From 051290e6c912108986c896916db087c71230a121 Mon Sep 17 00:00:00 2001 From: serxa Date: Mon, 24 Jun 2024 12:26:46 +0000 Subject: [PATCH 001/227] Add throughput introspection for scheduler queues --- src/Common/EventRateMeter.h | 18 ++++++++++++++++-- src/Common/Scheduler/ISchedulerNode.h | 14 ++++++++++++++ src/Common/Scheduler/Nodes/FairPolicy.h | 3 +-- src/Common/Scheduler/Nodes/FifoQueue.h | 3 +-- src/Common/Scheduler/Nodes/PriorityPolicy.h | 3 +-- .../Scheduler/Nodes/SemaphoreConstraint.h | 3 +-- .../Scheduler/Nodes/ThrottlerConstraint.h | 3 +-- src/Common/Scheduler/SchedulerRoot.h | 3 +-- src/Storages/System/StorageSystemScheduler.cpp | 2 ++ 9 files changed, 38 insertions(+), 14 deletions(-) diff --git a/src/Common/EventRateMeter.h b/src/Common/EventRateMeter.h index 3a21a80ce8b..4c38d1d9371 100644 --- a/src/Common/EventRateMeter.h +++ b/src/Common/EventRateMeter.h @@ -14,8 +14,9 @@ namespace DB class EventRateMeter { public: - explicit EventRateMeter(double now, double period_) + explicit EventRateMeter(double now, double period_, double step_ = 0.0) : period(period_) + , step(step_) , half_decay_time(period * std::numbers::ln2) // for `ExponentiallySmoothedAverage::sumWeights()` to be equal to `1/period` { reset(now); @@ -38,7 +39,16 @@ public: if (now - period <= start) // precise counting mode events = ExponentiallySmoothedAverage(events.value + count, now); else // exponential smoothing mode - events.add(count, now, half_decay_time); + { + // Adding events too often lead to low precision due to smoothing too often, so we buffer new events and add them in steps + step_count += count; + if (step_start + step <= now) + { + events.add(step_count, now, half_decay_time); + step_start = now; + step_count = 0; + } + } } /// Compute average event rate throughout `[now - period, now]` period. @@ -58,16 +68,20 @@ public: void reset(double now) { start = now; + step_start = now; events = ExponentiallySmoothedAverage(); data_points = 0; } private: const double period; + const double step; // duration of a step const double half_decay_time; double start; // Instant in past without events before it; when measurement started or reset ExponentiallySmoothedAverage events; // Estimated number of events in the last `period` size_t data_points = 0; + double step_start; // start instant of the last step + double step_count = 0.0; // number of events accumulated since step start }; } diff --git a/src/Common/Scheduler/ISchedulerNode.h b/src/Common/Scheduler/ISchedulerNode.h index 81b491b0eda..c051829e336 100644 --- a/src/Common/Scheduler/ISchedulerNode.h +++ b/src/Common/Scheduler/ISchedulerNode.h @@ -3,6 +3,8 @@ #include #include #include +#include +#include #include #include @@ -176,6 +178,14 @@ protected: /// Postponed to be handled in scheduler thread, so it is intended to be called from outside. void scheduleActivation(); + /// Helper for introspection metrics + void incrementDequeued(ResourceCost cost) + { + dequeued_requests++; + dequeued_cost += cost; + throughput.add(static_cast(clock_gettime_ns())/1e9, cost); + } + public: EventQueue * const event_queue; String basename; @@ -189,6 +199,10 @@ public: std::atomic dequeued_cost{0}; std::atomic canceled_cost{0}; std::atomic busy_periods{0}; + + /// Average dequeued_cost per second + /// WARNING: Should only be accessed from the scheduler thread, so that locking is not required + EventRateMeter throughput{static_cast(clock_gettime_ns())/1e9, 2, 1}; }; using SchedulerNodePtr = std::shared_ptr; diff --git a/src/Common/Scheduler/Nodes/FairPolicy.h b/src/Common/Scheduler/Nodes/FairPolicy.h index 0a4e55c253b..fba637e979e 100644 --- a/src/Common/Scheduler/Nodes/FairPolicy.h +++ b/src/Common/Scheduler/Nodes/FairPolicy.h @@ -188,8 +188,7 @@ public: if (request) { - dequeued_requests++; - dequeued_cost += request->cost; + incrementDequeued(request->cost); return {request, heap_size > 0}; } } diff --git a/src/Common/Scheduler/Nodes/FifoQueue.h b/src/Common/Scheduler/Nodes/FifoQueue.h index 9ec997c06d2..9fbc6d1ae65 100644 --- a/src/Common/Scheduler/Nodes/FifoQueue.h +++ b/src/Common/Scheduler/Nodes/FifoQueue.h @@ -59,8 +59,7 @@ public: if (requests.empty()) busy_periods++; queue_cost -= result->cost; - dequeued_requests++; - dequeued_cost += result->cost; + incrementDequeued(result->cost); return {result, !requests.empty()}; } diff --git a/src/Common/Scheduler/Nodes/PriorityPolicy.h b/src/Common/Scheduler/Nodes/PriorityPolicy.h index 22a5155cfeb..91dc95600d5 100644 --- a/src/Common/Scheduler/Nodes/PriorityPolicy.h +++ b/src/Common/Scheduler/Nodes/PriorityPolicy.h @@ -122,8 +122,7 @@ public: if (request) { - dequeued_requests++; - dequeued_cost += request->cost; + incrementDequeued(request->cost); return {request, !items.empty()}; } } diff --git a/src/Common/Scheduler/Nodes/SemaphoreConstraint.h b/src/Common/Scheduler/Nodes/SemaphoreConstraint.h index 10fce536f5d..92c6af9db18 100644 --- a/src/Common/Scheduler/Nodes/SemaphoreConstraint.h +++ b/src/Common/Scheduler/Nodes/SemaphoreConstraint.h @@ -81,8 +81,7 @@ public: child_active = child_now_active; if (!active()) busy_periods++; - dequeued_requests++; - dequeued_cost += request->cost; + incrementDequeued(request->cost); return {request, active()}; } diff --git a/src/Common/Scheduler/Nodes/ThrottlerConstraint.h b/src/Common/Scheduler/Nodes/ThrottlerConstraint.h index f4a5795bb2b..56866336f50 100644 --- a/src/Common/Scheduler/Nodes/ThrottlerConstraint.h +++ b/src/Common/Scheduler/Nodes/ThrottlerConstraint.h @@ -89,8 +89,7 @@ public: child_active = child_now_active; if (!active()) busy_periods++; - dequeued_requests++; - dequeued_cost += request->cost; + incrementDequeued(request->cost); return {request, active()}; } diff --git a/src/Common/Scheduler/SchedulerRoot.h b/src/Common/Scheduler/SchedulerRoot.h index 7af42fdbbea..5307aadc3cc 100644 --- a/src/Common/Scheduler/SchedulerRoot.h +++ b/src/Common/Scheduler/SchedulerRoot.h @@ -162,8 +162,7 @@ public: if (request == nullptr) // Possible in case of request cancel, just retry continue; - dequeued_requests++; - dequeued_cost += request->cost; + incrementDequeued(request->cost); return {request, current != nullptr}; } } diff --git a/src/Storages/System/StorageSystemScheduler.cpp b/src/Storages/System/StorageSystemScheduler.cpp index 339a59e88a5..b42c807d6fc 100644 --- a/src/Storages/System/StorageSystemScheduler.cpp +++ b/src/Storages/System/StorageSystemScheduler.cpp @@ -31,6 +31,7 @@ ColumnsDescription StorageSystemScheduler::getColumnsDescription() {"dequeued_requests", std::make_shared(), "The total number of resource requests dequeued from this node."}, {"canceled_requests", std::make_shared(), "The total number of resource requests canceled from this node."}, {"dequeued_cost", std::make_shared(), "The sum of costs (e.g. size in bytes) of all requests dequeued from this node."}, + {"throughput", std::make_shared(), "Current average throughput (dequeued cost per second)."}, {"canceled_cost", std::make_shared(), "The sum of costs (e.g. size in bytes) of all requests canceled from this node."}, {"busy_periods", std::make_shared(), "The total number of deactivations of this node."}, {"vruntime", std::make_shared(std::make_shared()), @@ -96,6 +97,7 @@ void StorageSystemScheduler::fillData(MutableColumns & res_columns, ContextPtr c res_columns[i++]->insert(node->dequeued_requests.load()); res_columns[i++]->insert(node->canceled_requests.load()); res_columns[i++]->insert(node->dequeued_cost.load()); + res_columns[i++]->insert(node->throughput.rate(static_cast(clock_gettime_ns())/1e9)); res_columns[i++]->insert(node->canceled_cost.load()); res_columns[i++]->insert(node->busy_periods.load()); From b0ac0327d4d38a918e9be8fc499038cc43cafb2c Mon Sep 17 00:00:00 2001 From: serxa Date: Tue, 25 Jun 2024 11:48:29 +0000 Subject: [PATCH 002/227] Fix bug in EventRateMeter It was relying on ExponentiallySmoothedCounter::get() which is designed for specific 1 second time interval between points. Now sum of weights is computed separatly in `duration` field, giving very accurate measurements independent of interval. --- src/Common/EventRateMeter.h | 52 ++++++---------- src/Common/ProgressIndication.h | 2 +- src/Common/tests/gtest_event_rate_meter.cpp | 68 +++++++++++++++++++++ 3 files changed, 86 insertions(+), 36 deletions(-) create mode 100644 src/Common/tests/gtest_event_rate_meter.cpp diff --git a/src/Common/EventRateMeter.h b/src/Common/EventRateMeter.h index 4c38d1d9371..b8a9112428f 100644 --- a/src/Common/EventRateMeter.h +++ b/src/Common/EventRateMeter.h @@ -4,8 +4,6 @@ #include -#include - namespace DB { @@ -14,10 +12,10 @@ namespace DB class EventRateMeter { public: - explicit EventRateMeter(double now, double period_, double step_ = 0.0) + explicit EventRateMeter(double now, double period_, size_t heating_ = 0) : period(period_) - , step(step_) - , half_decay_time(period * std::numbers::ln2) // for `ExponentiallySmoothedAverage::sumWeights()` to be equal to `1/period` + , max_interval(period * 10) + , heating(heating_) { reset(now); } @@ -30,25 +28,11 @@ public: { // Remove data for initial heating stage that can present at the beginning of a query. // Otherwise it leads to wrong gradual increase of average value, turning algorithm into not very reactive. - if (count != 0.0 && ++data_points < 5) - { - start = events.time; - events = ExponentiallySmoothedAverage(); - } + if (count != 0.0 && data_points++ <= heating) + reset(events.time, data_points); - if (now - period <= start) // precise counting mode - events = ExponentiallySmoothedAverage(events.value + count, now); - else // exponential smoothing mode - { - // Adding events too often lead to low precision due to smoothing too often, so we buffer new events and add them in steps - step_count += count; - if (step_start + step <= now) - { - events.add(step_count, now, half_decay_time); - step_start = now; - step_count = 0; - } - } + duration.add(std::min(max_interval, now - duration.time), now, period); + events.add(count, now, period); } /// Compute average event rate throughout `[now - period, now]` period. @@ -59,29 +43,27 @@ public: add(now, 0); if (unlikely(now <= start)) return 0; - if (now - period <= start) // precise counting mode - return events.value / (now - start); - else // exponential smoothing mode - return events.get(half_decay_time); // equals to `events.value / period` + + // We do not use .get() because sum of weights will anyway be canceled out (optimization) + return events.value / duration.value; } - void reset(double now) + void reset(double now, size_t data_points_ = 0) { start = now; - step_start = now; events = ExponentiallySmoothedAverage(); - data_points = 0; + duration = ExponentiallySmoothedAverage(); + data_points = data_points_; } private: const double period; - const double step; // duration of a step - const double half_decay_time; + const double max_interval; + const size_t heating; double start; // Instant in past without events before it; when measurement started or reset - ExponentiallySmoothedAverage events; // Estimated number of events in the last `period` + ExponentiallySmoothedAverage duration; // Current duration of a period + ExponentiallySmoothedAverage events; // Estimated number of events in last `duration` seconds size_t data_points = 0; - double step_start; // start instant of the last step - double step_count = 0.0; // number of events accumulated since step start }; } diff --git a/src/Common/ProgressIndication.h b/src/Common/ProgressIndication.h index a9965785889..d925077a072 100644 --- a/src/Common/ProgressIndication.h +++ b/src/Common/ProgressIndication.h @@ -91,7 +91,7 @@ private: bool write_progress_on_update = false; - EventRateMeter cpu_usage_meter{static_cast(clock_gettime_ns()), 2'000'000'000 /*ns*/}; // average cpu utilization last 2 second + EventRateMeter cpu_usage_meter{static_cast(clock_gettime_ns()), 2'000'000'000 /*ns*/, 4}; // average cpu utilization last 2 second, skip first 4 points HostToTimesMap hosts_data; /// In case of all of the above: /// - clickhouse-local diff --git a/src/Common/tests/gtest_event_rate_meter.cpp b/src/Common/tests/gtest_event_rate_meter.cpp new file mode 100644 index 00000000000..91ceec5eef7 --- /dev/null +++ b/src/Common/tests/gtest_event_rate_meter.cpp @@ -0,0 +1,68 @@ +#include + +#include + +#include + + +TEST(EventRateMeter, ExponentiallySmoothedAverage) +{ + double target = 100.0; + + // The test is only correct for timestep of 1 second because of + // how sum of weights is implemented inside `ExponentiallySmoothedAverage` + double time_step = 1.0; + + for (double half_decay_time : { 0.1, 1.0, 10.0, 100.0}) + { + DB::ExponentiallySmoothedAverage esa; + + int steps = static_cast(half_decay_time * 30 / time_step); + for (int i = 1; i <= steps; ++i) + esa.add(target * time_step, i * time_step, half_decay_time); + double measured = esa.get(half_decay_time); + ASSERT_LE(std::fabs(measured - target), 1e-5 * target); + } +} + +TEST(EventRateMeter, ConstantRate) +{ + double target = 100.0; + + for (double period : {0.1, 1.0, 10.0}) + { + for (double time_step : {0.001, 0.01, 0.1, 1.0}) + { + DB::EventRateMeter erm(0.0, period); + + int steps = static_cast(period * 30 / time_step); + for (int i = 1; i <= steps; ++i) + erm.add(i * time_step, target * time_step); + double measured = erm.rate(steps * time_step); + // std::cout << "T=" << period << " dt=" << time_step << " measured=" << measured << std::endl; + ASSERT_LE(std::fabs(measured - target), 1e-5 * target); + } + } +} + +TEST(EventRateMeter, PreciseStart) +{ + double target = 100.0; + + for (double period : {0.1, 1.0, 10.0}) + { + for (double time_step : {0.001, 0.01, 0.1, 1.0}) + { + DB::EventRateMeter erm(0.0, period); + + int steps = static_cast(period / time_step); + for (int i = 1; i <= steps; ++i) + { + erm.add(i * time_step, target * time_step); + double measured = erm.rate(i * time_step); + // std::cout << "T=" << period << " dt=" << time_step << " measured=" << measured << std::endl; + ASSERT_LE(std::fabs(measured - target), 1e-5 * target); + } + } + } +} From 849fb83c9770fedb937dc59df73c0cc172e115bf Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Tue, 9 Jul 2024 17:37:17 +0800 Subject: [PATCH 003/227] add function printf --- .../functions/string-replace-functions.md | 25 ++ src/Functions/printf.cpp | 308 ++++++++++++++++++ .../0_stateless/032010_printf.reference | 16 + tests/queries/0_stateless/032010_printf.sql | 24 ++ 4 files changed, 373 insertions(+) create mode 100644 src/Functions/printf.cpp create mode 100644 tests/queries/0_stateless/032010_printf.reference create mode 100644 tests/queries/0_stateless/032010_printf.sql diff --git a/docs/en/sql-reference/functions/string-replace-functions.md b/docs/en/sql-reference/functions/string-replace-functions.md index 8793ebdd1a3..177790c983e 100644 --- a/docs/en/sql-reference/functions/string-replace-functions.md +++ b/docs/en/sql-reference/functions/string-replace-functions.md @@ -223,3 +223,28 @@ SELECT translateUTF8('Münchener Straße', 'üß', 'us') AS res; │ Munchener Strase │ └──────────────────┘ ``` + +## printf + +The `printf` function formats the given string with the values (strings, integers, floating-points etc.) listed in the arguments, similar to printf function in C++. The format string can contain format specifiers starting with `%` character. Anything not contained in `%` and the following format specifier is considered literal text and copied verbatim into the output. Literal `%` character can be escaped by `%%`. + +**Syntax** + +``` sql +printf(format, arg1, arg2, ...) +``` + +**Example** + +Query: + +``` sql +select printf('%%%s %s %d', 'Hello', 'World', 2024); +``` + + +``` response +┌─printf('%%%s %s %d', 'Hello', 'World', 2024)─┐ +│ %Hello World 2024 │ +└──────────────────────────────────────────────┘ +``` diff --git a/src/Functions/printf.cpp b/src/Functions/printf.cpp new file mode 100644 index 00000000000..cb21d5e39ad --- /dev/null +++ b/src/Functions/printf.cpp @@ -0,0 +1,308 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +namespace DB +{ +namespace ErrorCodes +{ + extern const int ILLEGAL_COLUMN; + extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; + extern const int ILLEGAL_TYPE_OF_ARGUMENT; +} + +namespace +{ + +class FunctionPrintf : public IFunction +{ +private: + ContextPtr context; + FunctionOverloadResolverPtr function_concat; + + struct Instruction + { + std::string_view format; + size_t rows; + bool is_literal; /// format is literal string without any argument + ColumnWithTypeAndName input; /// Only used when is_literal is false + + ColumnWithTypeAndName execute() + { + if (is_literal) + return executeLiteral(format); + else if (isColumnConst(*input.column)) + return executeConstant(input); + else + return executeNonconstant(input); + } + + String toString() const + { + std::ostringstream oss; + oss << "format:" << format << ", rows:" << rows << ", is_literal:" << is_literal << ", input:" << input.dumpStructure() + << std::endl; + return oss.str(); + } + + private: + ColumnWithTypeAndName executeLiteral(std::string_view literal) + { + ColumnWithTypeAndName res; + auto str_col = ColumnString::create(); + str_col->insert(fmt::sprintf(literal)); + res.column = ColumnConst::create(std::move(str_col), rows); + res.type = std::make_shared(); + return res; + } + + ColumnWithTypeAndName executeConstant(const ColumnWithTypeAndName & arg) + { + ColumnWithTypeAndName tmp_arg = arg; + const auto & const_col = static_cast(*arg.column); + tmp_arg.column = const_col.getDataColumnPtr(); + + ColumnWithTypeAndName tmp_res = executeNonconstant(tmp_arg); + return ColumnWithTypeAndName{ColumnConst::create(tmp_res.column, arg.column->size()), tmp_res.type, tmp_res.name}; + } + + ColumnWithTypeAndName executeNonconstant(const ColumnWithTypeAndName & arg) + { + size_t size = arg.column->size(); + auto res_col = ColumnString::create(); + auto & res_str = static_cast(*res_col); + auto & res_offsets = res_str.getOffsets(); + auto & res_chars = res_str.getChars(); + res_offsets.reserve_exact(size); + res_chars.reserve(format.size() * size * 2); + + String s; + WhichDataType which(arg.type); + +#define EXECUTE_BY_TYPE(IS_TYPE, GET_TYPE) \ + else if (which.IS_TYPE()) \ + { \ + for (size_t i = 0; i < size; ++i) \ + { \ + auto a = arg.column->GET_TYPE(i); \ + s = fmt::sprintf(format, a); \ + res_str.insertData(s.data(), s.size()); \ + } \ + } + + if (false) + ; + EXECUTE_BY_TYPE(isNativeInt, getInt) + EXECUTE_BY_TYPE(isNativeUInt, getUInt) + EXECUTE_BY_TYPE(isFloat32, getFloat32) + EXECUTE_BY_TYPE(isFloat64, getFloat64) + else if (which.isStringOrFixedString()) + { + for (size_t i = 0; i < size; ++i) + { + auto a = arg.column->getDataAt(i).toView(); + s = fmt::sprintf(format, a); + res_str.insertData(s.data(), s.size()); + } + } + else throw Exception( + ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, + "The argument type of function {} is {}, but native numeric or string type is expected", + FunctionPrintf::name, + arg.type->getName()); +#undef EXECUTE_BY_TYPE + + ColumnWithTypeAndName res; + res.name = arg.name; + res.type = std::make_shared(); + res.column = std::move(res_col); + return res; + } + }; + +public: + static constexpr auto name = "printf"; + + static FunctionPtr create(ContextPtr context) { return std::make_shared(context); } + + explicit FunctionPrintf(ContextPtr context_) + : context(context_), function_concat(FunctionFactory::instance().get("concat", context)) { } + + String getName() const override { return name; } + + bool isVariadic() const override { return true; } + + bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return true; } + + size_t getNumberOfArguments() const override { return 0; } + + bool useDefaultImplementationForConstants() const override { return false; } + ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {0}; } + + DataTypePtr getReturnTypeImpl(const DataTypes & arguments) const override + { + if (arguments.empty()) + throw Exception( + ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, + "Number of arguments for function {} doesn't match: passed {}, should be at least 1", + getName(), + arguments.size()); + + /// First pattern argument must have string type + if (!isString(arguments[0])) + throw Exception( + ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, + "The first argument type of function {} is {}, but String type is expected", + getName(), + arguments[0]->getName()); + + for (size_t i = 1; i < arguments.size(); ++i) + { + if (!isNativeNumber(arguments[i]) && !isStringOrFixedString(arguments[i])) + throw Exception( + ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, + "The {}-th argument type of function {} is {}, but native numeric or string type is expected", + i + 1, + getName(), + arguments[i]->getName()); + } + return std::make_shared(); + } + + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override + { + const ColumnPtr & c0 = arguments[0].column; + const ColumnConst * c0_const_string = typeid_cast(&*c0); + if (!c0_const_string) + throw Exception(ErrorCodes::ILLEGAL_COLUMN, "First argument of function {} must be constant string", getName()); + + String format = c0_const_string->getValue(); + auto instructions = buildInstructions(format, arguments, input_rows_count); + + ColumnsWithTypeAndName concat_args(instructions.size()); + for (size_t i = 0; i < instructions.size(); ++i) + { + std::cout << "instruction[" << i << "]:" << instructions[i].toString() << std::endl; + concat_args[i] = instructions[i].execute(); + std::cout << "concat_args[" << i << "]:" << concat_args[i].dumpStructure() << std::endl; + } + + auto res = function_concat->build(concat_args)->execute(concat_args, std::make_shared(), input_rows_count); + return res; + } + +private: + std::vector buildInstructions(const String & format , const ColumnsWithTypeAndName & arguments, size_t input_rows_count) const + { + std::vector instructions; + instructions.reserve(arguments.size()); + + auto append_instruction = [&](const char * begin, const char * end, const ColumnWithTypeAndName & arg) + { + Instruction instr; + instr.rows = input_rows_count; + instr.format = std::string_view(begin, end - begin); + + size_t size = end - begin; + if (size > 1 && begin[0] == '%' and begin[1] != '%') + { + instr.is_literal = false; + instr.input = arg; + } + else + { + instr.is_literal = true; + } + instructions.emplace_back(std::move(instr)); + }; + + auto check_index_range = [&](size_t idx) + { + if (idx >= arguments.size()) + throw Exception( + ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, + "Number of arguments for function {} doesn't match: passed {}, but format is {}", + getName(), + arguments.size(), + format); + }; + + const char * begin = format.data(); + const char * end = format.data() + format.size(); + const char * curr = begin; + size_t idx = 0; + while (curr < end) + { + const char * tmp = curr; + bool is_first = curr == begin; /// If current instruction is the first one + bool is_literal = false; /// If current instruction is literal string without any argument + if (is_first) + { + if (*curr != '%') + is_literal = true; + else if (curr + 1 < end && *(curr + 1) == '%') + is_literal = true; + else + ++idx; /// Skip first argument if first instruction is not literal + } + + if (!is_literal) + ++curr; + + while (curr < end) + { + if (*curr != '%') + ++curr; + else if (curr + 1 < end && *(curr + 1) == '%') + curr += 2; + else + { + check_index_range(idx); + append_instruction(tmp, curr, arguments[idx]); + ++idx; + break; + } + } + + if (curr == end) + { + check_index_range(idx); + append_instruction(tmp, curr, arguments[idx]); + ++idx; + } + } + + /// Check if all arguments are used + if (idx != arguments.size()) + throw Exception( + ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, + "Number of arguments for function {} doesn't match: passed {}, but format is {}", + getName(), + arguments.size(), + format); + + return instructions; + } +}; + +} + +REGISTER_FUNCTION(Printf) +{ + factory.registerFunction(); +} + +} diff --git a/tests/queries/0_stateless/032010_printf.reference b/tests/queries/0_stateless/032010_printf.reference new file mode 100644 index 00000000000..58501cbd0fc --- /dev/null +++ b/tests/queries/0_stateless/032010_printf.reference @@ -0,0 +1,16 @@ +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 diff --git a/tests/queries/0_stateless/032010_printf.sql b/tests/queries/0_stateless/032010_printf.sql new file mode 100644 index 00000000000..58fe081e499 --- /dev/null +++ b/tests/queries/0_stateless/032010_printf.sql @@ -0,0 +1,24 @@ +-- Testing integer formats +select printf('%%d: %d', 123) = '%d: 123'; +select printf('%%i: %i', 123) = '%i: 123'; +select printf('%%u: %u', 123) = '%u: 123'; +select printf('%%o: %o', 123) = '%o: 173'; +select printf('%%x: %x', 123) = '%x: 7b'; +select printf('%%X: %X', 123) = '%X: 7B'; + +-- Testing floating point formats +select printf('%%f: %f', 123.456) = '%f: 123.456000'; +select printf('%%F: %F', 123.456) = '%F: 123.456000'; +select printf('%%e: %e', 123.456) = '%e: 1.234560e+02'; +select printf('%%E: %E', 123.456) = '%E: 1.234560E+02'; +select printf('%%g: %g', 123.456) = '%g: 123.456'; +select printf('%%G: %G', 123.456) = '%G: 123.456'; +select printf('%%a: %a', 123.456) = '%a: 0x1.edd2f1a9fbe77p+6'; +select printf('%%A: %A', 123.456) = '%A: 0X1.EDD2F1A9FBE77P+6'; + +-- Testing character formats +select printf('%%s: %s', 'abc') = '%s: abc'; + + +-- Testing the %% specifier +select printf('%%%%: %%') = '%%: %'; \ No newline at end of file From e0e3842772ead940f53346cc087ea54e5e6aa8fa Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Wed, 10 Jul 2024 10:15:33 +0800 Subject: [PATCH 004/227] support printf --- src/Functions/printf.cpp | 6 +++--- utils/check-style/aspell-ignore/en/aspell-dict.txt | 1 + 2 files changed, 4 insertions(+), 3 deletions(-) diff --git a/src/Functions/printf.cpp b/src/Functions/printf.cpp index cb21d5e39ad..b2a0143a3f8 100644 --- a/src/Functions/printf.cpp +++ b/src/Functions/printf.cpp @@ -52,7 +52,7 @@ private: String toString() const { - std::ostringstream oss; + std::ostringstream oss; // STYLE_CHECK_ALLOW_STD_STRING_STREAM oss << "format:" << format << ", rows:" << rows << ", is_literal:" << is_literal << ", input:" << input.dumpStructure() << std::endl; return oss.str(); @@ -195,9 +195,9 @@ public: ColumnsWithTypeAndName concat_args(instructions.size()); for (size_t i = 0; i < instructions.size(); ++i) { - std::cout << "instruction[" << i << "]:" << instructions[i].toString() << std::endl; + // std::cout << "instruction[" << i << "]:" << instructions[i].toString() << std::endl; concat_args[i] = instructions[i].execute(); - std::cout << "concat_args[" << i << "]:" << concat_args[i].dumpStructure() << std::endl; + // std::cout << "concat_args[" << i << "]:" << concat_args[i].dumpStructure() << std::endl; } auto res = function_concat->build(concat_args)->execute(concat_args, std::make_shared(), input_rows_count); diff --git a/utils/check-style/aspell-ignore/en/aspell-dict.txt b/utils/check-style/aspell-ignore/en/aspell-dict.txt index 943caf918d6..21a9b759466 100644 --- a/utils/check-style/aspell-ignore/en/aspell-dict.txt +++ b/utils/check-style/aspell-ignore/en/aspell-dict.txt @@ -2264,6 +2264,7 @@ prettyspacemonoblock prettyspacenoescapes prettyspacenoescapesmonoblock prewhere +printf privateKeyFile privateKeyPassphraseHandler prlimit From 8b1bc00e9a6462a6dce2946510d83a85bc69a139 Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Wed, 10 Jul 2024 10:52:53 +0800 Subject: [PATCH 005/227] fix style --- src/Functions/printf.cpp | 33 +++++++++++++++++---------------- 1 file changed, 17 insertions(+), 16 deletions(-) diff --git a/src/Functions/printf.cpp b/src/Functions/printf.cpp index b2a0143a3f8..c7c6bd228a7 100644 --- a/src/Functions/printf.cpp +++ b/src/Functions/printf.cpp @@ -9,19 +9,19 @@ #include #include +#include #include #include #include #include -#include namespace DB { namespace ErrorCodes { - extern const int ILLEGAL_COLUMN; - extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; - extern const int ILLEGAL_TYPE_OF_ARGUMENT; +extern const int ILLEGAL_COLUMN; +extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; +extern const int ILLEGAL_TYPE_OF_ARGUMENT; } namespace @@ -54,7 +54,7 @@ private: { std::ostringstream oss; // STYLE_CHECK_ALLOW_STD_STRING_STREAM oss << "format:" << format << ", rows:" << rows << ", is_literal:" << is_literal << ", input:" << input.dumpStructure() - << std::endl; + << std::endl; return oss.str(); } @@ -93,15 +93,15 @@ private: WhichDataType which(arg.type); #define EXECUTE_BY_TYPE(IS_TYPE, GET_TYPE) \ - else if (which.IS_TYPE()) \ - { \ - for (size_t i = 0; i < size; ++i) \ - { \ - auto a = arg.column->GET_TYPE(i); \ - s = fmt::sprintf(format, a); \ - res_str.insertData(s.data(), s.size()); \ - } \ - } + else if (which.IS_TYPE()) \ + { \ + for (size_t i = 0; i < size; ++i) \ + { \ + auto a = arg.column->GET_TYPE(i); \ + s = fmt::sprintf(format, a); \ + res_str.insertData(s.data(), s.size()); \ + } \ + } if (false) ; @@ -205,7 +205,8 @@ public: } private: - std::vector buildInstructions(const String & format , const ColumnsWithTypeAndName & arguments, size_t input_rows_count) const + std::vector + buildInstructions(const String & format, const ColumnsWithTypeAndName & arguments, size_t input_rows_count) const { std::vector instructions; instructions.reserve(arguments.size()); @@ -248,7 +249,7 @@ private: { const char * tmp = curr; bool is_first = curr == begin; /// If current instruction is the first one - bool is_literal = false; /// If current instruction is literal string without any argument + bool is_literal = false; /// If current instruction is literal string without any argument if (is_first) { if (*curr != '%') From 96d063bcc39712c5a21a8e51244a9e216af8536a Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Thu, 11 Jul 2024 12:10:44 +0800 Subject: [PATCH 006/227] renmae ut files --- src/Functions/printf.cpp | 118 +++++++++++------- ...erence => 03203_function_printf.reference} | 0 ...0_printf.sql => 03203_function_printf.sql} | 0 3 files changed, 73 insertions(+), 45 deletions(-) rename tests/queries/0_stateless/{032010_printf.reference => 03203_function_printf.reference} (100%) rename tests/queries/0_stateless/{032010_printf.sql => 03203_function_printf.sql} (100%) diff --git a/src/Functions/printf.cpp b/src/Functions/printf.cpp index c7c6bd228a7..247c4a65daf 100644 --- a/src/Functions/printf.cpp +++ b/src/Functions/printf.cpp @@ -40,7 +40,7 @@ private: bool is_literal; /// format is literal string without any argument ColumnWithTypeAndName input; /// Only used when is_literal is false - ColumnWithTypeAndName execute() + ColumnWithTypeAndName execute() const { if (is_literal) return executeLiteral(format); @@ -50,7 +50,7 @@ private: return executeNonconstant(input); } - String toString() const + [[maybe_unused]] String toString() const { std::ostringstream oss; // STYLE_CHECK_ALLOW_STD_STRING_STREAM oss << "format:" << format << ", rows:" << rows << ", is_literal:" << is_literal << ", input:" << input.dumpStructure() @@ -59,7 +59,7 @@ private: } private: - ColumnWithTypeAndName executeLiteral(std::string_view literal) + ColumnWithTypeAndName executeLiteral(std::string_view literal) const { ColumnWithTypeAndName res; auto str_col = ColumnString::create(); @@ -69,7 +69,7 @@ private: return res; } - ColumnWithTypeAndName executeConstant(const ColumnWithTypeAndName & arg) + ColumnWithTypeAndName executeConstant(const ColumnWithTypeAndName & arg) const { ColumnWithTypeAndName tmp_arg = arg; const auto & const_col = static_cast(*arg.column); @@ -79,57 +79,85 @@ private: return ColumnWithTypeAndName{ColumnConst::create(tmp_res.column, arg.column->size()), tmp_res.type, tmp_res.name}; } - ColumnWithTypeAndName executeNonconstant(const ColumnWithTypeAndName & arg) + template + bool executeNumber(const IColumn & column, ColumnString::Chars & res_chars, ColumnString::Offsets & res_offsets) const + { + const ColumnVector * concrete_column = checkAndGetColumn>(&column); + if (!concrete_column) + return false; + + String s; + size_t curr_offset = 0; + const auto & data = concrete_column->getData(); + for (size_t i = 0; i < data.size(); ++i) + { + T a = data[i]; + s = fmt::sprintf(format, static_cast>(a)); + memcpy(&res_chars[curr_offset], s.data(), s.size()); + res_chars[curr_offset + s.size()] = 0; + + curr_offset += s.size() + 1; + res_offsets[i] = curr_offset; + } + return true; + } + + template + bool executeString(const IColumn & column, ColumnString::Chars & res_chars, ColumnString::Offsets & res_offsets) const + { + const COLUMN * concrete_column = checkAndGetColumn(&column); + if (!concrete_column) + return false; + + String s; + size_t curr_offset = 0; + for (size_t i = 0; i < concrete_column->size(); ++i) + { + auto a = concrete_column->getDataAt(i).toView(); + s = fmt::sprintf(format, a); + memcpy(&res_chars[curr_offset], s.data(), s.size()); + res_chars[curr_offset + s.size()] = 0; + + curr_offset += s.size() + 1; + res_offsets[i] = curr_offset; + } + return true; + } + + ColumnWithTypeAndName executeNonconstant(const ColumnWithTypeAndName & arg) const { size_t size = arg.column->size(); auto res_col = ColumnString::create(); auto & res_str = static_cast(*res_col); auto & res_offsets = res_str.getOffsets(); auto & res_chars = res_str.getChars(); - res_offsets.reserve_exact(size); - res_chars.reserve(format.size() * size * 2); + res_offsets.resize_exact(size); + res_chars.reserve(format.size() * size); - String s; WhichDataType which(arg.type); - -#define EXECUTE_BY_TYPE(IS_TYPE, GET_TYPE) \ - else if (which.IS_TYPE()) \ - { \ - for (size_t i = 0; i < size; ++i) \ - { \ - auto a = arg.column->GET_TYPE(i); \ - s = fmt::sprintf(format, a); \ - res_str.insertData(s.data(), s.size()); \ - } \ - } - - if (false) - ; - EXECUTE_BY_TYPE(isNativeInt, getInt) - EXECUTE_BY_TYPE(isNativeUInt, getUInt) - EXECUTE_BY_TYPE(isFloat32, getFloat32) - EXECUTE_BY_TYPE(isFloat64, getFloat64) - else if (which.isStringOrFixedString()) + if (which.isNativeNumber() + && (executeNumber(*arg.column, res_chars, res_offsets) || executeNumber(*arg.column, res_chars, res_offsets) + || executeNumber(*arg.column, res_chars, res_offsets) + || executeNumber(*arg.column, res_chars, res_offsets) + || executeNumber(*arg.column, res_chars, res_offsets) || executeNumber(*arg.column, res_chars, res_offsets) + || executeNumber(*arg.column, res_chars, res_offsets) + || executeNumber(*arg.column, res_chars, res_offsets))) { - for (size_t i = 0; i < size; ++i) - { - auto a = arg.column->getDataAt(i).toView(); - s = fmt::sprintf(format, a); - res_str.insertData(s.data(), s.size()); - } + return {std::move(res_col), std::make_shared(), arg.name}; } - else throw Exception( - ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, - "The argument type of function {} is {}, but native numeric or string type is expected", - FunctionPrintf::name, - arg.type->getName()); -#undef EXECUTE_BY_TYPE - - ColumnWithTypeAndName res; - res.name = arg.name; - res.type = std::make_shared(); - res.column = std::move(res_col); - return res; + else if ( + which.isStringOrFixedString() + && (executeString(*arg.column, res_chars, res_offsets) + || executeString(*arg.column, res_chars, res_offsets))) + { + return {std::move(res_col), std::make_shared(), arg.name}; + } + else + throw Exception( + ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, + "The argument type of function {} is {}, but native numeric or string type is expected", + FunctionPrintf::name, + arg.type->getName()); } }; diff --git a/tests/queries/0_stateless/032010_printf.reference b/tests/queries/0_stateless/03203_function_printf.reference similarity index 100% rename from tests/queries/0_stateless/032010_printf.reference rename to tests/queries/0_stateless/03203_function_printf.reference diff --git a/tests/queries/0_stateless/032010_printf.sql b/tests/queries/0_stateless/03203_function_printf.sql similarity index 100% rename from tests/queries/0_stateless/032010_printf.sql rename to tests/queries/0_stateless/03203_function_printf.sql From 88851ddb569f9ae8c61420bde99d2ad5f3d76889 Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Thu, 11 Jul 2024 12:15:44 +0800 Subject: [PATCH 007/227] improve uts --- .../0_stateless/03203_function_printf.reference | 5 +++++ .../queries/0_stateless/03203_function_printf.sql | 14 ++++++++++++-- 2 files changed, 17 insertions(+), 2 deletions(-) diff --git a/tests/queries/0_stateless/03203_function_printf.reference b/tests/queries/0_stateless/03203_function_printf.reference index 58501cbd0fc..338ecb0183d 100644 --- a/tests/queries/0_stateless/03203_function_printf.reference +++ b/tests/queries/0_stateless/03203_function_printf.reference @@ -14,3 +14,8 @@ 1 1 1 +1 +1 +1 +1 +1 diff --git a/tests/queries/0_stateless/03203_function_printf.sql b/tests/queries/0_stateless/03203_function_printf.sql index 58fe081e499..c41cbf0b5e9 100644 --- a/tests/queries/0_stateless/03203_function_printf.sql +++ b/tests/queries/0_stateless/03203_function_printf.sql @@ -19,6 +19,16 @@ select printf('%%A: %A', 123.456) = '%A: 0X1.EDD2F1A9FBE77P+6'; -- Testing character formats select printf('%%s: %s', 'abc') = '%s: abc'; - -- Testing the %% specifier -select printf('%%%%: %%') = '%%: %'; \ No newline at end of file +select printf('%%%%: %%') = '%%: %'; + +-- Testing integer formats with precision +select printf('%%.5d: %.5d', 123) = '%.5d: 00123'; + +-- Testing floating point formats with precision +select printf('%%.2f: %.2f', 123.456) = '%.2f: 123.46'; +select printf('%%.2e: %.2e', 123.456) = '%.2e: 1.23e+02'; +select printf('%%.2g: %.2g', 123.456) = '%.2g: 1.2e+02'; + +-- Testing character formats with precision +select printf('%%.2s: %.2s', 'abc') = '%.2s: ab'; \ No newline at end of file From d988399aa68ad8a6bc412bfc48d9fdefe63c1657 Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Thu, 11 Jul 2024 14:32:24 +0800 Subject: [PATCH 008/227] fix failed uts --- src/Functions/printf.cpp | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/src/Functions/printf.cpp b/src/Functions/printf.cpp index 247c4a65daf..a890b886338 100644 --- a/src/Functions/printf.cpp +++ b/src/Functions/printf.cpp @@ -141,7 +141,9 @@ private: || executeNumber(*arg.column, res_chars, res_offsets) || executeNumber(*arg.column, res_chars, res_offsets) || executeNumber(*arg.column, res_chars, res_offsets) || executeNumber(*arg.column, res_chars, res_offsets) - || executeNumber(*arg.column, res_chars, res_offsets))) + || executeNumber(*arg.column, res_chars, res_offsets) + || executeNumber(*arg.column, res_chars, res_offsets) + || executeNumber(*arg.column, res_chars, res_offsets))) { return {std::move(res_col), std::make_shared(), arg.name}; } From 21ca5f2d65c936a2c5b5fbc8f3f0c40d0ce60a6a Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Thu, 11 Jul 2024 15:32:05 +0800 Subject: [PATCH 009/227] fix failed ut --- .../02415_all_new_functions_must_be_documented.reference | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference b/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference index a152066a460..873b6bbb660 100644 --- a/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference +++ b/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference @@ -562,6 +562,7 @@ positionCaseInsensitive positionCaseInsensitiveUTF8 positionUTF8 pow +printf proportionsZTest protocol queryID From 68aebce89f30eb7766c420d25e852e21e27dfe7d Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Mon, 15 Jul 2024 14:41:37 +0800 Subject: [PATCH 010/227] fix failed uts --- src/Functions/printf.cpp | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/src/Functions/printf.cpp b/src/Functions/printf.cpp index a890b886338..3efe854a53b 100644 --- a/src/Functions/printf.cpp +++ b/src/Functions/printf.cpp @@ -93,6 +93,8 @@ private: { T a = data[i]; s = fmt::sprintf(format, static_cast>(a)); + + res_chars.resize(curr_offset + s.size() + 1); memcpy(&res_chars[curr_offset], s.data(), s.size()); res_chars[curr_offset + s.size()] = 0; @@ -115,6 +117,8 @@ private: { auto a = concrete_column->getDataAt(i).toView(); s = fmt::sprintf(format, a); + + res_chars.resize(curr_offset + s.size() + 1); memcpy(&res_chars[curr_offset], s.data(), s.size()); res_chars[curr_offset + s.size()] = 0; From df636100d3643031ee8d2102b737e992a6e44a40 Mon Sep 17 00:00:00 2001 From: Samuele Guerrini Date: Mon, 8 Jul 2024 12:00:53 +0200 Subject: [PATCH 011/227] add entry in documentation for use_same_password_for_base_backup --- docs/en/operations/backup.md | 1 + 1 file changed, 1 insertion(+) diff --git a/docs/en/operations/backup.md b/docs/en/operations/backup.md index fc861e25e9f..248fdbc156f 100644 --- a/docs/en/operations/backup.md +++ b/docs/en/operations/backup.md @@ -85,6 +85,7 @@ The BACKUP and RESTORE statements take a list of DATABASE and TABLE names, a des - `password` for the file on disk - `base_backup`: the destination of the previous backup of this source. For example, `Disk('backups', '1.zip')` - `use_same_s3_credentials_for_base_backup`: whether base backup to S3 should inherit credentials from the query. Only works with `S3`. + - `use_same_password_for_base_backup`: whether base backup archive should inherit the password from the query. - `structure_only`: if enabled, allows to only backup or restore the CREATE statements without the data of tables - `storage_policy`: storage policy for the tables being restored. See [Using Multiple Block Devices for Data Storage](../engines/table-engines/mergetree-family/mergetree.md#table_engine-mergetree-multiple-volumes). This setting is only applicable to the `RESTORE` command. The specified storage policy applies only to tables with an engine from the `MergeTree` family. - `s3_storage_class`: the storage class used for S3 backup. For example, `STANDARD` From d8c68a27744dca3f68e4de08036148c2ecfd2ed6 Mon Sep 17 00:00:00 2001 From: Samuele Guerrini Date: Mon, 8 Jul 2024 12:00:53 +0200 Subject: [PATCH 012/227] working incremental password protected backups --- src/Backups/BackupFactory.h | 1 + src/Backups/BackupImpl.cpp | 13 +++++++++++-- src/Backups/BackupImpl.h | 7 +++++-- src/Backups/BackupSettings.cpp | 1 + src/Backups/BackupSettings.h | 3 +++ src/Backups/BackupsWorker.cpp | 2 ++ src/Backups/RestoreSettings.cpp | 1 + src/Backups/RestoreSettings.h | 3 +++ .../registerBackupEngineAzureBlobStorage.cpp | 6 ++++-- src/Backups/registerBackupEngineS3.cpp | 6 ++++-- src/Backups/registerBackupEnginesFileAndDisk.cpp | 6 ++++-- 11 files changed, 39 insertions(+), 10 deletions(-) diff --git a/src/Backups/BackupFactory.h b/src/Backups/BackupFactory.h index e13a9a12ca2..807b8516d49 100644 --- a/src/Backups/BackupFactory.h +++ b/src/Backups/BackupFactory.h @@ -41,6 +41,7 @@ public: bool allow_s3_native_copy = true; bool allow_azure_native_copy = true; bool use_same_s3_credentials_for_base_backup = false; + bool use_same_password_for_base_backup = false; bool azure_attempt_to_create_container = true; ReadSettings read_settings; WriteSettings write_settings; diff --git a/src/Backups/BackupImpl.cpp b/src/Backups/BackupImpl.cpp index 3f972c36e47..23f067a62f5 100644 --- a/src/Backups/BackupImpl.cpp +++ b/src/Backups/BackupImpl.cpp @@ -92,7 +92,8 @@ BackupImpl::BackupImpl( std::shared_ptr reader_, const ContextPtr & context_, bool is_internal_backup_, - bool use_same_s3_credentials_for_base_backup_) + bool use_same_s3_credentials_for_base_backup_, + bool use_same_password_for_base_backup_) : backup_info(backup_info_) , backup_name_for_logging(backup_info.toStringForLogging()) , use_archive(!archive_params_.archive_name.empty()) @@ -104,6 +105,7 @@ BackupImpl::BackupImpl( , version(INITIAL_BACKUP_VERSION) , base_backup_info(base_backup_info_) , use_same_s3_credentials_for_base_backup(use_same_s3_credentials_for_base_backup_) + , use_same_password_for_base_backup(use_same_password_for_base_backup_) , log(getLogger("BackupImpl")) { open(); @@ -120,7 +122,8 @@ BackupImpl::BackupImpl( const std::shared_ptr & coordination_, const std::optional & backup_uuid_, bool deduplicate_files_, - bool use_same_s3_credentials_for_base_backup_) + bool