diff --git a/contrib/poco b/contrib/poco index 7a2d304c215..6216cc01a10 160000 --- a/contrib/poco +++ b/contrib/poco @@ -1 +1 @@ -Subproject commit 7a2d304c21549427460428c9039009ef4bbfd899 +Subproject commit 6216cc01a107ce149863411ca29013a224f80343 diff --git a/dbms/programs/performance-test/PerformanceTest.cpp b/dbms/programs/performance-test/PerformanceTest.cpp index 45e9dfa89a7..ab55cd3d6cf 100644 --- a/dbms/programs/performance-test/PerformanceTest.cpp +++ b/dbms/programs/performance-test/PerformanceTest.cpp @@ -3,6 +3,8 @@ #include #include #include +#include +#include #include #include #include @@ -249,6 +251,54 @@ std::vector PerformanceTest::execute() runQueries(queries_with_indexes, statistics_by_run); } + + if (got_SIGINT) + { + return statistics_by_run; + } + + // Pull memory usage data from query log. The log is normally filled in + // background, so we have to flush it synchronously here to see all the + // previous queries. + { + NullBlockOutputStream null_output(Block{}); + RemoteBlockInputStream flush_log(connection, "system flush logs", + {} /* header */, context); + copyData(flush_log, null_output); + } + + for (auto & statistics : statistics_by_run) + { + if (statistics.query_id.empty()) + { + // We have statistics structs for skipped queries as well, so we + // have to filter them out. + continue; + } + + // We run some test queries several times, specifying the same query id, + // so this query to the log may return several records. Choose the + // last one, because this is when the query performance has stabilized. + RemoteBlockInputStream log_reader(connection, + "select memory_usage, query_start_time from system.query_log " + "where type = 2 and query_id = '" + statistics.query_id + "' " + "order by query_start_time desc", + {} /* header */, context); + + log_reader.readPrefix(); + Block block = log_reader.read(); + if (block.columns() == 0) + { + LOG_WARNING(log, "Query '" << statistics.query_id << "' is not found in query log."); + continue; + } + + auto column = block.getByName("memory_usage").column; + statistics.memory_usage = column->get64(0); + + log_reader.readSuffix(); + } + return statistics_by_run; } @@ -298,47 +348,6 @@ void PerformanceTest::runQueries( break; } } - - if (got_SIGINT) - { - return; - } - - // Pull memory usage data from query log. The log is normally filled in - // background, so we have to flush it synchronously here to see all the - // previous queries. - { - RemoteBlockInputStream flush_log(connection, "system flush logs", - {} /* header */, context); - flush_log.readPrefix(); - while (flush_log.read()); - flush_log.readSuffix(); - } - - for (auto & statistics : statistics_by_run) - { - RemoteBlockInputStream log_reader(connection, - "select memory_usage from system.query_log where type = 2 and query_id = '" - + statistics.query_id + "'", - {} /* header */, context); - - log_reader.readPrefix(); - Block block = log_reader.read(); - if (block.columns() == 0) - { - LOG_WARNING(log, "Query '" << statistics.query_id << "' is not found in query log."); - continue; - } - - assert(block.columns() == 1); - assert(block.getDataTypes()[0]->getName() == "UInt64"); - ColumnPtr column = block.getByPosition(0).column; - assert(column->size() == 1); - StringRef ref = column->getDataAt(0); - assert(ref.size == sizeof(UInt64)); - statistics.memory_usage = *reinterpret_cast(ref.data); - log_reader.readSuffix(); - } } diff --git a/dbms/src/DataStreams/CreatingSetsBlockInputStream.h b/dbms/src/DataStreams/CreatingSetsBlockInputStream.h index b37c45164b5..4a4ffef29d6 100644 --- a/dbms/src/DataStreams/CreatingSetsBlockInputStream.h +++ b/dbms/src/DataStreams/CreatingSetsBlockInputStream.h @@ -2,7 +2,7 @@ #include #include -#include /// SubqueriesForSets +#include namespace Poco { class Logger; } diff --git a/dbms/src/Dictionaries/LibraryDictionarySource.cpp b/dbms/src/Dictionaries/LibraryDictionarySource.cpp index b4de6506db1..2cb74b944d3 100644 --- a/dbms/src/Dictionaries/LibraryDictionarySource.cpp +++ b/dbms/src/Dictionaries/LibraryDictionarySource.cpp @@ -1,5 +1,6 @@ #include "LibraryDictionarySource.h" #include +#include #include #include #include @@ -134,7 +135,7 @@ LibraryDictionarySource::LibraryDictionarySource( ErrorCodes::FILE_DOESNT_EXIST); description.init(sample_block); library = std::make_shared(path, RTLD_LAZY -#if defined(RTLD_DEEPBIND) // Does not exists in freebsd +#if defined(RTLD_DEEPBIND) && !defined(ADDRESS_SANITIZER) // Does not exists in FreeBSD. Cannot work with Address Sanitizer. | RTLD_DEEPBIND #endif ); diff --git a/dbms/src/Formats/FormatFactory.cpp b/dbms/src/Formats/FormatFactory.cpp index 82fd22393d1..7fa4b59cf66 100644 --- a/dbms/src/Formats/FormatFactory.cpp +++ b/dbms/src/Formats/FormatFactory.cpp @@ -83,7 +83,17 @@ BlockInputStreamPtr FormatFactory::getInput( return std::make_shared(buf, sample, 0); if (!getCreators(name).input_processor_creator) - return getInput(name, buf, sample, context, max_block_size, rows_portion_size, std::move(callback)); + { + const auto & input_getter = getCreators(name).inout_creator; + if (!input_getter) + throw Exception("Format " + name + " is not suitable for input", ErrorCodes::FORMAT_IS_NOT_SUITABLE_FOR_INPUT); + + const Settings & settings = context.getSettingsRef(); + FormatSettings format_settings = getInputFormatSetting(settings); + + return input_getter( + buf, sample, context, max_block_size, rows_portion_size, callback ? callback : ReadCallback(), format_settings); + } auto format = getInputFormat(name, buf, sample, context, max_block_size, rows_portion_size, std::move(callback)); return std::make_shared(std::move(format)); @@ -106,13 +116,22 @@ BlockOutputStreamPtr FormatFactory::getOutput(const String & name, WriteBuffer & } if (!getCreators(name).output_processor_creator) - return getOutput(name, buf, sample, context); + { + const auto & output_getter = getCreators(name).output_creator; + if (!output_getter) + throw Exception("Format " + name + " is not suitable for output", ErrorCodes::FORMAT_IS_NOT_SUITABLE_FOR_OUTPUT); + + const Settings & settings = context.getSettingsRef(); + FormatSettings format_settings = getOutputFormatSetting(settings); + + /** Materialization is needed, because formats can use the functions `IDataType`, + * which only work with full columns. + */ + return std::make_shared( + output_getter(buf, sample, context, format_settings), sample); + } auto format = getOutputFormat(name, buf, sample, context); - - /** Materialization is needed, because formats can use the functions `IDataType`, - * which only work with full columns. - */ return std::make_shared(std::make_shared(format), sample); } diff --git a/dbms/src/Functions/currentUser.cpp b/dbms/src/Functions/currentUser.cpp new file mode 100644 index 00000000000..4501047fbb0 --- /dev/null +++ b/dbms/src/Functions/currentUser.cpp @@ -0,0 +1,54 @@ +#include +#include +#include +#include + + +namespace DB +{ + +class FunctionCurrentUser : public IFunction +{ + const String user_name; + +public: + static constexpr auto name = "currentUser"; + static FunctionPtr create(const Context & context) + { + return std::make_shared(context.getClientInfo().initial_user); + } + + explicit FunctionCurrentUser(const String & user_name_) : user_name{user_name_} + { + } + + String getName() const override + { + return name; + } + size_t getNumberOfArguments() const override + { + return 0; + } + + DataTypePtr getReturnTypeImpl(const DataTypes & /*arguments*/) const override + { + return std::make_shared(); + } + + bool isDeterministic() const override { return false; } + + void executeImpl(Block & block, const ColumnNumbers &, size_t result, size_t input_rows_count) override + { + block.getByPosition(result).column = DataTypeString().createColumnConst(input_rows_count, user_name); + } +}; + + +void registerFunctionCurrentUser(FunctionFactory & factory) +{ + factory.registerFunction(); + factory.registerAlias("user", FunctionCurrentUser::name, FunctionFactory::CaseInsensitive); +} + +} diff --git a/dbms/src/Functions/registerFunctionsMiscellaneous.cpp b/dbms/src/Functions/registerFunctionsMiscellaneous.cpp index 6d201d65bd3..5c86587cb84 100644 --- a/dbms/src/Functions/registerFunctionsMiscellaneous.cpp +++ b/dbms/src/Functions/registerFunctionsMiscellaneous.cpp @@ -6,6 +6,7 @@ namespace DB class FunctionFactory; void registerFunctionCurrentDatabase(FunctionFactory &); +void registerFunctionCurrentUser(FunctionFactory &); void registerFunctionHostName(FunctionFactory &); void registerFunctionVisibleWidth(FunctionFactory &); void registerFunctionToTypeName(FunctionFactory &); @@ -56,6 +57,7 @@ void registerFunctionConvertCharset(FunctionFactory &); void registerFunctionsMiscellaneous(FunctionFactory & factory) { registerFunctionCurrentDatabase(factory); + registerFunctionCurrentUser(factory); registerFunctionHostName(factory); registerFunctionVisibleWidth(factory); registerFunctionToTypeName(factory); diff --git a/dbms/src/Interpreters/ActionsVisitor.h b/dbms/src/Interpreters/ActionsVisitor.h index 9841c8e9df8..4d03f758f61 100644 --- a/dbms/src/Interpreters/ActionsVisitor.h +++ b/dbms/src/Interpreters/ActionsVisitor.h @@ -54,7 +54,6 @@ struct ScopeStack /// Collect ExpressionAction from AST. Returns PreparedSets and SubqueriesForSets too. -/// After AST is visited source ExpressionActions should be updated with popActionsLevel() method. class ActionsVisitor { public: @@ -63,9 +62,11 @@ public: PreparedSets & prepared_sets_, SubqueriesForSets & subqueries_for_sets_, bool no_subqueries_, bool only_consts_, bool no_storage_or_local_, std::ostream * ostr_ = nullptr); - void visit(const ASTPtr & ast); - - ExpressionActionsPtr popActionsLevel() { return actions_stack.popLevel(); } + void visit(const ASTPtr & ast, ExpressionActionsPtr & actions) + { + visit(ast); + actions = actions_stack.popLevel(); + } private: const Context & context; @@ -81,6 +82,7 @@ private: std::ostream * ostr; ScopeStack actions_stack; + void visit(const ASTPtr & ast); SetPtr makeSet(const ASTFunction * node, const