ls", f"{path_to_data}/store/100"])
+    assert "10000000-1000-4000-8000-000000000001" in store100
+    assert "10000000-1000-4000-8000-000000000002" in store100
+    assert "10000000-1000-4000-8000-000000000003" in store100
+    assert "d---------" in node1.exec_in_container(
+        ["ls", "-l", f"{path_to_data}/store/100"]
+    )
+
+    # Database is detached, nothing to clean up
+    store200 = node1.exec_in_container(["ls", f"{path_to_data}/store/200"])
+    assert "20000000-1000-4000-8000-000000000001" in store200
+    assert "20000000-1000-4000-8000-000000000002" in store200
+    assert "d---------" not in node1.exec_in_container(
+        ["ls", "-l", f"{path_to_data}/store/200"]
+    )
+
+    # Tables are detached, nothing to clean up
+    store300 = node1.exec_in_container(["ls", f"{path_to_data}/store/300"])
+    assert "30000000-1000-4000-8000-000000000001" in store300
+    assert "30000000-1000-4000-8000-000000000002" in store300
+    assert "30000000-1000-4000-8000-000000000003" in store300
+    assert "d---------" not in node1.exec_in_container(
+        ["ls", "-l", f"{path_to_data}/store/300"]
+    )
+
+    # Manually created garbage
+    store456 = node1.exec_in_container(["ls", f"{path_to_data}/store/456"])
+    assert "30000000-1000-4000-8000-000000000003" in store456
+    assert "45600000-1000-4000-8000-000000000003" in store456
+    assert "45600000-1000-4000-8000-000000000004" in store456
+    assert "testgarbage" in store456
+    assert "----------" in node1.exec_in_container(
+        ["ls", "-l", f"{path_to_data}/store/456"]
+    )
+
+    node1.wait_for_log_line(
+        "Removing unused directory", timeout=90, look_behind_lines=1000000
+    )
+    node1.wait_for_log_line(
+        "directories from store", timeout=90, look_behind_lines=1000000
+    )
+    node1.wait_for_log_line(
+        "Nothing to clean up from store/", timeout=90, look_behind_lines=1000000
+    )
+
+    store = node1.exec_in_container(["ls", f"{path_to_data}/store"])
+    assert "100" in store
+    assert "200" in store
+    assert "300" in store
+    assert "456" in store
+    assert "kek" not in store  # changed
+    assert "\n12\n" not in store  # changed
+    assert "d---------" not in node1.exec_in_container(
+        ["ls", "-l", f"{path_to_data}/store"]
+    )  # changed
+
+    # Metadata is removed, so store/100 contains garbage
+    store100 = node1.exec_in_container(["ls", f"{path_to_data}/store/100"])  # changed
+    assert "10000000-1000-4000-8000-000000000001" not in store100  # changed
+    assert "10000000-1000-4000-8000-000000000002" not in store100  # changed
+    assert "10000000-1000-4000-8000-000000000003" not in store100  # changed
+    assert "d---------" not in node1.exec_in_container(
+        ["ls", "-l", f"{path_to_data}/store/100"]
+    )  # changed
+
+    # Database is detached, nothing to clean up
+    store200 = node1.exec_in_container(["ls", f"{path_to_data}/store/200"])
+    assert "20000000-1000-4000-8000-000000000001" in store200
+    assert "20000000-1000-4000-8000-000000000002" in store200
+    assert "d---------" not in node1.exec_in_container(
+        ["ls", "-l", f"{path_to_data}/store/200"]
+    )
+
+    # Tables are detached, nothing to clean up
+    store300 = node1.exec_in_container(["ls", f"{path_to_data}/store/300"])
+    assert "30000000-1000-4000-8000-000000000001" in store300
+    assert "30000000-1000-4000-8000-000000000002" in store300
+    assert "30000000-1000-4000-8000-000000000003" in store300
+    assert "d---------" not in node1.exec_in_container(
+        ["ls", "-l", f"{path_to_data}/store/300"]
+    )
+
+    # Manually created garbage
+    store456 = node1.exec_in_container(["ls", f"{path_to_data}/store/456"])
+    assert "30000000-1000-4000-8000-000000000003" not in store456  # changed
+    assert "45600000-1000-4000-8000-000000000003" not in store456  # changed
+    assert "45600000-1000-4000-8000-000000000004" not in store456  # changed
+    assert "testgarbage" not in store456  # changed
+    assert "---------" not in node1.exec_in_container(
+        ["ls", "-l", f"{path_to_data}/store/456"]
+    )  # changed
+
+    node1.query("ATTACH TABLE db3.log2")
+    node1.query("ATTACH DATABASE db2")
+    node1.query("ATTACH TABLE db3.log")
+
+    assert "1\n" == node1.query("SELECT * FROM db3.log")
+    assert "1\n" == node1.query("SELECT * FROM db3.log2")
+    assert "1\n" == node1.query("SELECT * FROM db2.log")