use_same_s3_credentials_for_base_backup_, + bool use_same_password_for_base_backup_) : backup_info(backup_info_) , backup_name_for_logging(backup_info.toStringForLogging()) , use_archive(!archive_params_.archive_name.empty()) @@ -135,6 +138,7 @@ BackupImpl::BackupImpl( , base_backup_info(base_backup_info_) , deduplicate_files(deduplicate_files_) , use_same_s3_credentials_for_base_backup(use_same_s3_credentials_for_base_backup_) + , use_same_password_for_base_backup(use_same_password_for_base_backup_) , log(getLogger("BackupImpl")) { open(); @@ -258,6 +262,11 @@ std::shared_ptr BackupImpl::getBaseBackupUnlocked() const params.is_internal_backup = is_internal_backup; /// use_same_s3_credentials_for_base_backup should be inherited for base backups params.use_same_s3_credentials_for_base_backup = use_same_s3_credentials_for_base_backup; + /// use_same_password_for_base_backup should be inherited for base backups + params.use_same_password_for_base_backup = use_same_password_for_base_backup; + + if (params.use_same_password_for_base_backup) + params.password = archive_params.password; base_backup = BackupFactory::instance().createBackup(params); diff --git a/src/Backups/BackupImpl.h b/src/Backups/BackupImpl.h index 2b27e2ab090..d7846104c4c 100644 --- a/src/Backups/BackupImpl.h +++ b/src/Backups/BackupImpl.h @@ -41,7 +41,8 @@ public: std::shared_ptr reader_, const ContextPtr & context_, bool is_internal_backup_, - bool use_same_s3_credentials_for_base_backup_); + bool use_same_s3_credentials_for_base_backup_, + bool use_same_password_for_base_backup_); BackupImpl( const BackupInfo & backup_info_, @@ -53,7 +54,8 @@ public: const std::shared_ptr & coordination_, const std::optional & backup_uuid_, bool deduplicate_files_, - bool use_same_s3_credentials_for_base_backup_); + bool use_same_s3_credentials_for_base_backup_, + bool use_same_password_for_base_backup_); ~BackupImpl() override; @@ -153,6 +155,7 @@ private: bool writing_finalized = false; bool deduplicate_files = true; bool use_same_s3_credentials_for_base_backup = false; + bool use_same_password_for_base_backup = false; const LoggerPtr log; }; diff --git a/src/Backups/BackupSettings.cpp b/src/Backups/BackupSettings.cpp index e33880f88e3..37ddd344001 100644 --- a/src/Backups/BackupSettings.cpp +++ b/src/Backups/BackupSettings.cpp @@ -29,6 +29,7 @@ namespace ErrorCodes M(Bool, allow_s3_native_copy) \ M(Bool, allow_azure_native_copy) \ M(Bool, use_same_s3_credentials_for_base_backup) \ + M(Bool, use_same_password_for_base_backup) \ M(Bool, azure_attempt_to_create_container) \ M(Bool, read_from_filesystem_cache) \ M(UInt64, shard_num) \ diff --git a/src/Backups/BackupSettings.h b/src/Backups/BackupSettings.h index a6c4d5d7181..d8f48f6e1ac 100644 --- a/src/Backups/BackupSettings.h +++ b/src/Backups/BackupSettings.h @@ -50,6 +50,9 @@ struct BackupSettings /// Whether base backup to S3 should inherit credentials from the BACKUP query. bool use_same_s3_credentials_for_base_backup = false; + /// Wheter base backup archive should be unlocked using the same password as the incremental archive + bool use_same_password_for_base_backup = false; + /// Whether a new Azure container should be created if it does not exist (requires permissions at storage account level) bool azure_attempt_to_create_container = true; diff --git a/src/Backups/BackupsWorker.cpp b/src/Backups/BackupsWorker.cpp index 15a7d7c1eca..0614fb2da01 100644 --- a/src/Backups/BackupsWorker.cpp +++ b/src/Backups/BackupsWorker.cpp @@ -601,6 +601,7 @@ void BackupsWorker::doBackup( backup_create_params.allow_s3_native_copy = backup_settings.allow_s3_native_copy; backup_create_params.allow_azure_native_copy = backup_settings.allow_azure_native_copy; backup_create_params.use_same_s3_credentials_for_base_backup = backup_settings.use_same_s3_credentials_for_base_backup; + backup_create_params.use_same_password_for_base_backup = backup_settings.use_same_password_for_base_backup; backup_create_params.azure_attempt_to_create_container = backup_settings.azure_attempt_to_create_container; backup_create_params.read_settings = getReadSettingsForBackup(context, backup_settings); backup_create_params.write_settings = getWriteSettingsForBackup(context); @@ -912,6 +913,7 @@ void BackupsWorker::doRestore( backup_open_params.password = restore_settings.password; backup_open_params.allow_s3_native_copy = restore_settings.allow_s3_native_copy; backup_open_params.use_same_s3_credentials_for_base_backup = restore_settings.use_same_s3_credentials_for_base_backup; + backup_open_params.use_same_password_for_base_backup = restore_settings.use_same_password_for_base_backup; backup_open_params.read_settings = getReadSettingsForRestore(context); backup_open_params.write_settings = getWriteSettingsForRestore(context); backup_open_params.is_internal_backup = restore_settings.internal; diff --git a/src/Backups/RestoreSettings.cpp b/src/Backups/RestoreSettings.cpp index 7bbfd9ed751..a974fc11d00 100644 --- a/src/Backups/RestoreSettings.cpp +++ b/src/Backups/RestoreSettings.cpp @@ -164,6 +164,7 @@ namespace M(RestoreUDFCreationMode, create_function) \ M(Bool, allow_s3_native_copy) \ M(Bool, use_same_s3_credentials_for_base_backup) \ + M(Bool, use_same_password_for_base_backup) \ M(Bool, restore_broken_parts_as_detached) \ M(Bool, internal) \ M(String, host_id) \ diff --git a/src/Backups/RestoreSettings.h b/src/Backups/RestoreSettings.h index 06ecbc80aef..0fe5ee1a4bf 100644 --- a/src/Backups/RestoreSettings.h +++ b/src/Backups/RestoreSettings.h @@ -113,6 +113,9 @@ struct RestoreSettings /// Whether base backup from S3 should inherit credentials from the RESTORE query. bool use_same_s3_credentials_for_base_backup = false; + /// Wheter base backup archive should be unlocked using the same password as the incremental archive + bool use_same_password_for_base_backup = false; + /// If it's true RESTORE won't stop on broken parts while restoring, instead they will be restored as detached parts /// to the `detached` folder with names starting with `broken-from-backup'. bool restore_broken_parts_as_detached = false; diff --git a/src/Backups/registerBackupEngineAzureBlobStorage.cpp b/src/Backups/registerBackupEngineAzureBlobStorage.cpp index 626df99b00c..45f0386375a 100644 --- a/src/Backups/registerBackupEngineAzureBlobStorage.cpp +++ b/src/Backups/registerBackupEngineAzureBlobStorage.cpp @@ -141,7 +141,8 @@ void registerBackupEngineAzureBlobStorage(BackupFactory & factory) reader, params.context, params.is_internal_backup, - /* use_same_s3_credentials_for_base_backup*/ false); + /* use_same_s3_credentials_for_base_backup*/ false, + params.use_same_password_for_base_backup); } else { @@ -164,7 +165,8 @@ void registerBackupEngineAzureBlobStorage(BackupFactory & factory) params.backup_coordination, params.backup_uuid, params.deduplicate_files, - /* use_same_s3_credentials_for_base_backup */ false); + /* use_same_s3_credentials_for_base_backup */ false, + params.use_same_password_for_base_backup); } #else throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, "AzureBlobStorage support is disabled"); diff --git a/src/Backups/registerBackupEngineS3.cpp b/src/Backups/registerBackupEngineS3.cpp index 59ed9506af0..79e3e945557 100644 --- a/src/Backups/registerBackupEngineS3.cpp +++ b/src/Backups/registerBackupEngineS3.cpp @@ -120,7 +120,8 @@ void registerBackupEngineS3(BackupFactory & factory) reader, params.context, params.is_internal_backup, - params.use_same_s3_credentials_for_base_backup); + params.use_same_s3_credentials_for_base_backup, + params.use_same_password_for_base_backup); } else { @@ -144,7 +145,8 @@ void registerBackupEngineS3(BackupFactory & factory) params.backup_coordination, params.backup_uuid, params.deduplicate_files, - params.use_same_s3_credentials_for_base_backup); + params.use_same_s3_credentials_for_base_backup, + params.use_same_password_for_base_backup); } #else throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, "S3 support is disabled"); diff --git a/src/Backups/registerBackupEnginesFileAndDisk.cpp b/src/Backups/registerBackupEnginesFileAndDisk.cpp index 35263d39cba..c486f79a77a 100644 --- a/src/Backups/registerBackupEnginesFileAndDisk.cpp +++ b/src/Backups/registerBackupEnginesFileAndDisk.cpp @@ -178,7 +178,8 @@ void registerBackupEnginesFileAndDisk(BackupFactory & factory) reader, params.context, params.is_internal_backup, - params.use_same_s3_credentials_for_base_backup); + params.use_same_s3_credentials_for_base_backup, + params.use_same_password_for_base_backup); } else { @@ -197,7 +198,8 @@ void registerBackupEnginesFileAndDisk(BackupFactory & factory) params.backup_coordination, params.backup_uuid, params.deduplicate_files, - params.use_same_s3_credentials_for_base_backup); + params.use_same_s3_credentials_for_base_backup, + params.use_same_password_for_base_backup); } }; From 74de7833b8d429edcac9c4735143fc92349abdaf Mon Sep 17 00:00:00 2001 From: Samuele Guerrini Date: Mon, 8 Jul 2024 12:34:53 +0200 Subject: [PATCH 013/227] fix typo --- src/Backups/BackupSettings.h | 2 +- src/Backups/RestoreSettings.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/src/Backups/BackupSettings.h b/src/Backups/BackupSettings.h index d8f48f6e1ac..0abeb897db4 100644 --- a/src/Backups/BackupSettings.h +++ b/src/Backups/BackupSettings.h @@ -50,7 +50,7 @@ struct BackupSettings /// Whether base backup to S3 should inherit credentials from the BACKUP query. bool use_same_s3_credentials_for_base_backup = false; - /// Wheter base backup archive should be unlocked using the same password as the incremental archive + /// Whether base backup archive should be unlocked using the same password as the incremental archive bool use_same_password_for_base_backup = false; /// Whether a new Azure container should be created if it does not exist (requires permissions at storage account level) diff --git a/src/Backups/RestoreSettings.h b/src/Backups/RestoreSettings.h index 0fe5ee1a4bf..fe07a0a7208 100644 --- a/src/Backups/RestoreSettings.h +++ b/src/Backups/RestoreSettings.h @@ -113,7 +113,7 @@ struct RestoreSettings /// Whether base backup from S3 should inherit credentials from the RESTORE query. bool use_same_s3_credentials_for_base_backup = false; - /// Wheter base backup archive should be unlocked using the same password as the incremental archive + /// Whether base backup archive should be unlocked using the same password as the incremental archive bool use_same_password_for_base_backup = false; /// If it's true RESTORE won't stop on broken parts while restoring, instead they will be restored as detached parts From 12fb08648670dcf39659596ccb552f3462004fd8 Mon Sep 17 00:00:00 2001 From: Samuele Guerrini Date: Mon, 8 Jul 2024 16:25:07 +0200 Subject: [PATCH 014/227] added test for use_same_password_for_base_backup --- ...se_same_password_for_base_backup.reference | 21 ++++++++ ...ackup_use_same_password_for_base_backup.sh | 50 +++++++++++++++++++ 2 files changed, 71 insertions(+) create mode 100644 tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference create mode 100755 tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.sh diff --git a/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference b/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference new file mode 100644 index 00000000000..7354d50a7c0 --- /dev/null +++ b/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference @@ -0,0 +1,21 @@ +use_same_password_for_base_backup +base +BACKUP_CREATED +add_more_data_1 +inc_1 +BACKUP_CREATED +add_more_data_2 +inc_2 +BACKUP_CREATED +inc_2_bad +Couldn't unpack zip archive '02843_backup_use_same_password_for_base_backup_test_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE) +restore_inc_1 +RESTORED +restore_inc_2 +RESTORED +restore_inc_2_bad +Couldn't unpack zip archive '02843_backup_use_same_password_for_base_backup_test_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE) +count_inc_1 +20 +count_inc_2 +30 diff --git a/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.sh b/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.sh new file mode 100755 index 00000000000..a2b1a953e24 --- /dev/null +++ b/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.sh @@ -0,0 +1,50 @@ +#!/usr/bin/env bash +# Tags: no-fasttest + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +$CLICKHOUSE_CLIENT -nm -q " + DROP TABLE IF EXISTS data; + DROP TABLE IF EXISTS data_1; + DROP TABLE IF EXISTS data_2; + CREATE TABLE data (key Int) ENGINE=MergeTree() ORDER BY tuple(); + INSERT INTO data SELECT * from numbers(10); +" + +echo 'use_same_password_for_base_backup' +echo "base" +$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_base.zip') SETTINGS password='password';" | cut -f2 + +echo 'add_more_data_1' +$CLICKHOUSE_CLIENT -q "INSERT INTO data SELECT * FROM numbers(10,10);" + +echo "inc_1" +$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip') SETTINGS base_backup=Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_base.zip'),password='password',use_same_password_for_base_backup=1" | cut -f2 + +echo 'add_more_data_2' +$CLICKHOUSE_CLIENT -q "INSERT INTO data SELECT * FROM numbers(20,10);" + +echo "inc_2" +$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2.zip') SETTINGS base_backup=Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip'),password='password',use_same_password_for_base_backup=1" | cut -f2 + +echo "inc_2_bad" +$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2_bad.zip') SETTINGS base_backup=Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip'),password='password'" |& grep -m1 -o "Couldn't unpack zip archive '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE)" + +echo "restore_inc_1" +$CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_1 FROM Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip') SETTINGS password='password',use_same_password_for_base_backup=1" | cut -f2 + +echo "restore_inc_2" +$CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_2 FROM Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2.zip') SETTINGS password='password',use_same_password_for_base_backup=1" | cut -f2 + +echo "restore_inc_2_bad" +$CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_2 FROM Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2.zip') SETTINGS password='password'" |& grep -m1 -o "Couldn't unpack zip archive '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE)" + +echo "count_inc_1" +$CLICKHOUSE_CLIENT -q "SELECT COUNT(*) FROM data_1" | cut -f2 + +echo "count_inc_2" +$CLICKHOUSE_CLIENT -q "SELECT COUNT(*) FROM data_2" | cut -f2 + +exit 0 From 6f25aacd71948b682630889e75774eb4f2469aaa Mon Sep 17 00:00:00 2001 From: Samuele Guerrini Date: Tue, 9 Jul 2024 16:36:59 +0200 Subject: [PATCH 015/227] use CLICKHOUSE_TEST_NAME instead of CLICKHOUSE_TEST_UNIQUE_NAME for backup filename --- ...kup_use_same_password_for_base_backup.reference | 4 ++-- ...843_backup_use_same_password_for_base_backup.sh | 14 +++++++------- 2 files changed, 9 insertions(+), 9 deletions(-) diff --git a/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference b/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference index 7354d50a7c0..1a331cca46b 100644 --- a/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference +++ b/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference @@ -8,13 +8,13 @@ add_more_data_2 inc_2 BACKUP_CREATED inc_2_bad -Couldn't unpack zip archive '02843_backup_use_same_password_for_base_backup_test_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE) +Couldn't unpack zip archive '02843_backup_use_same_password_for_base_backup_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE) restore_inc_1 RESTORED restore_inc_2 RESTORED restore_inc_2_bad -Couldn't unpack zip archive '02843_backup_use_same_password_for_base_backup_test_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE) +Couldn't unpack zip archive '02843_backup_use_same_password_for_base_backup_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE) count_inc_1 20 count_inc_2 diff --git a/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.sh b/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.sh index a2b1a953e24..4c5bec3775c 100755 --- a/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.sh +++ b/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.sh @@ -15,31 +15,31 @@ $CLICKHOUSE_CLIENT -nm -q " echo 'use_same_password_for_base_backup' echo "base" -$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_base.zip') SETTINGS password='password';" | cut -f2 +$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_NAME}_base.zip') SETTINGS password='password';" | cut -f2 echo 'add_more_data_1' $CLICKHOUSE_CLIENT -q "INSERT INTO data SELECT * FROM numbers(10,10);" echo "inc_1" -$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip') SETTINGS base_backup=Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_base.zip'),password='password',use_same_password_for_base_backup=1" | cut -f2 +$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_NAME}_inc_1.zip') SETTINGS base_backup=Disk('backups', '${CLICKHOUSE_TEST_NAME}_base.zip'),password='password',use_same_password_for_base_backup=1" | cut -f2 echo 'add_more_data_2' $CLICKHOUSE_CLIENT -q "INSERT INTO data SELECT * FROM numbers(20,10);" echo "inc_2" -$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2.zip') SETTINGS base_backup=Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip'),password='password',use_same_password_for_base_backup=1" | cut -f2 +$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_NAME}_inc_2.zip') SETTINGS base_backup=Disk('backups', '${CLICKHOUSE_TEST_NAME}_inc_1.zip'),password='password',use_same_password_for_base_backup=1" | cut -f2 echo "inc_2_bad" -$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2_bad.zip') SETTINGS base_backup=Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip'),password='password'" |& grep -m1 -o "Couldn't unpack zip archive '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE)" +$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_NAME}_inc_2_bad.zip') SETTINGS base_backup=Disk('backups', '${CLICKHOUSE_TEST_NAME}_inc_1.zip'),password='password'" |& grep -m1 -o "Couldn't unpack zip archive '${CLICKHOUSE_TEST_NAME}_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE)" echo "restore_inc_1" -$CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_1 FROM Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip') SETTINGS password='password',use_same_password_for_base_backup=1" | cut -f2 +$CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_1 FROM Disk('backups', '${CLICKHOUSE_TEST_NAME}_inc_1.zip') SETTINGS password='password',use_same_password_for_base_backup=1" | cut -f2 echo "restore_inc_2" -$CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_2 FROM Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2.zip') SETTINGS password='password',use_same_password_for_base_backup=1" | cut -f2 +$CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_2 FROM Disk('backups', '${CLICKHOUSE_TEST_NAME}_inc_2.zip') SETTINGS password='password',use_same_password_for_base_backup=1" | cut -f2 echo "restore_inc_2_bad" -$CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_2 FROM Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2.zip') SETTINGS password='password'" |& grep -m1 -o "Couldn't unpack zip archive '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE)" +$CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_2 FROM Disk('backups', '${CLICKHOUSE_TEST_NAME}_inc_2.zip') SETTINGS password='password'" |& grep -m1 -o "Couldn't unpack zip archive '${CLICKHOUSE_TEST_NAME}_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE)" echo "count_inc_1" $CLICKHOUSE_CLIENT -q "SELECT COUNT(*) FROM data_1" | cut -f2 From cdd955f421e442cdb776e95e8e7ae8608bc8636e Mon Sep 17 00:00:00 2001 From: Samuele Guerrini Date: Wed, 10 Jul 2024 09:20:39 +0200 Subject: [PATCH 016/227] Revert "use CLICKHOUSE_TEST_NAME instead of CLICKHOUSE_TEST_UNIQUE_NAME for backup filename" This reverts commit 72f6368a2ede6a01390db770f1b9ddfa00d3f1fe. --- ...kup_use_same_password_for_base_backup.reference | 4 ++-- ...843_backup_use_same_password_for_base_backup.sh | 14 +++++++------- 2 files changed, 9 insertions(+), 9 deletions(-) diff --git a/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference b/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference index 1a331cca46b..7354d50a7c0 100644 --- a/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference +++ b/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference @@ -8,13 +8,13 @@ add_more_data_2 inc_2 BACKUP_CREATED inc_2_bad -Couldn't unpack zip archive '02843_backup_use_same_password_for_base_backup_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE) +Couldn't unpack zip archive '02843_backup_use_same_password_for_base_backup_test_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE) restore_inc_1 RESTORED restore_inc_2 RESTORED restore_inc_2_bad -Couldn't unpack zip archive '02843_backup_use_same_password_for_base_backup_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE) +Couldn't unpack zip archive '02843_backup_use_same_password_for_base_backup_test_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE) count_inc_1 20 count_inc_2 diff --git a/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.sh b/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.sh index 4c5bec3775c..a2b1a953e24 100755 --- a/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.sh +++ b/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.sh @@ -15,31 +15,31 @@ $CLICKHOUSE_CLIENT -nm -q " echo 'use_same_password_for_base_backup' echo "base" -$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_NAME}_base.zip') SETTINGS password='password';" | cut -f2 +$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_base.zip') SETTINGS password='password';" | cut -f2 echo 'add_more_data_1' $CLICKHOUSE_CLIENT -q "INSERT INTO data SELECT * FROM numbers(10,10);" echo "inc_1" -$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_NAME}_inc_1.zip') SETTINGS base_backup=Disk('backups', '${CLICKHOUSE_TEST_NAME}_base.zip'),password='password',use_same_password_for_base_backup=1" | cut -f2 +$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip') SETTINGS base_backup=Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_base.zip'),password='password',use_same_password_for_base_backup=1" | cut -f2 echo 'add_more_data_2' $CLICKHOUSE_CLIENT -q "INSERT INTO data SELECT * FROM numbers(20,10);" echo "inc_2" -$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_NAME}_inc_2.zip') SETTINGS base_backup=Disk('backups', '${CLICKHOUSE_TEST_NAME}_inc_1.zip'),password='password',use_same_password_for_base_backup=1" | cut -f2 +$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2.zip') SETTINGS base_backup=Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip'),password='password',use_same_password_for_base_backup=1" | cut -f2 echo "inc_2_bad" -$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_NAME}_inc_2_bad.zip') SETTINGS base_backup=Disk('backups', '${CLICKHOUSE_TEST_NAME}_inc_1.zip'),password='password'" |& grep -m1 -o "Couldn't unpack zip archive '${CLICKHOUSE_TEST_NAME}_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE)" +$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2_bad.zip') SETTINGS base_backup=Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip'),password='password'" |& grep -m1 -o "Couldn't unpack zip archive '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE)" echo "restore_inc_1" -$CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_1 FROM Disk('backups', '${CLICKHOUSE_TEST_NAME}_inc_1.zip') SETTINGS password='password',use_same_password_for_base_backup=1" | cut -f2 +$CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_1 FROM Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip') SETTINGS password='password',use_same_password_for_base_backup=1" | cut -f2 echo "restore_inc_2" -$CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_2 FROM Disk('backups', '${CLICKHOUSE_TEST_NAME}_inc_2.zip') SETTINGS password='password',use_same_password_for_base_backup=1" | cut -f2 +$CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_2 FROM Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2.zip') SETTINGS password='password',use_same_password_for_base_backup=1" | cut -f2 echo "restore_inc_2_bad" -$CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_2 FROM Disk('backups', '${CLICKHOUSE_TEST_NAME}_inc_2.zip') SETTINGS password='password'" |& grep -m1 -o "Couldn't unpack zip archive '${CLICKHOUSE_TEST_NAME}_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE)" +$CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_2 FROM Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2.zip') SETTINGS password='password'" |& grep -m1 -o "Couldn't unpack zip archive '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE)" echo "count_inc_1" $CLICKHOUSE_CLIENT -q "SELECT COUNT(*) FROM data_1" | cut -f2 From ba1c6fe3ef3a7f263939b2ae9e6c249d9c6b38c0 Mon Sep 17 00:00:00 2001 From: Samuele Guerrini Date: Wed, 10 Jul 2024 09:23:19 +0200 Subject: [PATCH 017/227] shorter grep output --- .../02843_backup_use_same_password_for_base_backup.reference | 4 ++-- .../02843_backup_use_same_password_for_base_backup.sh | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference b/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference index 7354d50a7c0..cbcb6b4cb7c 100644 --- a/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference +++ b/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference @@ -8,13 +8,13 @@ add_more_data_2 inc_2 BACKUP_CREATED inc_2_bad -Couldn't unpack zip archive '02843_backup_use_same_password_for_base_backup_test_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE) +_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE) restore_inc_1 RESTORED restore_inc_2 RESTORED restore_inc_2_bad -Couldn't unpack zip archive '02843_backup_use_same_password_for_base_backup_test_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE) +_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE) count_inc_1 20 count_inc_2 diff --git a/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.sh b/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.sh index a2b1a953e24..f2f1265c1a0 100755 --- a/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.sh +++ b/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.sh @@ -30,7 +30,7 @@ echo "inc_2" $CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2.zip') SETTINGS base_backup=Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip'),password='password',use_same_password_for_base_backup=1" | cut -f2 echo "inc_2_bad" -$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2_bad.zip') SETTINGS base_backup=Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip'),password='password'" |& grep -m1 -o "Couldn't unpack zip archive '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE)" +$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2_bad.zip') SETTINGS base_backup=Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip'),password='password'" |& grep -m1 -o "_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE)" echo "restore_inc_1" $CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_1 FROM Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip') SETTINGS password='password',use_same_password_for_base_backup=1" | cut -f2 @@ -39,7 +39,7 @@ echo "restore_inc_2" $CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_2 FROM Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2.zip') SETTINGS password='password',use_same_password_for_base_backup=1" | cut -f2 echo "restore_inc_2_bad" -$CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_2 FROM Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2.zip') SETTINGS password='password'" |& grep -m1 -o "Couldn't unpack zip archive '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE)" +$CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_2 FROM Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2.zip') SETTINGS password='password'" |& grep -m1 -o "_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE)" echo "count_inc_1" $CLICKHOUSE_CLIENT -q "SELECT COUNT(*) FROM data_1" | cut -f2 From 01c8faf190462d508d9cb1b7d342d6ccfbeeff88 Mon Sep 17 00:00:00 2001 From: Samuele Guerrini Date: Wed, 10 Jul 2024 11:15:33 +0200 Subject: [PATCH 018/227] Revert "shorter grep output" This reverts commit b0cbf1495dd0bcaba828706e895347e4ec550e29. --- .../02843_backup_use_same_password_for_base_backup.reference | 4 ++-- .../02843_backup_use_same_password_for_base_backup.sh | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference b/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference index cbcb6b4cb7c..7354d50a7c0 100644 --- a/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference +++ b/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference @@ -8,13 +8,13 @@ add_more_data_2 inc_2 BACKUP_CREATED inc_2_bad -_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE) +Couldn't unpack zip archive '02843_backup_use_same_password_for_base_backup_test_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE) restore_inc_1 RESTORED restore_inc_2 RESTORED restore_inc_2_bad -_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE) +Couldn't unpack zip archive '02843_backup_use_same_password_for_base_backup_test_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE) count_inc_1 20 count_inc_2 diff --git a/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.sh b/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.sh index f2f1265c1a0..a2b1a953e24 100755 --- a/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.sh +++ b/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.sh @@ -30,7 +30,7 @@ echo "inc_2" $CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2.zip') SETTINGS base_backup=Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip'),password='password',use_same_password_for_base_backup=1" | cut -f2 echo "inc_2_bad" -$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2_bad.zip') SETTINGS base_backup=Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip'),password='password'" |& grep -m1 -o "_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE)" +$CLICKHOUSE_CLIENT -q "BACKUP TABLE data TO Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2_bad.zip') SETTINGS base_backup=Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip'),password='password'" |& grep -m1 -o "Couldn't unpack zip archive '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE)" echo "restore_inc_1" $CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_1 FROM Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip') SETTINGS password='password',use_same_password_for_base_backup=1" | cut -f2 @@ -39,7 +39,7 @@ echo "restore_inc_2" $CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_2 FROM Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2.zip') SETTINGS password='password',use_same_password_for_base_backup=1" | cut -f2 echo "restore_inc_2_bad" -$CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_2 FROM Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2.zip') SETTINGS password='password'" |& grep -m1 -o "_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE)" +$CLICKHOUSE_CLIENT -q "RESTORE TABLE data AS data_2 FROM Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_2.zip') SETTINGS password='password'" |& grep -m1 -o "Couldn't unpack zip archive '${CLICKHOUSE_TEST_UNIQUE_NAME}_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE)" echo "count_inc_1" $CLICKHOUSE_CLIENT -q "SELECT COUNT(*) FROM data_1" | cut -f2 From 9dae370569054b4933b374c12dee6461c938fea8 Mon Sep 17 00:00:00 2001 From: Samuele Guerrini Date: Wed, 10 Jul 2024 11:23:36 +0200 Subject: [PATCH 019/227] fix db name in backup_use_same_password_for_base_backup test reference file --- .../02843_backup_use_same_password_for_base_backup.reference | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference b/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference index 7354d50a7c0..cdcf0532cd9 100644 --- a/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference +++ b/tests/queries/0_stateless/02843_backup_use_same_password_for_base_backup.reference @@ -8,13 +8,13 @@ add_more_data_2 inc_2 BACKUP_CREATED inc_2_bad -Couldn't unpack zip archive '02843_backup_use_same_password_for_base_backup_test_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE) +Couldn't unpack zip archive '02843_backup_use_same_password_for_base_backup_default_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE) restore_inc_1 RESTORED restore_inc_2 RESTORED restore_inc_2_bad -Couldn't unpack zip archive '02843_backup_use_same_password_for_base_backup_test_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE) +Couldn't unpack zip archive '02843_backup_use_same_password_for_base_backup_default_inc_1.zip': Password is required. (CANNOT_UNPACK_ARCHIVE) count_inc_1 20 count_inc_2 From 2e1f679ceb05afe4d5d813eb4048555c6311c3e1 Mon Sep 17 00:00:00 2001 From: Michael Stetsyuk Date: Tue, 16 Jul 2024 18:35:33 +0200 Subject: [PATCH 020/227] add S3DiskNoKeyErrors metric --- src/Common/CurrentMetrics.cpp | 2 ++ src/IO/S3/Client.cpp | 21 +++++++++++++++------ 2 files changed, 17 insertions(+), 6 deletions(-) diff --git a/src/Common/CurrentMetrics.cpp b/src/Common/CurrentMetrics.cpp index 7c97e73f278..2fedba0175b 100644 --- a/src/Common/CurrentMetrics.cpp +++ b/src/Common/CurrentMetrics.cpp @@ -306,6 +306,8 @@ \ M(FilteringMarksWithPrimaryKey, "Number of threads currently doing filtering of mark ranges by the primary key") \ M(FilteringMarksWithSecondaryKeys, "Number of threads currently doing filtering of mark ranges by secondary keys") \ + \ + M(S3DiskNoKeyErrors, "Number of no-key S3 disk errors") \ #ifdef APPLY_FOR_EXTERNAL_METRICS #define APPLY_FOR_METRICS(M) APPLY_FOR_BUILTIN_METRICS(M) APPLY_FOR_EXTERNAL_METRICS(M) diff --git a/src/IO/S3/Client.cpp b/src/IO/S3/Client.cpp index 3b958dea046..db20420db9f 100644 --- a/src/IO/S3/Client.cpp +++ b/src/IO/S3/Client.cpp @@ -24,6 +24,7 @@ #include #include +#include #include #include @@ -43,6 +44,11 @@ namespace ProfileEvents extern const Event TinyS3Clients; } +namespace CurrentMetrics +{ + extern const Metric S3DiskNoKeyErrors; +} + namespace DB { @@ -379,10 +385,10 @@ Model::HeadObjectOutcome Client::HeadObject(HeadObjectRequest & request) const request.overrideURI(std::move(*bucket_uri)); - /// The next call is NOT a recurcive call - /// This is a virtuall call Aws::S3::S3Client::HeadObject(const Model::HeadObjectRequest&) - return enrichErrorMessage( - HeadObject(static_cast(request))); + if (isClientForDisk()) + CurrentMetrics::add(CurrentMetrics::S3DiskNoKeyErrors); + + return enrichErrorMessage(std::move(result)); } /// For each request, we wrap the request functions from Aws::S3::Client with doRequest @@ -402,8 +408,11 @@ Model::ListObjectsOutcome Client::ListObjects(ListObjectsRequest & request) cons Model::GetObjectOutcome Client::GetObject(GetObjectRequest & request) const { - return enrichErrorMessage( - doRequest(request, [this](const Model::GetObjectRequest & req) { return GetObject(req); })); + auto resp = doRequest(request, [this](const Model::GetObjectRequest & req) { return GetObject(req); }); + if (!resp.IsSuccess() && isClientForDisk()) + CurrentMetrics::add(CurrentMetrics::S3DiskNoKeyErrors); + + return enrichErrorMessage(std::move(resp)); } Model::AbortMultipartUploadOutcome Client::AbortMultipartUpload(AbortMultipartUploadRequest & request) const From 2a893ed8cd10e6e3b7506f43b644b5037f96c49a Mon Sep 17 00:00:00 2001 From: Michael Stetsyuk Date: Mon, 22 Jul 2024 17:03:41 +0100 Subject: [PATCH 021/227] fxs --- src/IO/S3/Client.cpp | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/IO/S3/Client.cpp b/src/IO/S3/Client.cpp index db20420db9f..325c820f8bd 100644 --- a/src/IO/S3/Client.cpp +++ b/src/IO/S3/Client.cpp @@ -388,7 +388,8 @@ Model::HeadObjectOutcome Client::HeadObject(HeadObjectRequest & request) const if (isClientForDisk()) CurrentMetrics::add(CurrentMetrics::S3DiskNoKeyErrors); - return enrichErrorMessage(std::move(result)); + return enrichErrorMessage( + HeadObject(static_cast(request))); } /// For each request, we wrap the request functions from Aws::S3::Client with doRequest From 72f4919fdad5217f48bd83e51ce2d1f3f083087b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Mi=D1=81hael=20Stetsyuk?= <59827607+mstetsyuk@users.noreply.github.com> Date: Mon, 22 Jul 2024 21:33:47 +0100 Subject: [PATCH 022/227] Update src/Common/CurrentMetrics.cpp Co-authored-by: Sema Checherinda <104093494+CheSema@users.noreply.github.com> --- src/Common/CurrentMetrics.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Common/CurrentMetrics.cpp b/src/Common/CurrentMetrics.cpp index 2fedba0175b..39198147794 100644 --- a/src/Common/CurrentMetrics.cpp +++ b/src/Common/CurrentMetrics.cpp @@ -307,7 +307,7 @@ M(FilteringMarksWithPrimaryKey, "Number of threads currently doing filtering of mark ranges by the primary key") \ M(FilteringMarksWithSecondaryKeys, "Number of threads currently doing filtering of mark ranges by secondary keys") \ \ - M(S3DiskNoKeyErrors, "Number of no-key S3 disk errors") \ + M(S3DiskNoKeyErrors, "The number of `NoSuchKey` errors that occur when reading data from S3 cloud storage through ClickHouse disks.") \ #ifdef APPLY_FOR_EXTERNAL_METRICS #define APPLY_FOR_METRICS(M) APPLY_FOR_BUILTIN_METRICS(M) APPLY_FOR_EXTERNAL_METRICS(M) From aaf603035e31874d6d5bcd024d0f4040715baa72 Mon Sep 17 00:00:00 2001 From: Michael Stetsyuk Date: Tue, 23 Jul 2024 13:35:37 +0100 Subject: [PATCH 023/227] check error type --- src/IO/S3/Client.cpp | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/IO/S3/Client.cpp b/src/IO/S3/Client.cpp index 325c820f8bd..7196dfa9bdc 100644 --- a/src/IO/S3/Client.cpp +++ b/src/IO/S3/Client.cpp @@ -385,7 +385,7 @@ Model::HeadObjectOutcome Client::HeadObject(HeadObjectRequest & request) const request.overrideURI(std::move(*bucket_uri)); - if (isClientForDisk()) + if (isClientForDisk() && error.GetErrorType() == Aws::S3::S3Errors::NO_SUCH_KEY) CurrentMetrics::add(CurrentMetrics::S3DiskNoKeyErrors); return enrichErrorMessage( @@ -410,7 +410,7 @@ Model::ListObjectsOutcome Client::ListObjects(ListObjectsRequest & request) cons Model::GetObjectOutcome Client::GetObject(GetObjectRequest & request) const { auto resp = doRequest(request, [this](const Model::GetObjectRequest & req) { return GetObject(req); }); - if (!resp.IsSuccess() && isClientForDisk()) + if (!resp.IsSuccess() && isClientForDisk() && resp.GetError().GetErrorType() == Aws::S3::S3Errors::NO_SUCH_KEY) CurrentMetrics::add(CurrentMetrics::S3DiskNoKeyErrors); return enrichErrorMessage(std::move(resp)); From c2f85c6fd062dde095ee34178450dc94c245e691 Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Thu, 25 Jul 2024 17:43:02 +0800 Subject: [PATCH 024/227] support map type as first argument type --- .../functions/tuple-map-functions.md | 2 +- src/Functions/map.cpp | 131 +++++++++++------- .../0_stateless/01651_map_functions.reference | 3 + .../0_stateless/01651_map_functions.sql | 9 +- 4 files changed, 89 insertions(+), 56 deletions(-) diff --git a/docs/en/sql-reference/functions/tuple-map-functions.md b/docs/en/sql-reference/functions/tuple-map-functions.md index 24b356eca87..ae23387f6e5 100644 --- a/docs/en/sql-reference/functions/tuple-map-functions.md +++ b/docs/en/sql-reference/functions/tuple-map-functions.md @@ -62,7 +62,7 @@ Alias: `MAP_FROM_ARRAYS(keys, values)` **Arguments** -- `keys` — Array of keys to create the map from. [Array(T)](../data-types/array.md) where `T` can be any type supported by [Map](../data-types/map.md) as key type. +- `keys` — Array or map of keys to create the map from. [Array(T)](../data-types/array.md) where `T` can be any type supported by [Map](../data-types/map.md) as key type. - `values` - Array or map of values to create the map from. [Array](../data-types/array.md) or [Map](../data-types/map.md). **Returned value** diff --git a/src/Functions/map.cpp b/src/Functions/map.cpp index 66cd10a3f0b..5319390fb70 100644 --- a/src/Functions/map.cpp +++ b/src/Functions/map.cpp @@ -1,14 +1,17 @@ -#include +#include +#include +#include +#include +#include +#include +#include +#include +#include #include #include -#include -#include -#include -#include -#include -#include -#include +#include #include +#include #include @@ -178,22 +181,28 @@ public: getName(), arguments.size()); - /// The first argument should always be Array. - /// Because key type can not be nested type of Map, which is Tuple - DataTypePtr key_type; - if (const auto * keys_type = checkAndGetDataType(arguments[0].get())) - key_type = keys_type->getNestedType(); - else - throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "First argument for function {} must be an Array", getName()); + auto get_nested_type = [this](const DataTypePtr & type) -> DataTypePtr + { + DataTypePtr nested; + if (const auto * array_type = checkAndGetDataType(type.get())) + nested = array_type->getNestedType(); + else if (const auto * map_type = checkAndGetDataType(type.get())) + nested = std::make_shared(map_type->getKeyValueTypes()); + else + throw Exception( + ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, + "Argument types of function {} must be Array or Map, but {} is given", + getName(), + type->getName()); - DataTypePtr value_type; - if (const auto * value_array_type = checkAndGetDataType(arguments[1].get())) - value_type = value_array_type->getNestedType(); - else if (const auto * value_map_type = checkAndGetDataType(arguments[1].get())) - value_type = std::make_shared(value_map_type->getKeyValueTypes()); - else - throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Second argument for function {} must be Array or Map", getName()); + return nested; + }; + auto key_type = get_nested_type(arguments[0]); + auto value_type = get_nested_type(arguments[1]); + + /// Remove Nullable from key_type if needed for map key must not be Nullable + key_type = removeNullableOrLowCardinalityNullable(key_type); DataTypes key_value_types{key_type, value_type}; return std::make_shared(key_value_types); } @@ -201,44 +210,62 @@ public: ColumnPtr executeImpl( const ColumnsWithTypeAndName & arguments, const DataTypePtr & /* result_type */, size_t /* input_rows_count */) const override { - bool is_keys_const = isColumnConst(*arguments[0].column); - ColumnPtr holder_keys; - const ColumnArray * col_keys; - if (is_keys_const) + auto get_array_column = [this](const ColumnPtr & column) -> std::pair { - holder_keys = arguments[0].column->convertToFullColumnIfConst(); - col_keys = checkAndGetColumn(holder_keys.get()); - } - else + bool is_const = isColumnConst(*column); + ColumnPtr holder = is_const ? column->convertToFullColumnIfConst() : column; + + const ColumnArray * col_res = nullptr; + if (const auto * col_array = checkAndGetColumn(holder.get())) + col_res = col_array; + else if (const auto * col_map = checkAndGetColumn(holder.get())) + col_res = &col_map->getNestedColumn(); + else + throw Exception( + ErrorCodes::ILLEGAL_COLUMN, + "Argument columns of function {} must be Array or Map, but {} is given", + getName(), + holder->getName()); + + return {col_res, holder}; + }; + + auto [col_keys, key_holder] = get_array_column(arguments[0].column); + + /// Check if nested column of first argument contains NULL value in case its nested type is Nullable(T) type. + ColumnPtr data_keys = col_keys->getDataPtr(); + if (isColumnNullableOrLowCardinalityNullable(*data_keys)) { - col_keys = checkAndGetColumn(arguments[0].column.get()); + std::cout << "data keys is nullable" << std::endl; + const NullMap * null_map = nullptr; + if (const auto * nullable = checkAndGetColumn(data_keys.get())) + { + null_map = &nullable->getNullMapData(); + data_keys = nullable->getNestedColumnPtr(); + } + else if (const auto * low_cardinality = checkAndGetColumn(data_keys.get())) + { + if (const auto * nullable_dict = checkAndGetColumn(low_cardinality->getDictionaryPtr().get())) + { + null_map = &nullable_dict->getNullMapData(); + data_keys = ColumnLowCardinality::create(nullable_dict->getNestedColumnPtr(), low_cardinality->getIndexesPtr()); + } + } + + if (null_map && !memoryIsZero(null_map->data(), 0, null_map->size())) + throw Exception( + ErrorCodes::ILLEGAL_COLUMN, "The nested column of first argument in function {} must not contain NULLs", getName()); } - if (!col_keys) - throw Exception(ErrorCodes::ILLEGAL_COLUMN, "The first argument of function {} must be Array", getName()); - - bool is_values_const = isColumnConst(*arguments[1].column); - ColumnPtr holder_values; - if (is_values_const) - holder_values = arguments[1].column->convertToFullColumnIfConst(); - else - holder_values = arguments[1].column; - - const ColumnArray * col_values; - if (const auto * col_values_array = checkAndGetColumn(holder_values.get())) - col_values = col_values_array; - else if (const auto * col_values_map = checkAndGetColumn(holder_values.get())) - col_values = &col_values_map->getNestedColumn(); - else - throw Exception(ErrorCodes::ILLEGAL_COLUMN, "The second arguments of function {} must be Array or Map", getName()); - + auto [col_values, values_holder] = get_array_column(arguments[1].column); if (!col_keys->hasEqualOffsets(*col_values)) - throw Exception(ErrorCodes::SIZES_OF_ARRAYS_DONT_MATCH, "Two arguments for function {} must have equal sizes", getName()); + throw Exception(ErrorCodes::SIZES_OF_ARRAYS_DONT_MATCH, "Two arguments of function {} must have equal sizes", getName()); - const auto & data_keys = col_keys->getDataPtr(); const auto & data_values = col_values->getDataPtr(); const auto & offsets = col_keys->getOffsetsPtr(); - auto nested_column = ColumnArray::create(ColumnTuple::create(Columns{data_keys, data_values}), offsets); + std::cout << "before create array:" << "offsets:" << offsets->getName() << std::endl; + auto nested_column = ColumnArray::create(ColumnTuple::create(Columns{std::move(data_keys), data_values}), offsets); + std::cout << "after create array:" << "offsets:" << offsets->getName() << std::endl; return ColumnMap::create(nested_column); } }; diff --git a/tests/queries/0_stateless/01651_map_functions.reference b/tests/queries/0_stateless/01651_map_functions.reference index 471da5586b7..9114aa419b1 100644 --- a/tests/queries/0_stateless/01651_map_functions.reference +++ b/tests/queries/0_stateless/01651_map_functions.reference @@ -52,3 +52,6 @@ {1:4,2:5} {1:4,2:5} {1:4,2:5} +{1:3,2:4} +{1:3,2:4} +{1:3,2:4} {(1,3):'a',(2,4):'b'} diff --git a/tests/queries/0_stateless/01651_map_functions.sql b/tests/queries/0_stateless/01651_map_functions.sql index cf2460fce2c..4604ddd6db1 100644 --- a/tests/queries/0_stateless/01651_map_functions.sql +++ b/tests/queries/0_stateless/01651_map_functions.sql @@ -67,12 +67,15 @@ select mapFromArrays(['aa', 'bb'], [4, 5, 6]); -- { serverError SIZES_OF_ARRAYS_ select mapFromArrays([[1,2], [3,4]], [4, 5, 6]); -- { serverError SIZES_OF_ARRAYS_DONT_MATCH } select mapFromArrays(['a', 2], [4, 5]); -- { serverError NO_COMMON_TYPE} select mapFromArrays([1, 2], [4, 'a']); -- { serverError NO_COMMON_TYPE} +select mapFromArrays(['aa', 'bb'], map('a', 4)); -- { serverError SIZES_OF_ARRAYS_DONT_MATCH } +select mapFromArrays([1,null]::Array(Nullable(UInt8)), [3,4]); -- { serverError ILLEGAL_COLUMN } select mapFromArrays(['aa', 'bb'], map('a', 4, 'b', 5)); select mapFromArrays(['aa', 'bb'], materialize(map('a', 4, 'b', 5))) from numbers(2); -select mapFromArrays(map('a', 4, 'b', 4), ['aa', 'bb']) from numbers(2); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } -select mapFromArrays(['aa', 'bb'], map('a', 4)); -- { serverError SIZES_OF_ARRAYS_DONT_MATCH } - select mapFromArrays([toLowCardinality(1), toLowCardinality(2)], [4, 5]); select mapFromArrays([toLowCardinality(1), toLowCardinality(2)], materialize([4, 5])) from numbers(2); + +select mapFromArrays([1,2], [3,4]); +select mapFromArrays([1,2]::Array(Nullable(UInt8)), [3,4]); +select mapFromArrays([1,2], [3,4]) as x, mapFromArrays(x, ['a', 'b']); From e181ccd0173c46d31867097532f64df0be3944da Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Thu, 25 Jul 2024 17:53:51 +0800 Subject: [PATCH 025/227] update doc --- docs/en/sql-reference/functions/tuple-map-functions.