Block & sample_block); }; diff --git a/dbms/src/Interpreters/ExpressionAnalyzer.cpp b/dbms/src/Interpreters/ExpressionAnalyzer.cpp index 9405f7cedeb..8d4eecf2df8 100644 --- a/dbms/src/Interpreters/ExpressionAnalyzer.cpp +++ b/dbms/src/Interpreters/ExpressionAnalyzer.cpp @@ -58,6 +58,7 @@ #include #include #include +#include namespace DB { @@ -68,7 +69,6 @@ using LogAST = DebugASTLog; /// set to true to enable logs namespace ErrorCodes { extern const int UNKNOWN_IDENTIFIER; - extern const int ILLEGAL_AGGREGATION; extern const int EXPECTED_ALL_OR_ANY; } @@ -78,16 +78,15 @@ ExpressionAnalyzer::ExpressionAnalyzer( const Context & context_, const NameSet & required_result_columns_, size_t subquery_depth_, - bool do_global_, - const SubqueriesForSets & subqueries_for_sets_) - : ExpressionAnalyzerData(required_result_columns_, subqueries_for_sets_) + bool do_global) + : ExpressionAnalyzerData(required_result_columns_) , query(query_), context(context_), settings(context.getSettings()) - , subquery_depth(subquery_depth_), do_global(do_global_) + , subquery_depth(subquery_depth_) , syntax(syntax_analyzer_result_) { /// external_tables, subqueries_for_sets for global subqueries. /// Replaces global subqueries with the generated names of temporary tables that will be sent to remote servers. - initGlobalSubqueriesAndExternalTables(); + initGlobalSubqueriesAndExternalTables(do_global); /// has_aggregation, aggregation_keys, aggregate_descriptions, aggregated_columns. /// This analysis should be performed after processing global subqueries, because otherwise, @@ -116,9 +115,6 @@ void ExpressionAnalyzer::analyzeAggregation() auto * select_query = query->as(); - if (select_query && (select_query->groupBy() || select_query->having())) - has_aggregation = true; - ExpressionActionsPtr temp_actions = std::make_shared(sourceColumns(), context); if (select_query) @@ -150,11 +146,13 @@ void ExpressionAnalyzer::analyzeAggregation() } } - getAggregates(query, temp_actions); + has_aggregation = makeAggregateDescriptions(temp_actions); + if (select_query && (select_query->groupBy() || select_query->having())) + has_aggregation = true; if (has_aggregation) { - assertSelect(); + getSelectQuery(); /// assertSelect() /// Find out aggregation keys. if (select_query->groupBy()) @@ -223,7 +221,7 @@ void ExpressionAnalyzer::analyzeAggregation() } -void ExpressionAnalyzer::initGlobalSubqueriesAndExternalTables() +void ExpressionAnalyzer::initGlobalSubqueriesAndExternalTables(bool do_global) { /// Adds existing external tables (not subqueries) to the external_tables dictionary. ExternalTablesVisitor::Data tables_data{context, external_tables}; @@ -238,20 +236,6 @@ void ExpressionAnalyzer::initGlobalSubqueriesAndExternalTables() } -void ExpressionAnalyzer::makeSetsForIndex() -{ - const auto * select_query = query->as(); - - if (storage() && select_query && storage()->supportsIndexForIn()) - { - if (select_query->where()) - makeSetsForIndexImpl(select_query->where()); - if (select_query->prewhere()) - makeSetsForIndexImpl(select_query->prewhere()); - } -} - - void ExpressionAnalyzer::tryMakeSetForIndexFromSubquery(const ASTPtr & subquery_or_table_name) { auto set_key = PreparedSetKey::forSubquery(*subquery_or_table_name); @@ -277,8 +261,12 @@ void ExpressionAnalyzer::tryMakeSetForIndexFromSubquery(const ASTPtr & subquery_ } -void ExpressionAnalyzer::makeSetsForIndexImpl(const ASTPtr & node) +/// Perfomance optimisation for IN() if storage supports it. +void ExpressionAnalyzer::makeSetsForIndex(const ASTPtr & node) { + if (!node || !storage() || !storage()->supportsIndexForIn()) + return; + for (auto & child : node->children) { /// Don't descend into subqueries. @@ -290,15 +278,16 @@ void ExpressionAnalyzer::makeSetsForIndexImpl(const ASTPtr & node) if (func && func->name == "lambda") continue; - makeSetsForIndexImpl(child); + makeSetsForIndex(child); } const auto * func = node->as(); if (func && functionIsInOperator(func->name)) { const IAST & args = *func->arguments; + const ASTPtr & left_in_operand = args.children.at(0); - if (storage() && storage()->mayBenefitFromIndexForIn(args.children.at(0), context)) + if (storage()->mayBenefitFromIndexForIn(left_in_operand, context)) { const ASTPtr & arg = args.children.at(1); if (arg->as() || arg->as()) @@ -310,13 +299,13 @@ void ExpressionAnalyzer::makeSetsForIndexImpl(const ASTPtr & node) { NamesAndTypesList temp_columns = sourceColumns(); temp_columns.insert(temp_columns.end(), array_join_columns.begin(), array_join_columns.end()); - for (const auto & joined_column : columnsAddedByJoin()) - temp_columns.push_back(joined_column); + temp_columns.insert(temp_columns.end(), columnsAddedByJoin().begin(), columnsAddedByJoin().end()); + ExpressionActionsPtr temp_actions = std::make_shared(temp_columns, context); - getRootActions(func->arguments->children.at(0), true, temp_actions); + getRootActions(left_in_operand, true, temp_actions); Block sample_block_with_calculated_columns = temp_actions->getSampleBlock(); - if (sample_block_with_calculated_columns.has(args.children.at(0)->getColumnName())) + if (sample_block_with_calculated_columns.has(left_in_operand->getColumnName())) makeExplicitSet(func, sample_block_with_calculated_columns, true, context, settings.size_limits_for_set, prepared_sets); } @@ -331,71 +320,23 @@ void ExpressionAnalyzer::getRootActions(const ASTPtr & ast, bool no_subqueries, ActionsVisitor actions_visitor(context, settings.size_limits_for_set, subquery_depth, sourceColumns(), actions, prepared_sets, subqueries_for_sets, no_subqueries, only_consts, !isRemoteStorage(), log.stream()); - actions_visitor.visit(ast); - actions = actions_visitor.popActionsLevel(); + actions_visitor.visit(ast, actions); } -void ExpressionAnalyzer::getActionsFromJoinKeys(const ASTTableJoin & table_join, bool no_subqueries, ExpressionActionsPtr & actions) +bool ExpressionAnalyzer::makeAggregateDescriptions(ExpressionActionsPtr & actions) { - bool only_consts = false; - - LogAST log; - ActionsVisitor actions_visitor(context, settings.size_limits_for_set, subquery_depth, - sourceColumns(), actions, prepared_sets, subqueries_for_sets, - no_subqueries, only_consts, !isRemoteStorage(), log.stream()); - - if (table_join.using_expression_list) - actions_visitor.visit(table_join.using_expression_list); - else if (table_join.on_expression) + for (const ASTFunction * node : aggregates()) { - for (const auto & ast : analyzedJoin().key_asts_left) - actions_visitor.visit(ast); - } - - actions = actions_visitor.popActionsLevel(); -} - - -void ExpressionAnalyzer::getAggregates(const ASTPtr & ast, ExpressionActionsPtr & actions) -{ - const auto * select_query = query->as(); - - /// There can not be aggregate functions inside the WHERE and PREWHERE. - if (select_query && (ast.get() == select_query->where().get() || ast.get() == select_query->prewhere().get())) - { - assertNoAggregates(ast, "in WHERE or PREWHERE"); - return; - } - - /// If we are not analyzing a SELECT query, but a separate expression, then there can not be aggregate functions in it. - if (!select_query) - { - assertNoAggregates(ast, "in wrong place"); - return; - } - - const auto * node = ast->as(); - if (node && AggregateFunctionFactory::instance().isAggregateFunctionName(node->name)) - { - has_aggregation = true; AggregateDescription aggregate; aggregate.column_name = node->getColumnName(); - /// Make unique aggregate functions. - for (size_t i = 0; i < aggregate_descriptions.size(); ++i) - if (aggregate_descriptions[i].column_name == aggregate.column_name) - return; - const ASTs & arguments = node->arguments->children; aggregate.argument_names.resize(arguments.size()); DataTypes types(arguments.size()); for (size_t i = 0; i < arguments.size(); ++i) { - /// There can not be other aggregate functions within the aggregate functions. - assertNoAggregates(arguments[i], "inside another aggregate function"); - getRootActions(arguments[i], true, actions); const std::string & name = arguments[i]->getColumnName(); types[i] = actions->getSampleBlock().getByName(name).type; @@ -407,41 +348,24 @@ void ExpressionAnalyzer::getAggregates(const ASTPtr & ast, ExpressionActionsPtr aggregate_descriptions.push_back(aggregate); } - else - { - for (const auto & child : ast->children) - if (!child->as() && !child->as()) - getAggregates(child, actions); - } + + return !aggregates().empty(); } -void ExpressionAnalyzer::assertNoAggregates(const ASTPtr & ast, const char * description) -{ - const auto * node = ast->as(); - - if (node && AggregateFunctionFactory::instance().isAggregateFunctionName(node->name)) - throw Exception("Aggregate function " + node->getColumnName() - + " is found " + String(description) + " in query", ErrorCodes::ILLEGAL_AGGREGATION); - - for (const auto & child : ast->children) - if (!child->as() && !child->as()) - assertNoAggregates(child, description); -} - - -void ExpressionAnalyzer::assertSelect() const +const ASTSelectQuery * ExpressionAnalyzer::getSelectQuery() const { const auto * select_query = query->as(); - if (!select_query) throw Exception("Not a select query", ErrorCodes::LOGICAL_ERROR); + return select_query; } -void ExpressionAnalyzer::assertAggregation() const +const ASTSelectQuery * ExpressionAnalyzer::getAggregatingQuery() const { if (!has_aggregation) throw Exception("No aggregation", ErrorCodes::LOGICAL_ERROR); + return getSelectQuery(); } void ExpressionAnalyzer::initChain(ExpressionActionsChain & chain, const NamesAndTypesList & columns) const @@ -471,9 +395,7 @@ void ExpressionAnalyzer::addMultipleArrayJoinAction(ExpressionActionsPtr & actio bool ExpressionAnalyzer::appendArrayJoin(ExpressionActionsChain & chain, bool only_types) { - const auto * select_query = query->as(); - - assertSelect(); + const auto * select_query = getSelectQuery(); bool is_array_join_left; ASTPtr array_join_expression_list = select_query->array_join_expression_list(is_array_join_left); @@ -515,9 +437,7 @@ static void appendRequiredColumns( bool ExpressionAnalyzer::appendJoin(ExpressionActionsChain & chain, bool only_types) { - const auto * select_query = query->as(); - - assertSelect(); + const auto * select_query = getSelectQuery(); if (!select_query->join()) return false; @@ -538,9 +458,18 @@ bool ExpressionAnalyzer::appendJoin(ExpressionActionsChain & chain, bool only_ty throw Exception("Expected ANY or ALL in JOIN section, because setting (join_default_strictness) is empty", DB::ErrorCodes::EXPECTED_ALL_OR_ANY); } - const auto & table_to_join = join_element.table_expression->as(); + if (join_params.using_expression_list) + { + getRootActions(join_params.using_expression_list, only_types, step.actions); + } + else if (join_params.on_expression) + { + auto list = std::make_shared(); + list->children = analyzedJoin().key_asts_left; + getRootActions(list, only_types, step.actions); + } - getActionsFromJoinKeys(join_params, only_types, step.actions); + const auto & table_to_join = join_element.table_expression->as(); /// Two JOINs are not supported with the same subquery, but different USINGs. auto join_hash = join_element.getTreeHash(); @@ -626,9 +555,7 @@ bool ExpressionAnalyzer::appendJoin(ExpressionActionsChain & chain, bool only_ty bool ExpressionAnalyzer::appendPrewhere( ExpressionActionsChain & chain, bool only_types, const Names & additional_required_columns) { - const auto * select_query = query->as(); - - assertSelect(); + const auto * select_query = getSelectQuery(); if (!select_query->prewhere()) return false; @@ -701,9 +628,7 @@ bool ExpressionAnalyzer::appendPrewhere( bool ExpressionAnalyzer::appendWhere(ExpressionActionsChain & chain, bool only_types) { - const auto * select_query = query->as(); - - assertSelect(); + const auto * select_query = getSelectQuery(); if (!select_query->where()) return false; @@ -721,9 +646,7 @@ bool ExpressionAnalyzer::appendWhere(ExpressionActionsChain & chain, bool only_t bool ExpressionAnalyzer::appendGroupBy(ExpressionActionsChain & chain, bool only_types) { - const auto * select_query = query->as(); - - assertAggregation(); + const auto * select_query = getAggregatingQuery(); if (!select_query->groupBy()) return false; @@ -743,9 +666,7 @@ bool ExpressionAnalyzer::appendGroupBy(ExpressionActionsChain & chain, bool only void ExpressionAnalyzer::appendAggregateFunctionsArguments(ExpressionActionsChain & chain, bool only_types) { - const auto * select_query = query->as(); - - assertAggregation(); + const auto * select_query = getAggregatingQuery(); initChain(chain, sourceColumns()); ExpressionActionsChain::Step & step = chain.steps.back(); @@ -758,20 +679,27 @@ void ExpressionAnalyzer::appendAggregateFunctionsArguments(ExpressionActionsChai } } - getActionsBeforeAggregation(select_query->select(), step.actions, only_types); + /// Collect aggregates removing duplicates by node.getColumnName() + /// It's not clear why we recollect aggregates (for query parts) while we're able to use previously collected ones (for entire query) + /// @note The original recollection logic didn't remove duplicates. + GetAggregatesVisitor::Data data; + GetAggregatesVisitor(data).visit(select_query->select()); if (select_query->having()) - getActionsBeforeAggregation(select_query->having(), step.actions, only_types); + GetAggregatesVisitor(data).visit(select_query->having()); if (select_query->orderBy()) - getActionsBeforeAggregation(select_query->orderBy(), step.actions, only_types); + GetAggregatesVisitor(data).visit(select_query->orderBy()); + + /// TODO: data.aggregates -> aggregates() + for (const ASTFunction * node : data.aggregates) + for (auto & argument : node->arguments->children) + getRootActions(argument, only_types, step.actions); } bool ExpressionAnalyzer::appendHaving(ExpressionActionsChain & chain, bool only_types) { - const auto * select_query = query->as(); - - assertAggregation(); + const auto * select_query = getAggregatingQuery(); if (!select_query->having()) return false; @@ -787,9 +715,7 @@ bool ExpressionAnalyzer::appendHaving(ExpressionActionsChain & chain, bool only_ void ExpressionAnalyzer::appendSelect(ExpressionActionsChain & chain, bool only_types) { - const auto * select_query = query->as(); - - assertSelect(); + const auto * select_query = getSelectQuery(); initChain(chain, aggregated_columns); ExpressionActionsChain::Step & step = chain.steps.back(); @@ -802,9 +728,7 @@ void ExpressionAnalyzer::appendSelect(ExpressionActionsChain & chain, bool only_ bool ExpressionAnalyzer::appendOrderBy(ExpressionActionsChain & chain, bool only_types) { - const auto * select_query = query->as(); - - assertSelect(); + const auto * select_query = getSelectQuery(); if (!select_query->orderBy()) return false; @@ -828,9 +752,7 @@ bool ExpressionAnalyzer::appendOrderBy(ExpressionActionsChain & chain, bool only bool ExpressionAnalyzer::appendLimitBy(ExpressionActionsChain & chain, bool only_types) { - const auto * select_query = query->as(); - - assertSelect(); + const auto * select_query = getSelectQuery(); if (!select_query->limitBy()) return false; @@ -859,9 +781,7 @@ bool ExpressionAnalyzer::appendLimitBy(ExpressionActionsChain & chain, bool only void ExpressionAnalyzer::appendProjectResult(ExpressionActionsChain & chain) const { - const auto * select_query = query->as(); - - assertSelect(); + const auto * select_query = getSelectQuery(); initChain(chain, aggregated_columns); ExpressionActionsChain::Step & step = chain.steps.back(); @@ -892,19 +812,6 @@ void ExpressionAnalyzer::appendExpression(ExpressionActionsChain & chain, const } -void ExpressionAnalyzer::getActionsBeforeAggregation(const ASTPtr & ast, ExpressionActionsPtr & actions, bool no_subqueries) -{ - const auto * node = ast->as(); - - if (node && AggregateFunctionFactory::instance().isAggregateFunctionName(node->name)) - for (auto & argument : node->arguments->children) - getRootActions(argument, no_subqueries, actions); - else - for (auto & child : ast->children) - getActionsBeforeAggregation(child, actions, no_subqueries); -} - - ExpressionActionsPtr ExpressionAnalyzer::getActions(bool add_aliases, bool project_result) { ExpressionActionsPtr actions = std::make_shared(sourceColumns(), context); diff --git a/dbms/src/Interpreters/ExpressionAnalyzer.h b/dbms/src/Interpreters/ExpressionAnalyzer.h index 1959d98ca64..2e70662f196 100644 --- a/dbms/src/Interpreters/ExpressionAnalyzer.h +++ b/dbms/src/Interpreters/ExpressionAnalyzer.h @@ -2,9 +2,9 @@ #include #include -#include #include #include +#include #include #include @@ -51,10 +51,8 @@ struct ExpressionAnalyzerData Tables external_tables; protected: - ExpressionAnalyzerData(const NameSet & required_result_columns_, - const SubqueriesForSets & subqueries_for_sets_) - : required_result_columns(required_result_columns_), - subqueries_for_sets(subqueries_for_sets_) + ExpressionAnalyzerData(const NameSet & required_result_columns_) + : required_result_columns(required_result_columns_) {} }; @@ -91,8 +89,7 @@ public: const Context & context_, const NameSet & required_result_columns_ = {}, size_t subquery_depth_ = 0, - bool do_global_ = false, - const SubqueriesForSets & subqueries_for_set_ = {}); + bool do_global_ = false); /// Does the expression have aggregate functions or a GROUP BY or HAVING section. bool hasAggregation() const { return has_aggregation; } @@ -161,7 +158,7 @@ public: const ExpressionAnalyzerData & getAnalyzedData() const { return *this; } /// Create Set-s that we can from IN section to use the index on them. - void makeSetsForIndex(); + void makeSetsForIndex(const ASTPtr & node); bool hasGlobalSubqueries() { return has_global_subqueries; } @@ -170,7 +167,6 @@ private: const Context & context; const ExtractedSettings settings; size_t subquery_depth; - bool do_global; /// Do I need to prepare for execution global subqueries when analyzing the query. SyntaxAnalyzerResultPtr syntax; @@ -178,35 +174,30 @@ private: const AnalyzedJoin & analyzedJoin() const { return syntax->analyzed_join; } const NamesAndTypesList & sourceColumns() const { return syntax->required_source_columns; } const NamesAndTypesList & columnsAddedByJoin() const { return syntax->columns_added_by_join; } + const std::vector & aggregates() const { return syntax->aggregates; } /// Find global subqueries in the GLOBAL IN/JOIN sections. Fills in external_tables. - void initGlobalSubqueriesAndExternalTables(); + void initGlobalSubqueriesAndExternalTables(bool do_global); void addMultipleArrayJoinAction(ExpressionActionsPtr & actions, bool is_left) const; void addJoinAction(ExpressionActionsPtr & actions, bool only_types) const; - /// If ast is ASTSelectQuery with JOIN, add actions for JOIN key columns. - void getActionsFromJoinKeys(const ASTTableJoin & table_join, bool no_subqueries, ExpressionActionsPtr & actions); - void getRootActions(const ASTPtr & ast, bool no_subqueries, ExpressionActionsPtr & actions, bool only_consts = false); - void getActionsBeforeAggregation(const ASTPtr & ast, ExpressionActionsPtr & actions, bool no_subqueries); - /** Add aggregation keys to aggregation_keys, aggregate functions to aggregate_descriptions, * Create a set of columns aggregated_columns resulting after the aggregation, if any, * or after all the actions that are normally performed before aggregation. * Set has_aggregation = true if there is GROUP BY or at least one aggregate function. */ void analyzeAggregation(); - void getAggregates(const ASTPtr & ast, ExpressionActionsPtr & actions); - void assertNoAggregates(const ASTPtr & ast, const char * description); + bool makeAggregateDescriptions(ExpressionActionsPtr & actions); /// columns - the columns that are present before the transformations begin. void initChain(ExpressionActionsChain & chain, const NamesAndTypesList & columns) const; - void assertSelect() const; - void assertAggregation() const; + const ASTSelectQuery * getSelectQuery() const; + const ASTSelectQuery * getAggregatingQuery() const; /** * Create Set from a subquery or a table expression in the query. The created set is suitable for using the index. @@ -214,8 +205,6 @@ private: */ void tryMakeSetForIndexFromSubquery(const