tests/integration/test_system_merges/configs/user_overrides.xml

@@ -0,0 +1,7 @@
+<clickhouse>
+    <profiles>
+        <default>
+            <function_sleep_max_microseconds_per_block>10G</function_sleep_max_microseconds_per_block>
+        </default>
+    </profiles>
+</clickhouse>

tests/integration/test_system_merges/test.py

@@ -10,6 +10,7 @@ cluster = ClickHouseCluster(__file__)
 node1 = cluster.add_instance(
     "node1",
     main_configs=["configs/logs_config.xml"],
+    user_configs=["configs/user_overrides.xml"],
     with_zookeeper=True,
     macros={"shard": 0, "replica": 1},
 )
@@ -17,6 +18,7 @@ node1 = cluster.add_instance(
 node2 = cluster.add_instance(
     "node2",
     main_configs=["configs/logs_config.xml"],
+    user_configs=["configs/user_overrides.xml"],
     with_zookeeper=True,
     macros={"shard": 0, "replica": 2},
 )
@@ -183,10 +185,10 @@ def test_mutation_simple(started_cluster, replicated):
             starting_block, starting_block, starting_block + 1
         )
 
-        # ALTER will sleep for 3s * 3 (rows) = 9s
+        # ALTER will sleep for 9s
         def alter():
             node1.query(
-                f"ALTER TABLE {name} UPDATE a = 42 WHERE sleep(3) OR 1",
+                f"ALTER TABLE {name} UPDATE a = 42 WHERE sleep(9) OR 1",
                 settings=settings,
             )
 

tests/queries/0_stateless/01196_max_parser_depth.reference

@@ -1,5 +1,6 @@
+-- 1.
 Code: 167
+-- 2.
 Code: 167
-Code: 167
-Code: 167
+-- 3.
 Code: 167

tests/queries/0_stateless/01196_max_parser_depth.sh

@@ -6,6 +6,9 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
 
 { printf "select "; for _ in {1..1000}; do printf "coalesce(null, "; done; printf "1"; for _ in {1..1000}; do printf ")"; done; } > "${CLICKHOUSE_TMP}"/query
 
-cat "${CLICKHOUSE_TMP}"/query | $CLICKHOUSE_CLIENT 2>&1 | grep -o -F 'Code: 167'
-cat "${CLICKHOUSE_TMP}"/query | $CLICKHOUSE_LOCAL 2>&1 | grep -o -F 'Code: 167'
-cat "${CLICKHOUSE_TMP}"/query | $CLICKHOUSE_CURL --data-binary @- -vsS "$CLICKHOUSE_URL" 2>&1 | grep -o -F 'Code: 167'
+echo '-- 1.'
+cat "${CLICKHOUSE_TMP}"/query | $CLICKHOUSE_CLIENT 2>&1 | grep -o -m1 -F 'Code: 167'
+echo '-- 2.'
+cat "${CLICKHOUSE_TMP}"/query | $CLICKHOUSE_LOCAL 2>&1 | grep -o -m1 -F 'Code: 167'
+echo '-- 3.'
+cat "${CLICKHOUSE_TMP}"/query | $CLICKHOUSE_CURL --data-binary @- -vsS "$CLICKHOUSE_URL" 2>&1 | grep -o -m1 -F 'Code: 167'

tests/queries/0_stateless/02161_addressToLineWithInlines.sql

@@ -1,5 +1,4 @@
--- Tags: no-tsan, no-asan, no-ubsan, no-msan, no-debug, no-cpu-aarch64, disabled
--- Tag disabled: Parsing inlines may lead to "could not find abbreviation code" (FIXME)
+-- Tags: no-tsan, no-asan, no-ubsan, no-msan, no-debug, no-cpu-aarch64
 
 SET allow_introspection_functions = 0;
 SELECT addressToLineWithInlines(1); -- { serverError 446 }

tests/queries/0_stateless/02840_merge__table_or_filter.sql.j2

@@ -18,6 +18,11 @@ create view v2 as select * from d2;
 
 create table m as v1 engine=Merge(currentDatabase(), '^(v1|v2)$');
 