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/en/sql-reference/functions/tuple-map-functions.md b/docs/en/sql-reference/functions/tuple-map-functions.md index ae23387f6e5..db66188b1f5 100644 --- a/docs/en/sql-reference/functions/tuple-map-functions.md +++ b/docs/en/sql-reference/functions/tuple-map-functions.md @@ -62,7 +62,7 @@ Alias: `MAP_FROM_ARRAYS(keys, values)` **Arguments** -- `keys` — Array or map of keys to create the map from. [Array(T)](../data-types/array.md) where `T` can be any type supported by [Map](../data-types/map.md) as key type. +- `keys` — Array or map of keys to create the map from. [Array(T)](../data-types/array.md) where `T` can be any type supported by [Map](../data-types/map.md) as key type, or [Map](../data-types/map.md). - `values` - Array or map of values to create the map from. [Array](../data-types/array.md) or [Map](../data-types/map.md). **Returned value** From a32c702caa142d15bc3e5bc51ca90240d5d010a9 Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Thu, 25 Jul 2024 18:23:47 +0800 Subject: [PATCH 026/227] fix style --- src/Functions/map.cpp | 3 --- 1 file changed, 3 deletions(-) diff --git a/src/Functions/map.cpp b/src/Functions/map.cpp index 5319390fb70..a8e5f7ad90e 100644 --- a/src/Functions/map.cpp +++ b/src/Functions/map.cpp @@ -236,7 +236,6 @@ public: ColumnPtr data_keys = col_keys->getDataPtr(); if (isColumnNullableOrLowCardinalityNullable(*data_keys)) { - std::cout << "data keys is nullable" << std::endl; const NullMap * null_map = nullptr; if (const auto * nullable = checkAndGetColumn(data_keys.get())) { @@ -263,9 +262,7 @@ public: const auto & data_values = col_values->getDataPtr(); const auto & offsets = col_keys->getOffsetsPtr(); - std::cout << "before create array:" << "offsets:" << offsets->getName() << std::endl; auto nested_column = ColumnArray::create(ColumnTuple::create(Columns{std::move(data_keys), data_values}), offsets); - std::cout << "after create array:" << "offsets:" << offsets->getName() << std::endl; return ColumnMap::create(nested_column); } }; From cd06945a03df0f8dbec6ff82332236caf86fbff3 Mon Sep 17 00:00:00 2001 From: avogar Date: Thu, 25 Jul 2024 13:05:25 +0000 Subject: [PATCH 027/227] Fix crash with Variant + AggregateFunction type --- src/Columns/ColumnAggregateFunction.cpp | 6 +- ...ant_with_aggregate_function_type.reference | 6 ++ ...0_variant_with_aggregate_function_type.sql | 60 +++++++++++++++++++ 3 files changed, 71 insertions(+), 1 deletion(-) create mode 100644 tests/queries/0_stateless/03210_variant_with_aggregate_function_type.reference create mode 100644 tests/queries/0_stateless/03210_variant_with_aggregate_function_type.sql diff --git a/src/Columns/ColumnAggregateFunction.cpp b/src/Columns/ColumnAggregateFunction.cpp index e26fe790a8e..9934970c868 100644 --- a/src/Columns/ColumnAggregateFunction.cpp +++ b/src/Columns/ColumnAggregateFunction.cpp @@ -330,7 +330,11 @@ ColumnPtr ColumnAggregateFunction::filter(const Filter & filter, ssize_t result_ void ColumnAggregateFunction::expand(const Filter & mask, bool inverted) { - expandDataByMask(data, mask, inverted); + ensureOwnership(); + Arena & arena = createOrGetArena(); + char * default_ptr = arena.alignedAlloc(func->sizeOfData(), func->alignOfData()); + func->create(default_ptr); + expandDataByMask(data, mask, inverted, default_ptr); } ColumnPtr ColumnAggregateFunction::permute(const Permutation & perm, size_t limit) const diff --git a/tests/queries/0_stateless/03210_variant_with_aggregate_function_type.reference b/tests/queries/0_stateless/03210_variant_with_aggregate_function_type.reference new file mode 100644 index 00000000000..105e8e7d8bd --- /dev/null +++ b/tests/queries/0_stateless/03210_variant_with_aggregate_function_type.reference @@ -0,0 +1,6 @@ + 500 +fail 500 + 499 +fail 500 + 500 499 +fail 500 500 diff --git a/tests/queries/0_stateless/03210_variant_with_aggregate_function_type.sql b/tests/queries/0_stateless/03210_variant_with_aggregate_function_type.sql new file mode 100644 index 00000000000..cb9cdb0b456 --- /dev/null +++ b/tests/queries/0_stateless/03210_variant_with_aggregate_function_type.sql @@ -0,0 +1,60 @@ +SET allow_experimental_variant_type = 1; + +DROP TABLE IF EXISTS source; +CREATE TABLE source +( + Name String, + Value Int64 + +) ENGINE = MergeTree ORDER BY (); + +INSERT INTO source SELECT ['fail', 'success'][number % 2] as Name, number AS Value FROM numbers(1000); + +DROP TABLE IF EXISTS test_agg_variant; +CREATE TABLE test_agg_variant +( + Name String, + Value Variant(AggregateFunction(uniqExact, Int64), AggregateFunction(avg, Int64)) +) +ENGINE = MergeTree +ORDER BY (Name); + +INSERT INTO test_agg_variant +SELECT + Name, + t AS Value +FROM +( + SELECT + Name, + arrayJoin([ + uniqExactState(Value)::Variant(AggregateFunction(uniqExact, Int64), AggregateFunction(avg, Int64)), + avgState(Value)::Variant(AggregateFunction(uniqExact, Int64), AggregateFunction(avg, Int64)) + ]) AS t + FROM source + GROUP BY Name +); + +SELECT + Name, + uniqExactMerge(Value.`AggregateFunction(uniqExact, Int64)`) AS Value +FROM test_agg_variant +GROUP BY Name; + +SELECT + Name, + avgMerge(Value.`AggregateFunction(avg, Int64)`) AS Value +FROM test_agg_variant +GROUP BY Name; + +SELECT + Name, + uniqExactMerge(Value.`AggregateFunction(uniqExact, Int64)`) AS ValueUniq, + avgMerge(Value.`AggregateFunction(avg, Int64)`) AS ValueAvg +FROM test_agg_variant +GROUP BY Name; + + +DROP TABLE test_agg_variant; +DROP TABLE source; + From 5ea867231bafc01b4512989f351106b7afcc14af Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Thu, 25 Jul 2024 16:55:23 +0200 Subject: [PATCH 028/227] Read configuration for clickhouse-local from ~/.clickhouse-local Signed-off-by: Azat Khuzhin --- programs/local/LocalServer.cpp | 16 +++++++-- src/Common/Config/CMakeLists.txt | 1 + src/Common/Config/getLocalConfigPath.cpp | 46 ++++++++++++++++++++++++ src/Common/Config/getLocalConfigPath.h | 12 +++++++ 4 files changed, 73 insertions(+), 2 deletions(-) create mode 100644 src/Common/Config/getLocalConfigPath.cpp create mode 100644 src/Common/Config/getLocalConfigPath.h diff --git a/programs/local/LocalServer.cpp b/programs/local/LocalServer.cpp index 48e0cca7b73..ade4e0f49df 100644 --- a/programs/local/LocalServer.cpp +++ b/programs/local/LocalServer.cpp @@ -1,6 +1,7 @@ #include "LocalServer.h" #include +#include #include #include #include @@ -127,10 +128,21 @@ void LocalServer::initialize(Poco::Util::Application & self) { Poco::Util::Application::initialize(self); + const char * home_path_cstr = getenv("HOME"); // NOLINT(concurrency-mt-unsafe) + if (home_path_cstr) + home_path = home_path_cstr; + /// Load config files if exists - if (getClientConfiguration().has("config-file") || fs::exists("config.xml")) + std::string config_path; + if (getClientConfiguration().has("config-file")) + config_path = getClientConfiguration().getString("config-file"); + else if (config_path.empty() && fs::exists("config.xml")) + config_path = "config.xml"; + else if (config_path.empty()) + config_path = getLocalConfigPath(home_path).value_or(""); + + if (fs::exists(config_path)) { - const auto config_path = getClientConfiguration().getString("config-file", "config.xml"); ConfigProcessor config_processor(config_path, false, true); ConfigProcessor::setConfigPath(fs::path(config_path).parent_path()); auto loaded_config = config_processor.loadConfig(); diff --git a/src/Common/Config/CMakeLists.txt b/src/Common/Config/CMakeLists.txt index 09095ef5acc..2bd32b98bda 100644 --- a/src/Common/Config/CMakeLists.txt +++ b/src/Common/Config/CMakeLists.txt @@ -2,6 +2,7 @@ set (SRCS AbstractConfigurationComparison.cpp ConfigProcessor.cpp getClientConfigPath.cpp + getLocalConfigPath.cpp ConfigReloader.cpp YAMLParser.cpp ConfigHelper.cpp diff --git a/src/Common/Config/getLocalConfigPath.cpp b/src/Common/Config/getLocalConfigPath.cpp new file mode 100644 index 00000000000..afaa7f79026 --- /dev/null +++ b/src/Common/Config/getLocalConfigPath.cpp @@ -0,0 +1,46 @@ +#include + +#include +#include + + +namespace fs = std::filesystem; + +namespace DB +{ + +std::optional getLocalConfigPath(const std::string & home_path) +{ + std::string config_path; + bool found = false; + + std::vector names; + names.emplace_back("./clickhouse-local"); + if (!home_path.empty()) + names.emplace_back(home_path + "/.clickhouse-local/config"); + names.emplace_back("/etc/clickhouse-local/config"); + + for (const auto & name : names) + { + for (const auto & extension : {".xml", ".yaml", ".yml"}) + { + config_path = name + extension; + + std::error_code ec; + if (fs::exists(config_path, ec)) + { + found = true; + break; + } + } + if (found) + break; + } + + if (found) + return config_path; + + return std::nullopt; +} + +} diff --git a/src/Common/Config/getLocalConfigPath.h b/src/Common/Config/getLocalConfigPath.h new file mode 100644 index 00000000000..14625571d6c --- /dev/null +++ b/src/Common/Config/getLocalConfigPath.h @@ -0,0 +1,12 @@ +#pragma once + +#include +#include + +namespace DB +{ + +/// Return path to existing configuration file. +std::optional getLocalConfigPath(const std::string & home_path); + +} From 38c4251a0eebe285b3f4da3ec21d11421d0e00fa Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Sat, 27 Jul 2024 16:30:18 +0200 Subject: [PATCH 029/227] Cleanup get{Local,Client}ConfigPath() helpers Signed-off-by: Azat Khuzhin --- src/Common/Config/getClientConfigPath.cpp | 11 +---------- src/Common/Config/getLocalConfigPath.cpp | 11 +---------- 2 files changed, 2 insertions(+), 20 deletions(-) diff --git a/src/Common/Config/getClientConfigPath.cpp b/src/Common/Config/getClientConfigPath.cpp index a32ad1068bf..5e9bb16bc85 100644 --- a/src/Common/Config/getClientConfigPath.cpp +++ b/src/Common/Config/getClientConfigPath.cpp @@ -12,7 +12,6 @@ namespace DB std::optional getClientConfigPath(const std::string & home_path) { std::string config_path; - bool found = false; std::vector names; names.emplace_back("./clickhouse-client"); @@ -28,18 +27,10 @@ std::optional getClientConfigPath(const std::string & home_path) std::error_code ec; if (fs::exists(config_path, ec)) - { - found = true; - break; - } + return config_path; } - if (found) - break; } - if (found) - return config_path; - return std::nullopt; } diff --git a/src/Common/Config/getLocalConfigPath.cpp b/src/Common/Config/getLocalConfigPath.cpp index afaa7f79026..195de8aed03 100644 --- a/src/Common/Config/getLocalConfigPath.cpp +++ b/src/Common/Config/getLocalConfigPath.cpp @@ -12,7 +12,6 @@ namespace DB std::optional getLocalConfigPath(const std::string & home_path) { std::string config_path; - bool found = false; std::vector names; names.emplace_back("./clickhouse-local"); @@ -28,18 +27,10 @@ std::optional getLocalConfigPath(const std::string & home_path) std::error_code ec; if (fs::exists(config_path, ec)) - { - found = true; - break; - } + return config_path; } - if (found) - break; } - if (found) - return config_path; - return std::nullopt; } From 4606ec96d3a57003d3271124a98ab4194518e3af Mon Sep 17 00:00:00 2001 From: sakulali Date: Sat, 27 Jul 2024 22:17:53 +0800 Subject: [PATCH 030/227] Autodetect configuration file format if is not .xml, .yml or .yaml --- src/Common/Config/ConfigProcessor.cpp | 100 ++++++++++-------- src/Common/Config/ConfigProcessor.h | 2 + ..._clickhouse_client_config_format.reference | 8 +- .../01889_clickhouse_client_config_format.sh | 33 ++++-- 4 files changed, 89 insertions(+), 54 deletions(-) diff --git a/src/Common/Config/ConfigProcessor.cpp b/src/Common/Config/ConfigProcessor.cpp index 67d6036aa51..f0b83c035c8 100644 --- a/src/Common/Config/ConfigProcessor.cpp +++ b/src/Common/Config/ConfigProcessor.cpp @@ -138,9 +138,14 @@ static Node * getRootNode(Document * document) return XMLUtils::getRootNode(document); } +static size_t firstNonWhitespacePos(const std::string & s) +{ + return s.find_first_not_of(" \t\n\r"); +} + static bool allWhitespace(const std::string & s) { - return s.find_first_not_of(" \t\n\r") == std::string::npos; + return firstNonWhitespacePos(s) == std::string::npos; } static void deleteAttributesRecursive(Node * root) @@ -622,6 +627,52 @@ ConfigProcessor::Files ConfigProcessor::getConfigMergeFiles(const std::string & return files; } +XMLDocumentPtr ConfigProcessor::parseConfig(const std::string & config_path) +{ + fs::path p(config_path); + std::string extension = p.extension(); + boost::algorithm::to_lower(extension); + + if (extension == ".xml") + return dom_parser.parse(config_path); + else if (extension == ".yaml" || extension == ".yml") + return YAMLParser::parse(config_path); + else + { + /// Suppose non regular file parsed as XML, such as pipe: /dev/fd/X (regardless it has .xml extension or not) + if (!fs::is_regular_file(config_path)) + return dom_parser.parse(config_path); + + /// If the regular file begins with < it might be XML, otherwise it might be YAML. + bool maybe_xml = false; + { + std::ifstream file(config_path); + if (!file.is_open()) + throw Exception(ErrorCodes::CANNOT_LOAD_CONFIG, "Unknown format of '{}' config", config_path); + + std::string line; + while (std::getline(file, line)) + { + const size_t pos = firstNonWhitespacePos(line); + if (pos == std::string::npos) + continue; + + if (pos < line.size() && '<' == line[pos]) + { + maybe_xml = true; + } + + break; + } + } + + if (maybe_xml) + return dom_parser.parse(config_path); + else + return YAMLParser::parse(config_path); + } +} + XMLDocumentPtr ConfigProcessor::processConfig( bool * has_zk_includes, zkutil::ZooKeeperNodeCache * zk_node_cache, @@ -633,23 +684,7 @@ XMLDocumentPtr ConfigProcessor::processConfig( if (fs::exists(path)) { - fs::path p(path); - - std::string extension = p.extension(); - boost::algorithm::to_lower(extension); - - if (extension == ".yaml" || extension == ".yml") - { - config = YAMLParser::parse(path); - } - else if (extension == ".xml" || extension == ".conf" || extension.empty()) - { - config = dom_parser.parse(path); - } - else - { - throw Exception(ErrorCodes::CANNOT_LOAD_CONFIG, "Unknown format of '{}' config", path); - } + config = parseConfig(path); } else { @@ -673,20 +708,7 @@ XMLDocumentPtr ConfigProcessor::processConfig( LOG_DEBUG(log, "Merging configuration file '{}'.", merge_file); XMLDocumentPtr with; - - fs::path p(merge_file); - std::string extension = p.extension(); - boost::algorithm::to_lower(extension); - - if (extension == ".yaml" || extension == ".yml") - { - with = YAMLParser::parse(merge_file); - } - else - { - with = dom_parser.parse(merge_file); - } - + with = parseConfig(merge_file); if (!merge(config, with)) { LOG_DEBUG(log, "Merging bypassed - configuration file '{}' doesn't belong to configuration '{}' - merging root node name '{}' doesn't match '{}'", @@ -730,19 +752,7 @@ XMLDocumentPtr ConfigProcessor::processConfig( { LOG_DEBUG(log, "Including configuration file '{}'.", include_from_path); - fs::path p(include_from_path); - std::string extension = p.extension(); - boost::algorithm::to_lower(extension); - - if (extension == ".yaml" || extension == ".yml") - { - include_from = YAMLParser::parse(include_from_path); - } - else - { - include_from = dom_parser.parse(include_from_path); - } - + include_from = parseConfig(include_from_path); contributing_files.push_back(include_from_path); } diff --git a/src/Common/Config/ConfigProcessor.h b/src/Common/Config/ConfigProcessor.h index 5712c36d737..a9d1325b722 100644 --- a/src/Common/Config/ConfigProcessor.h +++ b/src/Common/Config/ConfigProcessor.h @@ -65,6 +65,8 @@ public: zkutil::ZooKeeperNodeCache * zk_node_cache = nullptr, const zkutil::EventPtr & zk_changed_event = nullptr); + XMLDocumentPtr parseConfig(const std::string & config_path); + /// These configurations will be used if there is no configuration file. static void registerEmbeddedConfig(std::string name, std::string_view content); diff --git a/tests/queries/0_stateless/01889_clickhouse_client_config_format.reference b/tests/queries/0_stateless/01889_clickhouse_client_config_format.reference index 202e32a583e..afe27ddb063 100644 --- a/tests/queries/0_stateless/01889_clickhouse_client_config_format.reference +++ b/tests/queries/0_stateless/01889_clickhouse_client_config_format.reference @@ -12,5 +12,9 @@ yml 2 yaml 2 -ini -Code: 347. Unknown format of '/config_default.ini' config. (CANNOT_LOAD_CONFIG) +autodetect xml (with leading whitespaces) +2 +autodetect xml (non leading whitespaces) +2 +autodetect yaml +2 diff --git a/tests/queries/0_stateless/01889_clickhouse_client_config_format.sh b/tests/queries/0_stateless/01889_clickhouse_client_config_format.sh index 4e970f17d3a..14d67f943f1 100755 --- a/tests/queries/0_stateless/01889_clickhouse_client_config_format.sh +++ b/tests/queries/0_stateless/01889_clickhouse_client_config_format.sh @@ -12,7 +12,9 @@ XML_config=$CLICKHOUSE_TMP/config_$CLICKHOUSE_DATABASE.XML conf_config=$CLICKHOUSE_TMP/config_$CLICKHOUSE_DATABASE.conf yml_config=$CLICKHOUSE_TMP/config_$CLICKHOUSE_DATABASE.yml yaml_config=$CLICKHOUSE_TMP/config_$CLICKHOUSE_DATABASE.yaml -ini_config=$CLICKHOUSE_TMP/config_$CLICKHOUSE_DATABASE.ini +autodetect_xml_with_leading_whitespace_config=$CLICKHOUSE_TMP/config_$CLICKHOUSE_DATABASE.config +autodetect_xml_non_leading_whitespace_config=$CLICKHOUSE_TMP/config_$CLICKHOUSE_DATABASE.cfg +autodetect_yaml_config=$CLICKHOUSE_TMP/config_$CLICKHOUSE_DATABASE.properties function cleanup() { @@ -22,7 +24,9 @@ function cleanup() rm "${conf_config:?}" rm "${yml_config:?}" rm "${yaml_config:?}" - rm "${ini_config:?}" + rm "${autodetect_xml_with_leading_whitespace_config:?}" + rm "${autodetect_xml_non_leading_whitespace_config:?}" + rm "${autodetect_yaml_config:?}" } trap cleanup EXIT @@ -52,9 +56,19 @@ EOL cat > "$yaml_config" < "$ini_config" < "$autodetect_xml_with_leading_whitespace_config" < + 2 + +EOL +cat > "$autodetect_xml_non_leading_whitespace_config" < + 2 + +EOL +cat > "$autodetect_yaml_config" <&1 |& sed -e "s#$CLICKHOUSE_TMP##" -e "s#DB::Exception: ##" + +echo 'autodetect xml (with leading whitespaces)' +$CLICKHOUSE_CLIENT --config "$autodetect_xml_with_leading_whitespace_config" -q "select getSetting('max_threads')" +echo 'autodetect xml (non leading whitespaces)' +$CLICKHOUSE_CLIENT --config "$autodetect_xml_non_leading_whitespace_config" -q "select getSetting('max_threads')" +echo 'autodetect yaml' +$CLICKHOUSE_CLIENT --config "$autodetect_yaml_config" -q "select getSetting('max_threads')" From 13c93cc90cc35e33cad71095fa0300c5f4f1753e Mon Sep 17 00:00:00 2001 From: Duc Canh Le Date: Sun, 28 Jul 2024 10:29:52 +0000 Subject: [PATCH 031/227] fix hung query when using empty tuple as lhs of function IN Signed-off-by: Duc Canh Le --- src/Analyzer/SetUtils.cpp | 6 ++++++ .../0_stateless/03210_empty_tuple_lhs_of_in.reference | 0 tests/queries/0_stateless/03210_empty_tuple_lhs_of_in.sql | 1 + 3 files changed, 7 insertions(+) create mode 100644 tests/queries/0_stateless/03210_empty_tuple_lhs_of_in.reference create mode 100644 tests/queries/0_stateless/03210_empty_tuple_lhs_of_in.sql diff --git a/src/Analyzer/SetUtils.cpp b/src/Analyzer/SetUtils.cpp index 0ecb3545225..e7d1a22f45a 100644 --- a/src/Analyzer/SetUtils.cpp +++ b/src/Analyzer/SetUtils.cpp @@ -43,6 +43,12 @@ size_t getCompoundTypeDepth(const IDataType & type) const auto & tuple_elements = assert_cast(*current_type).getElements(); if (!tuple_elements.empty()) current_type = tuple_elements.at(0).get(); + else + { + /// Special case: tuple with no element - tuple(). In this case, what's the compound type depth? + /// I'm not certain about the theoretical answer, but from experiment, 1 is the most reasonable choice. + return 1; + } ++result; } diff --git a/tests/queries/0_stateless/03210_empty_tuple_lhs_of_in.reference b/tests/queries/0_stateless/03210_empty_tuple_lhs_of_in.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03210_empty_tuple_lhs_of_in.sql b/tests/queries/0_stateless/03210_empty_tuple_lhs_of_in.sql new file mode 100644 index 00000000000..0fc5f072da5 --- /dev/null +++ b/tests/queries/0_stateless/03210_empty_tuple_lhs_of_in.sql @@ -0,0 +1 @@ +SELECT tuple() IN tuple(1) SETTINGS allow_experimental_map_type = 1; -- { serverError INCORRECT_ELEMENT_OF_SET } From b503bad51d4f1068b17be967c4e5f9b6393d490e Mon Sep 17 00:00:00 2001 From: flynn Date: Sun, 28 Jul 2024 14:42:14 +0000 Subject: [PATCH 032/227] Fix crash of uniq and uniqTheta with empty tuple argument --- src/AggregateFunctions/UniqVariadicHash.h | 3 +++ tests/queries/0_stateless/03208_uniq_with_empty_tuple.sql | 3 +++ 2 files changed, 6 insertions(+) create mode 100644 tests/queries/0_stateless/03208_uniq_with_empty_tuple.sql diff --git a/src/AggregateFunctions/UniqVariadicHash.h b/src/AggregateFunctions/UniqVariadicHash.h index 5bb245397d4..279feed8bc6 100644 --- a/src/AggregateFunctions/UniqVariadicHash.h +++ b/src/AggregateFunctions/UniqVariadicHash.h @@ -67,6 +67,9 @@ struct UniqVariadicHash { static UInt64 apply(size_t num_args, const IColumn ** columns, size_t row_num) { + if (!num_args) + return 0; + UInt64 hash; const auto & tuple_columns = assert_cast(columns[0])->getColumns(); diff --git a/tests/queries/0_stateless/03208_uniq_with_empty_tuple.sql b/tests/queries/0_stateless/03208_uniq_with_empty_tuple.sql new file mode 100644 index 00000000000..4c5705261f0 --- /dev/null +++ b/tests/queries/0_stateless/03208_uniq_with_empty_tuple.sql @@ -0,0 +1,3 @@ +-- https://github.com/ClickHouse/ClickHouse/issues/67303 +SELECT uniqTheta(tuple()); +SELECT uniq(tuple()); From 60acbe7720ce31a55e3f846536ad2e1572f541ea Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Mon, 29 Jul 2024 01:40:05 +0200 Subject: [PATCH 033/227] Fix stack overflow in too deep JSON --- src/Core/Settings.h | 1 + src/Core/SettingsChangesHistory.cpp | 1 + src/Formats/FormatFactory.cpp | 1 + src/Formats/FormatSettings.h | 1 + src/IO/ReadHelpers.cpp | 25 +++++++++++++------ .../0_stateless/03213_deep_json.reference | 0 tests/queries/0_stateless/03213_deep_json.sql | 5 ++++ 7 files changed, 27 insertions(+), 7 deletions(-) create mode 100644 tests/queries/0_stateless/03213_deep_json.reference create mode 100644 tests/queries/0_stateless/03213_deep_json.sql diff --git a/src/Core/Settings.h b/src/Core/Settings.h index 86e04b2ab4e..6c8e09f38f9 100644 --- a/src/Core/Settings.h +++ b/src/Core/Settings.h @@ -1126,6 +1126,7 @@ class IColumn; M(Bool, input_format_json_throw_on_bad_escape_sequence, true, "Throw an exception if JSON string contains bad escape sequence in JSON input formats. If disabled, bad escape sequences will remain as is in the data", 0) \ M(Bool, input_format_json_ignore_unnecessary_fields, true, "Ignore unnecessary fields and not parse them. Enabling this may not throw exceptions on json strings of invalid format or with duplicated fields", 0) \ M(Bool, input_format_json_case_insensitive_column_matching, false, "Ignore case when matching JSON keys with CH columns", 0) \ + M(UInt64, input_format_json_max_depth, 1000, "Maximum depth of a field in JSON. This is not a strict limit, it does not have to be applied precisely.", 0) \ M(Bool, input_format_try_infer_integers, true, "Try to infer integers instead of floats while schema inference in text formats", 0) \ M(Bool, input_format_try_infer_dates, true, "Try to infer dates from string fields while schema inference in text formats", 0) \ M(Bool, input_format_try_infer_datetimes, true, "Try to infer datetimes from string fields while schema inference in text formats", 0) \ diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index 9faf77e9087..b340091798c 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -57,6 +57,7 @@ String ClickHouseVersion::toString() const /// Note: please check if the key already exists to prevent duplicate entries. static std::initializer_list> settings_changes_history_initializer = { + {"24.8", {{"input_format_json_max_depth", 1000000, 1000, "It was unlimited in previous versions, but that was unsafe."}}}, {"24.7", {{"output_format_parquet_write_page_index", false, true, "Add a possibility to write page index into parquet files."}, {"output_format_binary_encode_types_in_binary_format", false, false, "Added new setting to allow to write type names in binary format in RowBinaryWithNamesAndTypes output format"}, {"input_format_binary_decode_types_in_binary_format", false, false, "Added new setting to allow to read type names in binary format in RowBinaryWithNamesAndTypes input format"}, diff --git a/src/Formats/FormatFactory.cpp b/src/Formats/FormatFactory.cpp index 501343d2ca9..695f15e3fdb 100644 --- a/src/Formats/FormatFactory.cpp +++ b/src/Formats/FormatFactory.cpp @@ -123,6 +123,7 @@ FormatSettings getFormatSettings(const ContextPtr & context, const Settings & se format_settings.import_nested_json = settings.input_format_import_nested_json; format_settings.input_allow_errors_num = settings.input_format_allow_errors_num; format_settings.input_allow_errors_ratio = settings.input_format_allow_errors_ratio; + format_settings.json.max_depth = settings.input_format_json_max_depth; format_settings.json.array_of_rows = settings.output_format_json_array_of_rows; format_settings.json.escape_forward_slashes = settings.output_format_json_escape_forward_slashes; format_settings.json.write_named_tuples_as_objects = settings.output_format_json_named_tuples_as_objects; diff --git a/src/Formats/FormatSettings.h b/src/Formats/FormatSettings.h index 72d1515b7e1..32676970011 100644 --- a/src/Formats/FormatSettings.h +++ b/src/Formats/FormatSettings.h @@ -205,6 +205,7 @@ struct FormatSettings struct JSON { + size_t max_depth = 1000; bool array_of_rows = false; bool quote_64bit_integers = true; bool quote_64bit_floats = false; diff --git a/src/IO/ReadHelpers.cpp b/src/IO/ReadHelpers.cpp index c771fced73a..bc9c7bb110c 100644 --- a/src/IO/ReadHelpers.cpp +++ b/src/IO/ReadHelpers.cpp @@ -3,6 +3,7 @@ #include #include #include +#include #include #include #include @@ -1494,10 +1495,20 @@ template bool readDateTimeTextFallback(time_t &, ReadBuffer &, const template -ReturnType skipJSONFieldImpl(ReadBuffer & buf, StringRef name_of_field, const FormatSettings::JSON & settings) +ReturnType skipJSONFieldImpl(ReadBuffer & buf, StringRef name_of_field, const FormatSettings::JSON & settings, size_t current_depth) { static constexpr bool throw_exception = std::is_same_v; + if (unlikely(current_depth > settings.max_depth)) + { + if constexpr (throw_exception) + throw Exception(ErrorCodes::INCORRECT_DATA, "JSON is too deep for key '{}'", name_of_field.toString()); + return ReturnType(false); + } + + if (unlikely(current_depth > 0 && current_depth % 1024 == 0)) + checkStackSize(); + if (buf.eof()) { if constexpr (throw_exception) @@ -1560,8 +1571,8 @@ ReturnType skipJSONFieldImpl(ReadBuffer & buf, StringRef name_of_field, const Fo while (true) { if constexpr (throw_exception) - skipJSONFieldImpl(buf, name_of_field, settings); - else if (!skipJSONFieldImpl(buf, name_of_field, settings)) + skipJSONFieldImpl(buf, name_of_field, settings, current_depth + 1); + else if (!skipJSONFieldImpl(buf, name_of_field, settings, current_depth + 1)) return ReturnType(false); skipWhitespaceIfAny(buf); @@ -1619,8 +1630,8 @@ ReturnType skipJSONFieldImpl(ReadBuffer & buf, StringRef name_of_field, const Fo skipWhitespaceIfAny(buf); if constexpr (throw_exception) - skipJSONFieldImpl(buf, name_of_field, settings); - else if (!skipJSONFieldImpl(buf, name_of_field, settings)) + skipJSONFieldImpl(buf, name_of_field, settings, current_depth + 1); + else if (!skipJSONFieldImpl(buf, name_of_field, settings, current_depth + 1)) return ReturnType(false); skipWhitespaceIfAny(buf); @@ -1659,12 +1670,12 @@ ReturnType skipJSONFieldImpl(ReadBuffer & buf, StringRef name_of_field, const Fo void skipJSONField(ReadBuffer & buf, StringRef name_of_field, const FormatSettings::JSON & settings) { - skipJSONFieldImpl(buf, name_of_field, settings); + skipJSONFieldImpl(buf, name_of_field, settings, 0); } bool trySkipJSONField(ReadBuffer & buf, StringRef name_of_field, const FormatSettings::JSON & settings) { - return skipJSONFieldImpl(buf, name_of_field, settings); + return skipJSONFieldImpl(buf, name_of_field, settings, 0); } diff --git a/tests/queries/0_stateless/03213_deep_json.reference b/tests/queries/0_stateless/03213_deep_json.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03213_deep_json.sql b/tests/queries/0_stateless/03213_deep_json.sql new file mode 100644 index 00000000000..4f79d99eb26 --- /dev/null +++ b/tests/queries/0_stateless/03213_deep_json.sql @@ -0,0 +1,5 @@ +-- The default limit works. +SELECT * FROM format("JSONCompactEachRow", 'x UInt32, y UInt32', REPEAT('[1,1,', 100000)) SETTINGS input_format_json_compact_allow_variable_number_of_columns = 1; -- { serverError TOO_DEEP_RECURSION } +-- Even if we relax the limit, it is also safe. +SET input_format_json_max_depth = 100000; +SELECT * FROM format("JSONCompactEachRow", 'x UInt32, y UInt32', REPEAT('[1,1,', 100000)) SETTINGS input_format_json_compact_allow_variable_number_of_columns = 1; -- { serverError TOO_DEEP_RECURSION } From 86f8fa5efbfd0c7bfe3f00ff37b85b2efe17cfe5 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Mon, 29 Jul 2024 01:40:16 +0200 Subject: [PATCH 034/227] Fix stack overflow in too deep JSON --- src/IO/ReadHelpers.cpp | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/IO/ReadHelpers.cpp b/src/IO/ReadHelpers.cpp index bc9c7bb110c..9559462e62b 100644 --- a/src/IO/ReadHelpers.cpp +++ b/src/IO/ReadHelpers.cpp @@ -10,7 +10,6 @@ #include #include #include -#include #include #include @@ -40,6 +39,7 @@ namespace ErrorCodes extern const int ATTEMPT_TO_READ_AFTER_EOF; extern const int LOGICAL_ERROR; extern const int BAD_ARGUMENTS; + extern const int TOO_DEEP_RECURSION; } template @@ -1502,7 +1502,7 @@ ReturnType skipJSONFieldImpl(ReadBuffer & buf, StringRef name_of_field, const Fo if (unlikely(current_depth > settings.max_depth)) { if constexpr (throw_exception) - throw Exception(ErrorCodes::INCORRECT_DATA, "JSON is too deep for key '{}'", name_of_field.toString()); + throw Exception(ErrorCodes::TOO_DEEP_RECURSION, "JSON is too deep for key '{}'", name_of_field.toString()); return ReturnType(false); } From 19ce50f959b65fa1c1fd82b32c5d59252ed72383 Mon Sep 17 00:00:00 2001 From: flynn Date: Mon, 29 Jul 2024 02:36:28 +0000 Subject: [PATCH 035/227] Add missing reference file --- tests/queries/0_stateless/03208_uniq_with_empty_tuple.reference | 2 ++ tests/queries/0_stateless/03208_uniq_with_empty_tuple.sql | 1 + 2 files changed, 3 insertions(+) create mode 100644 tests/queries/0_stateless/03208_uniq_with_empty_tuple.reference diff --git a/tests/queries/0_stateless/03208_uniq_with_empty_tuple.reference b/tests/queries/0_stateless/03208_uniq_with_empty_tuple.reference new file mode 100644 index 00000000000..6ed281c757a --- /dev/null +++ b/tests/queries/0_stateless/03208_uniq_with_empty_tuple.reference @@ -0,0 +1,2 @@ +1 +1 diff --git a/tests/queries/0_stateless/03208_uniq_with_empty_tuple.sql b/tests/queries/0_stateless/03208_uniq_with_empty_tuple.sql index 4c5705261f0..09eeaf6f370 100644 --- a/tests/queries/0_stateless/03208_uniq_with_empty_tuple.sql +++ b/tests/queries/0_stateless/03208_uniq_with_empty_tuple.sql @@ -1,3 +1,4 @@ +-- Tags: no-fasttest -- https://github.com/ClickHouse/ClickHouse/issues/67303 SELECT uniqTheta(tuple()); SELECT uniq(tuple()); From 56135d18cb40506fc500b497068121bc96e3a9ee Mon Sep 17 00:00:00 2001 From: avogar Date: Mon, 29 Jul 2024 11:23:01 +0000 Subject: [PATCH 036/227] Remove duplicated tests --- .../03036_dynamic_read_subcolumns.lib | 35 ------------------- .../03036_dynamic_read_subcolumns_1.reference | 19 ---------- .../03036_dynamic_read_subcolumns_1.sh | 21 ----------- .../03036_dynamic_read_subcolumns_2.reference | 19 ---------- .../03036_dynamic_read_subcolumns_2.sh | 21 ----------- .../03036_dynamic_read_subcolumns_3.reference | 19 ---------- .../03036_dynamic_read_subcolumns_3.sh | 21 ----------- 7 files changed, 155 deletions(-) delete mode 100755 tests/queries/0_stateless/03036_dynamic_read_subcolumns.lib delete mode 100644 tests/queries/0_stateless/03036_dynamic_read_subcolumns_1.reference delete mode 100755 tests/queries/0_stateless/03036_dynamic_read_subcolumns_1.sh delete mode 100644 tests/queries/0_stateless/03036_dynamic_read_subcolumns_2.reference delete mode 100755 tests/queries/0_stateless/03036_dynamic_read_subcolumns_2.sh delete mode 100644 tests/queries/0_stateless/03036_dynamic_read_subcolumns_3.reference delete mode 100755 tests/queries/0_stateless/03036_dynamic_read_subcolumns_3.sh diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns.lib b/tests/queries/0_stateless/03036_dynamic_read_subcolumns.lib deleted file mode 100755 index 4914051db82..00000000000 --- a/tests/queries/0_stateless/03036_dynamic_read_subcolumns.lib +++ /dev/null @@ -1,35 +0,0 @@ -#!/usr/bin/env bash - -function test() -{ - echo "test" - $CH_CLIENT -q "insert into test select number, number from numbers(100000) settings min_insert_block_size_rows=50000" - $CH_CLIENT -q "insert into test select number, 'str_' || toString(number) from numbers(100000, 100000) settings min_insert_block_size_rows=50000" - $CH_CLIENT -q "insert into test select number, arrayMap(x -> multiIf(number % 9 == 0, NULL, number % 9 == 3, 'str_' || toString(number), number), range(number % 10 + 1)) from numbers(200000, 100000) settings min_insert_block_size_rows=50000" - $CH_CLIENT -q "insert into test select number, NULL from numbers(300000, 100000) settings min_insert_block_size_rows=50000" - $CH_CLIENT -q "insert into test select number, multiIf(number % 4 == 3, 'str_' || toString(number), number % 4 == 2, NULL, number % 4 == 1, number, arrayMap(x -> multiIf(number % 9 == 0, NULL, number % 9 == 3, 'str_' || toString(number), number), range(number % 10 + 1))) from numbers(400000, 400000) settings min_insert_block_size_rows=50000" - $CH_CLIENT -q "insert into test select number, [range((number % 10 + 1)::UInt64)]::Array(Array(Dynamic)) from numbers(100000, 100000) settings min_insert_block_size_rows=50000" - - $CH_CLIENT -q "select distinct dynamicType(d) as type from test order by type" - $CH_CLIENT -q "select count() from test where dynamicType(d) == 'UInt64'" - $CH_CLIENT -q "select count() from test where d.UInt64 is not NULL" - $CH_CLIENT -q "select count() from test where dynamicType(d) == 'String'" - $CH_CLIENT -q "select count() from test where d.String is not NULL" - $CH_CLIENT -q "select count() from test where dynamicType(d) == 'Date'" - $CH_CLIENT -q "select count() from test where d.Date is not NULL" - $CH_CLIENT -q "select count() from test where dynamicType(d) == 'Array(Variant(String, UInt64))'" - $CH_CLIENT -q "select count() from test where not empty(d.\`Array(Variant(String, UInt64))\`)" - $CH_CLIENT -q "select count() from test where dynamicType(d) == 'Array(Array(Dynamic))'" - $CH_CLIENT -q "select count() from test where not empty(d.\`Array(Array(Dynamic))\`)" - $CH_CLIENT -q "select count() from test where d is NULL" - $CH_CLIENT -q "select count() from test where not empty(d.\`Tuple(a Array(Dynamic))\`.a.String)" - - $CH_CLIENT -q "select d, d.UInt64, d.String, d.\`Array(Variant(String, UInt64))\` from test format Null" - $CH_CLIENT -q "select d.UInt64, d.String, d.\`Array(Variant(String, UInt64))\` from test format Null" - $CH_CLIENT -q "select d.Int8, d.Date, d.\`Array(String)\` from test format Null" - $CH_CLIENT -q "select d, d.UInt64, d.Date, d.\`Array(Variant(String, UInt64))\`, d.\`Array(Variant(String, UInt64))\`.size0, d.\`Array(Variant(String, UInt64))\`.UInt64 from test format Null" - $CH_CLIENT -q "select d.UInt64, d.Date, d.\`Array(Variant(String, UInt64))\`, d.\`Array(Variant(String, UInt64))\`.size0, d.\`Array(Variant(String, UInt64))\`.UInt64, d.\`Array(Variant(String, UInt64))\`.String from test format Null" - $CH_CLIENT -q "select d, d.\`Tuple(a UInt64, b String)\`.a, d.\`Array(Dynamic)\`.\`Variant(String, UInt64)\`.UInt64, d.\`Array(Variant(String, UInt64))\`.UInt64 from test format Null" - $CH_CLIENT -q "select d.\`Array(Dynamic)\`.\`Variant(String, UInt64)\`.UInt64, d.\`Array(Dynamic)\`.size0, d.\`Array(Variant(String, UInt64))\`.UInt64 from test format Null" - $CH_CLIENT -q "select d.\`Array(Array(Dynamic))\`.size1, d.\`Array(Array(Dynamic))\`.UInt64, d.\`Array(Array(Dynamic))\`.\`Map(String, Tuple(a UInt64))\`.values.a from test format Null" -} diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_1.reference b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_1.reference deleted file mode 100644 index 0d51ecfac3b..00000000000 --- a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_1.reference +++ /dev/null @@ -1,19 +0,0 @@ -Memory -test -Array(Array(Dynamic)) -Array(Variant(String, UInt64)) -None -String -UInt64 -200000 -200000 -200000 -200000 -0 -0 -200000 -200000 -100000 -100000 -200000 -0 diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_1.sh b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_1.sh deleted file mode 100755 index aabba731816..00000000000 --- a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_1.sh +++ /dev/null @@ -1,21 +0,0 @@ -#!/usr/bin/env bash -# Tags: long - -CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) -# reset --log_comment -CLICKHOUSE_LOG_COMMENT= -# shellcheck source=../shell_config.sh -. "$CUR_DIR"/../shell_config.sh - - -# shellcheck source=./03036_dynamic_read_subcolumns.lib -. "$CUR_DIR"/03036_dynamic_read_subcolumns.lib - -CH_CLIENT="$CLICKHOUSE_CLIENT --allow_experimental_variant_type=1 --use_variant_as_common_type=1 --allow_experimental_dynamic_type=1" - -$CH_CLIENT -q "drop table if exists test;" - -echo "Memory" -$CH_CLIENT -q "create table test (id UInt64, d Dynamic) engine=Memory" -test -$CH_CLIENT -q "drop table test;" diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_2.reference b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_2.reference deleted file mode 100644 index 099b7574566..00000000000 --- a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_2.reference +++ /dev/null @@ -1,19 +0,0 @@ -MergeTree compact -test -Array(Array(Dynamic)) -Array(Variant(String, UInt64)) -None -String -UInt64 -200000 -200000 -200000 -200000 -0 -0 -200000 -200000 -100000 -100000 -200000 -0 diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_2.sh b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_2.sh deleted file mode 100755 index 872f4c20a98..00000000000 --- a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_2.sh +++ /dev/null @@ -1,21 +0,0 @@ -#!/usr/bin/env bash -# Tags: long - -CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) -# reset --log_comment -CLICKHOUSE_LOG_COMMENT= -# shellcheck source=../shell_config.sh -. "$CUR_DIR"/../shell_config.sh - - -# shellcheck source=./03036_dynamic_read_subcolumns.lib -. "$CUR_DIR"/03036_dynamic_read_subcolumns.lib - -CH_CLIENT="$CLICKHOUSE_CLIENT --allow_experimental_variant_type=1 --use_variant_as_common_type=1 --allow_experimental_dynamic_type=1" - -$CH_CLIENT -q "drop table if exists test;" - -echo "MergeTree compact" -$CH_CLIENT -q "create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000;" -test -$CH_CLIENT -q "drop table test;" diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_3.reference b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_3.reference deleted file mode 100644 index 35db4a22b4c..00000000000 --- a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_3.reference +++ /dev/null @@ -1,19 +0,0 @@ -MergeTree wide -test -Array(Array(Dynamic)) -Array(Variant(String, UInt64)) -None -String -UInt64 -200000 -200000 -200000 -200000 -0 -0 -200000 -200000 -100000 -100000 -200000 -0 diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_3.sh b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_3.sh deleted file mode 100755 index 96276c96add..00000000000 --- a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_3.sh +++ /dev/null @@ -1,21 +0,0 @@ -#!