ASTPtr & subquery_or_table_name); - void makeSetsForIndexImpl(const ASTPtr & node); - bool isRemoteStorage() const; }; diff --git a/dbms/src/Interpreters/GetAggregatesVisitor.h b/dbms/src/Interpreters/GetAggregatesVisitor.h new file mode 100644 index 00000000000..ba1501fc624 --- /dev/null +++ b/dbms/src/Interpreters/GetAggregatesVisitor.h @@ -0,0 +1,75 @@ +#pragma once + +#include +#include + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int ILLEGAL_AGGREGATION; +} + +class GetAggregatesMatcher +{ +public: + using Visitor = ConstInDepthNodeVisitor; + + struct Data + { + const char * assert_no_aggregates = nullptr; + std::unordered_set uniq_names; + std::vector aggregates; + }; + + static bool needChildVisit(const ASTPtr & node, const ASTPtr & child) + { + if (child->as() || child->as()) + return false; + if (auto * func = node->as()) + if (isAggregateFunction(func->name)) + return false; + return true; + } + + static void visit(const ASTPtr & ast, Data & data) + { + if (auto * func = ast->as()) + visit(*func, ast, data); + } + +private: + static void visit(const ASTFunction & node, const ASTPtr &, Data & data) + { + if (!isAggregateFunction(node.name)) + return; + + if (data.assert_no_aggregates) + throw Exception("Aggregate function " + node.getColumnName() + " is found " + String(data.assert_no_aggregates) + " in query", + ErrorCodes::ILLEGAL_AGGREGATION); + + String column_name = node.getColumnName(); + if (data.uniq_names.count(column_name)) + return; + + data.uniq_names.insert(column_name); + data.aggregates.push_back(&node); + } + + static bool isAggregateFunction(const String & name) + { + return AggregateFunctionFactory::instance().isAggregateFunctionName(name); + } +}; + +using GetAggregatesVisitor = GetAggregatesMatcher::Visitor; + + +inline void assertNoAggregates(const ASTPtr & ast, const char * description) +{ + GetAggregatesVisitor::Data data{description, {}, {}}; + GetAggregatesVisitor(data).visit(ast); +} + +} diff --git a/dbms/src/Interpreters/GlobalSubqueriesVisitor.h b/dbms/src/Interpreters/GlobalSubqueriesVisitor.h index 1622c27f62f..583ec026af0 100644 --- a/dbms/src/Interpreters/GlobalSubqueriesVisitor.h +++ b/dbms/src/Interpreters/GlobalSubqueriesVisitor.h @@ -7,7 +7,6 @@ #include #include #include -#include #include #include #include diff --git a/dbms/src/Interpreters/InterpreterSelectQuery.cpp b/dbms/src/Interpreters/InterpreterSelectQuery.cpp index 899686ced96..db65c8a95b1 100644 --- a/dbms/src/Interpreters/InterpreterSelectQuery.cpp +++ b/dbms/src/Interpreters/InterpreterSelectQuery.cpp @@ -790,7 +790,8 @@ void InterpreterSelectQuery::executeImpl(TPipeline & pipeline, const BlockInputS if (!dry_run) from_stage = storage->getQueryProcessingStage(context); - query_analyzer->makeSetsForIndex(); + query_analyzer->makeSetsForIndex(query.where()); + query_analyzer->makeSetsForIndex(query.prewhere()); auto optimize_prewhere = [&](auto & merge_tree) { diff --git a/dbms/src/Interpreters/PredicateExpressionsOptimizer.h b/dbms/src/Interpreters/PredicateExpressionsOptimizer.h index 4fa5cb20e2c..302c2e62a65 100644 --- a/dbms/src/Interpreters/PredicateExpressionsOptimizer.h +++ b/dbms/src/Interpreters/PredicateExpressionsOptimizer.h @@ -1,7 +1,6 @@ #pragma once #include "DatabaseAndTableWithAlias.h" -#include "ExpressionAnalyzer.h" #include #include diff --git a/dbms/src/Interpreters/SubqueryForSet.h b/dbms/src/Interpreters/SubqueryForSet.h index 1844f686f0a..79d32d836c6 100644 --- a/dbms/src/Interpreters/SubqueryForSet.h +++ b/dbms/src/Interpreters/SubqueryForSet.h @@ -12,7 +12,6 @@ class Join; using JoinPtr = std::shared_ptr; class InterpreterSelectWithUnionQuery; -struct AnalyzedJoin; /// Information on what to do when executing a subquery in the [GLOBAL] IN/JOIN section. diff --git a/dbms/src/Interpreters/SyntaxAnalyzer.cpp b/dbms/src/Interpreters/SyntaxAnalyzer.cpp index 282b19991b1..d31e1a886ac 100644 --- a/dbms/src/Interpreters/SyntaxAnalyzer.cpp +++ b/dbms/src/Interpreters/SyntaxAnalyzer.cpp @@ -14,6 +14,8 @@ #include #include #include +#include +#include /// getSmallestColumn() #include #include @@ -558,6 +560,30 @@ void checkJoin(const ASTTablesInSelectQueryElement * join) ErrorCodes::NOT_IMPLEMENTED); } +std::vector getAggregates(const ASTPtr & query) +{ + if (const auto * select_query = query->as()) + { + /// There can not be aggregate functions inside the WHERE and PREWHERE. + if (select_query->where()) + assertNoAggregates(select_query->where(), "in WHERE"); + if (select_query->prewhere()) + assertNoAggregates(select_query->prewhere(), "in PREWHERE"); + + GetAggregatesVisitor::Data data; + GetAggregatesVisitor(data).visit(query); + + /// There can not be other aggregate functions within the aggregate functions. + for (const ASTFunction * node : data.aggregates) + for (auto & arg : node->arguments->children) + assertNoAggregates(arg, "inside another aggregate function"); + return data.aggregates; + } + else + assertNoAggregates(query, "in wrong place"); + return {}; +} + } /// Calculate which columns are required to execute the expression. @@ -840,6 +866,7 @@ SyntaxAnalyzerResultPtr SyntaxAnalyzer::analyze( collectJoinedColumns(result.analyzed_join, *select_query, source_columns_set, result.aliases, settings.join_use_nulls); } + result.aggregates = getAggregates(query); result.collectUsedColumns(query, additional_source_columns); return std::make_shared(result); } diff --git a/dbms/src/Interpreters/SyntaxAnalyzer.h b/dbms/src/Interpreters/SyntaxAnalyzer.h index c7addb03526..a31dfef7e82 100644 --- a/dbms/src/Interpreters/SyntaxAnalyzer.h +++ b/dbms/src/Interpreters/SyntaxAnalyzer.h @@ -10,6 +10,8 @@ namespace DB NameSet removeDuplicateColumns(NamesAndTypesList & columns); +class ASTFunction; + struct SyntaxAnalyzerResult { StoragePtr storage; @@ -22,6 +24,7 @@ struct SyntaxAnalyzerResult NamesAndTypesList columns_added_by_join; Aliases aliases; + std::vector aggregates; /// Which column is needed to be ARRAY-JOIN'ed to get the specified. /// For example, for `SELECT s.v ... ARRAY JOIN a AS s` will get "s.v" -> "a.v". diff --git a/dbms/src/Processors/Executors/PipelineExecutor.h b/dbms/src/Processors/Executors/PipelineExecutor.h index e448d5e00dd..9c722cc69c3 100644 --- a/dbms/src/Processors/Executors/PipelineExecutor.h +++ b/dbms/src/Processors/Executors/PipelineExecutor.h @@ -177,7 +177,7 @@ private: /// Prepare processor with pid number. /// Check parents and children of current processor and push them to stacks if they also need to be prepared. /// If processor wants to be expanded, ExpandPipelineTask from thread_number's execution context will be used. - bool prepareProcessor(size_t pid, Stack & children, Stack & parents, size_t thread_number, bool async); + bool prepareProcessor(UInt64 pid, Stack & children, Stack & parents, size_t thread_number, bool async); void doExpandPipeline(ExpandPipelineTask * task, bool processing); void executeImpl(size_t num_threads); diff --git a/dbms/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp b/dbms/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp index ccbcd04857d..4430829501b 100644 --- a/dbms/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp +++ b/dbms/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp @@ -16,6 +16,7 @@ #include #include #include +#include /// Allow to use __uint128_t as a template parameter for boost::rational. // https://stackoverflow.com/questions/41198673/uint128-t-not-working-with-clang-and-libstdc diff --git a/dbms/src/Storages/MergeTree/RPNBuilder.h b/dbms/src/Storages/MergeTree/RPNBuilder.h index d5244c3285d..234f7b06ffe 100644 --- a/dbms/src/Storages/MergeTree/RPNBuilder.h +++ b/dbms/src/Storages/MergeTree/RPNBuilder.h @@ -4,7 +4,6 @@ #include #include #include -#include #include #include #include diff --git a/dbms/tests/queries/0_stateless/00990_function_current_user.reference b/dbms/tests/queries/0_stateless/00990_function_current_user.reference new file mode 100644 index 00000000000..f1f321b1ecd --- /dev/null +++ b/dbms/tests/queries/0_stateless/00990_function_current_user.reference @@ -0,0 +1,4 @@ +1 +1 +1 1 +1 diff --git a/dbms/tests/queries/0_stateless/00990_function_current_user.sql b/dbms/tests/queries/0_stateless/00990_function_current_user.sql new file mode 100644 index 00000000000..38bd815ecef --- /dev/null +++ b/dbms/tests/queries/0_stateless/00990_function_current_user.sql @@ -0,0 +1,5 @@ +-- since actual user name is unknown, have to perform just smoke tests +select currentUser() IS NOT NULL; +select length(currentUser()) > 0; +select currentUser() = user(), currentUser() = USER(); +select currentUser() = initial_user from system.processes where query like '%$!@#%'; diff --git a/dbms/tests/queries/0_stateless/00990_request_splitting.reference b/dbms/tests/queries/0_stateless/00990_request_splitting.