+{# -- FIXME:
+select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v1') or 0 or 0 settings {{ settings }};
+select _table, key from m where (value = 10 and _table = 'v3') or (value = 20 and _table = 'v3') or 0 or 0 settings {{ settings }};
+#}
+
 -- avoid reorder
 set max_threads=1;
 -- { echoOn }

tests/queries/0_stateless/02869_unicode_minus.reference

@@ -0,0 +1,2 @@
+-1
+-1

tests/queries/0_stateless/02869_unicode_minus.sql

@@ -0,0 +1,2 @@
+SELECT 1 − 2;
+SELECT −1;

tests/queries/0_stateless/02887_byteswap.reference

@@ -25,5 +25,9 @@
 -549755813889
 4039370097989451775
 128
+72057594037927936
+-2361183241434822606849
+1329227995784915872903807060280344576
+-43556142965880123323311949751266331066369
 0
 1

tests/queries/0_stateless/02887_byteswap.sql

@@ -33,6 +33,12 @@ SELECT byteSwap(-2147483649::Int64);
 SELECT byteSwap(-1242525266376::Int64);
 SELECT byteSwap(-9223372036854775808::Int64);
 
+SELECT byteSwap(18446744073709551616::UInt128);
+SELECT byteSwap(-9223372036854775809::Int128);
+
+SELECT byteSwap(340282366920938463463374607431768211456::UInt256);
+SELECT byteSwap(-170141183460469231731687303715884105729::Int256);
+
 -- Booleans are interpreted as UInt8
 SELECT byteSwap(false);
 SELECT byteSwap(true);
@@ -52,6 +58,4 @@ SELECT byteSwap(generateUUIDv4()); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT }
 SELECT byteSwap(toDecimal32(2, 4));  -- { serverError ILLEGAL_TYPE_OF_ARGUMENT }
 SELECT byteSwap(toFloat32(123.456));  -- { serverError NOT_IMPLEMENTED }
 SELECT byteSwap(toFloat64(123.456));  -- { serverError NOT_IMPLEMENTED }
-SELECT byteSwap(18446744073709551616::UInt128);  -- { serverError NOT_IMPLEMENTED }
-SELECT byteSwap(-9223372036854775809::Int128);  -- { serverError NOT_IMPLEMENTED }
 

tests/queries/0_stateless/02894_ast_depth_check.reference

@@ -0,0 +1 @@
+TOO_DEEP_AST

tests/queries/0_stateless/02894_ast_depth_check.sh

@@ -0,0 +1,7 @@
+#!/usr/bin/env bash
+
+CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
+# shellcheck source=../shell_config.sh
+. "$CURDIR"/../shell_config.sh
+
+$CLICKHOUSE_LOCAL --query "SELECT * FROM format('TSV', \$\$ a UInt8, x ALIAS ''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN''IN'' \$\$, '1')" 2>&1 | grep -oF 'TOO_DEEP_AST'

tests/queries/0_stateless/02896_multiple_OR.reference

@@ -0,0 +1,14 @@
+-- { echoOn }
+SELECT * FROM or_bug WHERE (key = 1) OR false OR false;
+1
+SELECT * FROM or_bug WHERE (key = 1) OR false;
+1
+SELECT * FROM or_bug WHERE (key = 1);
+1
+-- { echoOn }
+select * from forms where text_field like '%this%' or 0 = 1 or 0 = 1;
+5840ead423829c1eab29fa97	this is a test
+select * from forms where text_field like '%this%' or 0 = 1;
+5840ead423829c1eab29fa97	this is a test
+select * from forms where text_field like '%this%';
+5840ead423829c1eab29fa97	this is a test

tests/queries/0_stateless/02896_multiple_OR.sql

@@ -0,0 +1,28 @@
+-- https://github.com/ClickHouse/ClickHouse/pull/52653
+DROP TABLE IF EXISTS or_bug;
+CREATE TABLE or_bug (key UInt8) ENGINE=MergeTree ORDER BY key;
+INSERT INTO TABLE or_bug VALUES (0), (1);
+
+-- { echoOn }
+SELECT * FROM or_bug WHERE (key = 1) OR false OR false;
+SELECT * FROM or_bug WHERE (key = 1) OR false;
+SELECT * FROM or_bug WHERE (key = 1);
+-- { echoOff }
+
+-- https://github.com/ClickHouse/ClickHouse/issues/55288
+DROP TABLE IF EXISTS forms;
+CREATE TABLE forms
+(
+   `form_id` FixedString(24),
+   `text_field` String
+)
+ENGINE = MergeTree
+PRIMARY KEY form_id
+ORDER BY form_id;
+insert into forms values ('5840ead423829c1eab29fa97','this is a test');
+
+-- { echoOn }
+select * from forms where text_field like '%this%' or 0 = 1 or 0 = 1;
+select * from forms where text_field like '%this%' or 0 = 1;
+select * from forms where text_field like '%this%';
+-- { echoOff }