/usr/bin/env bash -# Tags: long - -CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) -# reset --log_comment -CLICKHOUSE_LOG_COMMENT= -# shellcheck source=../shell_config.sh -. "$CUR_DIR"/../shell_config.sh - - -# shellcheck source=./03036_dynamic_read_subcolumns.lib -. "$CUR_DIR"/03036_dynamic_read_subcolumns.lib - -CH_CLIENT="$CLICKHOUSE_CLIENT --allow_experimental_variant_type=1 --use_variant_as_common_type=1 --allow_experimental_dynamic_type=1" - -$CH_CLIENT -q "drop table if exists test;" - -echo "MergeTree wide" -$CH_CLIENT -q "create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1;" -test -$CH_CLIENT -q "drop table test;" From 161e794618a86f39ad4f5e6435306e384c85d01d Mon Sep 17 00:00:00 2001 From: kssenii Date: Mon, 29 Jul 2024 12:57:36 +0200 Subject: [PATCH 037/227] Add no-distributed-cache tag --- docker/test/stateless/run.sh | 4 ++++ tests/clickhouse-test | 10 ++++++++++ tests/queries/0_stateless/02995_index_9.sh | 2 +- 3 files changed, 15 insertions(+), 1 deletion(-) diff --git a/docker/test/stateless/run.sh b/docker/test/stateless/run.sh index f9f96c76d59..da17b82d91b 100755 --- a/docker/test/stateless/run.sh +++ b/docker/test/stateless/run.sh @@ -212,6 +212,10 @@ function run_tests() ADDITIONAL_OPTIONS+=('--shared-catalog') fi + if [[ "$USE_DISTRIBUTED_CACHE" -eq 1 ]]; then + ADDITIONAL_OPTIONS+=('--distributed-cache') + fi + if [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then ADDITIONAL_OPTIONS+=('--replicated-database') # Too many tests fail for DatabaseReplicated in parallel. diff --git a/tests/clickhouse-test b/tests/clickhouse-test index a29c786e998..f810a8714bd 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -681,6 +681,7 @@ class FailureReason(enum.Enum): BUILD = "not running for current build" NO_PARALLEL_REPLICAS = "smth in not supported with parallel replicas" SHARED_MERGE_TREE = "no-shared-merge-tree" + DISTRIBUTED_CACHE = "distributed-cache" # UNKNOWN reasons NO_REFERENCE = "no reference file" @@ -1191,6 +1192,9 @@ class TestCase: elif tags and ("no-replicated-database" in tags) and args.replicated_database: return FailureReason.REPLICATED_DB + elif tags and ("no-distributed-cache" in tags) and args.distributed_cache: + return FailureReason.DISTRIBUTED_CACHE + elif ( tags and ("atomic-database" in tags) @@ -3204,6 +3208,12 @@ def parse_args(): default=False, help="Run tests over s3 storage", ) + parser.add_argument( + "--distributed-cache", + action="store_true", + default=False, + help="Run tests with enabled distributed cache", + ) parser.add_argument( "--azure-blob-storage", action="store_true", diff --git a/tests/queries/0_stateless/02995_index_9.sh b/tests/queries/0_stateless/02995_index_9.sh index 76160c62aaa..46426f22185 100755 --- a/tests/queries/0_stateless/02995_index_9.sh +++ b/tests/queries/0_stateless/02995_index_9.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage +# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage, no-distributed-cache CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh From 8affc58c4d77eb490b6b969020d8d797bfb49cb9 Mon Sep 17 00:00:00 2001 From: kssenii Date: Mon, 29 Jul 2024 14:10:19 +0200 Subject: [PATCH 038/227] Tag more tests --- tests/queries/0_stateless/01603_read_with_backoff_bug.sql | 2 +- .../0_stateless/01605_adaptive_granularity_block_borders.sql | 2 +- tests/queries/0_stateless/02714_local_object_storage.sql | 2 ++ tests/queries/0_stateless/02770_async_buffer_ignore.sh | 2 +- 4 files changed, 5 insertions(+), 3 deletions(-) diff --git a/tests/queries/0_stateless/01603_read_with_backoff_bug.sql b/tests/queries/0_stateless/01603_read_with_backoff_bug.sql index 1cf52c0288b..3d467515b5c 100644 --- a/tests/queries/0_stateless/01603_read_with_backoff_bug.sql +++ b/tests/queries/0_stateless/01603_read_with_backoff_bug.sql @@ -1,4 +1,4 @@ --- Tags: no-tsan +-- Tags: no-tsan, no-distributed-cache -- Tag no-tsan: Too long for TSan set enable_filesystem_cache=0; diff --git a/tests/queries/0_stateless/01605_adaptive_granularity_block_borders.sql b/tests/queries/0_stateless/01605_adaptive_granularity_block_borders.sql index 9b96ce3e586..5f09dc423b2 100644 --- a/tests/queries/0_stateless/01605_adaptive_granularity_block_borders.sql +++ b/tests/queries/0_stateless/01605_adaptive_granularity_block_borders.sql @@ -1,4 +1,4 @@ --- Tags: no-random-merge-tree-settings, no-tsan, no-debug, no-object-storage +-- Tags: no-random-merge-tree-settings, no-tsan, no-debug, no-object-storage, no-distributed-cache -- no-tsan: too slow -- no-object-storage: for remote tables we use thread pool even when reading with one stream, so memory consumption is higher diff --git a/tests/queries/0_stateless/02714_local_object_storage.sql b/tests/queries/0_stateless/02714_local_object_storage.sql index 914eac1ddba..bfe3162df97 100644 --- a/tests/queries/0_stateless/02714_local_object_storage.sql +++ b/tests/queries/0_stateless/02714_local_object_storage.sql @@ -1,3 +1,5 @@ +-- Tags: no-distributed-cache + SET min_bytes_to_use_direct_io='1Gi'; -- It does not work (fixme) SET local_filesystem_read_method='pread'; -- ui_uring local_fs_method does not work here (fixme) diff --git a/tests/queries/0_stateless/02770_async_buffer_ignore.sh b/tests/queries/0_stateless/02770_async_buffer_ignore.sh index 6eaf655324d..37f002767d6 100755 --- a/tests/queries/0_stateless/02770_async_buffer_ignore.sh +++ b/tests/queries/0_stateless/02770_async_buffer_ignore.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-fasttest, no-random-settings +# Tags: no-fasttest, no-random-settings, no-distributed-cache CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh From 1333fe24645f36f2dd5f45a084d379de834e71a8 Mon Sep 17 00:00:00 2001 From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com> Date: Mon, 29 Jul 2024 16:09:51 +0200 Subject: [PATCH 039/227] Update 01603_read_with_backoff_bug.sql --- tests/queries/0_stateless/01603_read_with_backoff_bug.sql | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/01603_read_with_backoff_bug.sql b/tests/queries/0_stateless/01603_read_with_backoff_bug.sql index 3d467515b5c..8b97f3514b3 100644 --- a/tests/queries/0_stateless/01603_read_with_backoff_bug.sql +++ b/tests/queries/0_stateless/01603_read_with_backoff_bug.sql @@ -1,4 +1,4 @@ --- Tags: no-tsan, no-distributed-cache +-- Tags: long, no-tsan, no-distributed-cache -- Tag no-tsan: Too long for TSan set enable_filesystem_cache=0; From af63af8ff2d5f86aa22300b1e19c37b36307dac9 Mon Sep 17 00:00:00 2001 From: kssenii Date: Mon, 29 Jul 2024 16:44:19 +0200 Subject: [PATCH 040/227] Update a few more tests --- tests/queries/0_stateless/02150_index_hypothesis_race_long.sh | 1 + tests/queries/0_stateless/02995_index_2.sh | 2 +- tests/queries/0_stateless/02995_index_4.sh | 2 +- 3 files changed, 3 insertions(+), 2 deletions(-) diff --git a/tests/queries/0_stateless/02150_index_hypothesis_race_long.sh b/tests/queries/0_stateless/02150_index_hypothesis_race_long.sh index 114f60cc393..be7cfa78492 100755 --- a/tests/queries/0_stateless/02150_index_hypothesis_race_long.sh +++ b/tests/queries/0_stateless/02150_index_hypothesis_race_long.sh @@ -1,4 +1,5 @@ #!/usr/bin/env bash +# Tags: no-distributed-cache CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/02995_index_2.sh b/tests/queries/0_stateless/02995_index_2.sh index e7451c7ee4b..f2f466e8ff7 100755 --- a/tests/queries/0_stateless/02995_index_2.sh +++ b/tests/queries/0_stateless/02995_index_2.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage +# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage, no-distributed-cache CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/02995_index_4.sh b/tests/queries/0_stateless/02995_index_4.sh index 1a0458728f9..59be06ae17a 100755 --- a/tests/queries/0_stateless/02995_index_4.sh +++ b/tests/queries/0_stateless/02995_index_4.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage +# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage, no-distributed-cache CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh From 819196cc7469ed02677b7edbc63ebd5dea666456 Mon Sep 17 00:00:00 2001 From: Jordi Date: Mon, 29 Jul 2024 17:48:08 +0200 Subject: [PATCH 041/227] Add total marks and parts to ProfileEvents --- src/Common/ProfileEvents.cpp | 2 ++ src/Processors/QueryPlan/ReadFromMergeTree.cpp | 4 ++++ 2 files changed, 6 insertions(+) diff --git a/src/Common/ProfileEvents.cpp b/src/Common/ProfileEvents.cpp index 2b090136e2e..f8bd84df131 100644 --- a/src/Common/ProfileEvents.cpp +++ b/src/Common/ProfileEvents.cpp @@ -193,8 +193,10 @@ M(ReplicaPartialShutdown, "How many times Replicated table has to deinitialize its state due to session expiration in ZooKeeper. The state is reinitialized every time when ZooKeeper is available again.") \ \ M(SelectedParts, "Number of data parts selected to read from a MergeTree table.") \ + M(SelectedPartsTotal, "Number of total data parts before selecting which ones to read from a MergeTree table.") \ M(SelectedRanges, "Number of (non-adjacent) ranges in all data parts selected to read from a MergeTree table.") \ M(SelectedMarks, "Number of marks (index granules) selected to read from a MergeTree table.") \ + M(SelectedMarksTotal, "Number of total marks (index granules) before selecting which ones to read from a MergeTree table.") \ M(SelectedRows, "Number of rows SELECTed from all tables.") \ M(SelectedBytes, "Number of bytes (uncompressed; for columns as they stored in memory) SELECTed from all tables.") \ M(RowsReadByMainReader, "Number of rows read from MergeTree tables by the main reader (after PREWHERE step).") \ diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index a7ce92fe8e0..6e3a71660ad 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -120,8 +120,10 @@ bool restorePrewhereInputs(PrewhereInfo & info, const NameSet & inputs) namespace ProfileEvents { extern const Event SelectedParts; + extern const Event SelectedPartsTotal; extern const Event SelectedRanges; extern const Event SelectedMarks; + extern const Event SelectedMarksTotal; extern const Event SelectQueriesWithPrimaryKeyUsage; } @@ -1971,8 +1973,10 @@ void ReadFromMergeTree::initializePipeline(QueryPipelineBuilder & pipeline, cons } ProfileEvents::increment(ProfileEvents::SelectedParts, result.selected_parts); + ProfileEvents::increment(ProfileEvents::SelectedPartsTotal, result.total_parts); ProfileEvents::increment(ProfileEvents::SelectedRanges, result.selected_ranges); ProfileEvents::increment(ProfileEvents::SelectedMarks, result.selected_marks); + ProfileEvents::increment(ProfileEvents::SelectedMarksTotal, result.total_marks_pk); auto query_id_holder = MergeTreeDataSelectExecutor::checkLimits(data, result, context); From 28ae67187542d7ee3edf9dd30700d5589a2f8389 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?= Date: Mon, 29 Jul 2024 15:52:25 +0000 Subject: [PATCH 042/227] Print debug info --- .../integration/test_storage_s3_queue/test.py | 29 ++++++++++++------- 1 file changed, 18 insertions(+), 11 deletions(-) diff --git a/tests/integration/test_storage_s3_queue/test.py b/tests/integration/test_storage_s3_queue/test.py index bf3c28c5429..93234e635b4 100644 --- a/tests/integration/test_storage_s3_queue/test.py +++ b/tests/integration/test_storage_s3_queue/test.py @@ -1352,16 +1352,7 @@ def test_shards_distributed(started_cluster, mode, processing_threads): def get_count(node, table_name): return int(run_query(node, f"SELECT count() FROM {table_name}")) - for _ in range(30): - if ( - get_count(node, dst_table_name) + get_count(node_2, dst_table_name) - ) == total_rows: - break - time.sleep(1) - - if ( - get_count(node, dst_table_name) + get_count(node_2, dst_table_name) - ) != total_rows: + def print_debug_info(): processed_files = ( node.query( f""" @@ -1388,7 +1379,7 @@ select splitByChar('/', file_name)[-1] as file from system.s3queue where zookeep ) count = get_count(node, dst_table_name) + get_count(node_2, dst_table_name) - logging.debug(f"Processed rows: {count}/{files_to_generate}") + logging.debug(f"Processed rows: {count}/{total_rows}") info = node.query( f""" @@ -1425,6 +1416,18 @@ select splitByChar('/', file_name)[-1] as file from system.s3queue where zookeep logging.debug(f"Intersecting files: {intersection(files1, files2)}") + for _ in range(30): + if ( + get_count(node, dst_table_name) + get_count(node_2, dst_table_name) + ) == total_rows: + break + time.sleep(1) + + if ( + get_count(node, dst_table_name) + get_count(node_2, dst_table_name) + ) != total_rows: + print_debug_info() + assert False get_query = f"SELECT column1, column2, column3 FROM {dst_table_name}" @@ -1433,6 +1436,10 @@ select splitByChar('/', file_name)[-1] as file from system.s3queue where zookeep list(map(int, l.split())) for l in run_query(node_2, get_query).splitlines() ] + if len(res1) + len(res2) != total_rows or len(res1) <= 0 or len(res2) <= 0 or True: + logging.debug(f"res1 size: {len(res1)}, res2 size: {len(res2)}, total_rows: {total_rows}") + print_debug_info() + assert len(res1) + len(res2) == total_rows # Checking that all engines have made progress From f7f2a9872d3a723559779d0be0f2ef915e8cfd31 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Mon, 29 Jul 2024 16:02:06 +0000 Subject: [PATCH 043/227] Automatic style fix --- tests/integration/test_storage_s3_queue/test.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tests/integration/test_storage_s3_queue/test.py b/tests/integration/test_storage_s3_queue/test.py index 93234e635b4..0846872e716 100644 --- a/tests/integration/test_storage_s3_queue/test.py +++ b/tests/integration/test_storage_s3_queue/test.py @@ -1437,7 +1437,9 @@ select splitByChar('/', file_name)[-1] as file from system.s3queue where zookeep ] if len(res1) + len(res2) != total_rows or len(res1) <= 0 or len(res2) <= 0 or True: - logging.debug(f"res1 size: {len(res1)}, res2 size: {len(res2)}, total_rows: {total_rows}") + logging.debug( + f"res1 size: {len(res1)}, res2 size: {len(res2)}, total_rows: {total_rows}" + ) print_debug_info() assert len(res1) + len(res2) == total_rows From cb056cf3a5080cbff61f6efd070733ae2061d5b8 Mon Sep 17 00:00:00 2001 From: Peter Nguyen Date: Mon, 29 Jul 2024 12:05:44 -0700 Subject: [PATCH 044/227] Add camelCase aliases for percentRank() and denseRank() for percent_rank() and dense_rank() --- src/Processors/Transforms/WindowTransform.cpp | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/src/Processors/Transforms/WindowTransform.cpp b/src/Processors/Transforms/WindowTransform.cpp index 2b255c5120e..a1b46c8e36c 100644 --- a/src/Processors/Transforms/WindowTransform.cpp +++ b/src/Processors/Transforms/WindowTransform.cpp @@ -2721,20 +2721,24 @@ void registerWindowFunctions(AggregateFunctionFactory & factory) parameters); }, properties}, AggregateFunctionFactory::Case::Insensitive); - factory.registerFunction("dense_rank", {[](const std::string & name, + factory.registerFunction("denseRank", {[](const std::string & name, const DataTypes & argument_types, const Array & parameters, const Settings *) { return std::make_shared(name, argument_types, parameters); }, properties}, AggregateFunctionFactory::Case::Insensitive); - factory.registerFunction("percent_rank", {[](const std::string & name, + factory.registerAlias("dense_rank", "denseRank", AggregateFunctionFactory::Case::Sensitive); + + factory.registerFunction("percentRank", {[](const std::string & name, const DataTypes & argument_types, const Array & parameters, const Settings *) { return std::make_shared(name, argument_types, parameters); }, properties}, AggregateFunctionFactory::Case::Insensitive); + factory.registerAlias("percent_rank", "percentRank", AggregateFunctionFactory::Case::Sensitive); + factory.registerFunction("row_number", {[](const std::string & name, const DataTypes & argument_types, const Array & parameters, const Settings *) { From 9811a2e71b825a55c376edfb38303c817493cd9e Mon Sep 17 00:00:00 2001 From: Peter Nguyen Date: Sun, 28 Jul 2024 17:32:32 -0700 Subject: [PATCH 045/227] Add test 03213_denseRank_percentRank_alias --- ...3213_denseRank_percentRank_alias.reference | 45 ++++++++++++++ .../03213_denseRank_percentRank_alias.sql | 59 +++++++++++++++++++ 2 files changed, 104 insertions(+) create mode 100644 tests/queries/0_stateless/03213_denseRank_percentRank_alias.reference create mode 100644 tests/queries/0_stateless/03213_denseRank_percentRank_alias.sql diff --git a/tests/queries/0_stateless/03213_denseRank_percentRank_alias.reference b/tests/queries/0_stateless/03213_denseRank_percentRank_alias.reference new file mode 100644 index 00000000000..b49e179df68 --- /dev/null +++ b/tests/queries/0_stateless/03213_denseRank_percentRank_alias.reference @@ -0,0 +1,45 @@ +---- denseRank() ---- +0 0 0 1 1 1 1 +3 0 0 2 2 2 2 +1 0 1 3 3 3 3 +4 0 1 4 4 4 4 +2 0 2 5 5 5 5 +6 1 0 1 1 1 1 +9 1 0 2 2 2 2 +7 1 1 3 3 3 3 +5 1 2 4 4 4 4 +8 1 2 5 5 5 5 +12 2 0 1 1 1 1 +10 2 1 2 2 2 2 +13 2 1 3 3 3 3 +11 2 2 4 4 4 4 +14 2 2 5 5 5 5 +15 3 0 1 1 1 1 +18 3 0 2 2 2 2 +16 3 1 3 3 3 3 +19 3 1 4 4 4 4 +17 3 2 5 5 5 5 +21 4 0 1 1 1 1 +24 4 0 2 2 2 2 +22 4 1 3 3 3 3 +20 4 2 4 4 4 4 +23 4 2 5 5 5 5 +27 5 0 1 1 1 1 +25 5 1 2 2 2 2 +28 5 1 3 3 3 3 +26 5 2 4 4 4 4 +29 5 2 5 5 5 5 +30 6 0 1 1 1 1 +---- percentRank() ---- +Lenovo Thinkpad Laptop 700 1 0 +Sony VAIO Laptop 700 1 0 +Dell Vostro Laptop 800 3 0.6666666666666666 +HP Elite Laptop 1200 4 1 +Microsoft Lumia Smartphone 200 1 0 +HTC One Smartphone 400 2 0.3333333333333333 +Nexus Smartphone 500 3 0.6666666666666666 +iPhone Smartphone 900 4 1 +Kindle Fire Tablet 150 1 0 +Samsung Galaxy Tab Tablet 200 2 0.5 +iPad Tablet 700 3 1 +Others Unknow 200 1 0 diff --git a/tests/queries/0_stateless/03213_denseRank_percentRank_alias.sql b/tests/queries/0_stateless/03213_denseRank_percentRank_alias.sql new file mode 100644 index 00000000000..ff841294eb1 --- /dev/null +++ b/tests/queries/0_stateless/03213_denseRank_percentRank_alias.sql @@ -0,0 +1,59 @@ +-- https://github.com/ClickHouse/ClickHouse/issues/67042 +-- Reference generated using percent_rank() and dense_rank() + +-- From ClickHouse/tests/queries/0_stateless/01591_window_functions.sql (for deterministic query) +SELECT '---- denseRank() ----'; +select number, p, o, + count(*) over w, + rank() over w, + denseRank() over w, + row_number() over w +from (select number, intDiv(number, 5) p, mod(number, 3) o + from numbers(31) order by o, number) t +window w as (partition by p order by o, number) +order by p, o, number +settings max_block_size = 2; + +-- Modifed from ClickHouse/tests/queries/0_stateless/01592_window_functions.sql (for deterministic query) +SELECT '---- percentRank() ----'; + +drop table if exists product_groups; +drop table if exists products; + +CREATE TABLE product_groups ( + group_id Int64, + group_name String +) Engine = Memory; + +CREATE TABLE products ( + product_id Int64, + product_name String, + price DECIMAL(11, 2), + group_id Int64 +) Engine = Memory; + +INSERT INTO product_groups VALUES (1, 'Smartphone'),(2, 'Laptop'),(3, 'Tablet'); +INSERT INTO products (product_id,product_name, group_id,price) VALUES (1, 'Microsoft Lumia', 1, 200), (2, 'HTC One', 1, 400), (3, 'Nexus', 1, 500), (4, 'iPhone', 1, 900),(5, 'HP Elite', 2, 1200),(6, 'Lenovo Thinkpad', 2, 700),(7, 'Sony VAIO', 2, 700),(8, 'Dell Vostro', 2, 800),(9, 'iPad', 3, 700),(10, 'Kindle Fire', 3, 150),(11, 'Samsung Galaxy Tab', 3, 200); +INSERT INTO product_groups VALUES (4, 'Unknow'); +INSERT INTO products (product_id,product_name, group_id,price) VALUES (12, 'Others', 4, 200); + + +SELECT * +FROM +( + SELECT + product_name, + group_name, + price, + rank() OVER (PARTITION BY group_name ORDER BY price ASC) AS rank, + percentRank() OVER (PARTITION BY group_name ORDER BY price ASC) AS percent + FROM products + INNER JOIN product_groups USING (group_id) +) AS t +ORDER BY + group_name ASC, + price ASC, + product_name ASC; + +drop table product_groups; +drop table products; From 7a53a14940ae1be299305548f0d024de7f279fe3 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Tue, 30 Jul 2024 09:29:16 +0200 Subject: [PATCH 046/227] Update 03213_deep_json.sql --- tests/queries/0_stateless/03213_deep_json.sql | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/queries/0_stateless/03213_deep_json.sql b/tests/queries/0_stateless/03213_deep_json.sql index 4f79d99eb26..2a9476381ff 100644 --- a/tests/queries/0_stateless/03213_deep_json.sql +++ b/tests/queries/0_stateless/03213_deep_json.sql @@ -1,5 +1,5 @@ -- The default limit works. -SELECT * FROM format("JSONCompactEachRow", 'x UInt32, y UInt32', REPEAT('[1,1,', 100000)) SETTINGS input_format_json_compact_allow_variable_number_of_columns = 1; -- { serverError TOO_DEEP_RECURSION } +SELECT * FROM format("JSONCompactEachRow", 'x UInt32, y UInt32', REPEAT('[1,1,', 100000)) SETTINGS input_format_json_compact_allow_variable_number_of_columns = 1; -- { serverError TOO_DEEP_RECURSION, INCORRECT_DATA } -- Even if we relax the limit, it is also safe. SET input_format_json_max_depth = 100000; -SELECT * FROM format("JSONCompactEachRow", 'x UInt32, y UInt32', REPEAT('[1,1,', 100000)) SETTINGS input_format_json_compact_allow_variable_number_of_columns = 1; -- { serverError TOO_DEEP_RECURSION } +SELECT * FROM format("JSONCompactEachRow", 'x UInt32, y UInt32', REPEAT('[1,1,', 100000)) SETTINGS input_format_json_compact_allow_variable_number_of_columns = 1; -- { serverError TOO_DEEP_RECURSION, INCORRECT_DATA } From a70571762f7d73a7ecc94981e8086418ecfdeb3b Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Tue, 30 Jul 2024 10:08:35 +0200 Subject: [PATCH 047/227] Enable text_log by default --- programs/server/config.xml | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/programs/server/config.xml b/programs/server/config.xml index 94825a55f67..844aff8f668 100644 --- a/programs/server/config.xml +++ b/programs/server/config.xml @@ -1130,8 +1130,7 @@ 7500 - + system part_log
@@ -1143,9 +1142,9 @@ false - system text_log
@@ -1154,9 +1153,8 @@ 8192 524288 false - + trace - --> From 3e6a1b99e023eb3d592c72c17ae4913a9074b5af Mon Sep 17 00:00:00 2001 From: Antonio Andelic Date: Tue, 30 Jul 2024 11:05:45 +0200 Subject: [PATCH 048/227] Fix file/URI parsing with archive syntax --- src/IO/Archives/ArchiveUtils.cpp | 50 +++++++++++++++++++ src/IO/Archives/ArchiveUtils.h | 14 ++++++ src/IO/Archives/createArchiveReader.cpp | 13 ++--- src/IO/Archives/createArchiveWriter.cpp | 9 ++-- src/IO/S3/URI.cpp | 39 +++++++-------- src/IO/S3/URI.h | 3 +- src/Storages/StorageFile.cpp | 8 ++- src/TableFunctions/TableFunctionFile.cpp | 9 ++-- .../03214_parsing_archive_name_file.reference | 12 +++++ .../03214_parsing_archive_name_file.sh | 21 ++++++++ .../03215_parsing_archive_name_s3.reference | 2 + .../03215_parsing_archive_name_s3.sql | 6 +++ .../data_minio/::03215_archive.csv | 1 + .../data_minio/test :: 03215_archive.csv | 1 + .../data_minio/test::03215_archive.csv | 1 + 15 files changed, 145 insertions(+), 44 deletions(-) create mode 100644 src/IO/Archives/ArchiveUtils.cpp create mode 100644 tests/queries/0_stateless/03214_parsing_archive_name_file.reference create mode 100755 tests/queries/0_stateless/03214_parsing_archive_name_file.sh create mode 100644 tests/queries/0_stateless/03215_parsing_archive_name_s3.reference create mode 100644 tests/queries/0_stateless/03215_parsing_archive_name_s3.sql create mode 100644 tests/queries/0_stateless/data_minio/::03215_archive.csv create mode 100644 tests/queries/0_stateless/data_minio/test :: 03215_archive.csv create mode 100644 tests/queries/0_stateless/data_minio/test::03215_archive.csv diff --git a/src/IO/Archives/ArchiveUtils.cpp b/src/IO/Archives/ArchiveUtils.cpp new file mode 100644 index 00000000000..50009087de3 --- /dev/null +++ b/src/IO/Archives/ArchiveUtils.cpp @@ -0,0 +1,50 @@ +#include + +#include +#include + +namespace DB +{ + +namespace +{ + +using namespace std::literals; +constexpr std::array tar_extensions{".tar"sv, ".tar.gz"sv, ".tgz"sv, ".tar.zst"sv, ".tzst"sv, ".tar.xz"sv, ".tar.bz2"sv, ".tar.lzma"sv}; +constexpr std::array zip_extensions{".zip"sv, ".zipx"sv}; +constexpr std::array sevenz_extensiosns{".7z"sv}; + +bool hasSupportedExtension(std::string_view path, const auto & supported_extensions) +{ + for (auto supported_extension : supported_extensions) + { + if (path.ends_with(supported_extension)) + return true; + } + + return false; +} + +} + +bool hasSupportedTarExtension(std::string_view path) +{ + return hasSupportedExtension(path, tar_extensions); +} + +bool hasSupportedZipExtension(std::string_view path) +{ + return hasSupportedExtension(path, zip_extensions); +} + +bool hasSupported7zExtension(std::string_view path) +{ + return hasSupportedExtension(path, sevenz_extensiosns); +} + +bool hasSupportedArchiveExtension(std::string_view path) +{ + return hasSupportedTarExtension(path) || hasSupportedZipExtension(path) || hasSupported7zExtension(path); +} + +} diff --git a/src/IO/Archives/ArchiveUtils.h b/src/IO/Archives/ArchiveUtils.h index 1b66be005a2..cdb731d1d57 100644 --- a/src/IO/Archives/ArchiveUtils.h +++ b/src/IO/Archives/ArchiveUtils.h @@ -10,3 +10,17 @@ #include #include #endif + +#include + +namespace DB +{ + +bool hasSupportedTarExtension(std::string_view path); +bool hasSupportedZipExtension(std::string_view path); +bool hasSupported7zExtension(std::string_view path); + +bool hasSupportedArchiveExtension(std::string_view path); + + +} diff --git a/src/IO/Archives/createArchiveReader.cpp b/src/IO/Archives/createArchiveReader.cpp index 782602091ac..dfa098eede0 100644 --- a/src/IO/Archives/createArchiveReader.cpp +++ b/src/IO/Archives/createArchiveReader.cpp @@ -1,6 +1,7 @@ #include #include #include +#include #include @@ -12,7 +13,6 @@ extern const int CANNOT_UNPACK_ARCHIVE; extern const int SUPPORT_IS_DISABLED; } - std::shared_ptr createArchiveReader(const String & path_to_archive) { return createArchiveReader(path_to_archive, {}, 0); @@ -24,11 +24,7 @@ std::shared_ptr createArchiveReader( [[maybe_unused]] const std::function()> & archive_read_function, [[maybe_unused]] size_t archive_size) { - using namespace std::literals; - static constexpr std::array tar_extensions{ - ".tar"sv, ".tar.gz"sv, ".tgz"sv, ".tar.zst"sv, ".tzst"sv, ".tar.xz"sv, ".tar.bz2"sv, ".tar.lzma"sv}; - - if (path_to_archive.ends_with(".zip") || path_to_archive.ends_with(".zipx")) + if (hasSupportedZipExtension(path_to_archive)) { #if USE_MINIZIP return std::make_shared(path_to_archive, archive_read_function, archive_size); @@ -36,8 +32,7 @@ std::shared_ptr createArchiveReader( throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, "minizip library is disabled"); #endif } - else if (std::any_of( - tar_extensions.begin(), tar_extensions.end(), [&](const auto extension) { return path_to_archive.ends_with(extension); })) + else if (hasSupportedTarExtension(path_to_archive)) { #if USE_LIBARCHIVE return std::make_shared(path_to_archive, archive_read_function); @@ -45,7 +40,7 @@ std::shared_ptr createArchiveReader( throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, "libarchive library is disabled"); #endif } - else if (path_to_archive.ends_with(".7z")) + else if (hasSupported7zExtension(path_to_archive)) { #if USE_LIBARCHIVE return std::make_shared(path_to_archive); diff --git a/src/IO/Archives/createArchiveWriter.cpp b/src/IO/Archives/createArchiveWriter.cpp index 9a169587088..53be0a85a10 100644 --- a/src/IO/Archives/createArchiveWriter.cpp +++ b/src/IO/Archives/createArchiveWriter.cpp @@ -1,3 +1,4 @@ +#include #include #include #include @@ -24,10 +25,7 @@ std::shared_ptr createArchiveWriter(const String & path_to_archi std::shared_ptr createArchiveWriter(const String & path_to_archive, [[maybe_unused]] std::unique_ptr archive_write_buffer) { - using namespace std::literals; - static constexpr std::array tar_extensions{ - ".tar"sv, ".tar.gz"sv, ".tgz"sv, ".tar.bz2"sv, ".tar.lzma"sv, ".tar.zst"sv, ".tzst"sv, ".tar.xz"sv}; - if (path_to_archive.ends_with(".zip") || path_to_archive.ends_with(".zipx")) + if (hasSupportedZipExtension(path_to_archive)) { #if USE_MINIZIP return std::make_shared(path_to_archive, std::move(archive_write_buffer)); @@ -35,8 +33,7 @@ createArchiveWriter(const String & path_to_archive, [[maybe_unused]] std::unique throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, "minizip library is disabled"); #endif } - else if (std::any_of( - tar_extensions.begin(), tar_extensions.end(), [&](const auto extension) { return path_to_archive.ends_with(extension); })) + else if (hasSupportedTarExtension(path_to_archive)) { #if USE_LIBARCHIVE return std::make_shared(path_to_archive, std::move(archive_write_buffer)); diff --git a/src/IO/S3/URI.cpp b/src/IO/S3/URI.cpp index 4bf7a3ddf86..b9c400d2b98 100644 --- a/src/IO/S3/URI.cpp +++ b/src/IO/S3/URI.cpp @@ -6,6 +6,7 @@ #include #include #include +#include #include @@ -54,10 +55,7 @@ URI::URI(const std::string & uri_) static constexpr auto OSS = "OSS"; static constexpr auto EOS = "EOS"; - if (containsArchive(uri_)) - std::tie(uri_str, archive_pattern) = getPathToArchiveAndArchivePattern(uri_); - else - uri_str = uri_; + std::tie(uri_str, archive_pattern) = getURIAndArchivePattern(uri_); uri = Poco::URI(uri_str); std::unordered_map mapper; @@ -167,32 +165,29 @@ void URI::validateBucket(const String & bucket, const Poco::URI & uri) !uri.empty() ? " (" + uri.toString() + ")" : ""); } -bool URI::containsArchive(const std::string & source) +std::pair> URI::getURIAndArchivePattern(const std::string & source) { size_t pos = source.find("::"); - return (pos != std::string::npos); -} + if (pos == String::npos) + return {source, std::nullopt}; -std::pair URI::getPathToArchiveAndArchivePattern(const std::string & source) -{ - size_t pos = source.find("::"); - assert(pos != std::string::npos); + std::string_view path_to_archive_view = std::string_view{source}.substr(0, pos); + while (path_to_archive_view.ends_with(' ')) + path_to_archive_view.remove_suffix(1); - std::string path_to_archive = source.substr(0, pos); - while ((!path_to_archive.empty()) && path_to_archive.ends_with(' ')) - path_to_archive.pop_back(); + if (path_to_archive_view.empty() || !hasSupportedArchiveExtension(path_to_archive_view)) + return {source, std::nullopt}; - if (path_to_archive.empty()) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Path to archive is empty"); + auto archive_uri = path_to_archive_view; - std::string_view path_in_archive_view = std::string_view{source}.substr(pos + 2); - while (path_in_archive_view.front() == ' ') - path_in_archive_view.remove_prefix(1); + std::string_view archive_pattern_view = std::string_view{source}.substr(pos + 2); + while (archive_pattern_view.front() == ' ') + archive_pattern_view.remove_prefix(1); - if (path_in_archive_view.empty()) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Filename is empty"); + if (archive_pattern_view.empty()) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Archive pattern is empty"); - return {path_to_archive, std::string{path_in_archive_view}}; + return std::pair{std::string{archive_uri}, std::string{archive_pattern_view}}; } } diff --git a/src/IO/S3/URI.h b/src/IO/S3/URI.h index 363f98c46f5..e4bb0d9eae1 100644 --- a/src/IO/S3/URI.h +++ b/src/IO/S3/URI.h @@ -42,8 +42,7 @@ struct URI static void validateBucket(const std::string & bucket, const Poco::URI & uri); private: - bool containsArchive(const std::string & source); - std::pair getPathToArchiveAndArchivePattern(const std::string & source); + std::pair> getURIAndArchivePattern(const std::string & source); }; } diff --git a/src/Storages/StorageFile.cpp b/src/Storages/StorageFile.cpp index 98cd5c4dfa9..de56fcf66a0 100644 --- a/src/Storages/StorageFile.cpp +++ b/src/Storages/StorageFile.cpp @@ -25,6 +25,7 @@ #include #include #include +#include #include #include #include @@ -2247,8 +2248,11 @@ void StorageFile::parseFileSource(String source, String & filename, String & pat while (path_to_archive_view.ends_with(' ')) path_to_archive_view.remove_suffix(1); - if (path_to_archive_view.empty()) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Path to archive is empty"); + if (path_to_archive_view.empty() || !hasSupportedArchiveExtension(path_to_archive_view)) + { + filename = std::move(source); + return; + } path_to_archive = path_to_archive_view; diff --git a/src/TableFunctions/TableFunctionFile.cpp b/src/TableFunctions/TableFunctionFile.cpp index 1b6d86f8fa5..12b88ae2b14 100644 --- a/src/TableFunctions/TableFunctionFile.cpp +++ b/src/TableFunctions/TableFunctionFile.cpp @@ -63,9 +63,12 @@ std::optional TableFunctionFile::tryGetFormatFromFirstArgument() return FormatFactory::instance().tryGetFormatFromFileName(filename); } -StoragePtr TableFunctionFile::getStorage(const String & source, - const String & format_, const ColumnsDescription & columns, - ContextPtr global_context, const std::string & table_name, +StoragePtr TableFunctionFile::getStorage( + const String & source, + const String & format_, + const ColumnsDescription & columns, + ContextPtr global_context, + const std::string & table_name, const std::string & compression_method_) const { // For `file` table function, we are going to use format settings from the diff --git a/tests/queries/0_stateless/03214_parsing_archive_name_file.reference b/tests/queries/0_stateless/03214_parsing_archive_name_file.reference new file mode 100644 index 00000000000..243a7c8fd02 --- /dev/null +++ b/tests/queries/0_stateless/03214_parsing_archive_name_file.reference @@ -0,0 +1,12 @@ +::nonexistentfile.csv +1 +nonexistent::nonexistentfile.csv +1 +nonexistent :: nonexistentfile.csv +1 +nonexistent ::nonexistentfile.csv +1 +nonexistent.tar.gz :: nonexistentfile.csv +1 +nonexistent.zip:: nonexistentfile.csv +1 diff --git a/tests/queries/0_stateless/03214_parsing_archive_name_file.sh b/tests/queries/0_stateless/03214_parsing_archive_name_file.sh new file mode 100755 index 00000000000..32bf3246c84 --- /dev/null +++ b/tests/queries/0_stateless/03214_parsing_archive_name_file.sh @@ -0,0 +1,21 @@ +#!/usr/bin/env bash + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +function try_to_read_file() +{ + file_to_read=$1 + file_argument=$2 + + echo $file_argument + $CLICKHOUSE_LOCAL -q "SELECT * FROM file('$file_argument')" 2>&1 | rg -c "Cannot stat file.*$file_to_read" +} + +try_to_read_file "::nonexistentfile.csv" "::nonexistentfile.csv" +try_to_read_file "nonexistent::nonexistentfile.csv" "nonexistent::nonexistentfile.csv" +try_to_read_file "nonexistent :: nonexistentfile.csv" "nonexistent :: nonexistentfile.csv" +try_to_read_file "nonexistent ::nonexistentfile.csv" "nonexistent ::nonexistentfile.csv" +try_to_read_file "nonexistent.tar.gz" "nonexistent.tar.gz :: nonexistentfile.csv" +try_to_read_file "nonexistent.zip" "nonexistent.zip:: nonexistentfile.csv" diff --git a/tests/queries/0_stateless/03215_parsing_archive_name_s3.reference b/tests/queries/0_stateless/03215_parsing_archive_name_s3.reference new file mode 100644 index 00000000000..9dd925a7480 --- /dev/null +++ b/tests/queries/0_stateless/03215_parsing_archive_name_s3.reference @@ -0,0 +1,2 @@ +::03215_archive.csv test/::03215_archive.csv +test::03215_archive.csv test/test::03215_archive.csv diff --git a/tests/queries/0_stateless/03215_parsing_archive_name_s3.sql b/tests/queries/0_stateless/03215_parsing_archive_name_s3.sql new file mode 100644 index 00000000000..9d01f53c838 --- /dev/null +++ b/tests/queries/0_stateless/03215_parsing_archive_name_s3.sql @@ -0,0 +1,6 @@ +-- Tags: no-fasttest +-- Tag no-fasttest: Depends on AWS + +SELECT _file, _path FROM s3(s3_conn, filename='::03215_archive.csv') ORDER BY (_file, _path); +SELECT _file, _path FROM s3(s3_conn, filename='test :: 03215_archive.csv') ORDER BY (_file, _path); -- { serverError STD_EXCEPTION } +SELECT _file, _path FROM s3(s3_conn, filename='test::03215_archive.csv') ORDER BY (_file, _path); diff --git a/tests/queries/0_stateless/data_minio/::03215_archive.csv b/tests/queries/0_stateless/data_minio/::03215_archive.csv new file mode 100644 index 00000000000..d00491fd7e5 --- /dev/null +++ b/tests/queries/0_stateless/data_minio/::03215_archive.csv @@ -0,0 +1 @@ +1 diff --git a/tests/queries/0_stateless/data_minio/test :: 03215_archive.csv b/tests/queries/0_stateless/data_minio/test :: 03215_archive.csv new file mode 100644 index 00000000000..d00491fd7e5 --- /dev/null +++ b/tests/queries/0_stateless/data_minio/test :: 03215_archive.csv @@ -0,0 +1 @@ +1 diff --git a/tests/queries/0_stateless/data_minio/test::03215_archive.csv b/tests/queries/0_stateless/data_minio/test::03215_archive.csv new file mode 100644 index 00000000000..d00491fd7e5 --- /dev/null +++ b/tests/queries/0_stateless/data_minio/test::03215_archive.csv @@ -0,0 +1 @@ +1 From 4d4fc8fd6f0123613305423d861429f54222d23f Mon Sep 17 00:00:00 2001 From: Antonio Andelic Date: Tue, 30 Jul 2024 11:57:37 +0200 Subject: [PATCH 049/227] Add setting to disable archive path syntax --- src/Core/Settings.h | 1 + src/Core/SettingsChangesHistory.cpp | 1 + src/IO/S3/URI.cpp | 8 ++++++-- src/IO/S3/URI.h | 2 +- src/Storages/ObjectStorage/S3/Configuration.cpp | 8 ++++---- src/Storages/StorageFile.cpp | 14 ++++++++++++-- src/Storages/StorageFile.h | 2 +- src/TableFunctions/TableFunctionFile.cpp | 5 +++-- .../03214_parsing_archive_name_file.reference | 4 ++++ .../0_stateless/03214_parsing_archive_name_file.sh | 10 ++++++++-- .../03215_parsing_archive_name_s3.reference | 1 + .../0_stateless/03215_parsing_archive_name_s3.sql | 1 + .../data_minio/test.zip::03215_archive.csv | 1 + 13 files changed, 44 insertions(+), 14 deletions(-) create mode 100644 tests/queries/0_stateless/data_minio/test.zip::03215_archive.csv diff --git a/src/Core/Settings.h b/src/Core/Settings.h index 4fc2034b855..5114a8204cd 100644 --- a/src/Core/Settings.h +++ b/src/Core/Settings.h @@ -936,6 +936,7 @@ class IColumn; M(UInt64, parallel_replicas_min_number_of_rows_per_replica, 0, "Limit the number of replicas used in a query to (estimated rows to read / min_number_of_rows_per_replica). The max is still limited by 'max_parallel_replicas'", 0) \ M(Bool, parallel_replicas_prefer_local_join, true, "If true, and JOIN can be executed with parallel replicas algorithm, and all storages of right JOIN part are *MergeTree, local JOIN will be used instead of GLOBAL JOIN.", 0) \ M(UInt64, parallel_replicas_mark_segment_size, 128, "Parts virtually divided into segments to be distributed between replicas for parallel reading. This setting controls the size of these segments. Not recommended to change until you're absolutely sure in what you're doing", 0) \ + M(Bool, allow_archive_path_syntax, true, "File/S3 engines/table function will parse paths with '::' as ' :: ' if archive has correct extension", 0) \ \ M(Bool, allow_experimental_inverted_index, false, "If it is set to true, allow to use experimental inverted index.", 0) \ M(Bool, allow_experimental_full_text_index, false, "If it is set to true, allow to use experimental full-text index.", 0) \ diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index 9faf77e9087..8483a267237 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -124,6 +124,7 @@ static std::initializer_list mapper; diff --git a/src/IO/S3/URI.h b/src/IO/S3/URI.h index e4bb0d9eae1..80e2da96cd4 100644 --- a/src/IO/S3/URI.h +++ b/src/IO/S3/URI.h @@ -36,7 +36,7 @@ struct URI bool is_virtual_hosted_style; URI() = default; - explicit URI(const std::string & uri_); + explicit URI(const std::string & uri_, bool allow_archive_path_syntax = false); void addRegionToURI(const std::string & region); static void validateBucket(const std::string & bucket, const Poco::URI & uri); diff --git a/src/Storages/ObjectStorage/S3/Configuration.cpp b/src/Storages/ObjectStorage/S3/Configuration.cpp index 094ca069e7a..7542f59dcc4 100644 --- a/src/Storages/ObjectStorage/S3/Configuration.cpp +++ b/src/Storages/ObjectStorage/S3/Configuration.cpp @@ -142,14 +142,14 @@ ObjectStoragePtr StorageS3Configuration::createObjectStorage(ContextPtr context, void StorageS3Configuration::fromNamedCollection(const NamedCollection & collection, ContextPtr context) { - const auto settings = context->getSettingsRef(); + const auto & settings = context->getSettingsRef(); validateNamedCollection(collection, required_configuration_keys, optional_configuration_keys); auto filename = collection.getOrDefault("filename", ""); if (!filename.empty()) - url = S3::URI(std::filesystem::path(collection.get("url")) / filename); + url = S3::URI(std::filesystem::path(collection.get("url")) / filename, settings.allow_archive_path_syntax); else - url = S3::URI(collection.get("url")); + url = S3::URI(collection.get("url"), settings.allow_archive_path_syntax); auth_settings.access_key_id = collection.getOrDefault("access_key_id", ""); auth_settings.secret_access_key = collection.getOrDefault("secret_access_key", ""); @@ -330,7 +330,7 @@ void StorageS3Configuration::fromAST(ASTs & args, ContextPtr context, bool with_ } /// This argument is always the first - url = S3::URI(checkAndGetLiteralArgument(args[0], "url")); + url = S3::URI(checkAndGetLiteralArgument(args[0], "url"), context->getSettingsRef().allow_archive_path_syntax); if (engine_args_to_idx.contains("format")) { diff --git a/src/Storages/StorageFile.cpp b/src/Storages/StorageFile.cpp index de56fcf66a0..efb39f90053 100644 --- a/src/Storages/StorageFile.cpp +++ b/src/Storages/StorageFile.cpp @@ -2208,7 +2208,11 @@ void registerStorageFile(StorageFactory & factory) else if (type == Field::Types::UInt64) source_fd = static_cast(literal->value.get()); else if (type == Field::Types::String) - StorageFile::parseFileSource(literal->value.get(), source_path, storage_args.path_to_archive); + StorageFile::parseFileSource( + literal->value.get(), + source_path, + storage_args.path_to_archive, + factory_args.getLocalContext()->getSettingsRef().allow_archive_path_syntax); else throw Exception(ErrorCodes::BAD_ARGUMENTS, "Second argument must be path or file descriptor"); } @@ -2235,8 +2239,14 @@ SchemaCache & StorageFile::getSchemaCache(const ContextPtr & context) return schema_cache; } -void StorageFile::parseFileSource(String source, String & filename, String & path_to_archive) +void StorageFile::parseFileSource(String source, String & filename, String & path_to_archive, bool allow_archive_path_syntax) { + if (!allow_archive_path_syntax) + { + filename = std::move(source); + return; + } + size_t pos = source.find("::"); if (pos == String::npos) { diff --git a/src/Storages/StorageFile.h b/src/Storages/StorageFile.h index 895a8a663b8..bb969c1877c 100644 --- a/src/Storages/StorageFile.h +++ b/src/Storages/StorageFile.h @@ -128,7 +128,7 @@ public: static SchemaCache & getSchemaCache(const ContextPtr & context); - static void parseFileSource(String source, String & filename, String & path_to_archive); + static void parseFileSource(String source, String & filename, String & path_to_archive, bool allow_archive_path_syntax); static ArchiveInfo getArchiveInfo( const std::string & path_to_archive, diff --git a/src/TableFunctions/TableFunctionFile.cpp b/src/TableFunctions/TableFunctionFile.cpp index 12b88ae2b14..af327cfe54e 100644 --- a/src/TableFunctions/TableFunctionFile.cpp +++ b/src/TableFunctions/TableFunctionFile.cpp @@ -26,7 +26,7 @@ void TableFunctionFile::parseFirstArguments(const ASTPtr & arg, const ContextPtr if (context->getApplicationType() != Context::ApplicationType::LOCAL) { ITableFunctionFileLike::parseFirstArguments(arg, context); - StorageFile::parseFileSource(std::move(filename), filename, path_to_archive); + StorageFile::parseFileSource(std::move(filename), filename, path_to_archive, context->getSettingsRef().allow_archive_path_syntax); return; } @@ -42,7 +42,8 @@ void TableFunctionFile::parseFirstArguments(const ASTPtr & arg, const ContextPtr else if (filename == "stderr") fd = STDERR_FILENO; else - StorageFile::parseFileSource(std::move(filename), filename, path_to_archive); + StorageFile::parseFileSource( + std::move(filename), filename, path_to_archive, context->getSettingsRef().allow_archive_path_syntax); } else if (type == Field::Types::Int64 || type == Field::Types::UInt64) { diff --git a/tests/queries/0_stateless/03214_parsing_archive_name_file.reference b/tests/queries/0_stateless/03214_parsing_archive_name_file.reference index 243a7c8fd02..d793d26dfc3 100644 --- a/tests/queries/0_stateless/03214_parsing_archive_name_file.reference +++ b/tests/queries/0_stateless/03214_parsing_archive_name_file.reference @@ -10,3 +10,7 @@ nonexistent.tar.gz :: nonexistentfile.csv 1 nonexistent.zip:: nonexistentfile.csv 1 +nonexistent.tar.gz :: nonexistentfile.csv SETTINGS allow_archive_path_syntax=0 +1 +nonexistent.zip:: nonexistentfile.csv SETTINGS allow_archive_path_syntax=0 +1 diff --git a/tests/queries/0_stateless/03214_parsing_archive_name_file.sh b/tests/queries/0_stateless/03214_parsing_archive_name_file.sh index 32bf3246c84..2f77627f6be 100755 --- a/tests/queries/0_stateless/03214_parsing_archive_name_file.sh +++ b/tests/queries/0_stateless/03214_parsing_archive_name_file.sh @@ -8,14 +8,20 @@ function try_to_read_file() { file_to_read=$1 file_argument=$2 + settings=$3 - echo $file_argument - $CLICKHOUSE_LOCAL -q "SELECT * FROM file('$file_argument')" 2>&1 | rg -c "Cannot stat file.