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/dbms/tests/queries/0_stateless/00990_request_splitting.sql b/dbms/tests/queries/0_stateless/00990_request_splitting.sql new file mode 100644 index 00000000000..cc5b0ed2998 --- /dev/null +++ b/dbms/tests/queries/0_stateless/00990_request_splitting.sql @@ -0,0 +1 @@ +SELECT * FROM url('http://127.0.0.1:1337/? HTTP/1.1\r\nTest: test', CSV, 'column1 String'); -- { serverError 1000 } diff --git a/docs/en/database_engines/index.md b/docs/en/database_engines/index.md index 3190a79017f..dd802441be9 100644 --- a/docs/en/database_engines/index.md +++ b/docs/en/database_engines/index.md @@ -1,9 +1,11 @@ # Database Engines -Database engines provide working with tables. +Database engines allow you to work with tables. -By default, ClickHouse uses its native database engine which provides configurable [table engines](../operations/table_engines/index.md) and [SQL dialect](../query_language/syntax.md). +By default, ClickHouse uses its native database engine, which provides configurable [table engines](../operations/table_engines/index.md) and an [SQL dialect](../query_language/syntax.md). -Also you can use the following database engines: +You can also use the following database engines: - [MySQL](mysql.md) + +[Original article](https://clickhouse.yandex/docs/en/database_engines/) diff --git a/docs/en/database_engines/mysql.md b/docs/en/database_engines/mysql.md index fd5417ee27d..24352c1924c 100644 --- a/docs/en/database_engines/mysql.md +++ b/docs/en/database_engines/mysql.md @@ -1,10 +1,10 @@ # MySQL -Allows to connect to some database on remote MySQL server and perform `INSERT` and `SELECT` queries with tables to exchange data between ClickHouse and MySQL. +Allows to connect to databases on a remote MySQL server and perform `INSERT` and `SELECT` queries with tables to exchange data between ClickHouse and MySQL. -The `MySQL` database engine translate queries to the MySQL server, so you can perform operations such as `SHOW TABLES` or `SHOW CREATE TABLE`. +The `MySQL` database engine translate queries to the MySQL server so you can perform operations such as `SHOW TABLES` or `SHOW CREATE TABLE`. -You cannot perform with tables the following queries: +You cannot perform the following queries: - `ATTACH`/`DETACH` - `DROP` @@ -48,7 +48,7 @@ BINARY | [FixedString](../data_types/fixedstring.md) All other MySQL data types are converted into [String](../data_types/string.md). -[Nullable](../data_types/nullable.md) data type is supported. +[Nullable](../data_types/nullable.md) is supported. ## Examples of Use @@ -120,3 +120,5 @@ SELECT * FROM mysql_db.mysql_table │ 3 │ 4 │ └────────┴───────┘ ``` + +[Original article](https://clickhouse.yandex/docs/en/database_engines/mysql/) diff --git a/docs/en/interfaces/cli.md b/docs/en/interfaces/cli.md index 09c8baed5f5..9f9448f27c8 100644 --- a/docs/en/interfaces/cli.md +++ b/docs/en/interfaces/cli.md @@ -65,6 +65,31 @@ You can cancel a long query by pressing Ctrl+C. However, you will still need to The command-line client allows passing external data (external temporary tables) for querying. For more information, see the section "External data for query processing". +### Queries with Parameters {#cli-queries-with-parameters} + +You can create a query with parameters, and pass values for these parameters with the parameters of the client app. For example: + +```bash +clickhouse-client --param_parName="[1, 2]" -q "SELECT * FROM table WHERE a = {parName:Array(UInt16)}" +``` + +#### Syntax of a Query {#cli-queries-with-parameters-syntax} + +Format a query by the standard method. Values that you want to put into the query from the app parameters place in braces and format as follows: + +``` +{:} +``` + +- `name` — Identifier of a placeholder that should be used in app parameter as `--param_name = value`. +- `data type` — A data type of app parameter value. For example, data structure like `(integer, ('string', integer))` can have a data type `Tuple(UInt8, Tuple(String, UInt8))` (also you can use another [integer](../data_types/int_uint.md) types). + +#### Example + +```bash +clickhouse-client --param_tuple_in_tuple="(10, ('dt', 10))" -q "SELECT * FROM table WHERE val = {tuple_in_tuple:Tuple(UInt8, Tuple(String, UInt8))}" +``` + ## Configuring {#interfaces_cli_configuration} You can pass parameters to `clickhouse-client` (all parameters have a default value) using: diff --git a/docs/en/interfaces/http.md b/docs/en/interfaces/http.md index 37da364942c..ef32b101b71 100644 --- a/docs/en/interfaces/http.md +++ b/docs/en/interfaces/http.md @@ -244,5 +244,14 @@ curl -sS 'http://localhost:8123/?max_result_bytes=4000000&buffer_size=3000000&wa Use buffering to avoid situations where a query processing error occurred after the response code and HTTP headers were sent to the client. In this situation, an error message is written at the end of the response body, and on the client side, the error can only be detected at the parsing stage. +### Queries with Parameters {#cli-queries-with-parameters} + +You can create a query with parameters, and pass values for these parameters with the parameters of the HTTP request. For more information, see [CLI Formatted Queries](cli.md#cli-queries-with-parameters). + +### Example + +```bash +curl -sS "
?param_id=2¶m_phrase=test" -d "SELECT * FROM table WHERE int_column = {id:UInt8} and string_column = {phrase:String}" +``` [Original article](https://clickhouse.yandex/docs/en/interfaces/http_interface/) diff --git a/docs/en/operations/table_engines/mergetree.md b/docs/en/operations/table_engines/mergetree.md index 2a099a8947d..56cc99f5653 100644 --- a/docs/en/operations/table_engines/mergetree.md +++ b/docs/en/operations/table_engines/mergetree.md @@ -28,7 +28,7 @@ Main features: ## Creating a Table {#table_engine-mergetree-creating-a-table} -``` +```sql CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster] ( name1 [type1] [DEFAULT|MATERIALIZED|ALIAS expr1] [TTL expr1], @@ -47,7 +47,7 @@ CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster] For descriptions of request parameters, see the [request description](../../query_language/create.md). -**Query clauses** +### Query Clauses - `ENGINE` — Name and parameters of the engine. `ENGINE = MergeTree()`. The `MergeTree` engine does not have parameters. @@ -81,7 +81,7 @@ For descriptions of request parameters, see the [request description](../../quer - `merge_with_ttl_timeout` — Minimum delay in seconds before repeating a merge with TTL. Default value: 86400 (1 day). -**Example of setting the sections** +**Example of Sections Setting** ```sql ENGINE MergeTree() PARTITION BY toYYYYMM(EventDate) ORDER BY (CounterID, EventDate, intHash32(UserID)) SAMPLE BY intHash32(UserID) SETTINGS index_granularity=8192 @@ -107,7 +107,7 @@ CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster] ) ENGINE [=] MergeTree(date-column [, sampling_expression], (primary, key), index_granularity) ``` -**MergeTree() parameters** +**MergeTree() Parameters** - `date-column` — The name of a column of the [Date](../../data_types/date.md) type. ClickHouse automatically creates partitions by month based on this column. The partition names are in the `"YYYYMM"` format. - `sampling_expression` — An expression for sampling. @@ -312,16 +312,16 @@ Reading from a table is automatically parallelized. Determines the lifetime of values. -The `TTL` clause can be set for the whole table and for each individual column. If `TTL` is set for the whole table, individual `TTL` for columns are ignored. +The `TTL` clause can be set for the whole table and for each individual column. If both `TTL` are set, ClickHouse uses that `TTL` which expires earlier. - -The table must have the column of the [Date](../../data_types/date.md) or [DateTime](../../data_types/datetime.md) data type. This date column should be used in the `TTL` clause. You can only set lifetime of the data as an interval from the date column value. +The table must have the column in the [Date](../../data_types/date.md) or [DateTime](../../data_types/datetime.md) data type. To define the lifetime of data, use operations on this time column, for example: ``` -TTL date_time + interval +TTL time_column +TTL time_column + interval ``` -You can set the `interval` by any expression, returning the value of the `DateTime` data type. For example, you can use [time interval](../../query_language/operators.md#operators-datetime) operators. +To define `interval`, use [time interval](../../query_language/operators.md#operators-datetime) operators. ``` TTL date_time + INTERVAL 1 MONTH @@ -330,20 +330,20 @@ TTL date_time + INTERVAL 15 HOUR **Column TTL** -When the values in the column expire, ClickHouse replace them with the default values for the column data type. If all the column values in the data part become expired, ClickHouse deletes this column from the data part in a filesystem. +When the values in the column expire, ClickHouse replaces them with the default values for the column data type. If all the column values in the data part expire, ClickHouse deletes this column from the data part in a filesystem. -The `TTL` clause cannot be used for key columns. +The `TTL` clause can't be used for key columns. **Table TTL** -When some data in table expires, ClickHouse deletes all the corresponding rows. +When data in a table expires, ClickHouse deletes all corresponding rows. -**Cleaning up of Data** +**Removing Data** -Data with expired TTL is removed, when ClickHouse merges data parts. +Data with an expired TTL is removed when ClickHouse merges data parts. -When ClickHouse see that some data is expired, it performs off-schedule merge. To control frequency of such merges, you can set [merge_with_ttl_timeout](#mergetree_setting-merge_with_ttl_timeout). If it is too low, many off-schedule merges consume much resources. +When ClickHouse see that data is expired, it performs an off-schedule merge. To control the frequency of such merges, you can set [merge_with_ttl_timeout](#mergetree_setting-merge_with_ttl_timeout). If the value is too low, it will perform many off-schedule merges that may consume a lot of resources. -If you perform the `SELECT` query between merges you can get the expired data. To avoid it, use the [OPTIMIZE](../../query_language/misc.md#misc_operations-optimize) query before `SELECT`. +If you perform the `SELECT` query between merges, you may get expired data. To avoid it, use the [OPTIMIZE](../../query_language/misc.md#misc_operations-optimize) query before `SELECT`. [Original article](https://clickhouse.yandex/docs/en/operations/table_engines/mergetree/) diff --git a/docs/en/query_language/agg_functions/combinators.md b/docs/en/query_language/agg_functions/combinators.md index 133b4d489d7..d2cf85c3d75 100644 --- a/docs/en/query_language/agg_functions/combinators.md +++ b/docs/en/query_language/agg_functions/combinators.md @@ -44,5 +44,71 @@ Merges the intermediate aggregation states in the same way as the -Merge combina Converts an aggregate function for tables into an aggregate function for arrays that aggregates the corresponding array items and returns an array of results. For example, `sumForEach` for the arrays `[1, 2]`, `[3, 4, 5]`and`[6, 7]`returns the result `[10, 13, 5]` after adding