*$file_to_read" + echo $file_argument $settings + $CLICKHOUSE_LOCAL -q "SELECT * FROM file('$file_argument') $settings" 2>&1 | rg -c "Cannot stat file.*$file_to_read" } +# if archive extension is not detected for part before '::', path is taken as is try_to_read_file "::nonexistentfile.csv" "::nonexistentfile.csv" try_to_read_file "nonexistent::nonexistentfile.csv" "nonexistent::nonexistentfile.csv" try_to_read_file "nonexistent :: nonexistentfile.csv" "nonexistent :: nonexistentfile.csv" try_to_read_file "nonexistent ::nonexistentfile.csv" "nonexistent ::nonexistentfile.csv" +# if archive extension is detected for part before '::', path is split into archive and filename try_to_read_file "nonexistent.tar.gz" "nonexistent.tar.gz :: nonexistentfile.csv" try_to_read_file "nonexistent.zip" "nonexistent.zip:: nonexistentfile.csv" +# disabling archive syntax will always parse path as is +try_to_read_file "nonexistent.tar.gz :: nonexistentfile.csv" "nonexistent.tar.gz :: nonexistentfile.csv" "SETTINGS allow_archive_path_syntax=0" +try_to_read_file "nonexistent.zip:: nonexistentfile.csv" "nonexistent.zip:: nonexistentfile.csv" "SETTINGS allow_archive_path_syntax=0" diff --git a/tests/queries/0_stateless/03215_parsing_archive_name_s3.reference b/tests/queries/0_stateless/03215_parsing_archive_name_s3.reference index 9dd925a7480..b4804c82dc2 100644 --- a/tests/queries/0_stateless/03215_parsing_archive_name_s3.reference +++ b/tests/queries/0_stateless/03215_parsing_archive_name_s3.reference @@ -1,2 +1,3 @@ ::03215_archive.csv test/::03215_archive.csv test::03215_archive.csv test/test::03215_archive.csv +test.zip::03215_archive.csv test/test.zip::03215_archive.csv diff --git a/tests/queries/0_stateless/03215_parsing_archive_name_s3.sql b/tests/queries/0_stateless/03215_parsing_archive_name_s3.sql index 9d01f53c838..3a7ed0b864c 100644 --- a/tests/queries/0_stateless/03215_parsing_archive_name_s3.sql +++ b/tests/queries/0_stateless/03215_parsing_archive_name_s3.sql @@ -4,3 +4,4 @@ SELECT _file, _path FROM s3(s3_conn, filename='::03215_archive.csv') ORDER BY (_file, _path); SELECT _file, _path FROM s3(s3_conn, filename='test :: 03215_archive.csv') ORDER BY (_file, _path); -- { serverError STD_EXCEPTION } SELECT _file, _path FROM s3(s3_conn, filename='test::03215_archive.csv') ORDER BY (_file, _path); +SELECT _file, _path FROM s3(s3_conn, filename='test.zip::03215_archive.csv') ORDER BY (_file, _path) SETTINGS allow_archive_path_syntax=0; diff --git a/tests/queries/0_stateless/data_minio/test.zip::03215_archive.csv b/tests/queries/0_stateless/data_minio/test.zip::03215_archive.csv new file mode 100644 index 00000000000..d00491fd7e5 --- /dev/null +++ b/tests/queries/0_stateless/data_minio/test.zip::03215_archive.csv @@ -0,0 +1 @@ +1 From fb466287dac16801518547f34b42edbc16a57fae Mon Sep 17 00:00:00 2001 From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com> Date: Tue, 30 Jul 2024 12:13:47 +0200 Subject: [PATCH 050/227] Update 02150_index_hypothesis_race_long.sh --- tests/queries/0_stateless/02150_index_hypothesis_race_long.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/02150_index_hypothesis_race_long.sh b/tests/queries/0_stateless/02150_index_hypothesis_race_long.sh index be7cfa78492..c29b604d23d 100755 --- a/tests/queries/0_stateless/02150_index_hypothesis_race_long.sh +++ b/tests/queries/0_stateless/02150_index_hypothesis_race_long.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-distributed-cache +# Tags: long, no-distributed-cache CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh From 2530c5eb41a759baded5380a4e697c2e884c0abd Mon Sep 17 00:00:00 2001 From: Antonio Andelic Date: Tue, 30 Jul 2024 12:37:35 +0200 Subject: [PATCH 051/227] Fix tests --- tests/queries/0_stateless/02952_archive_parsing.reference | 0 tests/queries/0_stateless/02952_archive_parsing.sql | 1 - tests/queries/0_stateless/03214_parsing_archive_name_file.sh | 2 +- 3 files changed, 1 insertion(+), 2 deletions(-) delete mode 100644 tests/queries/0_stateless/02952_archive_parsing.reference delete mode 100644 tests/queries/0_stateless/02952_archive_parsing.sql diff --git a/tests/queries/0_stateless/02952_archive_parsing.reference b/tests/queries/0_stateless/02952_archive_parsing.reference deleted file mode 100644 index e69de29bb2d..00000000000 diff --git a/tests/queries/0_stateless/02952_archive_parsing.sql b/tests/queries/0_stateless/02952_archive_parsing.sql deleted file mode 100644 index 49b0223e6ec..00000000000 --- a/tests/queries/0_stateless/02952_archive_parsing.sql +++ /dev/null @@ -1 +0,0 @@ -SELECT * FROM file('::a'); -- { serverError BAD_ARGUMENTS } diff --git a/tests/queries/0_stateless/03214_parsing_archive_name_file.sh b/tests/queries/0_stateless/03214_parsing_archive_name_file.sh index 2f77627f6be..b54cbb10aa6 100755 --- a/tests/queries/0_stateless/03214_parsing_archive_name_file.sh +++ b/tests/queries/0_stateless/03214_parsing_archive_name_file.sh @@ -11,7 +11,7 @@ function try_to_read_file() settings=$3 echo $file_argument $settings - $CLICKHOUSE_LOCAL -q "SELECT * FROM file('$file_argument') $settings" 2>&1 | rg -c "Cannot stat file.*$file_to_read" + $CLICKHOUSE_LOCAL -q "SELECT * FROM file('$file_argument') $settings" 2>&1 | grep -c "Cannot stat file.*$file_to_read" } # if archive extension is not detected for part before '::', path is taken as is From 27a15bc5ace68acb0ba62791ca6e3d3f17ae569c Mon Sep 17 00:00:00 2001 From: avogar Date: Tue, 30 Jul 2024 10:39:35 +0000 Subject: [PATCH 052/227] Remove sh files --- .../03036_dynamic_read_subcolumns_1.sh | 19 ------------------- .../03036_dynamic_read_subcolumns_2.sh | 19 ------------------- .../03036_dynamic_read_subcolumns_3.sh | 19 ------------------- 3 files changed, 57 deletions(-) delete mode 100755 tests/queries/0_stateless/03036_dynamic_read_subcolumns_1.sh delete mode 100755 tests/queries/0_stateless/03036_dynamic_read_subcolumns_2.sh delete mode 100755 tests/queries/0_stateless/03036_dynamic_read_subcolumns_3.sh diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_1.sh b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_1.sh deleted file mode 100755 index ed548ae74e9..00000000000 --- a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_1.sh +++ /dev/null @@ -1,19 +0,0 @@ -#!/usr/bin/env bash -# Tags: long - -CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) -# shellcheck source=../shell_config.sh -. "$CUR_DIR"/../shell_config.sh - - -# shellcheck source=./03036_dynamic_read_subcolumns.lib -. "$CUR_DIR"/03036_dynamic_read_subcolumns.lib - -CH_CLIENT="$CLICKHOUSE_CLIENT --allow_experimental_variant_type=1 --use_variant_as_common_type=1 --allow_experimental_dynamic_type=1" - -$CH_CLIENT -q "drop table if exists test;" - -echo "Memory" -$CH_CLIENT -q "create table test (id UInt64, d Dynamic) engine=Memory" -test -$CH_CLIENT -q "drop table test;" diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_2.sh b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_2.sh deleted file mode 100755 index 95dafcf5832..00000000000 --- a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_2.sh +++ /dev/null @@ -1,19 +0,0 @@ -#!/usr/bin/env bash -# Tags: long - -CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) -# shellcheck source=../shell_config.sh -. "$CUR_DIR"/../shell_config.sh - - -# shellcheck source=./03036_dynamic_read_subcolumns.lib -. "$CUR_DIR"/03036_dynamic_read_subcolumns.lib - -CH_CLIENT="$CLICKHOUSE_CLIENT --allow_experimental_variant_type=1 --use_variant_as_common_type=1 --allow_experimental_dynamic_type=1" - -$CH_CLIENT -q "drop table if exists test;" - -echo "MergeTree compact" -$CH_CLIENT -q "create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000;" -test -$CH_CLIENT -q "drop table test;" diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_3.sh b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_3.sh deleted file mode 100755 index a3c2d93e568..00000000000 --- a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_3.sh +++ /dev/null @@ -1,19 +0,0 @@ -#!/usr/bin/env bash -# Tags: long - -CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) -# shellcheck source=../shell_config.sh -. "$CUR_DIR"/../shell_config.sh - - -# shellcheck source=./03036_dynamic_read_subcolumns.lib -. "$CUR_DIR"/03036_dynamic_read_subcolumns.lib - -CH_CLIENT="$CLICKHOUSE_CLIENT --allow_experimental_variant_type=1 --use_variant_as_common_type=1 --allow_experimental_dynamic_type=1" - -$CH_CLIENT -q "drop table if exists test;" - -echo "MergeTree wide" -$CH_CLIENT -q "create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1;" -test -$CH_CLIENT -q "drop table test;" From 6372fdee6d344bd87d58ce89fa069b55750c9aba Mon Sep 17 00:00:00 2001 From: kssenii Date: Tue, 30 Jul 2024 13:46:05 +0200 Subject: [PATCH 053/227] Update tests --- tests/queries/0_stateless/02995_index_3.sh | 2 +- tests/queries/0_stateless/03143_asof_join_ddb_long.sql | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/queries/0_stateless/02995_index_3.sh b/tests/queries/0_stateless/02995_index_3.sh index 506429e2696..219ae81154f 100755 --- a/tests/queries/0_stateless/02995_index_3.sh +++ b/tests/queries/0_stateless/02995_index_3.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage +# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage, no-distributed-cache CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/03143_asof_join_ddb_long.sql b/tests/queries/0_stateless/03143_asof_join_ddb_long.sql index 17a67511030..a927e4f1e1f 100644 --- a/tests/queries/0_stateless/03143_asof_join_ddb_long.sql +++ b/tests/queries/0_stateless/03143_asof_join_ddb_long.sql @@ -1,4 +1,4 @@ --- Tags: long +-- Tags: long, no-distributed-cache DROP TABLE IF EXISTS build; DROP TABLE IF EXISTS skewed_probe; From a1ececb24c0a6f21d3985f5a9f8a726befef78c3 Mon Sep 17 00:00:00 2001 From: avogar Date: Tue, 30 Jul 2024 11:51:48 +0000 Subject: [PATCH 054/227] Fix use-of-unitialized-value --- src/Columns/ColumnAggregateFunction.cpp | 33 ++++++++++++++++++++++--- 1 file changed, 30 insertions(+), 3 deletions(-) diff --git a/src/Columns/ColumnAggregateFunction.cpp b/src/Columns/ColumnAggregateFunction.cpp index 9934970c868..4bc48c62eb4 100644 --- a/src/Columns/ColumnAggregateFunction.cpp +++ b/src/Columns/ColumnAggregateFunction.cpp @@ -332,9 +332,36 @@ void ColumnAggregateFunction::expand(const Filter & mask, bool inverted) { ensureOwnership(); Arena & arena = createOrGetArena(); - char * default_ptr = arena.alignedAlloc(func->sizeOfData(), func->alignOfData()); - func->create(default_ptr); - expandDataByMask(data, mask, inverted, default_ptr); + + if (mask.size() < data.size()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Mask size should be no less than data size."); + + ssize_t from = data.size() - 1; + ssize_t index = mask.size() - 1; + data.resize(mask.size()); + while (index >= 0) + { + if (!!mask[index] ^ inverted) + { + if (from < 0) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Too many bytes in mask"); + + /// Copy only if it makes sense. + if (index != from) + data[index] = data[from]; + --from; + } + else + { + data[index] = arena.alignedAlloc(func->sizeOfData(), func->alignOfData()); + func->create(data[index]); + } + + --index; + } + + if (from != -1) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Not enough bytes in mask"); } ColumnPtr ColumnAggregateFunction::permute(const Permutation & perm, size_t limit) const From 8b52d7b711d54f1d4bb5b2f39bf4aea3966f64dc Mon Sep 17 00:00:00 2001 From: Michael Stetsyuk Date: Tue, 30 Jul 2024 13:35:19 +0100 Subject: [PATCH 055/227] fxs --- src/IO/S3/Client.cpp | 19 +++++++++---------- src/IO/S3/Client.h | 2 +- .../test_checking_s3_blobs_paranoid/test.py | 12 ++++++++++++ tests/integration/test_storage_delta/test.py | 12 ++++++++++++ 4 files changed, 34 insertions(+), 11 deletions(-) diff --git a/src/IO/S3/Client.cpp b/src/IO/S3/Client.cpp index 7196dfa9bdc..8f037ea71be 100644 --- a/src/IO/S3/Client.cpp +++ b/src/IO/S3/Client.cpp @@ -385,10 +385,9 @@ Model::HeadObjectOutcome Client::HeadObject(HeadObjectRequest & request) const request.overrideURI(std::move(*bucket_uri)); - if (isClientForDisk() && error.GetErrorType() == Aws::S3::S3Errors::NO_SUCH_KEY) - CurrentMetrics::add(CurrentMetrics::S3DiskNoKeyErrors); - - return enrichErrorMessage( + /// The next call is NOT a recurcive call + /// This is a virtuall call Aws::S3::S3Client::HeadObject(const Model::HeadObjectRequest&) + return processRequestResult( HeadObject(static_cast(request))); } @@ -409,11 +408,8 @@ Model::ListObjectsOutcome Client::ListObjects(ListObjectsRequest & request) cons Model::GetObjectOutcome Client::GetObject(GetObjectRequest & request) const { - auto resp = doRequest(request, [this](const Model::GetObjectRequest & req) { return GetObject(req); }); - if (!resp.IsSuccess() && isClientForDisk() && resp.GetError().GetErrorType() == Aws::S3::S3Errors::NO_SUCH_KEY) - CurrentMetrics::add(CurrentMetrics::S3DiskNoKeyErrors); - - return enrichErrorMessage(std::move(resp)); + return processRequestResult( + doRequest(request, [this](const Model::GetObjectRequest & req) { return GetObject(req); })); } Model::AbortMultipartUploadOutcome Client::AbortMultipartUpload(AbortMultipartUploadRequest & request) const @@ -699,11 +695,14 @@ Client::doRequestWithRetryNetworkErrors(RequestType & request, RequestFn request } template -RequestResult Client::enrichErrorMessage(RequestResult && outcome) const +RequestResult Client::processRequestResult(RequestResult && outcome) const { if (outcome.IsSuccess() || !isClientForDisk()) return std::forward(outcome); + if (outcome.GetError().GetErrorType() == Aws::S3::S3Errors::NO_SUCH_KEY) + CurrentMetrics::add(CurrentMetrics::S3DiskNoKeyErrors); + String enriched_message = fmt::format( "{} {}", outcome.GetError().GetMessage(), diff --git a/src/IO/S3/Client.h b/src/IO/S3/Client.h index 11cace4e1fd..e54953419e1 100644 --- a/src/IO/S3/Client.h +++ b/src/IO/S3/Client.h @@ -274,7 +274,7 @@ private: void insertRegionOverride(const std::string & bucket, const std::string & region) const; template - RequestResult enrichErrorMessage(RequestResult && outcome) const; + RequestResult processRequestResult(RequestResult && outcome) const; String initial_endpoint; std::shared_ptr credentials_provider; diff --git a/tests/integration/test_checking_s3_blobs_paranoid/test.py b/tests/integration/test_checking_s3_blobs_paranoid/test.py index 1ed70e20b79..dde636b5d29 100644 --- a/tests/integration/test_checking_s3_blobs_paranoid/test.py +++ b/tests/integration/test_checking_s3_blobs_paranoid/test.py @@ -689,3 +689,15 @@ def test_no_key_found_disk(cluster, broken_s3): "DB::Exception: The specified key does not exist. This error happened for S3 disk." in error ) + + s3_disk_no_key_errors_metric_value = int( + node.query( + """ + SELECT value + FROM system.metrics + WHERE metric = 'S3DiskNoKeyErrors' + """ + ).strip() + ) + + assert s3_disk_no_key_errors_metric_value > 0 diff --git a/tests/integration/test_storage_delta/test.py b/tests/integration/test_storage_delta/test.py index d3dd7cfe52a..67cc7cdd6da 100644 --- a/tests/integration/test_storage_delta/test.py +++ b/tests/integration/test_storage_delta/test.py @@ -452,6 +452,18 @@ def test_restart_broken(started_cluster): f"SELECT count() FROM {TABLE_NAME}" ) + s3_disk_no_key_errors_metric_value = int( + instance.query( + """ + SELECT value + FROM system.metrics + WHERE metric = 'S3DiskNoKeyErrors' + """ + ).strip() + ) + + assert s3_disk_no_key_errors_metric_value == 0 + minio_client.make_bucket(bucket) upload_directory(minio_client, bucket, f"/{TABLE_NAME}", "") From d69f6cccde7633214eba48c08d7647e4ea9a40da Mon Sep 17 00:00:00 2001 From: kssenii Date: Tue, 30 Jul 2024 15:08:26 +0200 Subject: [PATCH 056/227] Fix --- src/Common/ThreadPoolTaskTracker.cpp | 5 ++++- src/IO/WriteBufferFromS3.cpp | 9 ++++----- 2 files changed, 8 insertions(+), 6 deletions(-) diff --git a/src/Common/ThreadPoolTaskTracker.cpp b/src/Common/ThreadPoolTaskTracker.cpp index 61d34801f7a..1697a13f780 100644 --- a/src/Common/ThreadPoolTaskTracker.cpp +++ b/src/Common/ThreadPoolTaskTracker.cpp @@ -19,6 +19,10 @@ TaskTracker::TaskTracker(ThreadPoolCallbackRunnerUnsafe scheduler_, size_t TaskTracker::~TaskTracker() { + /// Tasks should be waited outside of dtor. + /// Important for WriteBufferFromS3/AzureBlobStorage, where TaskTracker is currently used. + chassert(finished_futures.empty() && futures.empty()); + safeWaitAll(); } @@ -170,4 +174,3 @@ bool TaskTracker::isAsync() const } } - diff --git a/src/IO/WriteBufferFromS3.cpp b/src/IO/WriteBufferFromS3.cpp index 3682e49b018..e702b4d35ad 100644 --- a/src/IO/WriteBufferFromS3.cpp +++ b/src/IO/WriteBufferFromS3.cpp @@ -277,12 +277,10 @@ WriteBufferFromS3::~WriteBufferFromS3() "The file might not be written to S3. " "{}.", getVerboseLogDetails()); - return; } - - /// That destructor could be call with finalized=false in case of exceptions - if (!finalized && !canceled) + else if (!finalized) { + /// That destructor could be call with finalized=false in case of exceptions LOG_INFO( log, "WriteBufferFromS3 is not finalized in destructor. " @@ -291,9 +289,10 @@ WriteBufferFromS3::~WriteBufferFromS3() getVerboseLogDetails()); } + /// Wait for all tasks, because they contain reference to this write buffer. task_tracker->safeWaitAll(); - if (!multipart_upload_id.empty() && !multipart_upload_finished) + if (!canceled && !multipart_upload_id.empty() && !multipart_upload_finished) { LOG_WARNING(log, "WriteBufferFromS3 was neither finished nor aborted, try to abort upload in destructor. {}.", getVerboseLogDetails()); tryToAbortMultipartUpload(); From 20faed85ca30c6352fd091e8d4d763fb98fe1311 Mon Sep 17 00:00:00 2001 From: Antonio Andelic Date: Tue, 30 Jul 2024 15:18:37 +0200 Subject: [PATCH 057/227] Remove useless file --- tests/queries/0_stateless/data_minio/test :: 03215_archive.csv | 1 - 1 file changed, 1 deletion(-) delete mode 100644 tests/queries/0_stateless/data_minio/test :: 03215_archive.csv diff --git a/tests/queries/0_stateless/data_minio/test :: 03215_archive.csv b/tests/queries/0_stateless/data_minio/test :: 03215_archive.csv deleted file mode 100644 index d00491fd7e5..00000000000 --- a/tests/queries/0_stateless/data_minio/test :: 03215_archive.csv +++ /dev/null @@ -1 +0,0 @@ -1 From 4aedb9d40298c1a3204bb72a3288ea711eb5e2f6 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Tue, 30 Jul 2024 23:02:22 +0200 Subject: [PATCH 058/227] Update test --- tests/integration/test_system_flush_logs/test.py | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/tests/integration/test_system_flush_logs/test.py b/tests/integration/test_system_flush_logs/test.py index 2022f9d4a89..713b327eb76 100644 --- a/tests/integration/test_system_flush_logs/test.py +++ b/tests/integration/test_system_flush_logs/test.py @@ -13,9 +13,8 @@ node = cluster.add_instance( ) system_logs = [ - # disabled by default - ("system.text_log", 0), # enabled by default + ("system.text_log", 1), ("system.query_log", 1), ("system.query_thread_log", 1), ("system.part_log", 1), From 4a4bd97b4b63e495f76273ea8c12045dc129d81b Mon Sep 17 00:00:00 2001 From: Peter Nguyen Date: Tue, 30 Jul 2024 16:50:37 -0600 Subject: [PATCH 059/227] Fix case sensitivity for percent_rank, dense_rank, and their aliases --- src/Processors/Transforms/WindowTransform.cpp | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/src/Processors/Transforms/WindowTransform.cpp b/src/Processors/Transforms/WindowTransform.cpp index a1b46c8e36c..5fad68e4968 100644 --- a/src/Processors/Transforms/WindowTransform.cpp +++ b/src/Processors/Transforms/WindowTransform.cpp @@ -2726,18 +2726,18 @@ void registerWindowFunctions(AggregateFunctionFactory & factory) { return std::make_shared(name, argument_types, parameters); - }, properties}, AggregateFunctionFactory::Case::Insensitive); + }, properties}); - factory.registerAlias("dense_rank", "denseRank", AggregateFunctionFactory::Case::Sensitive); + factory.registerAlias("dense_rank", "denseRank", AggregateFunctionFactory::Case::Insensitive); factory.registerFunction("percentRank", {[](const std::string & name, const DataTypes & argument_types, const Array & parameters, const Settings *) { return std::make_shared(name, argument_types, parameters); - }, properties}, AggregateFunctionFactory::Case::Insensitive); + }, properties}); - factory.registerAlias("percent_rank", "percentRank", AggregateFunctionFactory::Case::Sensitive); + factory.registerAlias("percent_rank", "percentRank", AggregateFunctionFactory::Case::Insensitive); factory.registerFunction("row_number", {[](const std::string & name, const DataTypes & argument_types, const Array & parameters, const Settings *) From 3b12fec141fda8cb2a3ef68ac96e6e58f1fd69e3 Mon Sep 17 00:00:00 2001 From: Peter Nguyen Date: Tue, 30 Jul 2024 17:05:48 -0600 Subject: [PATCH 060/227] Update dense_rank doc to mention the denseRank alias --- docs/en/sql-reference/window-functions/dense_rank.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/docs/en/sql-reference/window-functions/dense_rank.md b/docs/en/sql-reference/window-functions/dense_rank.md index d6445b68c55..2c8617fb668 100644 --- a/docs/en/sql-reference/window-functions/dense_rank.md +++ b/docs/en/sql-reference/window-functions/dense_rank.md @@ -12,6 +12,8 @@ The [rank](./rank.md) function provides the same behaviour, but with gaps in ran **Syntax** +Alias: `denseRank` (case-sensitive) + ```sql dense_rank (column_name) OVER ([[PARTITION BY grouping_column] [ORDER BY sorting_column] From 419a5e7f730dabe514becabc6c24ec5b87325e28 Mon Sep 17 00:00:00 2001 From: Peter Nguyen Date: Tue, 30 Jul 2024 17:17:01 -0600 Subject: [PATCH 061/227] Update window-functions doc with denseRank and percentRank aliases --- docs/en/sql-reference/window-functions/index.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/docs/en/sql-reference/window-functions/index.md b/docs/en/sql-reference/window-functions/index.md index 0c3e2ea1cb6..27d4bd763c7 100644 --- a/docs/en/sql-reference/window-functions/index.md +++ b/docs/en/sql-reference/window-functions/index.md @@ -23,8 +23,8 @@ ClickHouse supports the standard grammar for defining windows and window functio | `INTERVAL` syntax for `DateTime` `RANGE OFFSET` frame | ❌ (specify the number of seconds instead (`RANGE` works with any numeric type).) | | `GROUPS` frame | ❌ | | Calculating aggregate functions over a frame (`sum(value) over (order by time)`) | ✅ (All aggregate functions are supported) | -| `rank()`, `dense_rank()`, `row_number()` | ✅ | -| `percent_rank()` | ✅ Efficiently computes the relative standing of a value within a partition in a dataset. This function effectively replaces the more verbose and computationally intensive manual SQL calculation expressed as `ifNull((rank() OVER(PARTITION BY x ORDER BY y) - 1) / nullif(count(1) OVER(PARTITION BY x) - 1, 0), 0)`| +| `rank()`, `dense_rank()`, `row_number()` | ✅
Alias: `denseRank()` | +| `percent_rank()` | ✅ Efficiently computes the relative standing of a value within a partition in a dataset. This function effectively replaces the more verbose and computationally intensive manual SQL calculation expressed as `ifNull((rank() OVER(PARTITION BY x ORDER BY y) - 1) / nullif(count(1) OVER(PARTITION BY x) - 1, 0), 0)`
Alias: `percentRank()`| | `lag/lead(value, offset)` | ❌
You can use one of the following workarounds:
1) `any(value) over (.... rows between preceding and preceding)`, or `following` for `lead`
2) `lagInFrame/leadInFrame`, which are analogous, but respect the window frame. To get behavior identical to `lag/lead`, use `rows between unbounded preceding and unbounded following` | | ntile(buckets) | ✅
Specify window like, (partition by x order by y rows between unbounded preceding and unrounded following). | From bc312eb046db07d901e208cdc1bb0abb1df3eabd Mon Sep 17 00:00:00 2001 From: Antonio Andelic Date: Tue, 30 Jul 2024 21:27:50 +0200 Subject: [PATCH 062/227] Improve check --- src/IO/S3/URI.cpp | 32 ++++++++++++------- src/Storages/StorageFile.cpp | 18 +++++------ .../03215_parsing_archive_name_s3.sql | 2 +- 3 files changed, 30 insertions(+), 22 deletions(-) diff --git a/src/IO/S3/URI.cpp b/src/IO/S3/URI.cpp index 33a4939c810..fead18315d8 100644 --- a/src/IO/S3/URI.cpp +++ b/src/IO/S3/URI.cpp @@ -55,10 +55,10 @@ URI::URI(const std::string & uri_, bool allow_archive_path_syntax) static constexpr auto OSS = "OSS"; static constexpr auto EOS = "EOS"; - if (!allow_archive_path_syntax) - uri_str = uri_; - else + if (allow_archive_path_syntax) std::tie(uri_str, archive_pattern) = getURIAndArchivePattern(uri_); + else + uri_str = uri_; uri = Poco::URI(uri_str); @@ -176,22 +176,30 @@ std::pair> URI::getURIAndArchivePattern( return {source, std::nullopt}; std::string_view path_to_archive_view = std::string_view{source}.substr(0, pos); + bool contains_spaces_around_operator = false; while (path_to_archive_view.ends_with(' ')) + { + contains_spaces_around_operator = true; path_to_archive_view.remove_suffix(1); - - if (path_to_archive_view.empty() || !hasSupportedArchiveExtension(path_to_archive_view)) - return {source, std::nullopt}; - - auto archive_uri = path_to_archive_view; + } std::string_view archive_pattern_view = std::string_view{source}.substr(pos + 2); - while (archive_pattern_view.front() == ' ') + while (archive_pattern_view.starts_with(' ')) + { + contains_spaces_around_operator = true; archive_pattern_view.remove_prefix(1); + } - if (archive_pattern_view.empty()) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Archive pattern is empty"); + /// possible situations when the first part can be archive is only if one of the following is true: + /// - it contains supported extension + /// - it contains spaces after or before :: (URI cannot contain spaces) + /// - it contains characters that could mean glob expression + if (archive_pattern_view.empty() || path_to_archive_view.empty() + || (!contains_spaces_around_operator && !hasSupportedArchiveExtension(path_to_archive_view) + && path_to_archive_view.find_first_of("*?{") == std::string_view::npos)) + return {source, std::nullopt}; - return std::pair{std::string{archive_uri}, std::string{archive_pattern_view}}; + return std::pair{std::string{path_to_archive_view}, std::string{archive_pattern_view}}; } } diff --git a/src/Storages/StorageFile.cpp b/src/Storages/StorageFile.cpp index efb39f90053..8c079aa4600 100644 --- a/src/Storages/StorageFile.cpp +++ b/src/Storages/StorageFile.cpp @@ -2258,21 +2258,21 @@ void StorageFile::parseFileSource(String source, String & filename, String & pat while (path_to_archive_view.ends_with(' ')) path_to_archive_view.remove_suffix(1); - if (path_to_archive_view.empty() || !hasSupportedArchiveExtension(path_to_archive_view)) + std::string_view filename_view = std::string_view{source}.substr(pos + 2); + while (filename_view.starts_with(' ')) + filename_view.remove_prefix(1); + + /// possible situations when the first part can be archive is only if one of the following is true: + /// - it contains supported extension + /// - it contains characters that could mean glob expression + if (filename_view.empty() || path_to_archive_view.empty() + || (!hasSupportedArchiveExtension(path_to_archive_view) && path_to_archive_view.find_first_of("*?{") == std::string_view::npos)) { filename = std::move(source); return; } path_to_archive = path_to_archive_view; - - std::string_view filename_view = std::string_view{source}.substr(pos + 2); - while (filename_view.front() == ' ') - filename_view.remove_prefix(1); - - if (filename_view.empty()) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Filename is empty"); - filename = filename_view; } diff --git a/tests/queries/0_stateless/03215_parsing_archive_name_s3.sql b/tests/queries/0_stateless/03215_parsing_archive_name_s3.sql index 3a7ed0b864c..e34be475c5a 100644 --- a/tests/queries/0_stateless/03215_parsing_archive_name_s3.sql +++ b/tests/queries/0_stateless/03215_parsing_archive_name_s3.sql @@ -2,6 +2,6 @@ -- Tag no-fasttest: Depends on AWS SELECT _file, _path FROM s3(s3_conn, filename='::03215_archive.csv') ORDER BY (_file, _path); -SELECT _file, _path FROM s3(s3_conn, filename='test :: 03215_archive.csv') ORDER BY (_file, _path); -- { serverError STD_EXCEPTION } +SELECT _file, _path FROM s3(s3_conn, filename='test :: 03215_archive.csv') ORDER BY (_file, _path); -- { serverError S3_ERROR } SELECT _file, _path FROM s3(s3_conn, filename='test::03215_archive.csv') ORDER BY (_file, _path); SELECT _file, _path FROM s3(s3_conn, filename='test.zip::03215_archive.csv') ORDER BY (_file, _path) SETTINGS allow_archive_path_syntax=0; From f9f17fb61e2ab27f90434b5e3fc9081c061eaae4 Mon Sep 17 00:00:00 2001 From: Antonio Andelic Date: Wed, 31 Jul 2024 12:32:17 +0200 Subject: [PATCH 063/227] Fix reference --- .../0_stateless/03215_parsing_archive_name_s3.reference | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tests/queries/0_stateless/03215_parsing_archive_name_s3.reference b/tests/queries/0_stateless/03215_parsing_archive_name_s3.reference index b4804c82dc2..b27524812c7 100644 --- a/tests/queries/0_stateless/03215_parsing_archive_name_s3.reference +++ b/tests/queries/0_stateless/03215_parsing_archive_name_s3.reference @@ -1,3 +1,3 @@ -::03215_archive.csv test/::03215_archive.csv -test::03215_archive.csv test/test::03215_archive.csv -test.zip::03215_archive.csv test/test.zip::03215_archive.csv +::03215_archive.csv test/::03215_archive.csv +test::03215_archive.csv test/test::03215_archive.csv +test.zip::03215_archive.csv test/test.zip::03215_archive.csv From c81d3322b18b0eb4b45b91ac019a8c4f42d7518d Mon Sep 17 00:00:00 2001 From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com> Date: Wed, 31 Jul 2024 13:39:30 +0200 Subject: [PATCH 064/227] Update 02150_index_hypothesis_race_long.sh --- tests/queries/0_stateless/02150_index_hypothesis_race_long.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/02150_index_hypothesis_race_long.sh b/tests/queries/0_stateless/02150_index_hypothesis_race_long.sh index c29b604d23d..5c432350768 100755 --- a/tests/queries/0_stateless/02150_index_hypothesis_race_long.sh +++ b/tests/queries/0_stateless/02150_index_hypothesis_race_long.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-distributed-cache +# Tags: long, no-random-settings, no-distributed-cache CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh From d5f8540fbbcc7e5f8014212d67804412c7466d7d Mon Sep 17 00:00:00 2001 From: divanik Date: Wed, 31 Jul 2024 12:00:09 +0000 Subject: [PATCH 065/227] Fix --- tests/integration/helpers/cluster.py | 25 +++++++++++++------- tests/integration/helpers/retry_decorator.py | 7 ++++-- 2 files changed, 21 insertions(+), 11 deletions(-) diff --git a/tests/integration/helpers/cluster.py b/tests/integration/helpers/cluster.py index 7d80fbe90f8..7f0a9154be9 100644 --- a/tests/integration/helpers/cluster.py +++ b/tests/integration/helpers/cluster.py @@ -2692,11 +2692,13 @@ class ClickHouseCluster: images_pull_cmd = self.base_cmd + ["pull"] # sometimes dockerhub/proxy can be flaky - retry( - log_function=lambda exception: logging.info( - "Got exception pulling images: %s", exception - ), - )(run_and_check)(images_pull_cmd) + def logging_pulling_images(**kwargs): + if "exception" in kwargs: + logging.info( + "Got exception pulling images: %s", kwargs["exception"] + ) + + retry(log_function=logging_pulling_images)(run_and_check)(images_pull_cmd) if self.with_zookeeper_secure and self.base_zookeeper_cmd: logging.debug("Setup ZooKeeper Secure") @@ -2969,11 +2971,16 @@ class ClickHouseCluster: "Trying to create Azurite instance by command %s", " ".join(map(str, azurite_start_cmd)), ) - retry( - log_function=lambda exception: logging.info( + def logging_azurite_initialization(exception, retry_number, sleep_time): + logging.info( f"Azurite initialization failed with error: {exception}" - ), - )(run_and_check)(azurite_start_cmd) + ) + + retry( + log_function=logging_azurite_initialization, + )( + run_and_check + )(azurite_start_cmd) self.up_called = True logging.info("Trying to connect to Azurite") self.wait_azurite_to_start() diff --git a/tests/integration/helpers/retry_decorator.py b/tests/integration/helpers/retry_decorator.py index aaa040464c2..e7bafbe29c1 100644 --- a/tests/integration/helpers/retry_decorator.py +++ b/tests/integration/helpers/retry_decorator.py @@ -8,7 +8,7 @@ def retry( delay: float = 1, backoff: float = 1.5, jitter: float = 2, - log_function=lambda *args, **kwargs: None, + log_function=None, # should take **kwargs or arguments: `retry_number`, `exception` and `sleep_time` retriable_expections_list: List[Type[BaseException]] = [Exception], ): def inner(func): @@ -26,8 +26,11 @@ def retry( break if not should_retry or (retry == retries - 1): raise e - log_function(retry=retry, exception=e) sleep_time = current_delay + random.uniform(0, jitter) + if log_function is not None: + log_function( + retry_number=retry, exception=e, sleep_time=sleep_time + ) time.sleep(sleep_time) current_delay *= backoff From 406ac2279ecbfc24913548dfcf459c55dd450723 Mon Sep 17 00:00:00 2001 From: Dmitry Novik Date: Wed, 31 Jul 2024 14:48:33 +0200 Subject: [PATCH 066/227] Analyzer: Do not traverse unresolved subtrees --- src/Planner/findParallelReplicasQuery.cpp | 12 +++++------- src/Planner/findQueryForParallelReplicas.h | 2 +- ...5_analyzer_replace_with_dummy_tables.reference | 0 .../03215_analyzer_replace_with_dummy_tables.sql | 15 +++++++++++++++ 4 files changed, 21 insertions(+), 8 deletions(-) create mode 100644 tests/queries/0_stateless/03215_analyzer_replace_with_dummy_tables.reference create mode 100644 tests/queries/0_stateless/03215_analyzer_replace_with_dummy_tables.sql diff --git a/src/Planner/findParallelReplicasQuery.cpp b/src/Planner/findParallelReplicasQuery.cpp index c89a70be541..1140f30ad9c 100644 --- a/src/Planner/findParallelReplicasQuery.cpp +++ b/src/Planner/findParallelReplicasQuery.cpp @@ -113,13 +113,13 @@ std::stack getSupportingParallelReplicasQuery(const IQueryTre return res; } -class ReplaceTableNodeToDummyVisitor : public InDepthQueryTreeVisitor +class ReplaceTableNodeToDummyVisitor : public InDepthQueryTreeVisitorWithContext { public: - using Base = InDepthQueryTreeVisitor; + using Base = InDepthQueryTreeVisitorWithContext; using Base::Base; - void visitImpl(const QueryTreeNodePtr & node) + void enterImpl(QueryTreeNodePtr & node) { auto * table_node = node->as(); auto * table_function_node = node->as(); @@ -134,21 +134,19 @@ public: ColumnsDescription(storage_snapshot->getColumns(get_column_options)), storage_snapshot); - auto dummy_table_node = std::make_shared(std::move(storage_dummy), context); + auto dummy_table_node = std::make_shared(std::move(storage_dummy), getContext()); dummy_table_node->setAlias(node->getAlias()); replacement_map.emplace(node.get(), std::move(dummy_table_node)); } } - ContextPtr context; std::unordered_map replacement_map; }; QueryTreeNodePtr replaceTablesWithDummyTables(const QueryTreeNodePtr & query, const ContextPtr & context) { - ReplaceTableNodeToDummyVisitor visitor; - visitor.context = context; + ReplaceTableNodeToDummyVisitor visitor(context); visitor.visit(query); return query->cloneAndReplace(visitor.replacement_map); diff --git a/src/Planner/findQueryForParallelReplicas.h b/src/Planner/findQueryForParallelReplicas.h index f5dc69dfa0e..cdce4ad0b47 100644 --- a/src/Planner/findQueryForParallelReplicas.h +++ b/src/Planner/findQueryForParallelReplicas.h @@ -13,7 +13,7 @@ using QueryTreeNodePtr = std::shared_ptr; struct SelectQueryOptions; -/// Find a qury which can be executed with parallel replicas up to WithMergableStage. +/// Find a query which can be executed with parallel replicas up to WithMergableStage. /// Returned query will always contain some (>1) subqueries, possibly with joins. const QueryNode * findQueryForParallelReplicas(const QueryTreeNodePtr & query_tree_node, SelectQueryOptions & select_query_options); diff --git a/tests/queries/0_stateless/03215_analyzer_replace_with_dummy_tables.reference b/tests/queries/0_stateless/03215_analyzer_replace_with_dummy_tables.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03215_analyzer_replace_with_dummy_tables.sql b/tests/queries/0_stateless/03215_analyzer_replace_with_dummy_tables.sql new file mode 100644 index 00000000000..12d2bd627a7 --- /dev/null +++ b/tests/queries/0_stateless/03215_analyzer_replace_with_dummy_tables.sql @@ -0,0 +1,15 @@ +create table t (number UInt64) engine MergeTree order by number; + +SELECT 1 +FROM +( + SELECT number IN ( + SELECT number + FROM view( + SELECT number + FROM numbers(1) + ) + ) + FROM t +) +SETTINGS allow_experimental_parallel_reading_from_replicas = 1, max_parallel_replicas = 2, allow_experimental_analyzer = 1; From bca21bc6c735b710ed15061e41790066b5ffd843 Mon Sep 17 00:00:00 2001 From: sakulali Date: Wed, 31 Jul 2024 21:33:14 +0800 Subject: [PATCH 067/227] add test cases --- src/Common/Config/ConfigProcessor.cpp | 11 ++++------- ...89_clickhouse_client_config_format.reference | 4 ++++ .../01889_clickhouse_client_config_format.sh | 17 +++++++++++++++++ 3 files changed, 25 insertions(+), 7 deletions(-) diff --git a/src/Common/Config/ConfigProcessor.cpp b/src/Common/Config/ConfigProcessor.cpp index f0b83c035c8..c4b4a1d5e7e 100644 --- a/src/Common/Config/ConfigProcessor.cpp +++ b/src/Common/Config/ConfigProcessor.cpp @@ -654,22 +654,19 @@ XMLDocumentPtr ConfigProcessor::parseConfig(const std::string & config_path) while (std::getline(file, line)) { const size_t pos = firstNonWhitespacePos(line); - if (pos == std::string::npos) - continue; if (pos < line.size() && '<' == line[pos]) { maybe_xml = true; + break; } - - break; + else if (pos != std::string::npos) + break; } } - if (maybe_xml) return dom_parser.parse(config_path); - else - return YAMLParser::parse(config_path); + return YAMLParser::parse(config_path); } } diff --git a/tests/queries/0_stateless/01889_clickhouse_client_config_format.reference b/tests/queries/0_stateless/01889_clickhouse_client_config_format.reference index afe27ddb063..ef0d9ffc538 100644 --- a/tests/queries/0_stateless/01889_clickhouse_client_config_format.reference +++ b/tests/queries/0_stateless/01889_clickhouse_client_config_format.reference @@ -18,3 +18,7 @@ autodetect xml (non leading whitespaces) 2 autodetect yaml 2 +autodetect invalid xml +Code: 1000, e.code() = 0, SAXParseException: Invalid token in '/config_test.badxml', line 2 column 12, Stack trace (when copying this message, always include the lines below): +autodetect invalid yaml +Code: 585. Unable to parse YAML configuration file /config_test.badyaml, yaml-cpp: error at line 2, column 12: illegal map value. (CANNOT_PARSE_YAML) diff --git a/tests/queries/0_stateless/01889_clickhouse_client_config_format.sh b/tests/queries/0_stateless/01889_clickhouse_client_config_format.sh index 14d67f943f1..9a44ec0d5f5 100755 --- a/tests/queries/0_stateless/01889_clickhouse_client_config_format.sh +++ b/tests/queries/0_stateless/01889_clickhouse_client_config_format.sh @@ -15,6 +15,8 @@ yaml_config=$CLICKHOUSE_TMP/config_$CLICKHOUSE_DATABASE.yaml autodetect_xml_with_leading_whitespace_config=$CLICKHOUSE_TMP/config_$CLICKHOUSE_DATABASE.config autodetect_xml_non_leading_whitespace_config=$CLICKHOUSE_TMP/config_$CLICKHOUSE_DATABASE.cfg autodetect_yaml_config=$CLICKHOUSE_TMP/config_$CLICKHOUSE_DATABASE.properties +autodetect_invalid_xml_config=$CLICKHOUSE_TMP/config_$CLICKHOUSE_DATABASE.badxml +autodetect_invalid_yaml_config=$CLICKHOUSE_TMP/config_$CLICKHOUSE_DATABASE.badyaml function cleanup() { @@ -27,6 +29,8 @@ function cleanup() rm "${autodetect_xml_with_leading_whitespace_config:?}" rm "${autodetect_xml_non_leading_whitespace_config:?}" rm "${autodetect_yaml_config:?}" + rm "${autodetect_invalid_xml_config:?}" + rm "${autodetect_invalid_yaml_config:?}" } trap cleanup EXIT @@ -70,6 +74,15 @@ EOL cat > "$autodetect_yaml_config" < "$autodetect_invalid_xml_config" < + +EOL +cat > "$autodetect_invalid_yaml_config" <&1 |& sed -n '1p' | sed -e "s#$CLICKHOUSE_TMP##" -e "s#Poco::Exception. ##" +echo 'autodetect invalid yaml' +$CLICKHOUSE_CLIENT --config "$autodetect_invalid_yaml_config" -q "select getSetting('max_threads')" 2>&1 |& sed -n '1p' | sed -e "s#$CLICKHOUSE_TMP##" -e "s#DB::Exception: ##" \ No newline at end of file From 31c142a96d49fbe1b46b21e4cdad366546dc7864 Mon Sep 17 00:00:00 2001 From: Michael Stetsyuk Date: Wed, 31 Jul 2024 14:44:54 +0100 Subject: [PATCH 068/227] make it possible to rerun test_storage_delta and test_checking_s3_blobs_paranoid --- .../test_checking_s3_blobs_paranoid/test.py | 2 ++ tests/integration/test_storage_delta/test.py | 29 +++++++++++++++++++ 2 files changed, 31 insertions(+) diff --git a/tests/integration/test_checking_s3_blobs_paranoid/test.py b/tests/integration/test_checking_s3_blobs_paranoid/test.py index dde636b5d29..afe8449b44a 100644 --- a/tests/integration/test_checking_s3_blobs_paranoid/test.py +++ b/tests/integration/test_checking_s3_blobs_paranoid/test.py @@ -61,6 +61,7 @@ def test_upload_after_check_works(cluster, broken_s3): node.query( """ + DROP TABLE IF EXISTS s3_upload_after_check_works; CREATE TABLE s3_upload_after_check_works ( id Int64, data String @@ -631,6 +632,7 @@ def test_no_key_found_disk(cluster, broken_s3): node.query( """ + DROP TABLE IF EXISTS no_key_found_disk; CREATE TABLE no_key_found_disk ( id Int64 ) ENGINE=MergeTree() diff --git a/tests/integration/test_storage_delta/test.py b/tests/integration/test_storage_delta/test.py index 67cc7cdd6da..698becc18c4 100644 --- a/tests/integration/test_storage_delta/test.py +++ b/tests/integration/test_storage_delta/test.py @@ -52,6 +52,11 @@ def get_spark(): return builder.master("local").getOrCreate() +def remove_local_directory_contents(local_path): + for local_file in glob.glob(local_path + "/**"): + os.unlink(local_file) + + @pytest.fixture(scope="module") def started_cluster(): try: @@ -169,6 +174,9 @@ def test_single_log_file(started_cluster): inserted_data ) + os.unlink(parquet_data_path) + remove_local_directory_contents(f"/{TABLE_NAME}") + def test_partition_by(started_cluster): instance = started_cluster.instances["node1"] @@ -191,6 +199,7 @@ def test_partition_by(started_cluster): create_delta_table(instance, TABLE_NAME) assert int(instance.query(f"SELECT count() FROM {TABLE_NAME}")) == 10 + remove_local_directory_contents(f"/{TABLE_NAME}") def test_checkpoint(started_cluster): instance = started_cluster.instances["node1"] @@ -266,6 +275,9 @@ def test_checkpoint(started_cluster): ).strip() ) + remove_local_directory_contents(f"/{TABLE_NAME}") + spark.sql(f"DROP TABLE {TABLE_NAME}") + def test_multiple_log_files(started_cluster): instance = started_cluster.instances["node1"] @@ -304,6 +316,8 @@ def test_multiple_log_files(started_cluster): "SELECT number, toString(number + 1) FROM numbers(200)" ) + remove_local_directory_contents(f"/{TABLE_NAME}") + def test_metadata(started_cluster): instance = started_cluster.instances["node1"] @@ -337,6 +351,9 @@ def test_metadata(started_cluster): create_delta_table(instance, TABLE_NAME) assert int(instance.query(f"SELECT count() FROM {TABLE_NAME}")) == 100 + os.unlink(parquet_data_path) + remove_local_directory_contents(f"/{TABLE_NAME}") + def test_types(started_cluster): TABLE_NAME = "test_types" @@ -409,6 +426,9 @@ def test_types(started_cluster): ] ) + remove_local_directory_contents(f"/{result_file}") + spark.sql(f"DROP TABLE {TABLE_NAME}") + def test_restart_broken(started_cluster): instance = started_cluster.instances["node1"] @@ -470,6 +490,9 @@ def test_restart_broken(started_cluster): assert int(instance.query(f"SELECT count() FROM {TABLE_NAME}")) == 100 + os.unlink(parquet_data_path) + remove_local_directory_contents(f"/{TABLE_NAME}") + def test_restart_broken_table_function(started_cluster): instance = started_cluster.instances["node1"] @@ -524,6 +547,9 @@ def test_restart_broken_table_function(started_cluster): assert int(instance.query(f"SELECT count() FROM {TABLE_NAME}")) == 100 + os.unlink(parquet_data_path) + remove_local_directory_contents(f"/{TABLE_NAME}") + def test_partition_columns(started_cluster): instance = started_cluster.instances["node1"] @@ -721,3 +747,6 @@ SELECT * FROM deltaLake('http://{started_cluster.minio_ip}:{started_cluster.mini ) == 1 ) + + remove_local_directory_contents(f"/{TABLE_NAME}") + spark.sql(f"DROP TABLE {TABLE_NAME}") From 2a2dba63cc0182247754a5a4819cb89f21825bfd Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Wed, 31 Jul 2024 13:48:04 +0000 Subject: [PATCH 069/227] Automatic style fix --- tests/integration/helpers/cluster.py | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/integration/helpers/cluster.py b/tests/integration/helpers/cluster.py index 7f0a9154be9..2e38aec3512 100644 --- a/tests/integration/helpers/cluster.py +++ b/tests/integration/helpers/cluster.py @@ -2971,6 +2971,7 @@ class ClickHouseCluster: "Trying to create Azurite instance by command %s", " ".join(map(str, azurite_start_cmd)), ) + def logging_azurite_initialization(exception, retry_number, sleep_time): logging.info( f"Azurite initialization failed with error: {exception}" From 7dbd3d75340522195e7d08a725cf5ae116288c8e Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Wed, 31 Jul 2024 13:51:38 +0000 Subject: [PATCH 070/227] Automatic style fix --- tests/integration/test_storage_delta/test.py | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/integration/test_storage_delta/test.py b/tests/integration/test_storage_delta/test.py index 698becc18c4..e485bc90ee0 100644 --- a/tests/integration/test_storage_delta/test.py +++ b/tests/integration/test_storage_delta/test.py @@ -201,6 +201,7 @@ def test_partition_by(started_cluster): remove_local_directory_contents(f"/{TABLE_NAME}") + def test_checkpoint(started_cluster): instance = started_cluster.instances["node1"] spark = started_cluster.spark_session From d6de2be4395e1bcc62ab32ad1d5b02e9db080303 Mon Sep 17 00:00:00 2001 From: Dmitry Novik Date: Wed, 31 Jul 2024 16:08:18 +0200 Subject: [PATCH 071/227] Fix build --- src/Planner/findParallelReplicasQuery.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Planner/findParallelReplicasQuery.cpp b/src/Planner/findParallelReplicasQuery.cpp index 1140f30ad9c..39edb1e6516 100644 --- a/src/Planner/findParallelReplicasQuery.cpp +++ b/src/Planner/findParallelReplicasQuery.cpp @@ -144,7 +144,7 @@ public: std::unordered_map replacement_map; }; -QueryTreeNodePtr replaceTablesWithDummyTables(const QueryTreeNodePtr & query, const ContextPtr & context) +QueryTreeNodePtr replaceTablesWithDummyTables(QueryTreeNodePtr query, const ContextPtr & context) { ReplaceTableNodeToDummyVisitor visitor(context); visitor.visit(query); From 9ffbd8f5073e180592a494742d1dc3af4427b55f Mon Sep 17 00:00:00 2001 From: divanik Date: Wed, 31 Jul 2024 14:13:43 +0000 Subject: [PATCH 072/227] Possible fix --- .../03164_s3_settings_for_queries_and_merges.sql | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/tests/queries/0_stateless/03164_s3_settings_for_queries_and_merges.sql b/tests/queries/0_stateless/03164_s3_settings_for_queries_and_merges.sql index ac2070fbd76..e43c9ae7717 100644 --- a/tests/queries/0_stateless/03164_s3_settings_for_queries_and_merges.sql +++ b/tests/queries/0_stateless/03164_s3_settings_for_queries_and_merges.sql @@ -22,18 +22,18 @@ SELECT count() FROM t_compact_bytes_s3 WHERE NOT ignore(c2, c4); SYSTEM FLUSH LOGS; SELECT - ProfileEvents['S3ReadRequestsCount'], + ProfileEvents['S3ReadRequestsCount'] - ProfileEvents['S3ReadRequestsError'], ProfileEvents['ReadBufferFromS3Bytes'] < ProfileEvents['ReadCompressedBytes'] * 1.1 FROM system.query_log -WHERE event_date >= yesterday() AND type = 'QueryFinish' +WHERE type = 'QueryFinish' AND current_database = currentDatabase() AND query ilike '%INSERT INTO t_compact_bytes_s3 SELECT number, number, number%'; SELECT - ProfileEvents['S3ReadRequestsCount'], + ProfileEvents['S3ReadRequestsCount'] - ProfileEvents['S3ReadRequestsError'], ProfileEvents['ReadBufferFromS3Bytes'] < ProfileEvents['ReadCompressedBytes'] * 1.1 FROM system.query_log -WHERE event_date >= yesterday() AND type = 'QueryFinish' +WHERE type = 'QueryFinish' AND current_database = currentDatabase() AND query ilike '%OPTIMIZE TABLE t_compact_bytes_s3 FINAL%'; From 67f4792b77f2a2cf0de21ead6e95c3635d26aa88 Mon Sep 17 00:00:00 2001 From: divanik Date: Wed, 31 Jul 2024 14:16:40 +0000 Subject: [PATCH 073/227] Style check --- tests/integration/helpers/cluster.py | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/integration/helpers/cluster.py b/tests/integration/helpers/cluster.py index 7f0a9154be9..2e38aec3512 100644 --- a/tests/integration/helpers/cluster.py +++ b/tests/integration/helpers/cluster.py @@ -2971,6 +2971,7 @@ class ClickHouseCluster: "Trying to create Azurite instance by command %s", " ".join(map(str, azurite_start_cmd)), ) + def logging_azurite_initialization(exception, retry_number, sleep_time): logging.info( f"Azurite initialization failed with error: {exception}" From e31569a065d4c81cdea671727c39983d7f3a84e5 Mon Sep 17 00:00:00 2001 From: Dmitry Novik Date: Wed, 31 Jul 2024 16:32:37 +0200 Subject: [PATCH 074/227] Expect an unknown cluster --- .../0_stateless/03215_analyzer_replace_with_dummy_tables.sql | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/03215_analyzer_replace_with_dummy_tables.sql b/tests/queries/0_stateless/03215_analyzer_replace_with_dummy_tables.sql index 12d2bd627a7..6d084c2ac50 100644 --- a/tests/queries/0_stateless/03215_analyzer_replace_with_dummy_tables.sql +++ b/tests/queries/0_stateless/03215_analyzer_replace_with_dummy_tables.sql @@ -12,4 +12,4 @@ FROM ) FROM t ) -SETTINGS allow_experimental_parallel_reading_from_replicas = 1, max_parallel_replicas = 2, allow_experimental_analyzer = 1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 1, max_parallel_replicas = 2, allow_experimental_analyzer = 1; -- { serverError CLUSTER_DOESNT_EXIST } From 8c36fbf4eddeba9282b53f726976b55f62d3ee19 Mon Sep 17 00:00:00 2001 From: divanik Date: Wed, 31 Jul 2024 14:42:38 +0000 Subject: [PATCH 075/227] Remove unnecessary change --- access/quotas.list | Bin 0 -> 1 bytes access/roles.list | Bin 0 -> 1 bytes access/row_policies.list | Bin 0 -> 1 bytes access/settings_profiles.list | Bin 0 -> 1 bytes access/users.list | Bin 0 -> 1 bytes .../03164_s3_settings_for_queries_and_merges.sql | 4 ++-- 6 files changed, 2 insertions(+), 2 deletions(-) create mode 100644 access/quotas.list create mode 100644 access/roles.list create mode 100644 access/row_policies.list create mode 100644 access/settings_profiles.list create mode 100644 access/users.list diff --git a/access/quotas.list b/access/quotas.list new file mode 100644 index 0000000000000000000000000000000000000000..f76dd238ade08917e6712764a16a22005a50573d GIT binary patch literal 1 IcmZPo000310RR91 literal 0 HcmV?d00001 diff --git a/access/roles.list b/access/roles.list new file mode 100644 index 0000000000000000000000000000000000000000..f76dd238ade08917e6712764a16a22005a50573d GIT binary patch literal 1 IcmZPo000310RR91 literal 0 HcmV?d00001 diff --git a/access/row_policies.list b/access/row_policies.list new file mode 100644 index 0000000000000000000000000000000000000000..f76dd238ade08917e6712764a16a22005a50573d GIT binary patch literal 1 IcmZPo000310RR91 literal 0 HcmV?d00001 diff --git a/access/settings_profiles.list b/access/settings_profiles.list new file mode 100644 index 0000000000000000000000000000000000000000..f76dd238ade08917e6712764a16a22005a50573d GIT binary patch literal 1 IcmZPo000310RR91 literal 0 HcmV?d00001 diff --git a/access/users.list b/access/users.list new file mode 100644 index 0000000000000000000000000000000000000000..f76dd238ade08917e6712764a16a22005a50573d GIT binary patch literal 1 IcmZPo000310RR91 literal 0 HcmV?d00001 diff --git a/tests/queries/0_stateless/03164_s3_settings_for_queries_and_merges.sql b/tests/queries/0_stateless/03164_s3_settings_for_queries_and_merges.sql index e43c9ae7717..94e390537df 100644 --- a/tests/queries/0_stateless/03164_s3_settings_for_queries_and_merges.sql +++ b/tests/queries/0_stateless/03164_s3_settings_for_queries_and_merges.sql @@ -25,7 +25,7 @@ SELECT ProfileEvents['S3ReadRequestsCount'] - ProfileEvents['S3ReadRequestsError'], ProfileEvents['ReadBufferFromS3Bytes'] < ProfileEvents['ReadCompressedBytes'] * 1.1 FROM system.query_log -WHERE type = 'QueryFinish' +WHERE event_date >= yesterday() AND type = 'QueryFinish' AND current_database = currentDatabase() AND query ilike '%INSERT INTO t_compact_bytes_s3 SELECT number, number, number%'; @@ -33,7 +33,7 @@ SELECT ProfileEvents['S3ReadRequestsCount'] - ProfileEvents['S3ReadRequestsError'], ProfileEvents['ReadBufferFromS3Bytes'] < ProfileEvents['ReadCompressedBytes'] * 1.1 FROM system.query_log -WHERE type = 'QueryFinish' +WHERE event_date >= yesterday() AND type = 'QueryFinish' AND current_database = currentDatabase() AND query ilike '%OPTIMIZE TABLE t_compact_bytes_s3 FINAL%'; From 20ec27f9dc79d7ee81cd06f1587de83c8ce81441 Mon Sep 17 00:00:00 2001 From: divanik Date: Wed, 31 Jul 2024 14:46:53 +0000 Subject: [PATCH 076/227] Remove trach dir --- access/quotas.list | Bin 1 -> 0 bytes access/roles.list | Bin 1 -> 0 bytes access/row_policies.list | Bin 1 -> 0 bytes access/settings_profiles.list | Bin 1 -> 0 bytes access/users.list | Bin 1 -> 0 bytes 5 files changed, 0 insertions(+), 0 deletions(-) delete mode 100644 access/quotas.list delete mode 100644 access/roles.list delete mode 100644 access/row_policies.list delete mode 100644 access/settings_profiles.list delete mode 100644 access/users.list diff --git a/access/quotas.list b/access/quotas.list deleted file mode 100644 index f76dd238ade08917e6712764a16a22005a50573d..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 1 IcmZPo000310RR91 diff --git a/access/roles.list b/access/roles.list deleted file mode 100644 index f76dd238ade08917e6712764a16a22005a50573d..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 1 IcmZPo000310RR91 diff --git a/access/row_policies.list b/access/row_policies.list deleted file mode 100644 index f76dd238ade08917e6712764a16a22005a50573d..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 1 IcmZPo000310RR91 diff --git a/access/settings_profiles.list b/access/settings_profiles.list deleted file mode 100644 index f76dd238ade08917e6712764a16a22005a50573d..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 1 IcmZPo000310RR91 diff --git a/access/users.list b/access/users.list deleted file mode 100644 index f76dd238ade08917e6712764a16a22005a50573d..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 1 IcmZPo000310RR91 From 650737890299f8cad2c77ad46022ee0a37b284eb Mon Sep 17 00:00:00 2001 From: divanik Date: Wed, 31 Jul 2024 14:49:30 +0000 Subject: [PATCH 077/227] Fix erroe with profile event name --- .../0_stateless/03164_s3_settings_for_queries_and_merges.sql | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/queries/0_stateless/03164_s3_settings_for_queries_and_merges.sql b/tests/queries/0_stateless/03164_s3_settings_for_queries_and_merges.sql index 94e390537df..001ef382850 100644 --- a/tests/queries/0_stateless/03164_s3_settings_for_queries_and_merges.sql +++ b/tests/queries/0_stateless/03164_s3_settings_for_queries_and_merges.sql @@ -22,7 +22,7 @@ SELECT count() FROM t_compact_bytes_s3 WHERE NOT ignore(c2, c4); SYSTEM FLUSH LOGS; SELECT - ProfileEvents['S3ReadRequestsCount'] - ProfileEvents['S3ReadRequestsError'], + ProfileEvents['S3ReadRequestsCount'] - ProfileEvents['S3ReadRequestsErrors'], ProfileEvents['ReadBufferFromS3Bytes'] < ProfileEvents['ReadCompressedBytes'] * 1.1 FROM system.query_log WHERE event_date >= yesterday() AND type = 'QueryFinish' @@ -30,7 +30,7 @@ WHERE event_date >= yesterday() AND type = 'QueryFinish' AND query ilike '%INSERT INTO t_compact_bytes_s3 SELECT number, number, number%'; SELECT - ProfileEvents['S3ReadRequestsCount'] - ProfileEvents['S3ReadRequestsError'], + ProfileEvents['S3ReadRequestsCount'] - ProfileEvents['S3ReadRequestsErrors'], ProfileEvents['ReadBufferFromS3Bytes'] < ProfileEvents['ReadCompressedBytes'] * 1.1 FROM system.query_log WHERE event_date >= yesterday() AND type = 'QueryFinish' From 5ab8c0357a84d0265bdb43f43795b06317854772 Mon Sep 17 00:00:00 2001 From: kssenii Date: Wed, 31 Jul 2024 16:30:14 +0200 Subject: [PATCH 078/227] Fix potential busy loop in keepFreeSpaceRatioFunc --- src/Interpreters/Cache/FileCache.cpp | 30 +++++++++++++------ src/Interpreters/Cache/IFileCachePriority.h | 8 ++++- .../Cache/LRUFileCachePriority.cpp | 21 +++++++++++-- src/Interpreters/Cache/LRUFileCachePriority.h | 2 +- .../Cache/SLRUFileCachePriority.cpp | 15 ++++++---- .../Cache/SLRUFileCachePriority.h | 2 +- 6 files changed, 57 insertions(+), 21 deletions(-) diff --git a/src/Interpreters/Cache/FileCache.cpp b/src/Interpreters/Cache/FileCache.cpp index a88c0de2cfe..bf8dd24a1db 100644 --- a/src/Interpreters/Cache/FileCache.cpp +++ b/src/Interpreters/Cache/FileCache.cpp @@ -998,18 +998,19 @@ void FileCache::freeSpaceRatioKeepingThreadFunc() FileCacheReserveStat stat; EvictionCandidates eviction_candidates; - bool limits_satisfied = true; + IFileCachePriority::DesiredSizeStatus desired_size_status; try { /// Collect at most `keep_up_free_space_remove_batch` elements to evict, /// (we use batches to make sure we do not block cache for too long, /// by default the batch size is quite small). - limits_satisfied = main_priority->collectCandidatesForEviction( + desired_size_status = main_priority->collectCandidatesForEviction( desired_size, desired_elements_num, keep_up_free_space_remove_batch, stat, eviction_candidates, lock); #ifdef DEBUG_OR_SANITIZER_BUILD /// Let's make sure that we correctly processed the limits. - if (limits_satisfied && eviction_candidates.size() < keep_up_free_space_remove_batch) + if (desired_size_status == IFileCachePriority::DesiredSizeStatus::SUCCESS + && eviction_candidates.size() < keep_up_free_space_remove_batch) { const auto current_size = main_priority->getSize(lock); chassert(current_size >= stat.total_stat.releasable_size); @@ -1063,13 +1064,24 @@ void FileCache::freeSpaceRatioKeepingThreadFunc() watch.stop(); ProfileEvents::increment(ProfileEvents::FilesystemCacheFreeSpaceKeepingThreadWorkMilliseconds, watch.elapsedMilliseconds()); - LOG_TRACE(log, "Free space ratio keeping thread finished in {} ms", watch.elapsedMilliseconds()); + LOG_TRACE(log, "Free space ratio keeping thread finished in {} ms (status: {})", + watch.elapsedMilliseconds(), desired_size_status); [[maybe_unused]] bool scheduled = false; - if (limits_satisfied) - scheduled = keep_up_free_space_ratio_task->scheduleAfter(general_reschedule_ms); - else - scheduled = keep_up_free_space_ratio_task->schedule(); + switch (desired_size_status) + { + case IFileCachePriority::DesiredSizeStatus::SUCCESS: [[fallthrough]]; + case IFileCachePriority::DesiredSizeStatus::CANNOT_EVICT: + { + scheduled = keep_up_free_space_ratio_task->scheduleAfter(general_reschedule_ms); + break; + } + case IFileCachePriority::DesiredSizeStatus::REACHED_MAX_CANDIDATES_LIMIT: + { + scheduled = keep_up_free_space_ratio_task->schedule(); + break; + } + } chassert(scheduled); } @@ -1546,7 +1558,7 @@ void FileCache::applySettingsIfPossible(const FileCacheSettings & new_settings, FileCacheReserveStat stat; if (main_priority->collectCandidatesForEviction( new_settings.max_size, new_settings.max_elements, 0/* max_candidates_to_evict */, - stat, eviction_candidates, cache_lock)) + stat, eviction_candidates, cache_lock) == IFileCachePriority::DesiredSizeStatus::SUCCESS) { if (eviction_candidates.size() == 0) { diff --git a/src/Interpreters/Cache/IFileCachePriority.h b/src/Interpreters/Cache/IFileCachePriority.h index 5d8eb9dd54a..9885ab00f78 100644 --- a/src/Interpreters/Cache/IFileCachePriority.h +++ b/src/Interpreters/Cache/IFileCachePriority.h @@ -151,7 +151,13 @@ public: /// and `desired_elements_num` as current cache state. /// Collect no more than `max_candidates_to_evict` elements. /// Return `true` if the first condition is satisfied. - virtual bool collectCandidatesForEviction( + enum class DesiredSizeStatus + { + SUCCESS, + CANNOT_EVICT, + REACHED_MAX_CANDIDATES_LIMIT, + }; + virtual DesiredSizeStatus collectCandidatesForEviction( size_t desired_size, size_t desired_elements_count, size_t max_candidates_to_evict, diff --git a/src/Interpreters/Cache/LRUFileCachePriority.cpp b/src/Interpreters/Cache/LRUFileCachePriority.cpp index ec96eb14a8a..7970eaa3e13 100644 --- a/src/Interpreters/Cache/LRUFileCachePriority.cpp +++ b/src/Interpreters/Cache/LRUFileCachePriority.cpp @@ -323,7 +323,7 @@ bool LRUFileCachePriority::collectCandidatesForEviction( } } -bool LRUFileCachePriority::collectCandidatesForEviction( +IFileCachePriority::DesiredSizeStatus LRUFileCachePriority::collectCandidatesForEviction( size_t desired_size, size_t desired_elements_count, size_t max_candidates_to_evict, @@ -336,12 +336,24 @@ bool LRUFileCachePriority::collectCandidatesForEviction( return canFit(0, 0, stat.total_stat.releasable_size, stat.total_stat.releasable_count, lock, &desired_size, &desired_elements_count); }; + auto status = DesiredSizeStatus::CANNOT_EVICT; auto stop_condition = [&]() { - return desired_limits_satisfied() || (max_candidates_to_evict && res.size() >= max_candidates_to_evict); + if (desired_limits_satisfied()) + { + status = DesiredSizeStatus::SUCCESS; + return true; + } + if (max_candidates_to_evict && res.size() >= max_candidates_to_evict) + { + status = DesiredSizeStatus::REACHED_MAX_CANDIDATES_LIMIT; + return true; + } + return false; }; iterateForEviction(res, stat, stop_condition, lock); - return desired_limits_satisfied(); + chassert(status != DesiredSizeStatus::SUCCESS || stop_condition()); + return status; } void LRUFileCachePriority::iterateForEviction( @@ -350,6 +362,9 @@ void LRUFileCachePriority::iterateForEviction( StopConditionFunc stop_condition, const CachePriorityGuard::Lock & lock) { + if (stop_condition()) + return; + ProfileEvents::increment(ProfileEvents::FilesystemCacheEvictionTries); IterateFunc iterate_func = [&](LockedKey & locked_key, const FileSegmentMetadataPtr & segment_metadata) diff --git a/src/Interpreters/Cache/LRUFileCachePriority.h b/src/Interpreters/Cache/LRUFileCachePriority.h index e0691cade43..9bced106727 100644 --- a/src/Interpreters/Cache/LRUFileCachePriority.h +++ b/src/Interpreters/Cache/LRUFileCachePriority.h @@ -63,7 +63,7 @@ public: const UserID & user_id, const CachePriorityGuard::Lock &) override; - bool collectCandidatesForEviction( + DesiredSizeStatus collectCandidatesForEviction( size_t desired_size, size_t desired_elements_count, size_t max_candidates_to_evict, diff --git a/src/Interpreters/Cache/SLRUFileCachePriority.cpp b/src/Interpreters/Cache/SLRUFileCachePriority.cpp index 7a3fdf5160e..dc0df223cb0 100644 --- a/src/Interpreters/Cache/SLRUFileCachePriority.cpp +++ b/src/Interpreters/Cache/SLRUFileCachePriority.cpp @@ -256,7 +256,7 @@ bool SLRUFileCachePriority::collectCandidatesForEvictionInProtected( return true; } -bool SLRUFileCachePriority::collectCandidatesForEviction( +IFileCachePriority::DesiredSizeStatus SLRUFileCachePriority::collectCandidatesForEviction( size_t desired_size, size_t desired_elements_count, size_t max_candidates_to_evict, @@ -268,7 +268,7 @@ bool SLRUFileCachePriority::collectCandidatesForEviction( const auto desired_probationary_elements_num = getRatio(desired_elements_count, 1 - size_ratio); FileCacheReserveStat probationary_stat; - const bool probationary_limit_satisfied = probationary_queue.collectCandidatesForEviction( + const auto probationary_desired_size_status = probationary_queue.collectCandidatesForEviction( desired_probationary_size, desired_probationary_elements_num, max_candidates_to_evict, probationary_stat, res, lock); @@ -285,14 +285,14 @@ bool SLRUFileCachePriority::collectCandidatesForEviction( chassert(!max_candidates_to_evict || res.size() <= max_candidates_to_evict); chassert(res.size() == stat.total_stat.releasable_count); - if (max_candidates_to_evict && res.size() >= max_candidates_to_evict) - return probationary_limit_satisfied; + if (probationary_desired_size_status == DesiredSizeStatus::REACHED_MAX_CANDIDATES_LIMIT) + return probationary_desired_size_status; const auto desired_protected_size = getRatio(desired_size, size_ratio); const auto desired_protected_elements_num = getRatio(desired_elements_count, size_ratio); FileCacheReserveStat protected_stat; - const bool protected_limit_satisfied = protected_queue.collectCandidatesForEviction( + const auto protected_desired_size_status = protected_queue.collectCandidatesForEviction( desired_protected_size, desired_protected_elements_num, max_candidates_to_evict - res.size(), protected_stat, res, lock); @@ -306,7 +306,10 @@ bool SLRUFileCachePriority::collectCandidatesForEviction( desired_protected_size, desired_protected_elements_num, protected_queue.getStateInfoForLog(lock)); - return probationary_limit_satisfied && protected_limit_satisfied; + if (probationary_desired_size_status == DesiredSizeStatus::SUCCESS) + return protected_desired_size_status; + else + return probationary_desired_size_status; } void SLRUFileCachePriority::downgrade(IteratorPtr iterator, const CachePriorityGuard::Lock & lock) diff --git a/src/Interpreters/Cache/SLRUFileCachePriority.h b/src/Interpreters/Cache/SLRUFileCachePriority.h index 2102a0ec558..e6d20e0d0ee 100644 --- a/src/Interpreters/Cache/SLRUFileCachePriority.h +++ b/src/Interpreters/Cache/SLRUFileCachePriority.h @@ -58,7 +58,7 @@ public: const UserID & user_id, const CachePriorityGuard::Lock &) override; - bool collectCandidatesForEviction( + DesiredSizeStatus collectCandidatesForEviction( size_t desired_size, size_t desired_elements_count, size_t max_candidates_to_evict, From 743d63767a74f41b3628c52ccf166be773baecf2 Mon Sep 17 00:00:00 2001 From: serxa Date: Wed, 31 Jul 2024 17:06:49 +0000 Subject: [PATCH 079/227] fix AsyncLoader destruction race --- src/Common/AsyncLoader.cpp | 33 ++++++++++++++++++++------------- 1 file changed, 20 insertions(+), 13 deletions(-) diff --git a/src/Common/AsyncLoader.cpp b/src/Common/AsyncLoader.cpp index 6264eb03106..d40e320e741 100644 --- a/src/Common/AsyncLoader.cpp +++ b/src/Common/AsyncLoader.cpp @@ -218,20 +218,27 @@ AsyncLoader::~AsyncLoader() { // All `LoadTask` objects should be destructed before AsyncLoader destruction because they hold a reference. // To make sure we check for all pending jobs to be finished. - std::unique_lock lock{mutex}; - if (scheduled_jobs.empty() && finished_jobs.empty()) - return; + { + std::unique_lock lock{mutex}; + if (!scheduled_jobs.empty() || !finished_jobs.empty()) + { + std::vector scheduled; + std::vector finished; + scheduled.reserve(scheduled_jobs.size()); + finished.reserve(finished_jobs.size()); + for (const auto & [job, _] : scheduled_jobs) + scheduled.push_back(job->name); + for (const auto & job : finished_jobs) + finished.push_back(job->name); + LOG_ERROR(log, "Bug. Destruction with pending ({}) and finished ({}) load jobs.", fmt::join(scheduled, ", "), fmt::join(finished, ", ")); + abort(); + } + } - std::vector scheduled; - std::vector finished; - scheduled.reserve(scheduled_jobs.size()); - finished.reserve(finished_jobs.size()); - for (const auto & [job, _] : scheduled_jobs) - scheduled.push_back(job->name); - for (const auto & job : finished_jobs) - finished.push_back(job->name); - LOG_ERROR(log, "Bug. Destruction with pending ({}) and finished ({}) load jobs.", fmt::join(scheduled, ", "), fmt::join(finished, ", ")); - abort(); + // When all jobs are done we could still have finalizing workers. + // These workers could call updateCurrentPriorityAndSpawn() that scans all pools. + // We need to stop all of them before destructing any of them. + stop(); } void AsyncLoader::start() From 2b79da36c0701bb9ca392fddd9129a7e0e04ef3f Mon Sep 17 00:00:00 2001 From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com> Date: Wed, 31 Jul 2024 19:26:45 +0200 Subject: [PATCH 080/227] Update 01605_adaptive_granularity_block_borders.sql --- .../0_stateless/01605_adaptive_granularity_block_borders.sql | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/01605_adaptive_granularity_block_borders.sql b/tests/queries/0_stateless/01605_adaptive_granularity_block_borders.sql index 5f09dc423b2..f9b8bb1c1c6 100644 --- a/tests/queries/0_stateless/01605_adaptive_granularity_block_borders.sql +++ b/tests/queries/0_stateless/01605_adaptive_granularity_block_borders.sql @@ -1,4 +1,4 @@ --- Tags: no-random-merge-tree-settings, no-tsan, no-debug, no-object-storage, no-distributed-cache +-- Tags: long, no-random-merge-tree-settings, no-tsan, no-debug, no-object-storage, no-distributed-cache -- no-tsan: too slow -- no-object-storage: for remote tables we use thread pool even when reading with one stream, so memory consumption is higher From bfb0133f26f85e64f2d608595b653c20a594abca Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Wed, 31 Jul 2024 17:38:20 +0200 Subject: [PATCH 081/227] Minor change --- src/Databases/DatabaseOnDisk.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Databases/DatabaseOnDisk.cpp b/src/Databases/DatabaseOnDisk.cpp index f419f5811a1..734f354d9a5 100644 --- a/src/Databases/DatabaseOnDisk.cpp +++ b/src/Databases/DatabaseOnDisk.cpp @@ -313,7 +313,7 @@ void DatabaseOnDisk::detachTablePermanently(ContextPtr query_context, const Stri std::lock_guard lock(mutex); if (const auto it = snapshot_detached_tables.find(table_name); it == snapshot_detached_tables.end()) { - throw Exception(ErrorCodes::LOGICAL_ERROR, "Snapshot doesn't contain info about detached table={}", table_name); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Snapshot doesn't contain info about detached table `{}`", table_name); } else { From 2c6c5c5c94a2f4131d898b172d6af285eec07c2f Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Wed, 31 Jul 2024 19:31:22 +0200 Subject: [PATCH 082/227] Add an assertion --- src/Common/TimerDescriptor.cpp | 3 +++ 1 file changed, 3 insertions(+) diff --git a/src/Common/TimerDescriptor.cpp b/src/Common/TimerDescriptor.cpp index 9a171ae9487..716dcee2747 100644 --- a/src/Common/TimerDescriptor.cpp +++ b/src/Common/TimerDescriptor.cpp @@ -110,6 +110,9 @@ void TimerDescriptor::drain() const throw ErrnoException(ErrorCodes::CANNOT_READ_FROM_SOCKET, "Cannot readlink for a timer_fd {}", timer_fd); LOG_TRACE(log, "Received EINTR while trying to drain a TimerDescriptor, fd {}: {}", timer_fd, std::string_view(link_path, link_path_length)); + + /// Check that it's actually a timerfd. + chassert(std::string_view(link_path, link_path_length).contains("timerfd")); continue; } From 26a81e366d722314bbda12d5232fa49117ef4498 Mon Sep 17 00:00:00 2001 From: shiyer7474 Date: Wed, 31 Jul 2024 18:05:53 +0000 Subject: [PATCH 083/227] Fix the serialization of parameters for parameterized views Removed the call to convertFieldToString() and added datatype specific serialization code. Parameterized view substitution was broken for multiple datatypes when parameter value was a function or expression returning datatype instance. Testcase added to cover Date/Date32/UUID/IP datatypes. --- src/Analyzer/Resolve/QueryAnalyzer.cpp | 10 +- ...zed_view_with_non_literal_params.reference | 31 ++++++ ...meterized_view_with_non_literal_params.sql | 97 +++++++++++++++++++ 3 files changed, 137 insertions(+), 1 deletion(-) create mode 100644 tests/queries/0_stateless/03209_parameterized_view_with_non_literal_params.reference create mode 100644 tests/queries/0_stateless/03209_parameterized_view_with_non_literal_params.sql diff --git a/src/Analyzer/Resolve/QueryAnalyzer.cpp b/src/Analyzer/Resolve/QueryAnalyzer.cpp index 767d5c11075..6113a38d463 100644 --- a/src/Analyzer/Resolve/QueryAnalyzer.cpp +++ b/src/Analyzer/Resolve/QueryAnalyzer.cpp @@ -4546,7 +4546,15 @@ void QueryAnalyzer::resolveTableFunction(QueryTreeNodePtr & table_function_node, resolveExpressionNode(nodes[1], scope, /* allow_lambda_expression */false, /* allow_table_function */false); if (auto * constant = nodes[1]->as()) { - view_params[identifier_node->getIdentifier().getFullName()] = convertFieldToString(constant->getValue()); + /// Serialize the constant value using datatype specific + /// interfaces to match the deserialization in ReplaceQueryParametersVistor. + WriteBufferFromOwnString buf; + auto constval = constant->getValue(); + auto realtype = constant->getResultType(); + auto tempcol = realtype->createColumn(); + tempcol->insert(constval); + realtype->getDefaultSerialization()->serializeTextEscaped(*tempcol, 0, buf, {}); + view_params[identifier_node->getIdentifier().getFullName()] = buf.str(); } } } diff --git a/tests/queries/0_stateless/03209_parameterized_view_with_non_literal_params.reference b/tests/queries/0_stateless/03209_parameterized_view_with_non_literal_params.reference new file mode 100644 index 00000000000..e4e6c313b85 --- /dev/null +++ b/tests/queries/0_stateless/03209_parameterized_view_with_non_literal_params.reference @@ -0,0 +1,31 @@ +Test with Date parameter +1 +2 +1 +3 +3 +3 +2 +Test with Date32 parameter +1 +2 +1 +3 +5 +3 +4 +Test with UUID parameter +4 +3 +3 +1 +2 +Test with 2 parameters +1 +1 +3 +3 +Test with IPv4 +1 +2 +3 diff --git a/tests/queries/0_stateless/03209_parameterized_view_with_non_literal_params.sql b/tests/queries/0_stateless/03209_parameterized_view_with_non_literal_params.sql new file mode 100644 index 00000000000..55795c7a785 --- /dev/null +++ b/tests/queries/0_stateless/03209_parameterized_view_with_non_literal_params.sql @@ -0,0 +1,97 @@ + +select 'Test with Date parameter'; + +drop table if exists date_table_pv; +create table date_table_pv (id Int32, dt Date) engine = Memory(); + +insert into date_table_pv values(1, today()); +insert into date_table_pv values(2, yesterday()); +insert into date_table_pv values(3, toDate('1974-04-07')); + +drop view if exists date_pv; +create view date_pv as select * from date_table_pv where dt = {dtparam:Date}; + +select id from date_pv(dtparam=today()); +select id from date_pv(dtparam=yesterday()); +select id from date_pv(dtparam=yesterday()+1); +select id from date_pv(dtparam='1974-04-07'); +select id from date_pv(dtparam=toDate('1974-04-07')); +select id from date_pv(dtparam=toString(toDate('1974-04-07'))); +select id from date_pv(dtparam=toDate('1975-04-07')); +select id from date_pv(dtparam=(select dt from date_table_pv where id = 2)); + +select 'Test with Date32 parameter'; + +drop table if exists date32_table_pv; +create table date32_table_pv (id Int32, dt Date32) engine = Memory(); + +insert into date32_table_pv values(1, today()); +insert into date32_table_pv values(2, yesterday()); +insert into date32_table_pv values(3, toDate32('2199-12-31')); +insert into date32_table_pv values(4, toDate32('1950-12-25')); +insert into date32_table_pv values(5, toDate32('1900-01-01')); + +drop view if exists date32_pv; +create view date32_pv as select * from date32_table_pv where dt = {dtparam:Date32}; + +select id from date32_pv(dtparam=today()); +select id from date32_pv(dtparam=yesterday()); +select id from date32_pv(dtparam=yesterday()+1); +select id from date32_pv(dtparam='2199-12-31'); +select id from date32_pv(dtparam=toDate32('1900-01-01')); +select id from date32_pv(dtparam=(select dt from date32_table_pv where id = 3)); +select id from date32_pv(dtparam=(select dt from date32_table_pv where id = 4)); + + +select 'Test with UUID parameter'; +drop table if exists uuid_table_pv; +create table uuid_table_pv (id Int32, uu UUID) engine = Memory(); + +insert into uuid_table_pv values(1, generateUUIDv4()); +insert into uuid_table_pv values(2, generateUUIDv7()); +insert into uuid_table_pv values(3, toUUID('11111111-2222-3333-4444-555555555555')); +insert into uuid_table_pv select 4, serverUUID(); + + +drop view if exists uuid_pv; +create view uuid_pv as select * from uuid_table_pv where uu = {uuidparam:UUID}; +select id from uuid_pv(uuidparam=serverUUID()); +select id from uuid_pv(uuidparam=toUUID('11111111-2222-3333-4444-555555555555')); +select id from uuid_pv(uuidparam='11111111-2222-3333-4444-555555555555'); +select id from uuid_pv(uuidparam=(select uu from uuid_table_pv where id = 1)); +select id from uuid_pv(uuidparam=(select uu from uuid_table_pv where id = 2)); +-- generateUUIDv4() is not constant foldable, hence cannot be used as parameter value +select id from uuid_pv(uuidparam=generateUUIDv4()); -- { serverError UNKNOWN_QUERY_PARAMETER } +-- But nested "select generateUUIDv4()" works! +select id from uuid_pv(uuidparam=(select generateUUIDv4())); + +select 'Test with 2 parameters'; + +drop view if exists date_pv2; +create view date_pv2 as select * from date_table_pv where dt = {dtparam:Date} and id = {intparam:Int32}; +select id from date_pv2(dtparam=today(),intparam=1); +select id from date_pv2(dtparam=today(),intparam=length('A')); +select id from date_pv2(dtparam='1974-04-07',intparam=length('AAA')); +select id from date_pv2(dtparam=toDate('1974-04-07'),intparam=length('BBB')); + +select 'Test with IPv4'; + +drop table if exists ipv4_table_pv; +create table ipv4_table_pv (id Int32, ipaddr IPv4) ENGINE = Memory(); +insert into ipv4_table_pv values (1, '116.106.34.242'); +insert into ipv4_table_pv values (2, '116.106.34.243'); +insert into ipv4_table_pv values (3, '116.106.34.244'); + +drop view if exists ipv4_pv; +create view ipv4_pv as select * from ipv4_table_pv where ipaddr = {ipv4param:IPv4}; +select id from ipv4_pv(ipv4param='116.106.34.242'); +select id from ipv4_pv(ipv4param=toIPv4('116.106.34.243')); +select id from ipv4_pv(ipv4param=(select ipaddr from ipv4_table_pv where id=3)); + +drop view date_pv; +drop view date_pv2; +drop view uuid_pv; +drop view ipv4_pv; +drop table date_table_pv; +drop table uuid_table_pv; +drop table ipv4_table_pv; From dde274f6fad979aa94ea31395b0434c81f72328a Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Wed, 31 Jul 2024 18:08:14 +0000 Subject: [PATCH 084/227] Re-enable ICU on s390/x --- contrib/icu-cmake/CMakeLists.txt | 4 +--- contrib/icudata | 2 +- 2 files changed, 2 insertions(+), 4 deletions(-) diff --git a/contrib/icu-cmake/CMakeLists.txt b/contrib/icu-cmake/CMakeLists.txt index f9d05f7fe97..adeaa7dcf33 100644 --- a/contrib/icu-cmake/CMakeLists.txt +++ b/contrib/icu-cmake/CMakeLists.txt @@ -4,9 +4,7 @@ else () option(ENABLE_ICU "Enable ICU" 0) endif () -# Temporarily disabled s390x because the ICU build links a blob (icudt71b_dat.S) and our friends from IBM did not explain how they generated -# the blob on s390x: https://github.com/ClickHouse/icudata/pull/2#issuecomment-2226957255 -if (NOT ENABLE_ICU OR ARCH_S390X) +if (NOT ENABLE_ICU) message(STATUS "Not using ICU") return() endif() diff --git a/contrib/icudata b/contrib/icudata index d345d6ac22f..4904951339a 160000 --- a/contrib/icudata +++ b/contrib/icudata @@ -1 +1 @@ -Subproject commit d345d6ac22f381c882420de9053d30ae1ff38d75 +Subproject commit 4904951339a70b4814d2d3723436b20d079cb01b From e2af1766eb1ea0ee2f6b862f53a0d3c13f53365b Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Wed, 31 Jul 2024 20:25:28 +0200 Subject: [PATCH 085/227] init --- src/Functions/DateTimeTransforms.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/src/Functions/DateTimeTransforms.h b/src/Functions/DateTimeTransforms.h index a7bd398cdaa..fe26c5cf353 100644 --- a/src/Functions/DateTimeTransforms.h +++ b/src/Functions/DateTimeTransforms.h @@ -381,11 +381,13 @@ struct ToStartOfWeekImpl static UInt16 execute(Int64 t, UInt8 week_mode, const DateLUTImpl & time_zone) { - return time_zone.toFirstDayNumOfWeek(time_zone.toDayNum(t), week_mode); + const auto & res = time_zone.toFirstDayNumOfWeek(time_zone.toDayNum(t), week_mode); + return res >= 0 ? res : 0; } static UInt16 execute(UInt32 t, UInt8 week_mode, const DateLUTImpl & time_zone) { - return time_zone.toFirstDayNumOfWeek(time_zone.toDayNum(t), week_mode); + const auto & res = time_zone.toFirstDayNumOfWeek(time_zone.toDayNum(t), week_mode); + return res >= 0 ? res : 0; } static UInt16 execute(Int32 d, UInt8 week_mode, const DateLUTImpl & time_zone) { From d0c643180f408d84a5f10a917f413248b9267202 Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Wed, 31 Jul 2024 20:29:36 +0200 Subject: [PATCH 086/227] add tests --- .../03215_toStartOfWeek_with_dateTime64_fix.reference | 2 ++ .../0_stateless/03215_toStartOfWeek_with_dateTime64_fix.sql | 2 ++ 2 files changed, 4 insertions(+) create mode 100644 tests/queries/0_stateless/03215_toStartOfWeek_with_dateTime64_fix.reference create mode 100644 tests/queries/0_stateless/03215_toStartOfWeek_with_dateTime64_fix.sql diff --git a/tests/queries/0_stateless/03215_toStartOfWeek_with_dateTime64_fix.reference b/tests/queries/0_stateless/03215_toStartOfWeek_with_dateTime64_fix.reference new file mode 100644 index 00000000000..fd698107f22 --- /dev/null +++ b/tests/queries/0_stateless/03215_toStartOfWeek_with_dateTime64_fix.reference @@ -0,0 +1,2 @@ +1970-01-01 +1970-01-01 diff --git a/tests/queries/0_stateless/03215_toStartOfWeek_with_dateTime64_fix.sql b/tests/queries/0_stateless/03215_toStartOfWeek_with_dateTime64_fix.sql new file mode 100644 index 00000000000..0f00a52cb86 --- /dev/null +++ b/tests/queries/0_stateless/03215_toStartOfWeek_with_dateTime64_fix.sql @@ -0,0 +1,2 @@ +SELECT toStartOfWeek(toDateTime64('1970-02-01', 6)); +SELECT toStartOfWeek(toDateTime('1970-01-01')); From 636c3f642340de6e5ca4892481ca156cb236a4cd Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Wed, 31 Jul 2024 20:31:22 +0200 Subject: [PATCH 087/227] Update DateTimeTransforms.h --- src/Functions/DateTimeTransforms.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/Functions/DateTimeTransforms.h b/src/Functions/DateTimeTransforms.h index fe26c5cf353..1970ec3bdb0 100644 --- a/src/Functions/DateTimeTransforms.h +++ b/src/Functions/DateTimeTransforms.h @@ -382,12 +382,12 @@ struct ToStartOfWeekImpl static UInt16 execute(Int64 t, UInt8 week_mode, const DateLUTImpl & time_zone) { const auto & res = time_zone.toFirstDayNumOfWeek(time_zone.toDayNum(t), week_mode); - return res >= 0 ? res : 0; + return std::max(res, 0); } static UInt16 execute(UInt32 t, UInt8 week_mode, const DateLUTImpl & time_zone) { const auto & res = time_zone.toFirstDayNumOfWeek(time_zone.toDayNum(t), week_mode); - return res >= 0 ? res : 0; + return std::max(res, 0); } static UInt16 execute(Int32 d, UInt8 week_mode, const DateLUTImpl & time_zone) { From 34ca3128ed0de60e03a105c43dc0924541f4a2c1 Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Wed, 31 Jul 2024 20:36:13 +0200 Subject: [PATCH 088/227] Update DateTimeTransforms.h --- src/Functions/DateTimeTransforms.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/Functions/DateTimeTransforms.h b/src/Functions/DateTimeTransforms.h index 1970ec3bdb0..46fb3bb9f57 100644 --- a/src/Functions/DateTimeTransforms.h +++ b/src/Functions/DateTimeTransforms.h @@ -381,12 +381,12 @@ struct ToStartOfWeekImpl static UInt16 execute(Int64 t, UInt8 week_mode, const DateLUTImpl & time_zone) { - const auto & res = time_zone.toFirstDayNumOfWeek(time_zone.toDayNum(t), week_mode); + const auto res = time_zone.toFirstDayNumOfWeek(time_zone.toDayNum(t), week_mode); return std::max(res, 0); } static UInt16 execute(UInt32 t, UInt8 week_mode, const DateLUTImpl & time_zone) { - const auto & res = time_zone.toFirstDayNumOfWeek(time_zone.toDayNum(t), week_mode); + const auto res = time_zone.toFirstDayNumOfWeek(time_zone.toDayNum(t), week_mode); return std::max(res, 0); } static UInt16 execute(Int32 d, UInt8 week_mode, const DateLUTImpl & time_zone) From cd3c6c3ae807a321d94079087eb1fb29f4764549 Mon Sep 17 00:00:00 2001 From: sakulali Date: Thu, 1 Aug 2024 03:37:45 +0800 Subject: [PATCH 089/227] try to fix flaky test 01889_clickhouse_client_config_format --- .../01889_clickhouse_client_config_format.reference | 2 +- .../0_stateless/01889_clickhouse_client_config_format.sh | 6 ++++-- 2 files changed, 5 insertions(+), 3 deletions(-) diff --git a/tests/queries/0_stateless/01889_clickhouse_client_config_format.reference b/tests/queries/0_stateless/01889_clickhouse_client_config_format.reference index ef0d9ffc538..2575200e6fa 100644 --- a/tests/queries/0_stateless/01889_clickhouse_client_config_format.reference +++ b/tests/queries/0_stateless/01889_clickhouse_client_config_format.reference @@ -19,6 +19,6 @@ autodetect xml (non leading whitespaces) autodetect yaml 2 autodetect invalid xml -Code: 1000, e.code() = 0, SAXParseException: Invalid token in '/config_test.badxml', line 2 column 12, Stack trace (when copying this message, always include the lines below): +Correct: invalid xml parsed with exception autodetect invalid yaml Code: 585. Unable to parse YAML configuration file /config_test.badyaml, yaml-cpp: error at line 2, column 12: illegal map value. (CANNOT_PARSE_YAML) diff --git a/tests/queries/0_stateless/01889_clickhouse_client_config_format.sh b/tests/queries/0_stateless/01889_clickhouse_client_config_format.sh index 9a44ec0d5f5..58fd6852116 100755 --- a/tests/queries/0_stateless/01889_clickhouse_client_config_format.sh +++ b/tests/queries/0_stateless/01889_clickhouse_client_config_format.sh @@ -108,7 +108,9 @@ echo 'autodetect xml (non leading whitespaces)' $CLICKHOUSE_CLIENT --config "$autodetect_xml_non_leading_whitespace_config" -q "select getSetting('max_threads')" echo 'autodetect yaml' $CLICKHOUSE_CLIENT --config "$autodetect_yaml_config" -q "select getSetting('max_threads')" + +# Error code is 1000 (Poco::Exception). It is not ignored. echo 'autodetect invalid xml' -$CLICKHOUSE_CLIENT --config "$autodetect_invalid_xml_config" -q "select getSetting('max_threads')" 2>&1 |& sed -n '1p' | sed -e "s#$CLICKHOUSE_TMP##" -e "s#Poco::Exception. ##" +$CLICKHOUSE_CLIENT --config "$autodetect_invalid_xml_config" -q "select getSetting('max_threads')" 2>&1 |& grep -q "Code: 1000" && echo "Correct: invalid xml parsed with exception" || echo 'Fail: expected error code 1000 but got other' echo 'autodetect invalid yaml' -$CLICKHOUSE_CLIENT --config "$autodetect_invalid_yaml_config" -q "select getSetting('max_threads')" 2>&1 |& sed -n '1p' | sed -e "s#$CLICKHOUSE_TMP##" -e "s#DB::Exception: ##" \ No newline at end of file +$CLICKHOUSE_CLIENT --config "$autodetect_invalid_yaml_config" -q "select getSetting('max_threads')" 2>&1 |& sed -e "s#$CLICKHOUSE_TMP##" -e "s#DB::Exception: ##" \ No newline at end of file From cc27c254abd4b6fd8f64b47e0bdf6195041bd5ef Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Wed, 31 Jul 2024 17:38:20 +0200 Subject: [PATCH 090/227] Minor change --- src/Databases/DatabaseOnDisk.