together the corresponding array items. +## -Resample + +Allows to divide data by groups, and then separately aggregates the data in those groups. Groups are created by splitting the values of one of the columns into intervals. + +``` +Resample(start, end, step)(, resampling_key) +``` + +**Parameters** + +- `start` — Starting value of the whole required interval for the values of `resampling_key`. +- `stop` — Ending value of the whole required interval for the values of `resampling_key`. The whole interval doesn't include the `stop` value `[start, stop)`. +- `step` — Step for separating the whole interval by subintervals. The `aggFunction` is executed over each of those subintervals independently. +- `resampling_key` — Column, which values are used for separating data by intervals. +- `aggFunction_params` — Parameters of `aggFunction`. + + +**Returned values** + +- Array of `aggFunction` results for each of subintervals. + +**Example** + +Consider the `people` table with the following data: + +```text +┌─name───┬─age─┬─wage─┐ +│ John │ 16 │ 10 │ +│ Alice │ 30 │ 15 │ +│ Mary │ 35 │ 8 │ +│ Evelyn │ 48 │ 11.5 │ +│ David │ 62 │ 9.9 │ +│ Brian │ 60 │ 16 │ +└────────┴─────┴──────┘ +``` + +Let's get the names of the persons which age lies in the intervals of `[30,60)` and `[60,75)`. As we use integer representation of age, then there are ages of `[30, 59]` and `[60,74]`. + +For aggregating names into the array, we use the aggregate function [groupArray](reference.md#agg_function-grouparray). It takes a single argument. For our case, it is the `name` column. The `groupArrayResample` function should use the `age` column to aggregate names by age. To define required intervals, we pass the `(30, 75, 30)` arguments into the `groupArrayResample` function. + +```sql +SELECT groupArrayResample(30, 75, 30)(name, age) from people +``` +```text +┌─groupArrayResample(30, 75, 30)(name, age)─────┐ +│ [['Alice','Mary','Evelyn'],['David','Brian']] │ +└───────────────────────────────────────────────┘ +``` + +Consider the results. + +`Jonh` is out of the sample because he is too young. Other people are distributed according to the specified age intervals. + +Now, let's count the total number of people and their average wage in the specified age intervals. + +```sql +SELECT + countResample(30, 75, 30)(name, age) AS amount, + avgResample(30, 75, 30)(wage, age) AS avg_wage +FROM people +``` +```text +┌─amount─┬─avg_wage──────────────────┐ +│ [3,2] │ [11.5,12.949999809265137] │ +└────────┴───────────────────────────┘ +``` [Original article](https://clickhouse.yandex/docs/en/query_language/agg_functions/combinators/) diff --git a/docs/en/query_language/agg_functions/reference.md b/docs/en/query_language/agg_functions/reference.md index 350803f5aef..3e3e1e2cc01 100644 --- a/docs/en/query_language/agg_functions/reference.md +++ b/docs/en/query_language/agg_functions/reference.md @@ -650,7 +650,7 @@ The function takes a variable number of parameters. Parameters can be `Tuple`, ` - [uniqHLL12](#agg_function-uniqhll12) -## groupArray(x), groupArray(max_size)(x) +## groupArray(x), groupArray(max_size)(x) {#agg_function-grouparray} Creates an array of argument values. Values can be added to the array in any (indeterminate) order. diff --git a/docs/en/query_language/create.md b/docs/en/query_language/create.md index 81d7982eb00..e40a452cfe0 100644 --- a/docs/en/query_language/create.md +++ b/docs/en/query_language/create.md @@ -10,10 +10,10 @@ CREATE DATABASE [IF NOT EXISTS] db_name [ON CLUSTER cluster] [ENGINE = engine(.. - `IF NOT EXISTS` - If the `db_name` database already exists then: + If the `db_name` database already exists, then ClickHouse doesn't create a new database and: - - If clause is specified, ClickHouse doesn't create a new database and doesn't throw an exception. - - If clause is not specified, then ClickHouse doesn't create a new database and throw and exception. + - Doesn't throw an exception if clause is specified. + - Throws an exception if clause isn't specified. - `ON CLUSTER` @@ -23,7 +23,7 @@ CREATE DATABASE [IF NOT EXISTS] db_name [ON CLUSTER cluster] [ENGINE = engine(.. - [MySQL](../database_engines/mysql.md) - Allows to retrieve data from the remote MySQL server. + Allows you to retrieve data from the remote MySQL server. By default, ClickHouse uses its own [database engine](../database_engines/index.md). diff --git a/docs/en/query_language/functions/other_functions.md b/docs/en/query_language/functions/other_functions.md index 57fa8acfee3..cc5fadcdd98 100644 --- a/docs/en/query_language/functions/other_functions.md +++ b/docs/en/query_language/functions/other_functions.md @@ -102,6 +102,9 @@ Sleeps 'seconds' seconds on each row. You can specify an integer or a floating-p Returns the name of the current database. You can use this function in table engine parameters in a CREATE TABLE query where you need to specify the database. +## currentUser() +Returns the login of authorized user (initiator of query execution). + ## isFinite(x) Accepts Float32 and Float64 and returns UInt8 equal to 1 if the argument is not infinite and not a NaN, otherwise 0. diff --git a/docs/ru/database_engines/index.md b/docs/ru/database_engines/index.md index 3190a79017f..52aefe1bba2 100644 --- a/docs/ru/database_engines/index.md +++ b/docs/ru/database_engines/index.md @@ -1,9 +1,10 @@ -# Database Engines +# Движки баз данных -Database engines provide working with tables. +Движки баз данных обеспечивают работу с таблицами. -By default, ClickHouse uses its native database engine which provides configurable [table engines](../operations/table_engines/index.md) and [SQL dialect](../query_language/syntax.md). +По умолчанию ClickHouse использует собственный движок баз данных, который поддерживает конфигурируемые [движки таблиц](../operations/table_engines/index.md) и [диалект SQL](../query_language/syntax.md). -Also you can use the following database engines: +Также можно использовать следующие движки баз данных: - [MySQL](mysql.md) + diff --git a/docs/ru/database_engines/mysql.md b/docs/ru/database_engines/mysql.md index fd5417ee27d..acfb71d839c 100644 --- a/docs/ru/database_engines/mysql.md +++ b/docs/ru/database_engines/mysql.md @@ -1,10 +1,10 @@ # MySQL -Allows to connect to some database on remote MySQL server and perform `INSERT` and `SELECT` queries with tables to exchange data between ClickHouse and MySQL. +Позволяет подключаться к базам данных на удалённом MySQL сервере и выполнять запросы `INSERT` и `SELECT` для обмена данными между ClickHouse и MySQL. -The `MySQL` database engine translate queries to the MySQL server, so you can perform operations such as `SHOW TABLES` or `SHOW CREATE TABLE`. +Движок баз данных `MySQL` транслирует запросы при передаче на сервер MySQL, что позволяет выполнять и другие виды запросов, например `SHOW TABLES` или `SHOW CREATE TABLE`. -You cannot perform with tables the following queries: +Не поддерживаемые виды запросов: - `ATTACH`/`DETACH` - `DROP` @@ -12,48 +12,45 @@ You cannot perform with tables the following queries: - `CREATE TABLE` - `ALTER` +## Создание базы данных -## Creating a Database - -``` sql +```sql CREATE DATABASE [IF NOT EXISTS] db_name [ON CLUSTER cluster] ENGINE = MySQL('host:port', 'database', 'user', 'password') ``` -**Engine Parameters** +**Параметры движка** -- `host:port` — MySQL server address. -- `database` — Remote database name. -- `user` — MySQL user. -- `password` — User password. +- `host:port` — адрес сервера MySQL. +- `database` — имя базы данных на удалённом сервере. +- `user` — пользователь MySQL. +- `password` — пароль пользователя. +## Поддержка типов данных -## Data Types Support +| MySQL | ClickHouse | +| ------ | ------------ | +| UNSIGNED TINYINT | [UInt8](../data_types/int_uint.md) | +| TINYINT | [Int8](../data_types/int_uint.md) | +| UNSIGNED SMALLINT | [UInt16](../data_types/int_uint.md) | +| SMALLINT | [Int16](../data_types/int_uint.md) | +| UNSIGNED INT, UNSIGNED MEDIUMINT | [UInt32](../data_types/int_uint.md) | +| INT, MEDIUMINT | [Int32](../data_types/int_uint.md) | +| UNSIGNED BIGINT | [UInt64](../data_types/int_uint.md) | +| BIGINT | [Int64](../data_types/int_uint.md) | +| FLOAT | [Float32](../data_types/float.md) | +| DOUBLE | [Float64](../data_types/float.md) | +| DATE | [Date](../data_types/date.md) | +| DATETIME, TIMESTAMP | [DateTime](../data_types/datetime.md) | +| BINARY | [FixedString](../data_types/fixedstring.md) | -MySQL | ClickHouse -------|------------ -UNSIGNED TINYINT | [UInt8](../data_types/int_uint.md) -TINYINT | [Int8](../data_types/int_uint.md) -UNSIGNED SMALLINT | [UInt16](../data_types/int_uint.md) -SMALLINT | [Int16](../data_types/int_uint.md) -UNSIGNED INT, UNSIGNED MEDIUMINT | [UInt32](../data_types/int_uint.md) -INT, MEDIUMINT | [Int32](../data_types/int_uint.md) -UNSIGNED BIGINT | [UInt64](../data_types/int_uint.md) -BIGINT | [Int64](../data_types/int_uint.md) -FLOAT | [Float32](../data_types/float.md) -DOUBLE | [Float64](../data_types/float.md) -DATE | [Date](../data_types/date.md) -DATETIME, TIMESTAMP | [DateTime](../data_types/datetime.md) -BINARY | [FixedString](../data_types/fixedstring.md) +Все прочие типы данных преобразуются в [String](../data_types/string.md). -All other MySQL data types are converted into [String](../data_types/string.md). +[Nullable](../data_types/nullable.md) поддержан. -[Nullable](../data_types/nullable.md) data type is supported. +## Примеры использования - -## Examples of Use - -Table in MySQL: +Таблица в MySQL: ``` mysql> USE test; @@ -77,14 +74,16 @@ mysql> select * from mysql_table; 1 row in set (0,00 sec) ``` -Database in ClickHouse, exchanging data with the MySQL server: +База данных в ClickHouse, позволяющая обмениваться данными с сервером MySQL: ```sql CREATE DATABASE mysql_db ENGINE = MySQL('localhost:3306', 'test', 'my_user', 'user_password') ``` + ```sql SHOW DATABASES ``` + ```text ┌─name─────┐ │ default │ @@ -92,31 +91,39 @@ SHOW DATABASES │ system │ └──────────┘ ``` + ```sql SHOW TABLES FROM mysql_db ``` + ```text ┌─name─────────┐ │ mysql_table │ └──────────────┘ ``` + ```sql SELECT * FROM mysql_db.mysql_table ``` + ```text ┌─int_id─┬─value─┐ │ 1 │ 2 │ └────────┴───────┘ ``` + ```sql INSERT INTO mysql_db.mysql_table VALUES (3,4) ``` + ```sql SELECT * FROM mysql_db.mysql_table ``` + ```text ┌─int_id─┬─value─┐ │ 1 │ 2 │ │ 3 │ 4 │ └────────┴───────┘ ``` + diff --git a/docs/ru/operations/table_engines/mergetree.md b/docs/ru/operations/table_engines/mergetree.md index 84ca187fd6a..d47336c2593 100644 --- a/docs/ru/operations/table_engines/mergetree.md +++ b/docs/ru/operations/table_engines/mergetree.md @@ -40,14 +40,15 @@ CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster] [ORDER BY expr] [PRIMARY KEY expr] [SAMPLE BY expr] +[TTL expr] [SETTINGS name=value, ...] ``` Описание параметров запроса смотрите в [описании запроса](../../query_language/create.md). -**Секции запроса** +### Секции запроса -- `ENGINE` — Имя и параметры движка. `ENGINE = MergeTree()`. `MergeTree` не имеет параметров. +- `ENGINE` — имя и параметры движка. `ENGINE = MergeTree()`. `MergeTree` не имеет параметров. - `PARTITION BY` — [ключ партиционирования](custom_partitioning_key.md). @@ -66,22 +67,22 @@ CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster] Если используется выражение для сэмплирования, то первичный ключ должен содержать его. Пример: `SAMPLE BY intHash32(UserID) ORDER BY (CounterID, EventDate, intHash32(UserID))`. -- `TTL` - выражение для задания времени хранения строк. +- `TTL` — выражение, определяющее длительность хранения строк. - Оно должно зависеть от стобца типа `Date` или `DateTime` и в качестве результата вычислять столбец типа `Date` или `DateTime`. Пример: - `TTL date + INTERVAL 1 DAY` + Должно зависеть от столбца `Date` или `DateTime` и возвращать столбец `Date` или `DateTime`. Пример:`TTL date + INTERVAL 1 DAY` - Подробнее смотрите в [TTL для стоблцов и таблиц](mergetree.md) + Дополнительные сведения смотрите в разделе [TTL для столбцов и таблиц](mergetree.md) - `SETTINGS` — дополнительные параметры, регулирующие поведение `MergeTree`: - `index_granularity` — гранулярность индекса. Число строк данных между «засечками» индекса. По умолчанию — 8192. Список всех доступных параметров можно посмотреть в [MergeTreeSettings.h](https://github.com/yandex/ClickHouse/blob/master/dbms/src/Storages/MergeTree/MergeTreeSettings.h). - `min_merge_bytes_to_use_direct_io` — минимальный объем данных, необходимый для прямого (небуферизованного) чтения/записи (direct I/O) на диск. При слиянии частей данных ClickHouse вычисляет общий объем хранения всех данных, подлежащих слиянию. Если общий объем хранения всех данных для чтения превышает `min_bytes_to_use_direct_io` байт, тогда ClickHouse использует флаг `O_DIRECT` при чтении данных с диска. Если `min_merge_bytes_to_use_direct_io = 0`, тогда прямой ввод-вывод отключен. Значение по умолчанию: `10 * 1024 * 1024 * 1024` байт. + - `merge_with_ttl_timeout` - Минимальное время в секундах для повторного выполнения слияний с TTL. По умолчанию - 86400 (1 день). **Пример задания секций** -``` +```sql ENGINE MergeTree() PARTITION BY toYYYYMM(EventDate) ORDER BY (CounterID, EventDate, intHash32(UserID)) SAMPLE BY intHash32(UserID) SETTINGS index_granularity=8192 ``` @@ -109,7 +110,8 @@ CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster] - `date-column` — имя столбца с типом [Date](../../data_types/date.md). На основе этого столбца ClickHouse автоматически создаёт партиции по месяцам. Имена партиций имеют формат `"YYYYMM"`. - `sampling_expression` — выражение для сэмплирования. -- `(primary, key)` — первичный ключ. Тип — [Tuple()](../../data_types/tuple.md- `index_granularity` — гранулярность индекса. Число строк данных между «засечками» индекса. Для большинства задач подходит значение 8192. +- `(primary, key)` — первичный ключ. Тип — [Tuple()](../../data_types/tuple.md) +- `index_granularity` — гранулярность индекса. Число строк данных между «засечками» индекса. Для большинства задач подходит значение 8192. **Пример** @@ -118,6 +120,7 @@ MergeTree(EventDate, intHash32(UserID), (CounterID, EventDate, intHash32(UserID) ``` Движок `MergeTree` сконфигурирован таким же образом, как и в примере выше для основного способа конфигурирования движка. + ## Хранение данных @@ -298,13 +301,42 @@ INDEX b (u64 * length(str), i32 + f64 * 100, date, str) TYPE set(100) GRANULARIT Чтения из таблицы автоматически распараллеливаются. +## TTL для столбцов и таблиц {#table_engine-mergetree-ttl} -## TTL для столбцов и таблиц +Определяет время жизни значений. -Данные с истекшим TTL удаляются во время слияний. +Секция `TTL` может быть установлена как для всей таблицы, так и для каждого отдельного столбца. Если установлены оба `TTL`, то ClickHouse использует тот, что истекает раньше. -Если TTL указан для столбца, то когда он истекает, значение заменяется на значение по умолчанию. Если все значения столбца обнулены в куске, то данные этого столбца удаляются с диска в куске. Если TTL указан для таблицы, то когда он истекает, удаляется строка. +Таблица должна иметь столбец типа [Date](../../data_types/date.md) или [DateTime](../../data_types/datetime.md). Для установки времени жизни данных, следует использовать операцию со столбцом с временем, например: -Когда истекает TTL для какого-нибудь значения или строки в куске, назначается внеочередное слияние. Чтобы контролировать частоту слияний с TTL, вы можете задать настройку `merge_with_ttl_timeout`. Если ее значение слишком мало, то будет происходить слишком много внеочередных слияний и мало обычных, вследствие чего может ухудшиться производительность. +``` +TTL time_column +TTL time_column + interval +``` + +Чтобы задать `interval`, используйте операторы [интервала времени](../../query_language/operators.md#operators-datetime). + +``` +TTL date_time + INTERVAL 1 MONTH +TTL date_time + INTERVAL 15 HOUR +``` + +**TTL столбца** + +Когда срок действия значений в столбце истечет, ClickHouse заменит их значениями по умолчанию для типа данных столбца. Если срок действия всех значений столбцов в части данных истек, ClickHouse удаляет столбец из куска данных в файловой системе. + +Секцию `TTL` нельзя использовать для ключевых столбцов. + +**TTL таблицы** + +Когда некоторые данные в таблице устаревают, ClickHouse удаляет все соответствующие строки. + +**Удаление данных** + +Данные с истекшим TTL удаляются, когда ClickHouse мёржит куски данных. + +Когда ClickHouse видит, что некоторые данные устарели, он выполняет внеплановые мёржи. Для управление частотой подобных мёржей, можно задать настройку [merge_with_ttl_timeout](#mergetree_setting-merge_with_ttl_timeout). Если её значение слишком низкое, придется выполнять много внеплановых мёржей, которые могут начать потреблять значительную долю ресурсов сервера. + +Если вы выполните запрос `SELECT` между слияниями вы можете получить устаревшие данные. Чтобы избежать этого используйте запрос [OPTIMIZE](../../query_language/misc.md#misc_operations-optimize) перед `SELECT`. [Оригинальная статья](https://clickhouse.yandex/docs/ru/operations/table_engines/mergetree/) diff --git a/docs/ru/query_language/create.md b/docs/ru/query_language/create.md index be75bb9ce51..ebbef603390 100644 --- a/docs/ru/query_language/create.md +++ b/docs/ru/query_language/create.md @@ -1,13 +1,30 @@ -## CREATE DATABASE +## CREATE DATABASE {#query_language-create-database} -Создание базы данных db\_name. +Создает базу данных. ```sql -CREATE DATABASE [IF NOT EXISTS] db_name [ON CLUSTER cluster] +CREATE DATABASE [IF NOT EXISTS] db_name [ON CLUSTER cluster] [ENGINE = engine(...)] ``` -`База данных` - это просто директория для таблиц. -Если написано `IF NOT EXISTS`, то запрос не будет возвращать ошибку, если база данных уже существует. +### Секции + +- `IF NOT EXISTS` + + Если база данных с именем `db_name` уже существует, то ClickHouse не создаёт базу данных и: + - Не генерирует исключение, если секция указана. + - Генерирует исключение, если секция не указана. + +- `ON CLUSTER` + + ClickHouse создаёт базу данных `db_name` на всех серверах указанного кластера. + +- `ENGINE` + + - [MySQL](../database_engines/mysql.md) + + Позволяет получать данные с удаленного сервера MySQL. + + По умолчанию ClickHouse использует собственный [движок баз данных](../database_engines/index.md). ## CREATE TABLE {#create-table-query} @@ -48,7 +65,7 @@ CREATE TABLE [IF NOT EXISTS] [db.]table_name ENGINE = engine AS SELECT ... Во всех случаях, если указано `IF NOT EXISTS`, то запрос не будет возвращать ошибку, если таблица уже существует. В этом случае, запрос будет ничего не делать. -После секции `ENGINE` в запросе могут использоваться и другие секции в зависимости от движка. Подробную документацию по созданию таблиц смотрите в описаниях [движков](../operations/table_engines/index.md#table_engines). +После секции `ENGINE` в запросе могут использоваться и другие секции в зависимости от движка. Подробную документацию по созданию таблиц смотрите в описаниях [движков таблиц](../operations/table_engines/index.md#table_engines). ### Значения по умолчанию {#create-default-values} @@ -88,11 +105,7 @@ CREATE TABLE [IF NOT EXISTS] [db.]table_name ENGINE = engine AS SELECT ... ### Выражение для TTL -Может быть указано только для таблиц семейства MergeTree. Выражение для указания времени хранения значений. Оно должно зависеть от стобца типа `Date` или `DateTime` и в качестве результата вычислять столбец типа `Date` или `DateTime`. Пример: - `TTL date + INTERVAL 1 DAY` - -Нельзя указывать TTL для ключевых столбцов. Подробнее смотрите в [TTL для стоблцов и таблиц](../operations/table_engines/mergetree.md) - +Определяет время хранения значений. Может быть указано только для таблиц семейства MergeTree. Подробнее смотрите в [TTL для столбцов и таблиц](../operations/table_engines/mergetree.md#table_engine-mergetree-ttl). ## Форматы сжатия для колонок diff --git a/docs/ru/query_language/functions/other_functions.md b/docs/ru/query_language/functions/other_functions.md index 1637c7bda93..00242eeb8c7 100644 --- a/docs/ru/query_language/functions/other_functions.md +++ b/docs/ru/query_language/functions/other_functions.md @@ -97,6 +97,9 @@ SELECT visibleWidth(NULL) Возвращает имя текущей базы данных. Эта функция может использоваться в параметрах движка таблицы в запросе CREATE TABLE там, где нужно указать базу данных. +## currentUser() +Возвращает логин пользователя, от имени которого исполняется запрос. В случае распределённого запроса, возвращается имя пользователя - инициатора запроса. + ## isFinite(x) Принимает Float32 или Float64 и возвращает UInt8, равный 1, если аргумент не бесконечный и не NaN, иначе 0. diff --git a/docs/ru/query_language/operators.md b/docs/ru/query_language/operators.md index a493e555af7..de38bb5b193 100644 --- a/docs/ru/query_language/operators.md +++ b/docs/ru/query_language/operators.md @@ -65,13 +65,13 @@ `a GLOBAL NOT IN ...` - функция `globalNotIn(a, b)` -## Оператор для работы с датами и временем +## Оператор для работы с датами и временем {#operators-datetime} ``` sql EXTRACT(part FROM date); ``` -Позволяет извлечь отдельные части из переданной даты. Например, можно получить месяц из даты, или минуты из времени. +Позволяет извлечь отдельные части из переданной даты. Например, можно получить месяц из даты, или минуты из времени. В параметре `part` указывается, какой фрагмент даты нужно получить. Доступные значения: @@ -99,8 +99,8 @@ SELECT EXTRACT(YEAR FROM toDate('2017-06-15')); ``` sql CREATE TABLE test.Orders ( - OrderId UInt64, - OrderName String, + OrderId UInt64, + OrderName String, OrderDate DateTime ) ENGINE = Log; @@ -110,11 +110,11 @@ ENGINE = Log; INSERT INTO test.Orders VALUES (1, 'Jarlsberg Cheese', toDateTime('2008-10-11 13:23:44')); ``` ``` sql -SELECT - toYear(OrderDate) AS OrderYear, - toMonth(OrderDate) AS OrderMonth, - toDayOfMonth(OrderDate) AS OrderDay, - toHour(OrderDate) AS OrderHour, +SELECT + toYear(OrderDate) AS OrderYear, + toMonth(OrderDate) AS OrderMonth, + toDayOfMonth(OrderDate) AS OrderDay, + toHour(OrderDate) AS OrderHour, toMinute(OrderDate) AS OrderMinute, toSecond(OrderDate) AS OrderSecond FROM test.Orders; diff --git a/docs/toc_zh.yml b/docs/toc_zh.yml index 6f9abc9b36c..d140ec88d64 100644 --- a/docs/toc_zh.yml +++ b/docs/toc_zh.yml @@ -47,26 +47,26 @@ nav: - 'AggregateFunction(name, types_of_arguments...)': 'data_types/nested_data_structures/aggregatefunction.md' - 'Tuple(T1, T2, ...)': 'data_types/tuple.md' - 'Nullable': 'data_types/nullable.md' - - 'Nested data structures': + - '嵌套数据结构': - 'hidden': 'data_types/nested_data_structures/index.md' - 'Nested(Name1 Type1, Name2 Type2, ...)': 'data_types/nested_data_structures/nested.md' - - 'Special data types': + - '特殊数据类型': - 'hidden': 'data_types/special_data_types/index.md' - 'Expression': 'data_types/special_data_types/expression.md' - 'Set': 'data_types/special_data_types/set.md' - 'Nothing': 'data_types/special_data_types/nothing.md' - - 'Domains': - - 'Overview': 'data_types/domains/overview.md' + - 'Domain类型': + - '介绍': 'data_types/domains/overview.md' - 'IPv4': 'data_types/domains/ipv4.md' - 'IPv6': 'data_types/domains/ipv6.md' -- 'Database Engines': - - 'Introduction': 'database_engines/index.md' +- '数据库引擎': + - '介绍': 'database_engines/index.md' - 'MySQL': 'database_engines/mysql.md' -- 'Table Engines': - - 'Introduction': 'operations/table_engines/index.md' - - 'MergeTree Family': +- '表引擎': + - '介绍': 'operations/table_engines/index.md' + - 'MergeTree': - 'MergeTree': 'operations/table_engines/mergetree.md' - 'Data Replication': 'operations/table_engines/replication.md' - 'Custom Partitioning Key': 'operations/table_engines/custom_partitioning_key.md' @@ -76,17 +76,17 @@ nav: - 'CollapsingMergeTree': 'operations/table_engines/collapsingmergetree.md' - 'VersionedCollapsingMergeTree': 'operations/table_engines/versionedcollapsingmergetree.md' - 'GraphiteMergeTree': 'operations/table_engines/graphitemergetree.md' - - 'Log Family': - - 'Introduction': 'operations/table_engines/log_family.md' + - 'Log': + - '介绍': 'operations/table_engines/log_family.md' - 'StripeLog': 'operations/table_engines/stripelog.md' - 'Log': 'operations/table_engines/log.md' - 'TinyLog': 'operations/table_engines/tinylog.md' - - 'Integrations': + - '外部表引擎': - 'Kafka': 'operations/table_engines/kafka.md' - 'MySQL': 'operations/table_engines/mysql.md' - 'JDBC': 'operations/table_engines/jdbc.md' - 'ODBC': 'operations/table_engines/odbc.md' - - 'Special': + - '其他表引擎': - 'Distributed': 'operations/table_engines/distributed.md' - 'External data': 'operations/table_engines/external_data.md' - 'Dictionary': 'operations/table_engines/dictionary.md' @@ -108,8 +108,8 @@ nav: - 'CREATE': 'query_language/create.md' - 'ALTER': 'query_language/alter.md' - 'SYSTEM': 'query_language/system.md' - - 'Other kinds of queries': 'query_language/misc.md' - - 'Functions': + - '其他类型的查询': 'query_language/misc.md' + - '函数': - '介绍': 'query_language/functions/index.md' - '算术函数': 'query_language/functions/arithmetic_functions.md' - '比较函数': 'query_language/functions/comparison_functions.md' @@ -142,13 +142,13 @@ nav: - 'Nullable处理函数': 'query_language/functions/functions_for_nulls.md' - '机器学习函数': 'query_language/functions/machine_learning_functions.md' - '其他函数': 'query_language/functions/other_functions.md' - - 'Aggregate functions': - - 'Introduction': 'query_language/agg_functions/index.md' - - 'Function reference': 'query_language/agg_functions/reference.md' - - 'Aggregate function combinators': 'query_language/agg_functions/combinators.md' - - 'Parametric aggregate functions': 'query_language/agg_functions/parametric_functions.md' - - 'Table functions': - - 'Introduction': 'query_language/table_functions/index.md' + - '聚合函数': + - '介绍': 'query_language/agg_functions/index.md' + - '函数列表': 'query_language/agg_functions/reference.md' + - '聚合函数组合子': 'query_language/agg_functions/combinators.md' + - '参数化聚合函数': 'query_language/agg_functions/parametric_functions.md' + - '表引擎函数': + - '介绍': 'query_language/table_functions/index.md' - 'file': 'query_language/table_functions/file.md' - 'merge': 'query_language/table_functions/merge.md' - 'numbers': 'query_language/table_functions/numbers.md' @@ -157,44 +157,44 @@ nav: - 'mysql': 'query_language/table_functions/mysql.md' - 'jdbc': 'query_language/table_functions/jdbc.md' - 'odbc': 'query_language/table_functions/odbc.md' - - 'Dictionaries': - - 'Introduction': 'query_language/dicts/index.md' - - 'External dictionaries': - - 'General description': 'query_language/dicts/external_dicts.md' - - 'Configuring an external dictionary': 'query_language/dicts/external_dicts_dict.md' - - 'Storing dictionaries in memory': 'query_language/dicts/external_dicts_dict_layout.md' - - 'Dictionary updates': 'query_language/dicts/external_dicts_dict_lifetime.md' - - 'Sources of external dictionaries': 'query_language/dicts/external_dicts_dict_sources.md' - - 'Dictionary key and fields': 'query_language/dicts/external_dicts_dict_structure.md' - - 'Internal dictionaries': 'query_language/dicts/internal_dicts.md' - - 'Operators': 'query_language/operators.md' - - 'General syntax': 'query_language/syntax.md' + - '字典': + - '介绍': 'query_language/dicts/index.md' + - '外部字典': + - '介绍': 'query_language/dicts/external_dicts.md' + - '配置外部字典': 'query_language/dicts/external_dicts_dict.md' + - '字典的内存布局': 'query_language/dicts/external_dicts_dict_layout.md' + - '字典的刷新策略': 'query_language/dicts/external_dicts_dict_lifetime.md' + - '字典的外部数据源': 'query_language/dicts/external_dicts_dict_sources.md' + - '字典的键和字段值': 'query_language/dicts/external_dicts_dict_structure.md' + - '内部字典': 'query_language/dicts/internal_dicts.md' + - '操作符': 'query_language/operators.md' + - '语法说明': 'query_language/syntax.md' - '运维': - - 'Introduction': 'operations/index.md' - - 'Requirements': 'operations/requirements.md' - - 'Monitoring': 'operations/monitoring.md' - - 'Troubleshooting': 'operations/troubleshooting.md' - - 'Usage recommendations': 'operations/tips.md' - - 'ClickHouse Update': 'operations/update.md' - - 'Access rights': 'operations/access_rights.md' - - 'Data backup': 'operations/backup.md' - - 'Configuration files': 'operations/configuration_files.md' - - 'Quotas': 'operations/quotas.md' - - 'System tables': 'operations/system_tables.md' - - 'Server configuration parameters': - - 'Introduction': 'operations/server_settings/index.md' - - 'Server settings': 'operations/server_settings/settings.md' - - 'Settings': - - 'Introduction': 'operations/settings/index.md' - - 'Permissions for queries': 'operations/settings/permissions_for_queries.md' - - 'Restrictions on query complexity': 'operations/settings/query_complexity.md' - - 'Settings': 'operations/settings/settings.md' - - 'Settings profiles': 'operations/settings/settings_profiles.md' - - 'User Settings': 'operations/settings/settings_users.md' - - 'Constraints on Settings': 'operations/settings/constraints_on_settings.md' - - 'Utilities': - - 'Overview': 'operations/utils/index.md' + - '介绍': 'operations/index.md' + - '环境要求': 'operations/requirements.md' + - '监控': 'operations/monitoring.md' + - '故障排查': 'operations/troubleshooting.md' + - '使用建议': 'operations/tips.md' + - '版本升级': 'operations/update.md' + - '访问权限控制': 'operations/access_rights.md' + - '数据备份': 'operations/backup.md' + - '配置文件': 'operations/configuration_files.md' + - '配额': 'operations/quotas.md' + - '系统表': 'operations/system_tables.md' + - 'Server参数配置': + - '介绍': 'operations/server_settings/index.md' + - 'Server参数说明': 'operations/server_settings/settings.md' + - 'Settings配置': + - '介绍': 'operations/settings/index.md' + - '查询权限管理': 'operations/settings/permissions_for_queries.md' + - '查询复杂性的限制': 'operations/settings/query_complexity.md' + - 'Setting列表': 'operations/settings/settings.md' + - 'Setting配置组': 'operations/settings/settings_profiles.md' + - '用户配置': 'operations/settings/settings_users.md' + - 'Settings的约束': 'operations/settings/constraints_on_settings.md' + - '常用工具': + - '介绍': 'operations/utils/index.md' - 'clickhouse-copier': 'operations/utils/clickhouse-copier.md' - 'clickhouse-local': 'operations/utils/clickhouse-local.md'