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Databases/DatabaseOnDisk.cpp b/src/Databases/DatabaseOnDisk.cpp index f419f5811a1..734f354d9a5 100644 --- a/src/Databases/DatabaseOnDisk.cpp +++ b/src/Databases/DatabaseOnDisk.cpp @@ -313,7 +313,7 @@ void DatabaseOnDisk::detachTablePermanently(ContextPtr query_context, const Stri std::lock_guard lock(mutex); if (const auto it = snapshot_detached_tables.find(table_name); it == snapshot_detached_tables.end()) { - throw Exception(ErrorCodes::LOGICAL_ERROR, "Snapshot doesn't contain info about detached table={}", table_name); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Snapshot doesn't contain info about detached table `{}`", table_name); } else { From 867784d55c989943d0c79eb9179b01e878fabcbe Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Wed, 31 Jul 2024 22:48:16 +0200 Subject: [PATCH 091/227] Update DateTimeTransforms.h --- src/Functions/DateTimeTransforms.h | 1 + 1 file changed, 1 insertion(+) diff --git a/src/Functions/DateTimeTransforms.h b/src/Functions/DateTimeTransforms.h index 46fb3bb9f57..ce7da406e9a 100644 --- a/src/Functions/DateTimeTransforms.h +++ b/src/Functions/DateTimeTransforms.h @@ -1,5 +1,6 @@ #pragma once +#include #include #include #include From 9cb52bd1381ad3e0929062801df7c4b542cf1117 Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Wed, 31 Jul 2024 23:11:35 +0200 Subject: [PATCH 092/227] fix build --- src/Functions/DateTimeTransforms.h | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/src/Functions/DateTimeTransforms.h b/src/Functions/DateTimeTransforms.h index ce7da406e9a..15f1b9580f3 100644 --- a/src/Functions/DateTimeTransforms.h +++ b/src/Functions/DateTimeTransforms.h @@ -1,6 +1,5 @@ #pragma once -#include #include #include #include @@ -25,7 +24,7 @@ namespace DB static constexpr auto millisecond_multiplier = 1'000; static constexpr auto microsecond_multiplier = 1'000'000; -static constexpr auto nanosecond_multiplier = 1'000'000'000; +static constexpr auto nanosecond_multiplier = 1'000'000'000; static constexpr FormatSettings::DateTimeOverflowBehavior default_date_time_overflow_behavior = FormatSettings::DateTimeOverflowBehavior::Ignore; @@ -382,12 +381,12 @@ struct ToStartOfWeekImpl static UInt16 execute(Int64 t, UInt8 week_mode, const DateLUTImpl & time_zone) { - const auto res = time_zone.toFirstDayNumOfWeek(time_zone.toDayNum(t), week_mode); + const int res = time_zone.toFirstDayNumOfWeek(time_zone.toDayNum(t), week_mode); return std::max(res, 0); } static UInt16 execute(UInt32 t, UInt8 week_mode, const DateLUTImpl & time_zone) { - const auto res = time_zone.toFirstDayNumOfWeek(time_zone.toDayNum(t), week_mode); + const int res = time_zone.toFirstDayNumOfWeek(time_zone.toDayNum(t), week_mode); return std::max(res, 0); } static UInt16 execute(Int32 d, UInt8 week_mode, const DateLUTImpl & time_zone) From 6403f3f545bee153ffaf4ce5bda6fcde33ef88d2 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Wed, 31 Jul 2024 23:29:54 +0200 Subject: [PATCH 093/227] Miscellaneous --- src/Common/Epoll.cpp | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/src/Common/Epoll.cpp b/src/Common/Epoll.cpp index 49c86222cf0..ef7c6e143a0 100644 --- a/src/Common/Epoll.cpp +++ b/src/Common/Epoll.cpp @@ -19,7 +19,7 @@ Epoll::Epoll() : events_count(0) { epoll_fd = epoll_create1(0); if (epoll_fd == -1) - throw DB::ErrnoException(DB::ErrorCodes::EPOLL_ERROR, "Cannot open epoll descriptor"); + throw ErrnoException(ErrorCodes::EPOLL_ERROR, "Cannot open epoll descriptor"); } Epoll::Epoll(Epoll && other) noexcept : epoll_fd(other.epoll_fd), events_count(other.events_count.load()) @@ -47,7 +47,7 @@ void Epoll::add(int fd, void * ptr, uint32_t events) ++events_count; if (epoll_ctl(epoll_fd, EPOLL_CTL_ADD, fd, &event) == -1) - throw DB::ErrnoException(DB::ErrorCodes::EPOLL_ERROR, "Cannot add new descriptor to epoll"); + throw ErrnoException(ErrorCodes::EPOLL_ERROR, "Cannot add new descriptor to epoll"); } void Epoll::remove(int fd) @@ -55,7 +55,7 @@ void Epoll::remove(int fd) --events_count; if (epoll_ctl(epoll_fd, EPOLL_CTL_DEL, fd, nullptr) == -1) - throw DB::ErrnoException(DB::ErrorCodes::EPOLL_ERROR, "Cannot remove descriptor from epoll"); + throw ErrnoException(ErrorCodes::EPOLL_ERROR, "Cannot remove descriptor from epoll"); } size_t Epoll::getManyReady(int max_events, epoll_event * events_out, int timeout) const @@ -82,7 +82,7 @@ size_t Epoll::getManyReady(int max_events, epoll_event * events_out, int timeout continue; } else - throw DB::ErrnoException(DB::ErrorCodes::EPOLL_ERROR, "Error in epoll_wait"); + throw ErrnoException(ErrorCodes::EPOLL_ERROR, "Error in epoll_wait"); } else break; From 0e36db543762cb146aa6c233d4536fd62c6101b2 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?= Date: Wed, 31 Jul 2024 22:09:04 +0000 Subject: [PATCH 094/227] Do not check the value of `num_messages_read` because it is not reliable librdkafka usually polls both messages on the first SELECT, but sometimes only one. If it polls only one message first, then it will read two messages before rebalancing at the second SELECT from the table. This means it usually reads a single message twice (thus num_messages_read = 4 is usually fine as 1 discarded message + 3 actually consumed message). But when only one message is read in the first SELECT, then 2 messages are discarded, thus num_messages_read will be 5 as 2 discarded message + 3 actually consumed messages. --- tests/integration/test_storage_kafka/test.py | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/tests/integration/test_storage_kafka/test.py b/tests/integration/test_storage_kafka/test.py index 8393e88db88..37457e00701 100644 --- a/tests/integration/test_storage_kafka/test.py +++ b/tests/integration/test_storage_kafka/test.py @@ -4771,7 +4771,7 @@ def test_system_kafka_consumers_rebalance(kafka_cluster, max_retries=15): assignments.current_offset, if(length(exceptions.time)>0, exceptions.time[1]::String, 'never') as last_exception_time_, if(length(exceptions.text)>0, exceptions.text[1], 'no exception') as last_exception_, - stable_timestamp(last_poll_time) as last_poll_time_, num_messages_read, stable_timestamp(last_commit_time) as last_commit_time_, + stable_timestamp(last_poll_time) as last_poll_time_, stable_timestamp(last_commit_time) as last_commit_time_, num_commits, stable_timestamp(last_rebalance_time) as last_rebalance_time_, num_rebalance_revocations, num_rebalance_assignments, is_currently_used FROM system.kafka_consumers WHERE database='test' and table IN ('kafka', 'kafka2') format Vertical; @@ -4791,7 +4791,6 @@ assignments.current_offset: [2] last_exception_time_: never last_exception_: no exception last_poll_time_: now -num_messages_read: 4 last_commit_time_: now num_commits: 2 last_rebalance_time_: now @@ -4810,7 +4809,6 @@ assignments.current_offset: [2] last_exception_time_: never last_exception_: no exception last_poll_time_: now -num_messages_read: 1 last_commit_time_: now num_commits: 1 last_rebalance_time_: never From 587d50380ad3d1cadc56bda0fa700e3441c16ab4 Mon Sep 17 00:00:00 2001 From: sakulali Date: Thu, 1 Aug 2024 09:12:00 +0800 Subject: [PATCH 095/227] fix clickhouse-test reference --- .../0_stateless/01889_clickhouse_client_config_format.reference | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/01889_clickhouse_client_config_format.reference b/tests/queries/0_stateless/01889_clickhouse_client_config_format.reference index 2575200e6fa..149315ad9d5 100644 --- a/tests/queries/0_stateless/01889_clickhouse_client_config_format.reference +++ b/tests/queries/0_stateless/01889_clickhouse_client_config_format.reference @@ -21,4 +21,4 @@ autodetect yaml autodetect invalid xml Correct: invalid xml parsed with exception autodetect invalid yaml -Code: 585. Unable to parse YAML configuration file /config_test.badyaml, yaml-cpp: error at line 2, column 12: illegal map value. (CANNOT_PARSE_YAML) +Code: 585. Unable to parse YAML configuration file /config_default.badyaml, yaml-cpp: error at line 2, column 12: illegal map value. (CANNOT_PARSE_YAML) From 5564489cca1c14e95e7c543e03c508849abaf079 Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Thu, 1 Aug 2024 15:31:54 +0800 Subject: [PATCH 096/227] change as request --- .../functions/tuple-map-functions.md | 18 +++++++++++--- src/Functions/map.cpp | 24 ++++++++++--------- .../0_stateless/01651_map_functions.reference | 4 ++++ .../0_stateless/01651_map_functions.sql | 7 +++++- 4 files changed, 38 insertions(+), 15 deletions(-) diff --git a/docs/en/sql-reference/functions/tuple-map-functions.md b/docs/en/sql-reference/functions/tuple-map-functions.md index db66188b1f5..d670ed42a2a 100644 --- a/docs/en/sql-reference/functions/tuple-map-functions.md +++ b/docs/en/sql-reference/functions/tuple-map-functions.md @@ -43,7 +43,7 @@ Result: ## mapFromArrays -Creates a map from an array of keys and an array of values. +Creates a map from an array or map of keys and an array or map of values. The function is a convenient alternative to syntax `CAST([...], 'Map(key_type, value_type)')`. For example, instead of writing @@ -62,8 +62,8 @@ Alias: `MAP_FROM_ARRAYS(keys, values)` **Arguments** -- `keys` — Array or map of keys to create the map from. [Array(T)](../data-types/array.md) where `T` can be any type supported by [Map](../data-types/map.md) as key type, or [Map](../data-types/map.md). -- `values` - Array or map of values to create the map from. [Array](../data-types/array.md) or [Map](../data-types/map.md). +- `keys` — Array or map of keys to create the map from [Array](../data-types/array.md) or [Map](../data-types/map.md). If `keys` is an array, we accept `Array(Nullable(T))` or `Array(LowCardinality(Nullable(T)))` as its type as long as it doesn't contain NULL value. +- `values` - Array or map of values to create the map from [Array](../data-types/array.md) or [Map](../data-types/map.md). **Returned value** @@ -99,6 +99,18 @@ Result: └───────────────────────────────────────────────────────┘ ``` +```sql +SELECT mapFromArrays(map('a', 1, 'b', 2, 'c', 3), [1, 2, 3]) +``` + +Result: + +``` +┌─mapFromArrays(map('a', 1, 'b', 2, 'c', 3), [1, 2, 3])─┐ +│ {('a',1):1,('b',2):2,('c',3):3} │ +└───────────────────────────────────────────────────────┘ +``` + ## extractKeyValuePairs Converts a string of key-value pairs to a [Map(String, String)](../data-types/map.md). diff --git a/src/Functions/map.cpp b/src/Functions/map.cpp index a8e5f7ad90e..738c61164a3 100644 --- a/src/Functions/map.cpp +++ b/src/Functions/map.cpp @@ -23,6 +23,7 @@ namespace ErrorCodes extern const int ILLEGAL_TYPE_OF_ARGUMENT; extern const int SIZES_OF_ARRAYS_DONT_MATCH; extern const int ILLEGAL_COLUMN; + extern const int BAD_ARGUMENTS; } namespace @@ -157,7 +158,7 @@ private: bool use_variant_as_common_type = false; }; -/// mapFromArrays(keys, values) is a function that allows you to make key-value pair from a pair of arrays +/// mapFromArrays(keys, values) is a function that allows you to make key-value pair from a pair of arrays or maps class FunctionMapFromArrays : public IFunction { public: @@ -181,13 +182,13 @@ public: getName(), arguments.size()); - auto get_nested_type = [this](const DataTypePtr & type) -> DataTypePtr + auto get_nested_type = [&](const DataTypePtr & type) { DataTypePtr nested; - if (const auto * array_type = checkAndGetDataType(type.get())) - nested = array_type->getNestedType(); - else if (const auto * map_type = checkAndGetDataType(type.get())) - nested = std::make_shared(map_type->getKeyValueTypes()); + if (const auto * type_as_array = checkAndGetDataType(type.get())) + nested = type_as_array->getNestedType(); + else if (const auto * type_as_map = checkAndGetDataType(type.get())) + nested = std::make_shared(type_as_map->getKeyValueTypes()); else throw Exception( ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, @@ -201,8 +202,9 @@ public: auto key_type = get_nested_type(arguments[0]); auto value_type = get_nested_type(arguments[1]); - /// Remove Nullable from key_type if needed for map key must not be Nullable + /// We accept Array(Nullable(T)) or Array(LowCardinality(Nullable(T))) as key types as long as the actual array doesn't contain NULL value(this is checked in executeImpl). key_type = removeNullableOrLowCardinalityNullable(key_type); + DataTypes key_value_types{key_type, value_type}; return std::make_shared(key_value_types); } @@ -210,7 +212,7 @@ public: ColumnPtr executeImpl( const ColumnsWithTypeAndName & arguments, const DataTypePtr & /* result_type */, size_t /* input_rows_count */) const override { - auto get_array_column = [this](const ColumnPtr & column) -> std::pair + auto get_array_column = [&](const ColumnPtr & column) -> std::pair { bool is_const = isColumnConst(*column); ColumnPtr holder = is_const ? column->convertToFullColumnIfConst() : column; @@ -231,8 +233,9 @@ public: }; auto [col_keys, key_holder] = get_array_column(arguments[0].column); + auto [col_values, values_holder] = get_array_column(arguments[1].column); - /// Check if nested column of first argument contains NULL value in case its nested type is Nullable(T) type. + /// Nullable(T) or LowCardinality(Nullable(T)) are okay as nested key types but actual NULL values are not okay. ColumnPtr data_keys = col_keys->getDataPtr(); if (isColumnNullableOrLowCardinalityNullable(*data_keys)) { @@ -253,10 +256,9 @@ public: if (null_map && !memoryIsZero(null_map->data(), 0, null_map->size())) throw Exception( - ErrorCodes::ILLEGAL_COLUMN, "The nested column of first argument in function {} must not contain NULLs", getName()); + ErrorCodes::BAD_ARGUMENTS, "The nested column of first argument in function {} must not contain NULLs", getName()); } - auto [col_values, values_holder] = get_array_column(arguments[1].column); if (!col_keys->hasEqualOffsets(*col_values)) throw Exception(ErrorCodes::SIZES_OF_ARRAYS_DONT_MATCH, "Two arguments of function {} must have equal sizes", getName()); diff --git a/tests/queries/0_stateless/01651_map_functions.reference b/tests/queries/0_stateless/01651_map_functions.reference index 9114aa419b1..e336a02318d 100644 --- a/tests/queries/0_stateless/01651_map_functions.reference +++ b/tests/queries/0_stateless/01651_map_functions.reference @@ -55,3 +55,7 @@ {1:3,2:4} {1:3,2:4} {1:3,2:4} {(1,3):'a',(2,4):'b'} +{(1,'a'):'c',(2,'b'):'d'} +{(1,'a'):'c',(2,'b'):'d'} +{(1,'a'):'c',(2,'b'):'d'} +{(1,'a'):'c',(2,'b'):'d'} diff --git a/tests/queries/0_stateless/01651_map_functions.sql b/tests/queries/0_stateless/01651_map_functions.sql index 4604ddd6db1..dc93a38b265 100644 --- a/tests/queries/0_stateless/01651_map_functions.sql +++ b/tests/queries/0_stateless/01651_map_functions.sql @@ -68,7 +68,7 @@ select mapFromArrays([[1,2], [3,4]], [4, 5, 6]); -- { serverError SIZES_OF_ARRAY select mapFromArrays(['a', 2], [4, 5]); -- { serverError NO_COMMON_TYPE} select mapFromArrays([1, 2], [4, 'a']); -- { serverError NO_COMMON_TYPE} select mapFromArrays(['aa', 'bb'], map('a', 4)); -- { serverError SIZES_OF_ARRAYS_DONT_MATCH } -select mapFromArrays([1,null]::Array(Nullable(UInt8)), [3,4]); -- { serverError ILLEGAL_COLUMN } +select mapFromArrays([1,null]::Array(Nullable(UInt8)), [3,4]); -- { serverError BAD_ARGUMENTS } select mapFromArrays(['aa', 'bb'], map('a', 4, 'b', 5)); select mapFromArrays(['aa', 'bb'], materialize(map('a', 4, 'b', 5))) from numbers(2); @@ -79,3 +79,8 @@ select mapFromArrays([toLowCardinality(1), toLowCardinality(2)], materialize([4, select mapFromArrays([1,2], [3,4]); select mapFromArrays([1,2]::Array(Nullable(UInt8)), [3,4]); select mapFromArrays([1,2], [3,4]) as x, mapFromArrays(x, ['a', 'b']); + +select mapFromArrays(map(1, 'a', 2, 'b'), array('c', 'd')); +select mapFromArrays(materialize(map(1, 'a', 2, 'b')), array('c', 'd')); +select mapFromArrays(map(1, 'a', 2, 'b'), materialize(array('c', 'd'))); +select mapFromArrays(materialize(map(1, 'a', 2, 'b')), materialize(array('c', 'd'))); From 681441e170202bc3963fb3fa1d7b7785192dbd2e Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Thu, 1 Aug 2024 16:01:39 +0800 Subject: [PATCH 097/227] fix style --- src/Functions/map.cpp | 14 ++++++-------- 1 file changed, 6 insertions(+), 8 deletions(-) diff --git a/src/Functions/map.cpp b/src/Functions/map.cpp index 738c61164a3..6e389f39dec 100644 --- a/src/Functions/map.cpp +++ b/src/Functions/map.cpp @@ -192,7 +192,7 @@ public: else throw Exception( ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, - "Argument types of function {} must be Array or Map, but {} is given", + "Arguments of function {} must be Array or Map, but {} is given", getName(), type->getName()); @@ -275,10 +275,7 @@ public: static constexpr auto name = "mapUpdate"; static FunctionPtr create(ContextPtr) { return std::make_shared(); } - String getName() const override - { - return name; - } + String getName() const override { return name; } size_t getNumberOfArguments() const override { return 2; } @@ -287,9 +284,11 @@ public: DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override { if (arguments.size() != 2) - throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, + throw Exception( + ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Number of arguments for function {} doesn't match: passed {}, should be 2", - getName(), arguments.size()); + getName(), + arguments.size()); const auto * left = checkAndGetDataType(arguments[0].type.get()); const auto * right = checkAndGetDataType(arguments[1].type.get()); @@ -405,7 +404,6 @@ public: return ColumnMap::create(nested_column); } }; - } REGISTER_FUNCTION(Map) From bababe2dd2fe490e877f9c4ac03d979bd739376a Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Mon, 29 Jul 2024 15:07:07 +0000 Subject: [PATCH 098/227] Fix remainders I --- src/Functions/FunctionsMultiStringSearch.h | 8 +++--- src/Functions/MultiMatchAnyImpl.h | 27 ++++++++++--------- src/Functions/MultiSearchFirstIndexImpl.h | 16 +++++------ src/Functions/MultiSearchFirstPositionImpl.h | 18 ++++++------- src/Functions/MultiSearchImpl.h | 18 ++++++------- .../0_stateless/00927_disable_hyperscan.sql | 2 +- 6 files changed, 45 insertions(+), 44 deletions(-) diff --git a/src/Functions/FunctionsMultiStringSearch.h b/src/Functions/FunctionsMultiStringSearch.h index 03db2651fd0..6bcc8581a38 100644 --- a/src/Functions/FunctionsMultiStringSearch.h +++ b/src/Functions/FunctionsMultiStringSearch.h @@ -81,7 +81,7 @@ public: return Impl::getReturnType(); } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { const ColumnPtr & haystack_ptr = arguments[0].column; const ColumnPtr & needles_ptr = arguments[1].column; @@ -110,13 +110,15 @@ public: col_haystack_vector->getChars(), col_haystack_vector->getOffsets(), col_needles_const->getValue(), vec_res, offsets_res, - allow_hyperscan, max_hyperscan_regexp_length, max_hyperscan_regexp_total_length, reject_expensive_hyperscan_regexps); + allow_hyperscan, max_hyperscan_regexp_length, max_hyperscan_regexp_total_length, reject_expensive_hyperscan_regexps, + input_rows_count); else Impl::vectorVector( col_haystack_vector->getChars(), col_haystack_vector->getOffsets(), col_needles_vector->getData(), col_needles_vector->getOffsets(), vec_res, offsets_res, - allow_hyperscan, max_hyperscan_regexp_length, max_hyperscan_regexp_total_length, reject_expensive_hyperscan_regexps); + allow_hyperscan, max_hyperscan_regexp_length, max_hyperscan_regexp_total_length, reject_expensive_hyperscan_regexps, + input_rows_count); // the combination of const haystack + const needle is not implemented because // useDefaultImplementationForConstants() == true makes upper layers convert both to diff --git a/src/Functions/MultiMatchAnyImpl.h b/src/Functions/MultiMatchAnyImpl.h index 20b2150048b..06d9eded9c9 100644 --- a/src/Functions/MultiMatchAnyImpl.h +++ b/src/Functions/MultiMatchAnyImpl.h @@ -66,9 +66,10 @@ struct MultiMatchAnyImpl bool allow_hyperscan, size_t max_hyperscan_regexp_length, size_t max_hyperscan_regexp_total_length, - bool reject_expensive_hyperscan_regexps) + bool reject_expensive_hyperscan_regexps, + size_t input_rows_count) { - vectorConstant(haystack_data, haystack_offsets, needles_arr, res, offsets, std::nullopt, allow_hyperscan, max_hyperscan_regexp_length, max_hyperscan_regexp_total_length, reject_expensive_hyperscan_regexps); + vectorConstant(haystack_data, haystack_offsets, needles_arr, res, offsets, std::nullopt, allow_hyperscan, max_hyperscan_regexp_length, max_hyperscan_regexp_total_length, reject_expensive_hyperscan_regexps, input_rows_count); } static void vectorConstant( @@ -81,7 +82,8 @@ struct MultiMatchAnyImpl bool allow_hyperscan, size_t max_hyperscan_regexp_length, size_t max_hyperscan_regexp_total_length, - bool reject_expensive_hyperscan_regexps) + bool reject_expensive_hyperscan_regexps, + size_t input_rows_count) { if (!allow_hyperscan) throw Exception(ErrorCodes::FUNCTION_NOT_ALLOWED, "Hyperscan functions are disabled, because setting 'allow_hyperscan' is set to 0"); @@ -101,7 +103,7 @@ struct MultiMatchAnyImpl throw Exception(ErrorCodes::HYPERSCAN_CANNOT_SCAN_TEXT, "Regular expression evaluation in vectorscan will be too slow. To ignore this error, disable setting 'reject_expensive_hyperscan_regexps'."); } - res.resize(haystack_offsets.size()); + res.resize(input_rows_count); if (needles_arr.empty()) { @@ -133,9 +135,8 @@ struct MultiMatchAnyImpl /// Once we hit the callback, there is no need to search for others. return 1; }; - const size_t haystack_offsets_size = haystack_offsets.size(); UInt64 offset = 0; - for (size_t i = 0; i < haystack_offsets_size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { UInt64 length = haystack_offsets[i] - offset - 1; /// vectorscan restriction. @@ -186,9 +187,10 @@ struct MultiMatchAnyImpl bool allow_hyperscan, size_t max_hyperscan_regexp_length, size_t max_hyperscan_regexp_total_length, - bool reject_expensive_hyperscan_regexps) + bool reject_expensive_hyperscan_regexps, + size_t input_rows_count) { - vectorVector(haystack_data, haystack_offsets, needles_data, needles_offsets, res, offsets, std::nullopt, allow_hyperscan, max_hyperscan_regexp_length, max_hyperscan_regexp_total_length, reject_expensive_hyperscan_regexps); + vectorVector(haystack_data, haystack_offsets, needles_data, needles_offsets, res, offsets, std::nullopt, allow_hyperscan, max_hyperscan_regexp_length, max_hyperscan_regexp_total_length, reject_expensive_hyperscan_regexps, input_rows_count); } static void vectorVector( @@ -202,12 +204,13 @@ struct MultiMatchAnyImpl bool allow_hyperscan, size_t max_hyperscan_regexp_length, size_t max_hyperscan_regexp_total_length, - bool reject_expensive_hyperscan_regexps) + bool reject_expensive_hyperscan_regexps, + size_t input_rows_count) { if (!allow_hyperscan) throw Exception(ErrorCodes::FUNCTION_NOT_ALLOWED, "Hyperscan functions are disabled, because setting 'allow_hyperscan' is set to 0"); - res.resize(haystack_offsets.size()); + res.resize(input_rows_count); #if USE_VECTORSCAN size_t prev_haystack_offset = 0; size_t prev_needles_offset = 0; @@ -216,7 +219,7 @@ struct MultiMatchAnyImpl std::vector needles; - for (size_t i = 0; i < haystack_offsets.size(); ++i) + for (size_t i = 0; i < input_rows_count; ++i) { needles.reserve(needles_offsets[i] - prev_needles_offset); @@ -306,7 +309,7 @@ struct MultiMatchAnyImpl std::vector needles; - for (size_t i = 0; i < haystack_offsets.size(); ++i) + for (size_t i = 0; i < input_rows_count; ++i) { const auto * const cur_haystack_data = &haystack_data[prev_haystack_offset]; const size_t cur_haystack_length = haystack_offsets[i] - prev_haystack_offset - 1; diff --git a/src/Functions/MultiSearchFirstIndexImpl.h b/src/Functions/MultiSearchFirstIndexImpl.h index 36a5fd514d9..5b34dbfe601 100644 --- a/src/Functions/MultiSearchFirstIndexImpl.h +++ b/src/Functions/MultiSearchFirstIndexImpl.h @@ -33,7 +33,8 @@ struct MultiSearchFirstIndexImpl bool /*allow_hyperscan*/, size_t /*max_hyperscan_regexp_length*/, size_t /*max_hyperscan_regexp_total_length*/, - bool /*reject_expensive_hyperscan_regexps*/) + bool /*reject_expensive_hyperscan_regexps*/, + size_t input_rows_count) { // For performance of Volnitsky search, it is crucial to save only one byte for pattern number. if (needles_arr.size() > std::numeric_limits::max()) @@ -48,14 +49,13 @@ struct MultiSearchFirstIndexImpl auto searcher = Impl::createMultiSearcherInBigHaystack(needles); - const size_t haystack_size = haystack_offsets.size(); - res.resize(haystack_size); + res.resize(input_rows_count); size_t iteration = 0; while (searcher.hasMoreToSearch()) { size_t prev_haystack_offset = 0; - for (size_t j = 0; j < haystack_size; ++j) + for (size_t j = 0; j < input_rows_count; ++j) { const auto * haystack = &haystack_data[prev_haystack_offset]; const auto * haystack_end = haystack + haystack_offsets[j] - prev_haystack_offset - 1; @@ -80,10 +80,10 @@ struct MultiSearchFirstIndexImpl bool /*allow_hyperscan*/, size_t /*max_hyperscan_regexp_length*/, size_t /*max_hyperscan_regexp_total_length*/, - bool /*reject_expensive_hyperscan_regexps*/) + bool /*reject_expensive_hyperscan_regexps*/, + size_t input_rows_count) { - const size_t haystack_size = haystack_offsets.size(); - res.resize(haystack_size); + res.resize(input_rows_count); size_t prev_haystack_offset = 0; size_t prev_needles_offset = 0; @@ -92,7 +92,7 @@ struct MultiSearchFirstIndexImpl std::vector needles; - for (size_t i = 0; i < haystack_size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { needles.reserve(needles_offsets[i] - prev_needles_offset); diff --git a/src/Functions/MultiSearchFirstPositionImpl.h b/src/Functions/MultiSearchFirstPositionImpl.h index ccdd82a0ee5..06bf7aa94d8 100644 --- a/src/Functions/MultiSearchFirstPositionImpl.h +++ b/src/Functions/MultiSearchFirstPositionImpl.h @@ -33,7 +33,8 @@ struct MultiSearchFirstPositionImpl bool /*allow_hyperscan*/, size_t /*max_hyperscan_regexp_length*/, size_t /*max_hyperscan_regexp_total_length*/, - bool /*reject_expensive_hyperscan_regexps*/) + bool /*reject_expensive_hyperscan_regexps*/, + size_t input_rows_count) { // For performance of Volnitsky search, it is crucial to save only one byte for pattern number. if (needles_arr.size() > std::numeric_limits::max()) @@ -52,14 +53,13 @@ struct MultiSearchFirstPositionImpl }; auto searcher = Impl::createMultiSearcherInBigHaystack(needles); - const size_t haystack_size = haystack_offsets.size(); - res.resize(haystack_size); + res.resize(input_rows_count); size_t iteration = 0; while (searcher.hasMoreToSearch()) { size_t prev_haystack_offset = 0; - for (size_t j = 0; j < haystack_size; ++j) + for (size_t j = 0; j < input_rows_count; ++j) { const auto * haystack = &haystack_data[prev_haystack_offset]; const auto * haystack_end = haystack + haystack_offsets[j] - prev_haystack_offset - 1; @@ -89,10 +89,10 @@ struct MultiSearchFirstPositionImpl bool /*allow_hyperscan*/, size_t /*max_hyperscan_regexp_length*/, size_t /*max_hyperscan_regexp_total_length*/, - bool /*reject_expensive_hyperscan_regexps*/) + bool /*reject_expensive_hyperscan_regexps*/, + size_t input_rows_count) { - const size_t haystack_size = haystack_offsets.size(); - res.resize(haystack_size); + res.resize(input_rows_count); size_t prev_haystack_offset = 0; size_t prev_needles_offset = 0; @@ -106,14 +106,12 @@ struct MultiSearchFirstPositionImpl return 1 + Impl::countChars(reinterpret_cast(start), reinterpret_cast(end)); }; - for (size_t i = 0; i < haystack_size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { needles.reserve(needles_offsets[i] - prev_needles_offset); for (size_t j = prev_needles_offset; j < needles_offsets[i]; ++j) - { needles.emplace_back(needles_data_string.getDataAt(j).toView()); - } auto searcher = Impl::createMultiSearcherInBigHaystack(needles); // sub-optimal diff --git a/src/Functions/MultiSearchImpl.h b/src/Functions/MultiSearchImpl.h index 467cc96a95f..909425f5a93 100644 --- a/src/Functions/MultiSearchImpl.h +++ b/src/Functions/MultiSearchImpl.h @@ -33,7 +33,8 @@ struct MultiSearchImpl bool /*allow_hyperscan*/, size_t /*max_hyperscan_regexp_length*/, size_t /*max_hyperscan_regexp_total_length*/, - bool /*reject_expensive_hyperscan_regexps*/) + bool /*reject_expensive_hyperscan_regexps*/, + size_t input_rows_count) { // For performance of Volnitsky search, it is crucial to save only one byte for pattern number. if (needles_arr.size() > std::numeric_limits::max()) @@ -48,14 +49,13 @@ struct MultiSearchImpl auto searcher = Impl::createMultiSearcherInBigHaystack(needles); - const size_t haystack_size = haystack_offsets.size(); - res.resize(haystack_size); + res.resize(input_rows_count); size_t iteration = 0; while (searcher.hasMoreToSearch()) { size_t prev_haystack_offset = 0; - for (size_t j = 0; j < haystack_size; ++j) + for (size_t j = 0; j < input_rows_count; ++j) { const auto * haystack = &haystack_data[prev_haystack_offset]; const auto * haystack_end = haystack + haystack_offsets[j] - prev_haystack_offset - 1; @@ -79,10 +79,10 @@ struct MultiSearchImpl bool /*allow_hyperscan*/, size_t /*max_hyperscan_regexp_length*/, size_t /*max_hyperscan_regexp_total_length*/, - bool /*reject_expensive_hyperscan_regexps*/) + bool /*reject_expensive_hyperscan_regexps*/, + size_t input_rows_count) { - const size_t haystack_size = haystack_offsets.size(); - res.resize(haystack_size); + res.resize(input_rows_count); size_t prev_haystack_offset = 0; size_t prev_needles_offset = 0; @@ -91,14 +91,12 @@ struct MultiSearchImpl std::vector needles; - for (size_t i = 0; i < haystack_size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { needles.reserve(needles_offsets[i] - prev_needles_offset); for (size_t j = prev_needles_offset; j < needles_offsets[i]; ++j) - { needles.emplace_back(needles_data_string.getDataAt(j).toView()); - } const auto * const haystack = &haystack_data[prev_haystack_offset]; const size_t haystack_length = haystack_offsets[i] - prev_haystack_offset - 1; diff --git a/tests/queries/0_stateless/00927_disable_hyperscan.sql b/tests/queries/0_stateless/00927_disable_hyperscan.sql index c07848a4fcc..24ec7a35adb 100644 --- a/tests/queries/0_stateless/00927_disable_hyperscan.sql +++ b/tests/queries/0_stateless/00927_disable_hyperscan.sql @@ -1,4 +1,4 @@ --- Tags: no-debug +-- Tags: no-fasttest SET allow_hyperscan = 1; From 0f850952fa304bcc63f7008d4bae54c47ea0564d Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Tue, 30 Jul 2024 09:52:25 +0000 Subject: [PATCH 099/227] Fix remainders II --- .../FunctionsMultiStringFuzzySearch.h | 8 +++--- src/Functions/MultiMatchAllIndicesImpl.h | 27 +++++++++++-------- 2 files changed, 21 insertions(+), 14 deletions(-) diff --git a/src/Functions/FunctionsMultiStringFuzzySearch.h b/src/Functions/FunctionsMultiStringFuzzySearch.h index a92a6570279..8346380c35d 100644 --- a/src/Functions/FunctionsMultiStringFuzzySearch.h +++ b/src/Functions/FunctionsMultiStringFuzzySearch.h @@ -71,7 +71,7 @@ public: return Impl::getReturnType(); } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { const ColumnPtr & haystack_ptr = arguments[0].column; const ColumnPtr & edit_distance_ptr = arguments[1].column; @@ -114,14 +114,16 @@ public: col_needles_const->getValue(), vec_res, offsets_res, edit_distance, - allow_hyperscan, max_hyperscan_regexp_length, max_hyperscan_regexp_total_length, reject_expensive_hyperscan_regexps); + allow_hyperscan, max_hyperscan_regexp_length, max_hyperscan_regexp_total_length, reject_expensive_hyperscan_regexps, + input_rows_count); else Impl::vectorVector( col_haystack_vector->getChars(), col_haystack_vector->getOffsets(), col_needles_vector->getData(), col_needles_vector->getOffsets(), vec_res, offsets_res, edit_distance, - allow_hyperscan, max_hyperscan_regexp_length, max_hyperscan_regexp_total_length, reject_expensive_hyperscan_regexps); + allow_hyperscan, max_hyperscan_regexp_length, max_hyperscan_regexp_total_length, reject_expensive_hyperscan_regexps, + input_rows_count); // the combination of const haystack + const needle is not implemented because // useDefaultImplementationForConstants() == true makes upper layers convert both to diff --git a/src/Functions/MultiMatchAllIndicesImpl.h b/src/Functions/MultiMatchAllIndicesImpl.h index 3e9c8fba215..3aeac808880 100644 --- a/src/Functions/MultiMatchAllIndicesImpl.h +++ b/src/Functions/MultiMatchAllIndicesImpl.h @@ -52,9 +52,10 @@ struct MultiMatchAllIndicesImpl bool allow_hyperscan, size_t max_hyperscan_regexp_length, size_t max_hyperscan_regexp_total_length, - bool reject_expensive_hyperscan_regexps) + bool reject_expensive_hyperscan_regexps, + size_t input_rows_count) { - vectorConstant(haystack_data, haystack_offsets, needles_arr, res, offsets, std::nullopt, allow_hyperscan, max_hyperscan_regexp_length, max_hyperscan_regexp_total_length, reject_expensive_hyperscan_regexps); + vectorConstant(haystack_data, haystack_offsets, needles_arr, res, offsets, std::nullopt, allow_hyperscan, max_hyperscan_regexp_length, max_hyperscan_regexp_total_length, reject_expensive_hyperscan_regexps, input_rows_count); } static void vectorConstant( @@ -67,7 +68,8 @@ struct MultiMatchAllIndicesImpl bool allow_hyperscan, size_t max_hyperscan_regexp_length, size_t max_hyperscan_regexp_total_length, - bool reject_expensive_hyperscan_regexps) + bool reject_expensive_hyperscan_regexps, + size_t input_rows_count) { if (!allow_hyperscan) throw Exception(ErrorCodes::FUNCTION_NOT_ALLOWED, "Hyperscan functions are disabled, because setting 'allow_hyperscan' is set to 0"); @@ -87,7 +89,7 @@ struct MultiMatchAllIndicesImpl throw Exception(ErrorCodes::HYPERSCAN_CANNOT_SCAN_TEXT, "Regular expression evaluation in vectorscan will be too slow. To ignore this error, disable setting 'reject_expensive_hyperscan_regexps'."); } - offsets.resize(haystack_offsets.size()); + offsets.resize(input_rows_count); if (needles_arr.empty()) { @@ -114,9 +116,8 @@ struct MultiMatchAllIndicesImpl static_cast*>(context)->push_back(id); return 0; }; - const size_t haystack_offsets_size = haystack_offsets.size(); UInt64 offset = 0; - for (size_t i = 0; i < haystack_offsets_size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { UInt64 length = haystack_offsets[i] - offset - 1; /// vectorscan restriction. @@ -146,6 +147,7 @@ struct MultiMatchAllIndicesImpl (void)max_hyperscan_regexp_length; (void)max_hyperscan_regexp_total_length; (void)reject_expensive_hyperscan_regexps; + (void)input_rows_count; throw Exception(ErrorCodes::NOT_IMPLEMENTED, "multi-search all indices is not implemented when vectorscan is off"); #endif // USE_VECTORSCAN } @@ -160,9 +162,10 @@ struct MultiMatchAllIndicesImpl bool allow_hyperscan, size_t max_hyperscan_regexp_length, size_t max_hyperscan_regexp_total_length, - bool reject_expensive_hyperscan_regexps) + bool reject_expensive_hyperscan_regexps, + size_t input_rows_count) { - vectorVector(haystack_data, haystack_offsets, needles_data, needles_offsets, res, offsets, std::nullopt, allow_hyperscan, max_hyperscan_regexp_length, max_hyperscan_regexp_total_length, reject_expensive_hyperscan_regexps); + vectorVector(haystack_data, haystack_offsets, needles_data, needles_offsets, res, offsets, std::nullopt, allow_hyperscan, max_hyperscan_regexp_length, max_hyperscan_regexp_total_length, reject_expensive_hyperscan_regexps, input_rows_count); } static void vectorVector( @@ -176,12 +179,13 @@ struct MultiMatchAllIndicesImpl bool allow_hyperscan, size_t max_hyperscan_regexp_length, size_t max_hyperscan_regexp_total_length, - bool reject_expensive_hyperscan_regexps) + bool reject_expensive_hyperscan_regexps, + size_t input_rows_count) { if (!allow_hyperscan) throw Exception(ErrorCodes::FUNCTION_NOT_ALLOWED, "Hyperscan functions are disabled, because setting 'allow_hyperscan' is set to 0"); #if USE_VECTORSCAN - offsets.resize(haystack_offsets.size()); + offsets.resize(input_rows_count); size_t prev_haystack_offset = 0; size_t prev_needles_offset = 0; @@ -189,7 +193,7 @@ struct MultiMatchAllIndicesImpl std::vector needles; - for (size_t i = 0; i < haystack_offsets.size(); ++i) + for (size_t i = 0; i < input_rows_count; ++i) { needles.reserve(needles_offsets[i] - prev_needles_offset); @@ -271,6 +275,7 @@ struct MultiMatchAllIndicesImpl (void)max_hyperscan_regexp_length; (void)max_hyperscan_regexp_total_length; (void)reject_expensive_hyperscan_regexps; + (void)input_rows_count; throw Exception(ErrorCodes::NOT_IMPLEMENTED, "multi-search all indices is not implemented when vectorscan is off"); #endif // USE_VECTORSCAN } From 99760ad7e4b1b7d4ee86dc2f5aa4df98ead9e260 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Tue, 30 Jul 2024 17:17:40 +0000 Subject: [PATCH 100/227] Fix remainders, pt. III --- src/Functions/FunctionsStringSearchToString.h | 4 ++-- src/Functions/FunctionsVisitParam.h | 5 +++-- src/Functions/URL/extractURLParameter.cpp | 7 ++++--- src/Functions/extract.cpp | 7 ++++--- 4 files changed, 13 insertions(+), 10 deletions(-) diff --git a/src/Functions/FunctionsStringSearchToString.h b/src/Functions/FunctionsStringSearchToString.h index 978a84de472..c889cf062a3 100644 --- a/src/Functions/FunctionsStringSearchToString.h +++ b/src/Functions/FunctionsStringSearchToString.h @@ -60,7 +60,7 @@ public: return std::make_shared(); } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { const ColumnPtr column = arguments[0].column; const ColumnPtr column_needle = arguments[1].column; @@ -75,7 +75,7 @@ public: ColumnString::Chars & vec_res = col_res->getChars(); ColumnString::Offsets & offsets_res = col_res->getOffsets(); - Impl::vector(col->getChars(), col->getOffsets(), col_needle->getValue(), vec_res, offsets_res); + Impl::vector(col->getChars(), col->getOffsets(), col_needle->getValue(), vec_res, offsets_res, input_rows_count); return col_res; } diff --git a/src/Functions/FunctionsVisitParam.h b/src/Functions/FunctionsVisitParam.h index 5e13fbbad5c..bcaaf0a1c20 100644 --- a/src/Functions/FunctionsVisitParam.h +++ b/src/Functions/FunctionsVisitParam.h @@ -168,11 +168,12 @@ struct ExtractParamToStringImpl { static void vector(const ColumnString::Chars & haystack_data, const ColumnString::Offsets & haystack_offsets, std::string needle, - ColumnString::Chars & res_data, ColumnString::Offsets & res_offsets) + ColumnString::Chars & res_data, ColumnString::Offsets & res_offsets, + size_t input_rows_count) { /// Constant 5 is taken from a function that performs a similar task FunctionsStringSearch.h::ExtractImpl res_data.reserve(haystack_data.size() / 5); - res_offsets.resize(haystack_offsets.size()); + res_offsets.resize(input_rows_count); /// We are looking for a parameter simply as a substring of the form "name" needle = "\"" + needle + "\":"; diff --git a/src/Functions/URL/extractURLParameter.cpp b/src/Functions/URL/extractURLParameter.cpp index f75875e0200..590c2779d9c 100644 --- a/src/Functions/URL/extractURLParameter.cpp +++ b/src/Functions/URL/extractURLParameter.cpp @@ -10,10 +10,11 @@ struct ExtractURLParameterImpl static void vector(const ColumnString::Chars & data, const ColumnString::Offsets & offsets, std::string pattern, - ColumnString::Chars & res_data, ColumnString::Offsets & res_offsets) + ColumnString::Chars & res_data, ColumnString::Offsets & res_offsets, + size_t input_rows_count) { res_data.reserve(data.size() / 5); - res_offsets.resize(offsets.size()); + res_offsets.resize(input_rows_count); pattern += '='; const char * param_str = pattern.c_str(); @@ -22,7 +23,7 @@ struct ExtractURLParameterImpl ColumnString::Offset prev_offset = 0; ColumnString::Offset res_offset = 0; - for (size_t i = 0; i < offsets.size(); ++i) + for (size_t i = 0; i < input_rows_count; ++i) { ColumnString::Offset cur_offset = offsets[i]; diff --git a/src/Functions/extract.cpp b/src/Functions/extract.cpp index 6bbdaff0e3f..c78ee9898b7 100644 --- a/src/Functions/extract.cpp +++ b/src/Functions/extract.cpp @@ -16,10 +16,11 @@ struct ExtractImpl const ColumnString::Offsets & offsets, const std::string & pattern, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t input_rows_count) { res_data.reserve(data.size() / 5); - res_offsets.resize(offsets.size()); + res_offsets.resize(input_rows_count); const OptimizedRegularExpression regexp = Regexps::createRegexp(pattern); @@ -29,7 +30,7 @@ struct ExtractImpl size_t prev_offset = 0; size_t res_offset = 0; - for (size_t i = 0; i < offsets.size(); ++i) + for (size_t i = 0; i < input_rows_count; ++i) { size_t cur_offset = offsets[i]; From 0452768983f379d201e1871ffce63452430bd0c6 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Wed, 31 Jul 2024 08:10:01 +0000 Subject: [PATCH 101/227] Fix remainders, pt. IV --- src/Functions/CountSubstringsImpl.h | 17 ++++----- src/Functions/FunctionsStringSearch.h | 17 ++++++--- src/Functions/FunctionsVisitParam.h | 3 +- src/Functions/HasTokenImpl.h | 5 ++- src/Functions/MatchImpl.h | 54 +++++++++++++-------------- src/Functions/MultiMatchAnyImpl.h | 2 +- src/Functions/PositionImpl.h | 22 +++++------ 7 files changed, 62 insertions(+), 58 deletions(-) diff --git a/src/Functions/CountSubstringsImpl.h b/src/Functions/CountSubstringsImpl.h index 9ff3e4e1f2a..8e91bc3aeb4 100644 --- a/src/Functions/CountSubstringsImpl.h +++ b/src/Functions/CountSubstringsImpl.h @@ -37,7 +37,8 @@ struct CountSubstringsImpl const std::string & needle, const ColumnPtr & start_pos, PaddedPODArray & res, - [[maybe_unused]] ColumnUInt8 * res_null) + [[maybe_unused]] ColumnUInt8 * res_null, + size_t /*input_rows_count*/) { /// `res_null` serves as an output parameter for implementing an XYZOrNull variant. assert(!res_null); @@ -150,7 +151,8 @@ struct CountSubstringsImpl const ColumnString::Offsets & needle_offsets, const ColumnPtr & start_pos, PaddedPODArray & res, - [[maybe_unused]] ColumnUInt8 * res_null) + [[maybe_unused]] ColumnUInt8 * res_null, + size_t input_rows_count) { /// `res_null` serves as an output parameter for implementing an XYZOrNull variant. assert(!res_null); @@ -158,9 +160,7 @@ struct CountSubstringsImpl ColumnString::Offset prev_haystack_offset = 0; ColumnString::Offset prev_needle_offset = 0; - size_t size = haystack_offsets.size(); - - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { size_t needle_size = needle_offsets[i] - prev_needle_offset - 1; size_t haystack_size = haystack_offsets[i] - prev_haystack_offset - 1; @@ -207,7 +207,8 @@ struct CountSubstringsImpl const ColumnString::Offsets & needle_offsets, const ColumnPtr & start_pos, PaddedPODArray & res, - [[maybe_unused]] ColumnUInt8 * res_null) + [[maybe_unused]] ColumnUInt8 * res_null, + size_t input_rows_count) { /// `res_null` serves as an output parameter for implementing an XYZOrNull variant. assert(!res_null); @@ -215,9 +216,7 @@ struct CountSubstringsImpl /// NOTE You could use haystack indexing. But this is a rare case. ColumnString::Offset prev_needle_offset = 0; - size_t size = needle_offsets.size(); - - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { res[i] = 0; auto start = start_pos != nullptr ? std::max(start_pos->getUInt(i), UInt64(1)) : UInt64(1); diff --git a/src/Functions/FunctionsStringSearch.h b/src/Functions/FunctionsStringSearch.h index fba6336ebff..7ec0076e395 100644 --- a/src/Functions/FunctionsStringSearch.h +++ b/src/Functions/FunctionsStringSearch.h @@ -163,7 +163,7 @@ public: return return_type; } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t /*input_rows_count*/) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const override { const ColumnPtr & column_haystack = (argument_order == ArgumentOrder::HaystackNeedle) ? arguments[0].column : arguments[1].column; const ColumnPtr & column_needle = (argument_order == ArgumentOrder::HaystackNeedle) ? arguments[1].column : arguments[0].column; @@ -236,7 +236,8 @@ public: col_needle_vector->getOffsets(), column_start_pos, vec_res, - null_map.get()); + null_map.get(), + input_rows_count); else if (col_haystack_vector && col_needle_const) Impl::vectorConstant( col_haystack_vector->getChars(), @@ -244,7 +245,8 @@ public: col_needle_const->getValue(), column_start_pos, vec_res, - null_map.get()); + null_map.get(), + input_rows_count); else if (col_haystack_vector_fixed && col_needle_vector) Impl::vectorFixedVector( col_haystack_vector_fixed->getChars(), @@ -253,14 +255,16 @@ public: col_needle_vector->getOffsets(), column_start_pos, vec_res, - null_map.get()); + null_map.get(), + input_rows_count); else if (col_haystack_vector_fixed && col_needle_const) Impl::vectorFixedConstant( col_haystack_vector_fixed->getChars(), col_haystack_vector_fixed->getN(), col_needle_const->getValue(), vec_res, - null_map.get()); + null_map.get(), + input_rows_count); else if (col_haystack_const && col_needle_vector) Impl::constantVector( col_haystack_const->getValue(), @@ -268,7 +272,8 @@ public: col_needle_vector->getOffsets(), column_start_pos, vec_res, - null_map.get()); + null_map.get(), + input_rows_count); else throw Exception( ErrorCodes::ILLEGAL_COLUMN, diff --git a/src/Functions/FunctionsVisitParam.h b/src/Functions/FunctionsVisitParam.h index bcaaf0a1c20..a77fa740f9c 100644 --- a/src/Functions/FunctionsVisitParam.h +++ b/src/Functions/FunctionsVisitParam.h @@ -93,7 +93,8 @@ struct ExtractParamImpl std::string needle, const ColumnPtr & start_pos, PaddedPODArray & res, - [[maybe_unused]] ColumnUInt8 * res_null) + [[maybe_unused]] ColumnUInt8 * res_null, + size_t /*input_rows_count*/) { /// `res_null` serves as an output parameter for implementing an XYZOrNull variant. assert(!res_null); diff --git a/src/Functions/HasTokenImpl.h b/src/Functions/HasTokenImpl.h index a4ff49859cc..4943bf708c5 100644 --- a/src/Functions/HasTokenImpl.h +++ b/src/Functions/HasTokenImpl.h @@ -35,12 +35,13 @@ struct HasTokenImpl const std::string & pattern, const ColumnPtr & start_pos, PaddedPODArray & res, - ColumnUInt8 * res_null) + ColumnUInt8 * res_null, + size_t input_rows_count) { if (start_pos != nullptr) throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Function '{}' does not support start_pos argument", name); - if (haystack_offsets.empty()) + if (input_rows_count == 0) return; const UInt8 * const begin = haystack_data.data(); diff --git a/src/Functions/MatchImpl.h b/src/Functions/MatchImpl.h index 55b2fee5400..ceac753fe79 100644 --- a/src/Functions/MatchImpl.h +++ b/src/Functions/MatchImpl.h @@ -127,17 +127,17 @@ struct MatchImpl const String & needle, [[maybe_unused]] const ColumnPtr & start_pos_, PaddedPODArray & res, - [[maybe_unused]] ColumnUInt8 * res_null) + [[maybe_unused]] ColumnUInt8 * res_null, + size_t input_rows_count) { /// `res_null` serves as an output parameter for implementing an XYZOrNull variant. assert(!res_null); - const size_t haystack_size = haystack_offsets.size(); - - assert(haystack_size == res.size()); + assert(res.size() == haystack_offsets.size()); + assert(res.size() == input_rows_count); assert(start_pos_ == nullptr); - if (haystack_offsets.empty()) + if (input_rows_count == 0) return; /// Shortcut for the silly but practical case that the pattern matches everything/nothing independently of the haystack: @@ -202,11 +202,11 @@ struct MatchImpl if (required_substring.empty()) { if (!regexp.getRE2()) /// An empty regexp. Always matches. - memset(res.data(), !negate, haystack_size * sizeof(res[0])); + memset(res.data(), !negate, input_rows_count * sizeof(res[0])); else { size_t prev_offset = 0; - for (size_t i = 0; i < haystack_size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { const bool match = regexp.getRE2()->Match( {reinterpret_cast(&haystack_data[prev_offset]), haystack_offsets[i] - prev_offset - 1}, @@ -291,16 +291,16 @@ struct MatchImpl size_t N, const String & needle, PaddedPODArray & res, - [[maybe_unused]] ColumnUInt8 * res_null) + [[maybe_unused]] ColumnUInt8 * res_null, + size_t input_rows_count) { /// `res_null` serves as an output parameter for implementing an XYZOrNull variant. assert(!res_null); - const size_t haystack_size = haystack.size() / N; + assert(res.size() == haystack.size() / N); + assert(res.size() == input_rows_count); - assert(haystack_size == res.size()); - - if (haystack.empty()) + if (input_rows_count == 0) return; /// Shortcut for the silly but practical case that the pattern matches everything/nothing independently of the haystack: @@ -370,11 +370,11 @@ struct MatchImpl if (required_substring.empty()) { if (!regexp.getRE2()) /// An empty regexp. Always matches. - memset(res.data(), !negate, haystack_size * sizeof(res[0])); + memset(res.data(), !negate, input_rows_count * sizeof(res[0])); else { size_t offset = 0; - for (size_t i = 0; i < haystack_size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { const bool match = regexp.getRE2()->Match( {reinterpret_cast(&haystack[offset]), N}, @@ -464,18 +464,18 @@ struct MatchImpl const ColumnString::Offsets & needle_offset, [[maybe_unused]] const ColumnPtr & start_pos_, PaddedPODArray & res, - [[maybe_unused]] ColumnUInt8 * res_null) + [[maybe_unused]] ColumnUInt8 * res_null, + size_t input_rows_count) { /// `res_null` serves as an output parameter for implementing an XYZOrNull variant. assert(!res_null); - const size_t haystack_size = haystack_offsets.size(); - - assert(haystack_size == needle_offset.size()); - assert(haystack_size == res.size()); + assert(haystack_offsets.size() == needle_offset.size()); + assert(res.size() == haystack_offsets.size()); + assert(res.size() == input_rows_count); assert(start_pos_ == nullptr); - if (haystack_offsets.empty()) + if (input_rows_count == 0) return; String required_substr; @@ -488,7 +488,7 @@ struct MatchImpl Regexps::LocalCacheTable cache; Regexps::RegexpPtr regexp; - for (size_t i = 0; i < haystack_size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { const auto * const cur_haystack_data = &haystack_data[prev_haystack_offset]; const size_t cur_haystack_length = haystack_offsets[i] - prev_haystack_offset - 1; @@ -573,15 +573,15 @@ struct MatchImpl const ColumnString::Offsets & needle_offset, [[maybe_unused]] const ColumnPtr & start_pos_, PaddedPODArray & res, - [[maybe_unused]] ColumnUInt8 * res_null) + [[maybe_unused]] ColumnUInt8 * res_null, + size_t input_rows_count) { /// `res_null` serves as an output parameter for implementing an XYZOrNull variant. assert(!res_null); - const size_t haystack_size = haystack.size()/N; - - assert(haystack_size == needle_offset.size()); - assert(haystack_size == res.size()); + assert(res.size() == input_rows_count); + assert(res.size() == haystack.size() / N); + assert(res.size() == needle_offset.size()); assert(start_pos_ == nullptr); if (haystack.empty()) @@ -597,7 +597,7 @@ struct MatchImpl Regexps::LocalCacheTable cache; Regexps::RegexpPtr regexp; - for (size_t i = 0; i < haystack_size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { const auto * const cur_haystack_data = &haystack[prev_haystack_offset]; const size_t cur_haystack_length = N; diff --git a/src/Functions/MultiMatchAnyImpl.h b/src/Functions/MultiMatchAnyImpl.h index 06d9eded9c9..fda752cbacc 100644 --- a/src/Functions/MultiMatchAnyImpl.h +++ b/src/Functions/MultiMatchAnyImpl.h @@ -165,7 +165,7 @@ struct MultiMatchAnyImpl memset(accum.data(), 0, accum.size()); for (size_t j = 0; j < needles.size(); ++j) { - MatchImpl::vectorConstant(haystack_data, haystack_offsets, String(needles[j].data(), needles[j].size()), nullptr, accum, nullptr); + MatchImpl::vectorConstant(haystack_data, haystack_offsets, String(needles[j].data(), needles[j].size()), nullptr, accum, nullptr, input_rows_count); for (size_t i = 0; i < res.size(); ++i) { if constexpr (FindAny) diff --git a/src/Functions/PositionImpl.h b/src/Functions/PositionImpl.h index eeb9d8b6a59..e525b5fab57 100644 --- a/src/Functions/PositionImpl.h +++ b/src/Functions/PositionImpl.h @@ -193,7 +193,8 @@ struct PositionImpl const std::string & needle, const ColumnPtr & start_pos, PaddedPODArray & res, - [[maybe_unused]] ColumnUInt8 * res_null) + [[maybe_unused]] ColumnUInt8 * res_null, + size_t input_rows_count) { /// `res_null` serves as an output parameter for implementing an XYZOrNull variant. assert(!res_null); @@ -214,13 +215,12 @@ struct PositionImpl } ColumnString::Offset prev_offset = 0; - size_t rows = haystack_offsets.size(); if (const ColumnConst * start_pos_const = typeid_cast(&*start_pos)) { /// Needle is empty and start_pos is constant UInt64 start = std::max(start_pos_const->getUInt(0), static_cast(1)); - for (size_t i = 0; i < rows; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { size_t haystack_size = Impl::countChars( reinterpret_cast(pos), reinterpret_cast(pos + haystack_offsets[i] - prev_offset - 1)); @@ -234,7 +234,7 @@ struct PositionImpl else { /// Needle is empty and start_pos is not constant - for (size_t i = 0; i < rows; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { size_t haystack_size = Impl::countChars( reinterpret_cast(pos), reinterpret_cast(pos + haystack_offsets[i] - prev_offset - 1)); @@ -359,7 +359,8 @@ struct PositionImpl const ColumnString::Offsets & needle_offsets, const ColumnPtr & start_pos, PaddedPODArray & res, - [[maybe_unused]] ColumnUInt8 * res_null) + [[maybe_unused]] ColumnUInt8 * res_null, + size_t input_rows_count) { /// `res_null` serves as an output parameter for implementing an XYZOrNull variant. assert(!res_null); @@ -367,9 +368,7 @@ struct PositionImpl ColumnString::Offset prev_haystack_offset = 0; ColumnString::Offset prev_needle_offset = 0; - size_t size = haystack_offsets.size(); - - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { size_t needle_size = needle_offsets[i] - prev_needle_offset - 1; size_t haystack_size = haystack_offsets[i] - prev_haystack_offset - 1; @@ -423,7 +422,8 @@ struct PositionImpl const ColumnString::Offsets & needle_offsets, const ColumnPtr & start_pos, PaddedPODArray & res, - [[maybe_unused]] ColumnUInt8 * res_null) + [[maybe_unused]] ColumnUInt8 * res_null, + size_t input_rows_count) { /// `res_null` serves as an output parameter for implementing an XYZOrNull variant. assert(!res_null); @@ -431,9 +431,7 @@ struct PositionImpl /// NOTE You could use haystack indexing. But this is a rare case. ColumnString::Offset prev_needle_offset = 0; - size_t size = needle_offsets.size(); - - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { size_t needle_size = needle_offsets[i] - prev_needle_offset - 1; From 59e51de6f8df35a0df0e08e8b3aacd3ba40c4cbd Mon Sep 17 00:00:00 2001 From: vdimir Date: Wed, 31 Jul 2024 14:58:01 +0000 Subject: [PATCH 102/227] Fix join on tuple with NULLs --- .../Passes/ComparisonTupleEliminationPass.cpp | 10 +++++++++- ...on_tuple_comparison_elimination_bug.reference | 8 ++++++++ ..._join_on_tuple_comparison_elimination_bug.sql | 16 ++++++++++++++++ 3 files changed, 33 insertions(+), 1 deletion(-) create mode 100644 tests/queries/0_stateless/03214_join_on_tuple_comparison_elimination_bug.reference create mode 100644 tests/queries/0_stateless/03214_join_on_tuple_comparison_elimination_bug.sql diff --git a/src/Analyzer/Passes/ComparisonTupleEliminationPass.cpp b/src/Analyzer/Passes/ComparisonTupleEliminationPass.cpp index 180470952cd..76dc8ab94b4 100644 --- a/src/Analyzer/Passes/ComparisonTupleEliminationPass.cpp +++ b/src/Analyzer/Passes/ComparisonTupleEliminationPass.cpp @@ -11,6 +11,7 @@ #include #include #include +#include #include namespace DB @@ -25,8 +26,15 @@ public: using Base = InDepthQueryTreeVisitorWithContext; using Base::Base; - static bool needChildVisit(QueryTreeNodePtr &, QueryTreeNodePtr & child) + static bool needChildVisit(QueryTreeNodePtr & parent, QueryTreeNodePtr & child) { + if (parent->getNodeType() == QueryTreeNodeType::JOIN) + { + /// In JOIN ON section comparison of tuples works a bit differently. + /// For example we can join on tuple(NULL) = tuple(NULL), join algorithms consider only NULLs on the top level. + if (parent->as().getJoinExpression().get() == child.get()) + return false; + } return child->getNodeType() != QueryTreeNodeType::TABLE_FUNCTION; } diff --git a/tests/queries/0_stateless/03214_join_on_tuple_comparison_elimination_bug.reference b/tests/queries/0_stateless/03214_join_on_tuple_comparison_elimination_bug.reference new file mode 100644 index 00000000000..c90a64e4a47 --- /dev/null +++ b/tests/queries/0_stateless/03214_join_on_tuple_comparison_elimination_bug.reference @@ -0,0 +1,8 @@ +1 +\N +1 +\N +1 +\N +1 +\N diff --git a/tests/queries/0_stateless/03214_join_on_tuple_comparison_elimination_bug.sql b/tests/queries/0_stateless/03214_join_on_tuple_comparison_elimination_bug.sql new file mode 100644 index 00000000000..7ef98f88cc7 --- /dev/null +++ b/tests/queries/0_stateless/03214_join_on_tuple_comparison_elimination_bug.sql @@ -0,0 +1,16 @@ +DROP TABLE IF EXISTS a; +DROP TABLE IF EXISTS b; + +CREATE TABLE a (key Nullable(String)) ENGINE = MergeTree ORDER BY tuple(); +INSERT INTO a VALUES (NULL), ('1'); + +CREATE TABLE b (key Nullable(String)) ENGINE = MergeTree ORDER BY tuple(); +INSERT INTO b VALUES (NULL), ('1'); + +SELECT a.key FROM a LEFT SEMI JOIN b ON tuple(a.key) = tuple(b.key) ORDER BY a.key; +SELECT a.key FROM a LEFT SEMI JOIN b ON a.key IS NOT DISTINCT FROM b.key ORDER BY a.key; +SELECT a.key FROM a LEFT SEMI JOIN b ON tuple(a.key) = tuple(b.key) ORDER BY a.key; +SELECT a.key FROM a LEFT ANY JOIN b ON tuple(a.key) = tuple(b.key) ORDER BY a.key; + +DROP TABLE IF EXISTS a; +DROP TABLE IF EXISTS b; From 8fc77bec6639c8f0361858d9e031f3cb1175ae30 Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Thu, 1 Aug 2024 13:14:14 +0200 Subject: [PATCH 103/227] fix tests --- ...403_enable_extended_results_for_datetime_functions.reference | 2 +- .../0_stateless/03215_toStartOfWeek_with_dateTime64_fix.sql | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/queries/0_stateless/02403_enable_extended_results_for_datetime_functions.reference b/tests/queries/0_stateless/02403_enable_extended_results_for_datetime_functions.reference index c830d790000..147e368b5c9 100644 --- a/tests/queries/0_stateless/02403_enable_extended_results_for_datetime_functions.reference +++ b/tests/queries/0_stateless/02403_enable_extended_results_for_datetime_functions.reference @@ -64,7 +64,7 @@ toStartOfMonth;toDateTime64;false 2099-07-07 type;toStartOfMonth;toDateTime64;false Date toStartOfWeek;toDate32;false 2099-07-07 type;toStartOfWeek;toDate32;false Date -toStartOfWeek;toDateTime64;false 2099-07-07 +toStartOfWeek;toDateTime64;false 1970-01-01 type;toStartOfWeek;toDateTime64;false Date toMonday;toDate32;false 2099-07-08 type;toMonday;toDate32;false Date diff --git a/tests/queries/0_stateless/03215_toStartOfWeek_with_dateTime64_fix.sql b/tests/queries/0_stateless/03215_toStartOfWeek_with_dateTime64_fix.sql index 0f00a52cb86..1769d96aa8d 100644 --- a/tests/queries/0_stateless/03215_toStartOfWeek_with_dateTime64_fix.sql +++ b/tests/queries/0_stateless/03215_toStartOfWeek_with_dateTime64_fix.sql @@ -1,2 +1,2 @@ -SELECT toStartOfWeek(toDateTime64('1970-02-01', 6)); +SELECT toStartOfWeek(toDateTime64('1970-01-01', 6)); SELECT toStartOfWeek(toDateTime('1970-01-01')); From d0b514f12843fd09f18888666aa98170a52fb9d5 Mon Sep 17 00:00:00 2001 From: kssenii Date: Thu, 1 Aug 2024 13:27:09 +0200 Subject: [PATCH 104/227] Review fixes --- src/Interpreters/Cache/FileCache.cpp | 12 ++++++------ src/Interpreters/Cache/IFileCachePriority.h | 6 +++--- src/Interpreters/Cache/LRUFileCachePriority.cpp | 10 +++++----- src/Interpreters/Cache/LRUFileCachePriority.h | 2 +- src/Interpreters/Cache/SLRUFileCachePriority.cpp | 6 +++--- src/Interpreters/Cache/SLRUFileCachePriority.h | 2 +- 6 files changed, 19 insertions(+), 19 deletions(-) diff --git a/src/Interpreters/Cache/FileCache.cpp b/src/Interpreters/Cache/FileCache.cpp index bf8dd24a1db..1f79c7d9032 100644 --- a/src/Interpreters/Cache/FileCache.cpp +++ b/src/Interpreters/Cache/FileCache.cpp @@ -998,7 +998,7 @@ void FileCache::freeSpaceRatioKeepingThreadFunc() FileCacheReserveStat stat; EvictionCandidates eviction_candidates; - IFileCachePriority::DesiredSizeStatus desired_size_status; + IFileCachePriority::CollectStatus desired_size_status; try { /// Collect at most `keep_up_free_space_remove_batch` elements to evict, @@ -1009,7 +1009,7 @@ void FileCache::freeSpaceRatioKeepingThreadFunc() #ifdef DEBUG_OR_SANITIZER_BUILD /// Let's make sure that we correctly processed the limits. - if (desired_size_status == IFileCachePriority::DesiredSizeStatus::SUCCESS + if (desired_size_status == IFileCachePriority::CollectStatus::SUCCESS && eviction_candidates.size() < keep_up_free_space_remove_batch) { const auto current_size = main_priority->getSize(lock); @@ -1070,13 +1070,13 @@ void FileCache::freeSpaceRatioKeepingThreadFunc() [[maybe_unused]] bool scheduled = false; switch (desired_size_status) { - case IFileCachePriority::DesiredSizeStatus::SUCCESS: [[fallthrough]]; - case IFileCachePriority::DesiredSizeStatus::CANNOT_EVICT: + case IFileCachePriority::CollectStatus::SUCCESS: [[fallthrough]]; + case IFileCachePriority::CollectStatus::CANNOT_EVICT: { scheduled = keep_up_free_space_ratio_task->scheduleAfter(general_reschedule_ms); break; } - case IFileCachePriority::DesiredSizeStatus::REACHED_MAX_CANDIDATES_LIMIT: + case IFileCachePriority::CollectStatus::REACHED_MAX_CANDIDATES_LIMIT: { scheduled = keep_up_free_space_ratio_task->schedule(); break; @@ -1558,7 +1558,7 @@ void FileCache::applySettingsIfPossible(const FileCacheSettings & new_settings, FileCacheReserveStat stat; if (main_priority->collectCandidatesForEviction( new_settings.max_size, new_settings.max_elements, 0/* max_candidates_to_evict */, - stat, eviction_candidates, cache_lock) == IFileCachePriority::DesiredSizeStatus::SUCCESS) + stat, eviction_candidates, cache_lock) == IFileCachePriority::CollectStatus::SUCCESS) { if (eviction_candidates.size() == 0) { diff --git a/src/Interpreters/Cache/IFileCachePriority.h b/src/Interpreters/Cache/IFileCachePriority.h index 9885ab00f78..6970d02473a 100644 --- a/src/Interpreters/Cache/IFileCachePriority.h +++ b/src/Interpreters/Cache/IFileCachePriority.h @@ -150,14 +150,14 @@ public: /// Collect eviction candidates sufficient to have `desired_size` /// and `desired_elements_num` as current cache state. /// Collect no more than `max_candidates_to_evict` elements. - /// Return `true` if the first condition is satisfied. - enum class DesiredSizeStatus + /// Return SUCCESS status if the first condition is satisfied. + enum class CollectStatus { SUCCESS, CANNOT_EVICT, REACHED_MAX_CANDIDATES_LIMIT, }; - virtual DesiredSizeStatus collectCandidatesForEviction( + virtual CollectStatus collectCandidatesForEviction( size_t desired_size, size_t desired_elements_count, size_t max_candidates_to_evict, diff --git a/src/Interpreters/Cache/LRUFileCachePriority.cpp b/src/Interpreters/Cache/LRUFileCachePriority.cpp index 7970eaa3e13..0e0170c76e3 100644 --- a/src/Interpreters/Cache/LRUFileCachePriority.cpp +++ b/src/Interpreters/Cache/LRUFileCachePriority.cpp @@ -323,7 +323,7 @@ bool LRUFileCachePriority::collectCandidatesForEviction( } } -IFileCachePriority::DesiredSizeStatus LRUFileCachePriority::collectCandidatesForEviction( +IFileCachePriority::CollectStatus LRUFileCachePriority::collectCandidatesForEviction( size_t desired_size, size_t desired_elements_count, size_t max_candidates_to_evict, @@ -336,23 +336,23 @@ IFileCachePriority::DesiredSizeStatus LRUFileCachePriority::collectCandidatesFor return canFit(0, 0, stat.total_stat.releasable_size, stat.total_stat.releasable_count, lock, &desired_size, &desired_elements_count); }; - auto status = DesiredSizeStatus::CANNOT_EVICT; + auto status = CollectStatus::CANNOT_EVICT; auto stop_condition = [&]() { if (desired_limits_satisfied()) { - status = DesiredSizeStatus::SUCCESS; + status = CollectStatus::SUCCESS; return true; } if (max_candidates_to_evict && res.size() >= max_candidates_to_evict) { - status = DesiredSizeStatus::REACHED_MAX_CANDIDATES_LIMIT; + status = CollectStatus::REACHED_MAX_CANDIDATES_LIMIT; return true; } return false; }; iterateForEviction(res, stat, stop_condition, lock); - chassert(status != DesiredSizeStatus::SUCCESS || stop_condition()); + chassert(status != CollectStatus::SUCCESS || stop_condition()); return status; } diff --git a/src/Interpreters/Cache/LRUFileCachePriority.h b/src/Interpreters/Cache/LRUFileCachePriority.h index 9bced106727..0ca62b19d37 100644 --- a/src/Interpreters/Cache/LRUFileCachePriority.h +++ b/src/Interpreters/Cache/LRUFileCachePriority.h @@ -63,7 +63,7 @@ public: const UserID & user_id, const CachePriorityGuard::Lock &) override; - DesiredSizeStatus collectCandidatesForEviction( + CollectStatus collectCandidatesForEviction( size_t desired_size, size_t desired_elements_count, size_t max_candidates_to_evict, diff --git a/src/Interpreters/Cache/SLRUFileCachePriority.cpp b/src/Interpreters/Cache/SLRUFileCachePriority.cpp index dc0df223cb0..f5ea519d7d4 100644 --- a/src/Interpreters/Cache/SLRUFileCachePriority.cpp +++ b/src/Interpreters/Cache/SLRUFileCachePriority.cpp @@ -256,7 +256,7 @@ bool SLRUFileCachePriority::collectCandidatesForEvictionInProtected( return true; } -IFileCachePriority::DesiredSizeStatus SLRUFileCachePriority::collectCandidatesForEviction( +IFileCachePriority::CollectStatus SLRUFileCachePriority::collectCandidatesForEviction( size_t desired_size, size_t desired_elements_count, size_t max_candidates_to_evict, @@ -285,7 +285,7 @@ IFileCachePriority::DesiredSizeStatus SLRUFileCachePriority::collectCandidatesFo chassert(!max_candidates_to_evict || res.size() <= max_candidates_to_evict); chassert(res.size() == stat.total_stat.releasable_count); - if (probationary_desired_size_status == DesiredSizeStatus::REACHED_MAX_CANDIDATES_LIMIT) + if (probationary_desired_size_status == CollectStatus::REACHED_MAX_CANDIDATES_LIMIT) return probationary_desired_size_status; const auto desired_protected_size = getRatio(desired_size, size_ratio); @@ -306,7 +306,7 @@ IFileCachePriority::DesiredSizeStatus SLRUFileCachePriority::collectCandidatesFo desired_protected_size, desired_protected_elements_num, protected_queue.getStateInfoForLog(lock)); - if (probationary_desired_size_status == DesiredSizeStatus::SUCCESS) + if (probationary_desired_size_status == CollectStatus::SUCCESS) return protected_desired_size_status; else return probationary_desired_size_status; diff --git a/src/Interpreters/Cache/SLRUFileCachePriority.h b/src/Interpreters/Cache/SLRUFileCachePriority.h index e6d20e0d0ee..23bc8c0908b 100644 --- a/src/Interpreters/Cache/SLRUFileCachePriority.h +++ b/src/Interpreters/Cache/SLRUFileCachePriority.h @@ -58,7 +58,7 @@ public: const UserID & user_id, const CachePriorityGuard::Lock &) override; - DesiredSizeStatus collectCandidatesForEviction( + CollectStatus collectCandidatesForEviction( size_t desired_size, size_t desired_elements_count, size_t max_candidates_to_evict, From 2c018d6f4d4d6fa1c04e91306b60b6e85d8e468f Mon Sep 17 00:00:00 2001 From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com> Date: Thu, 1 Aug 2024 13:38:41 +0200 Subject: [PATCH 105/227] Update cluster.py --- tests/integration/helpers/cluster.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/integration/helpers/cluster.py b/tests/integration/helpers/cluster.py index 90b28a4cda3..acf033de46d 100644 --- a/tests/integration/helpers/cluster.py +++ b/tests/integration/helpers/cluster.py @@ -248,7 +248,7 @@ def check_rabbitmq_is_available(rabbitmq_id, cookie): ), stdout=subprocess.PIPE, ) - p.communicate() + p.wait(timeout=60) return p.returncode == 0 From 53d93b177e371b8dcc13e404f48209e3df456fe4 Mon Sep 17 00:00:00 2001 From: vdimir Date: Thu, 1 Aug 2024 11:56:07 +0000 Subject: [PATCH 106/227] Fix 02434_cancel_insert_when_client_dies --- .../02434_cancel_insert_when_client_dies.sh | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/tests/queries/0_stateless/02434_cancel_insert_when_client_dies.sh b/tests/queries/0_stateless/02434_cancel_insert_when_client_dies.sh index f85aaed7716..45f4194104e 100755 --- a/tests/queries/0_stateless/02434_cancel_insert_when_client_dies.sh +++ b/tests/queries/0_stateless/02434_cancel_insert_when_client_dies.sh @@ -10,22 +10,26 @@ export DATA_FILE="$CLICKHOUSE_TMP/deduptest.tsv" export TEST_MARK="02434_insert_${CLICKHOUSE_DATABASE}_" $CLICKHOUSE_CLIENT -q 'select * from numbers(5000000) format TSV' > $DATA_FILE -$CLICKHOUSE_CLIENT -q 'create table dedup_test(A Int64) Engine = MergeTree order by A settings non_replicated_deduplication_window=1000;' +$CLICKHOUSE_CLIENT -q "create table dedup_test(A Int64) Engine = MergeTree order by A + settings non_replicated_deduplication_window=1000 + , merge_tree_clear_old_temporary_directories_interval_seconds = 1 + ;" $CLICKHOUSE_CLIENT -q "create table dedup_dist(A Int64) Engine = Distributed('test_cluster_one_shard_two_replicas', currentDatabase(), dedup_test)" function insert_data { - SETTINGS="query_id=$ID&max_insert_block_size=110000&min_insert_block_size_rows=110000" + # send_logs_level: https://github.com/ClickHouse/ClickHouse/issues/67599 + SETTINGS="query_id=$ID&max_insert_block_size=110000&min_insert_block_size_rows=110000&send_logs_level=fatal" # max_block_size=10000, so external table will contain smaller blocks that will be squashed on insert-select (more chances to catch a bug on query cancellation) TRASH_SETTINGS="query_id=$ID&input_format_parallel_parsing=0&max_threads=1&max_insert_threads=1&max_insert_block_size=110000&max_block_size=10000&min_insert_block_size_bytes=0&min_insert_block_size_rows=110000&max_insert_block_size=110000" TYPE=$(( RANDOM % 5 )) if [[ "$TYPE" -eq 0 ]]; then # client will send 10000-rows blocks, server will squash them into 110000-rows blocks (more chances to catch a bug on query cancellation) - $CLICKHOUSE_CLIENT --max_block_size=10000 --max_insert_block_size=10000 --query_id="$ID" \ + $CLICKHOUSE_CLIENT --allow_repeated_settings --send_logs_level=fatal --max_block_size=10000 --max_insert_block_size=10000 --query_id="$ID" \ -q 'insert into dedup_test settings max_insert_block_size=110000, min_insert_block_size_rows=110000 format TSV' < $DATA_FILE elif [[ "$TYPE" -eq 1 ]]; then - $CLICKHOUSE_CLIENT --max_block_size=10000 --max_insert_block_size=10000 --query_id="$ID" --prefer_localhost_replica="$(( RANDOM % 2))" \ + $CLICKHOUSE_CLIENT --allow_repeated_settings --send_logs_level=fatal --max_block_size=10000 --max_insert_block_size=10000 --query_id="$ID" --prefer_localhost_replica="$(( RANDOM % 2))" \ -q 'insert into dedup_dist settings max_insert_block_size=110000, min_insert_block_size_rows=110000 format TSV' < $DATA_FILE elif [[ "$TYPE" -eq 2 ]]; then $CLICKHOUSE_CURL -sS -X POST --data-binary @- "$CLICKHOUSE_URL&$SETTINGS&query=insert+into+dedup_test+format+TSV" < $DATA_FILE From 029deaeee8431d0ef6f2a460c2bd8631c8025254 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Thu, 1 Aug 2024 14:13:11 +0200 Subject: [PATCH 107/227] Fix 02910_bad_logs_level_in_local in fast tests! --- tests/queries/0_stateless/02910_bad_logs_level_in_local.sh | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/queries/0_stateless/02910_bad_logs_level_in_local.sh b/tests/queries/0_stateless/02910_bad_logs_level_in_local.sh index badf7232a95..b5de10bf191 100755 --- a/tests/queries/0_stateless/02910_bad_logs_level_in_local.sh +++ b/tests/queries/0_stateless/02910_bad_logs_level_in_local.sh @@ -1,14 +1,14 @@ #!/usr/bin/expect -f log_user 0 -set timeout 60 +set timeout 30 match_max 100000 spawn bash -c "clickhouse-local" expect ":) " send -- "SET send_logs_level = 't'\r" -expect "Exception on client:" +expect "Unexpected value of LogsLevel:" {} timeout {exit 1} expect ":) " send -- "exit\r" expect eof From 1d85f9b1cba3c8fe168286a660d3c0a4fd471a95 Mon Sep 17 00:00:00 2001 From: Michael Stetsyuk Date: Thu, 1 Aug 2024 14:42:58 +0100 Subject: [PATCH 108/227] fix remove_local_directory_contents --- tests/integration/test_storage_delta/test.py | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/tests/integration/test_storage_delta/test.py b/tests/integration/test_storage_delta/test.py index e485bc90ee0..384b8296f66 100644 --- a/tests/integration/test_storage_delta/test.py +++ b/tests/integration/test_storage_delta/test.py @@ -52,9 +52,13 @@ def get_spark(): return builder.master("local").getOrCreate() -def remove_local_directory_contents(local_path): - for local_file in glob.glob(local_path + "/**"): - os.unlink(local_file) +def remove_local_directory_contents(full_path): + for path in glob.glob(f"{full_path}/**"): + if os.path.isfile(path): + os.unlink(path) + else: + remove_local_directory_contents(path) + os.rmdir(path) @pytest.fixture(scope="module") From 0978441a0261c6003c7a9f4661ac87138e909622 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Wed, 31 Jul 2024 17:38:20 +0200 Subject: [PATCH 109/227] Minor change --- src/Databases/DatabaseOnDisk.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Databases/DatabaseOnDisk.cpp b/src/Databases/DatabaseOnDisk.cpp index f419f5811a1..734f354d9a5 100644 --- a/src/Databases/DatabaseOnDisk.cpp +++ b/src/Databases/DatabaseOnDisk.cpp @@ -313,7 +313,7 @@ void DatabaseOnDisk::detachTablePermanently(ContextPtr query_context, const Stri std::lock_guard lock(mutex); if (const auto it = snapshot_detached_tables.find(table_name); it == snapshot_detached_tables.end()) { - throw Exception(ErrorCodes::LOGICAL_ERROR, "Snapshot doesn't contain info about detached table={}", table_name); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Snapshot doesn't contain info about detached table `{}`", table_name); } else { From d683fb05a009ed3f58c0e11fc329c3783f934369 Mon Sep 17 00:00:00 2001 From: kssenii Date: Thu, 1 Aug 2024 19:05:30 +0200 Subject: [PATCH 110/227] Fix --- .../IO/CachedOnDiskReadBufferFromFile.cpp | 17 ++++++--- src/Interpreters/Cache/FileCache.cpp | 35 ++++++++++++------- src/Interpreters/Cache/FileCache.h | 4 ++- src/Interpreters/Cache/FileSegment.cpp | 9 ++++- src/Interpreters/Cache/FileSegment.h | 4 ++- tests/config/config.d/storage_conf.xml | 3 +- 6 files changed, 51 insertions(+), 21 deletions(-) diff --git a/src/Disks/IO/CachedOnDiskReadBufferFromFile.cpp b/src/Disks/IO/CachedOnDiskReadBufferFromFile.cpp index c928d25c7b8..b471f3fc58f 100644 --- a/src/Disks/IO/CachedOnDiskReadBufferFromFile.cpp +++ b/src/Disks/IO/CachedOnDiskReadBufferFromFile.cpp @@ -135,8 +135,11 @@ bool CachedOnDiskReadBufferFromFile::nextFileSegmentsBatch() else { CreateFileSegmentSettings create_settings(FileSegmentKind::Regular); - file_segments = cache->getOrSet(cache_key, file_offset_of_buffer_end, size, file_size.value(), create_settings, settings.filesystem_cache_segments_batch_size, user); + file_segments = cache->getOrSet( + cache_key, file_offset_of_buffer_end, size, file_size.value(), + create_settings, settings.filesystem_cache_segments_batch_size, user); } + return !file_segments->empty(); } @@ -158,8 +161,8 @@ void CachedOnDiskReadBufferFromFile::initialize() LOG_TEST( log, - "Having {} file segments to read: {}, current offset: {}", - file_segments->size(), file_segments->toString(), file_offset_of_buffer_end); + "Having {} file segments to read: {}, current read range: [{}, {})", + file_segments->size(), file_segments->toString(), file_offset_of_buffer_end, read_until_position); initialized = true; } @@ -1043,6 +1046,10 @@ bool CachedOnDiskReadBufferFromFile::nextImplStep() if (file_segments->size() == 1) { size_t remaining_size_to_read = std::min(current_read_range.right, read_until_position - 1) - file_offset_of_buffer_end + 1; + + LOG_TEST(log, "Remaining size to read: {}, read: {}. Resizing buffer to {}", + remaining_size_to_read, size, nextimpl_working_buffer_offset + std::min(size, remaining_size_to_read)); + size = std::min(size, remaining_size_to_read); chassert(implementation_buffer->buffer().size() >= nextimpl_working_buffer_offset + size); implementation_buffer->buffer().resize(nextimpl_working_buffer_offset + size); @@ -1055,8 +1062,8 @@ bool CachedOnDiskReadBufferFromFile::nextImplStep() chassert( file_offset_of_buffer_end <= read_until_position, - fmt::format("Expected {} <= {} (size: {}, read range: {})", - file_offset_of_buffer_end, read_until_position, size, current_read_range.toString())); + fmt::format("Expected {} <= {} (size: {}, read range: {}, hold file segments: {} ({}))", + file_offset_of_buffer_end, read_until_position, size, current_read_range.toString(), file_segments->size(), file_segments->toString(true))); } swap(*implementation_buffer); diff --git a/src/Interpreters/Cache/FileCache.cpp b/src/Interpreters/Cache/FileCache.cpp index a88c0de2cfe..0a03f5dcc7d 100644 --- a/src/Interpreters/Cache/FileCache.cpp +++ b/src/Interpreters/Cache/FileCache.cpp @@ -316,14 +316,14 @@ FileSegments FileCache::getImpl(const LockedKey & locked_key, const FileSegment: return result; } -std::vector FileCache::splitRange(size_t offset, size_t size) +std::vector FileCache::splitRange(size_t offset, size_t size, size_t aligned_size) { assert(size > 0); std::vector ranges; size_t current_pos = offset; size_t end_pos_non_included = offset + size; - size_t remaining_size = size; + size_t remaining_size = aligned_size; FileSegments file_segments; const size_t max_size = max_file_segment_size.load(); @@ -343,17 +343,20 @@ FileSegments FileCache::splitRangeIntoFileSegments( LockedKey & locked_key, size_t offset, size_t size, + size_t aligned_size, FileSegment::State state, size_t file_segments_limit, const CreateFileSegmentSettings & create_settings) { - assert(size > 0); + chassert(size > 0); + chassert(size <= aligned_size); + /// We take `size` as a soft limit and `aligned_size` as a hard limit. auto current_pos = offset; auto end_pos_non_included = offset + size; size_t current_file_segment_size; - size_t remaining_size = size; + size_t remaining_size = aligned_size; FileSegments file_segments; const size_t max_size = max_file_segment_size.load(); @@ -369,6 +372,8 @@ FileSegments FileCache::splitRangeIntoFileSegments( current_pos += current_file_segment_size; } + chassert(file_segments.size() == file_segments_limit || file_segments.back()->range().contains(offset + size - 1), + fmt::format("Offset: {}, size: {}, file segments: {}", offset, size, toString(file_segments))); return file_segments; } @@ -376,6 +381,7 @@ void FileCache::fillHolesWithEmptyFileSegments( LockedKey & locked_key, FileSegments & file_segments, const FileSegment::Range & range, + size_t non_aligned_right_offset, size_t file_segments_limit, bool fill_with_detached_file_segments, const CreateFileSegmentSettings & create_settings) @@ -442,7 +448,7 @@ void FileCache::fillHolesWithEmptyFileSegments( } else { - auto ranges = splitRange(current_pos, hole_size); + auto ranges = splitRange(current_pos, hole_size, hole_size); FileSegments hole; for (const auto & r : ranges) { @@ -479,7 +485,7 @@ void FileCache::fillHolesWithEmptyFileSegments( chassert(!file_segments_limit || file_segments.size() < file_segments_limit); - if (current_pos <= range.right) + if (current_pos <= non_aligned_right_offset) { /// ________] -- requested range /// _____] @@ -487,6 +493,7 @@ void FileCache::fillHolesWithEmptyFileSegments( /// segmentN auto hole_size = range.right - current_pos + 1; + auto non_aligned_size = non_aligned_right_offset - current_pos + 1; if (fill_with_detached_file_segments) { @@ -497,7 +504,7 @@ void FileCache::fillHolesWithEmptyFileSegments( } else { - auto ranges = splitRange(current_pos, hole_size); + auto ranges = splitRange(current_pos, non_aligned_size, hole_size); FileSegments hole; for (const auto & r : ranges) { @@ -542,7 +549,7 @@ FileSegmentsHolderPtr FileCache::set( else { file_segments = splitRangeIntoFileSegments( - *locked_key, offset, size, FileSegment::State::EMPTY, /* file_segments_limit */0, create_settings); + *locked_key, offset, size, size, FileSegment::State::EMPTY, /* file_segments_limit */0, create_settings); } return std::make_unique(std::move(file_segments)); @@ -659,9 +666,13 @@ FileCache::getOrSet( } } + chassert(range.left >= aligned_offset); + if (file_segments.empty()) { - file_segments = splitRangeIntoFileSegments(*locked_key, range.left, range.size(), FileSegment::State::EMPTY, file_segments_limit, create_settings); + file_segments = splitRangeIntoFileSegments( + *locked_key, range.left, /* size */offset + size - range.left, /* aligned_size */range.size(), + FileSegment::State::EMPTY, file_segments_limit, create_settings); } else { @@ -669,9 +680,9 @@ FileCache::getOrSet( chassert(file_segments.back()->range().left <= range.right); fillHolesWithEmptyFileSegments( - *locked_key, file_segments, range, file_segments_limit, /* fill_with_detached */false, create_settings); + *locked_key, file_segments, range, offset + size - 1, file_segments_limit, /* fill_with_detached */false, create_settings); - if (!file_segments.front()->range().contains(offset)) + if (!file_segments.front()->range().contains(range.left)) { throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected {} to include {} " "(end offset: {}, aligned offset: {}, aligned end offset: {})", @@ -713,7 +724,7 @@ FileSegmentsHolderPtr FileCache::get( } fillHolesWithEmptyFileSegments( - *locked_key, file_segments, range, file_segments_limit, /* fill_with_detached */true, CreateFileSegmentSettings{}); + *locked_key, file_segments, range, offset + size - 1, file_segments_limit, /* fill_with_detached */true, CreateFileSegmentSettings{}); chassert(!file_segments_limit || file_segments.size() <= file_segments_limit); return std::make_unique(std::move(file_segments)); diff --git a/src/Interpreters/Cache/FileCache.h b/src/Interpreters/Cache/FileCache.h index 527fd9d5edf..3f7eec73b56 100644 --- a/src/Interpreters/Cache/FileCache.h +++ b/src/Interpreters/Cache/FileCache.h @@ -263,7 +263,7 @@ private: /// Split range into subranges by max_file_segment_size, /// each subrange size must be less or equal to max_file_segment_size. - std::vector splitRange(size_t offset, size_t size); + std::vector splitRange(size_t offset, size_t size, size_t aligned_size); /// Split range into subranges by max_file_segment_size (same as in splitRange()) /// and create a new file segment for each subrange. @@ -273,6 +273,7 @@ private: LockedKey & locked_key, size_t offset, size_t size, + size_t aligned_size, FileSegment::State state, size_t file_segments_limit, const CreateFileSegmentSettings & create_settings); @@ -281,6 +282,7 @@ private: LockedKey & locked_key, FileSegments & file_segments, const FileSegment::Range & range, + size_t non_aligned_right_offset, size_t file_segments_limit, bool fill_with_detached_file_segments, const CreateFileSegmentSettings & settings); diff --git a/src/Interpreters/Cache/FileSegment.cpp b/src/Interpreters/Cache/FileSegment.cpp index 1664a91b694..c46fb978ae4 100644 --- a/src/Interpreters/Cache/FileSegment.cpp +++ b/src/Interpreters/Cache/FileSegment.cpp @@ -1008,7 +1008,12 @@ FileSegment & FileSegmentsHolder::add(FileSegmentPtr && file_segment) return *file_segments.back(); } -String FileSegmentsHolder::toString() +String FileSegmentsHolder::toString(bool with_state) +{ + return DB::toString(file_segments, with_state); +} + +String toString(const FileSegments & file_segments, bool with_state) { String ranges; for (const auto & file_segment : file_segments) @@ -1018,6 +1023,8 @@ String FileSegmentsHolder::toString() ranges += file_segment->range().toString(); if (file_segment->isUnbound()) ranges += "(unbound)"; + if (with_state) + ranges += "(" + FileSegment::stateToString(file_segment->state()) + ")"; } return ranges; } diff --git a/src/Interpreters/Cache/FileSegment.h b/src/Interpreters/Cache/FileSegment.h index d6b37b60dc1..25ffb880b45 100644 --- a/src/Interpreters/Cache/FileSegment.h +++ b/src/Interpreters/Cache/FileSegment.h @@ -291,7 +291,7 @@ struct FileSegmentsHolder : private boost::noncopyable size_t size() const { return file_segments.size(); } - String toString(); + String toString(bool with_state = false); void popFront() { completeAndPopFrontImpl(); } @@ -317,4 +317,6 @@ private: using FileSegmentsHolderPtr = std::unique_ptr; +String toString(const FileSegments & file_segments, bool with_state = false); + } diff --git a/tests/config/config.d/storage_conf.xml b/tests/config/config.d/storage_conf.xml index 7a9b579c00a..4daa64b520d 100644 --- a/tests/config/config.d/storage_conf.xml +++ b/tests/config/config.d/storage_conf.xml @@ -19,7 +19,8 @@ cache s3_disk s3_cache/ - 104857600 + 100Mi + 5Mi 1 100 LRU From fc72742e37ae78f47b3e55a969e088f5c372ee36 Mon Sep 17 00:00:00 2001 From: sakulali Date: Fri, 2 Aug 2024 02:00:20 +0800 Subject: [PATCH 111/227] ping CI From 67b11300e45f6e24c3515a978d23a9bc998a666e Mon Sep 17 00:00:00 2001 From: Michael Kolupaev Date: Thu, 1 Aug 2024 21:05:49 +0000 Subject: [PATCH 112/227] Fix crash in KeyCondition::cloneASTWithInversionPushDown() caused by type change --- src/Storages/MergeTree/KeyCondition.cpp | 82 +++++++++---------- .../03215_key_condition_bug.reference | 1 + .../0_stateless/03215_key_condition_bug.sql | 3 + 3 files changed, 44 insertions(+), 42 deletions(-) create mode 100644 tests/queries/0_stateless/03215_key_condition_bug.reference create mode 100644 tests/queries/0_stateless/03215_key_condition_bug.sql diff --git a/src/Storages/MergeTree/KeyCondition.cpp b/src/Storages/MergeTree/KeyCondition.cpp index 69bffac9160..eaf9f0af623 100644 --- a/src/Storages/MergeTree/KeyCondition.cpp +++ b/src/Storages/MergeTree/KeyCondition.cpp @@ -566,6 +566,7 @@ static const ActionsDAG::Node & cloneASTWithInversionPushDown( } const ActionsDAG::Node * res = nullptr; + bool handled_inversion = false; switch (node.type) { @@ -582,7 +583,7 @@ static const ActionsDAG::Node & cloneASTWithInversionPushDown( /// Re-generate column name for constant. /// DAG form query (with enabled analyzer) uses suffixes for constants, like 1_UInt8. /// DAG from PK does not use it. This breaks matching by column name sometimes. - /// Ideally, we should not compare manes, but DAG subtrees instead. + /// Ideally, we should not compare names, but DAG subtrees instead. name = ASTLiteral(column_const->getDataColumn()[0]).getColumnName(); else name = node.result_name; @@ -593,9 +594,9 @@ static const ActionsDAG::Node & cloneASTWithInversionPushDown( case (ActionsDAG::ActionType::ALIAS): { /// Ignore aliases - const auto & alias = cloneASTWithInversionPushDown(*node.children.front(), inverted_dag, to_inverted, context, need_inversion); - to_inverted[&node] = &alias; - return alias; + res = &cloneASTWithInversionPushDown(*node.children.front(), inverted_dag, to_inverted, context, need_inversion); + handled_inversion = true; + break; } case (ActionsDAG::ActionType::ARRAY_JOIN): { @@ -608,20 +609,10 @@ static const ActionsDAG::Node & cloneASTWithInversionPushDown( auto name = node.function_base->getName(); if (name == "not") { - const auto & arg = cloneASTWithInversionPushDown(*node.children.front(), inverted_dag, to_inverted, context, !need_inversion); - to_inverted[&node] = &arg; - return arg; + res = &cloneASTWithInversionPushDown(*node.children.front(), inverted_dag, to_inverted, context, !need_inversion); + handled_inversion = true; } - - if (name == "materialize") - { - /// Ignore materialize - const auto & arg = cloneASTWithInversionPushDown(*node.children.front(), inverted_dag, to_inverted, context, need_inversion); - to_inverted[&node] = &arg; - return arg; - } - - if (name == "indexHint") + else if (name == "indexHint") { ActionsDAG::NodeRawConstPtrs children; if (const auto * adaptor = typeid_cast(node.function_base.get())) @@ -636,12 +627,10 @@ static const ActionsDAG::Node & cloneASTWithInversionPushDown( } } - const auto & func = inverted_dag.addFunction(node.function_base, children, ""); - to_inverted[&node] = &func; - return func; + res = &inverted_dag.addFunction(node.function_base, children, ""); + handled_inversion = true; } - - if (need_inversion && (name == "and" || name == "or")) + else if (need_inversion && (name == "and" || name == "or")) { ActionsDAG::NodeRawConstPtrs children(node.children); @@ -659,34 +648,43 @@ static const ActionsDAG::Node & cloneASTWithInversionPushDown( /// We match columns by name, so it is important to fill name correctly. /// So, use empty string to make it automatically. - const auto & func = inverted_dag.addFunction(function_builder, children, ""); - to_inverted[&node] = &func; - return func; + res = &inverted_dag.addFunction(function_builder, children, ""); + handled_inversion = true; } - - ActionsDAG::NodeRawConstPtrs children(node.children); - - for (auto & arg : children) - arg = &cloneASTWithInversionPushDown(*arg, inverted_dag, to_inverted, context, false); - - auto it = inverse_relations.find(name); - if (it != inverse_relations.end()) + else { - const auto & func_name = need_inversion ? it->second : it->first; - auto function_builder = FunctionFactory::instance().get(func_name, context); - const auto & func = inverted_dag.addFunction(function_builder, children, ""); - to_inverted[&node] = &func; - return func; - } + ActionsDAG::NodeRawConstPtrs children(node.children); - res = &inverted_dag.addFunction(node.function_base, children, ""); - chassert(res->result_type == node.result_type); + for (auto & arg : children) + arg = &cloneASTWithInversionPushDown(*arg, inverted_dag, to_inverted, context, false); + + auto it = inverse_relations.find(name); + if (it != inverse_relations.end()) + { + const auto & func_name = need_inversion ? it->second : it->first; + auto function_builder = FunctionFactory::instance().get(func_name, context); + res = &inverted_dag.addFunction(function_builder, children, ""); + handled_inversion = true; + } + else + { + res = &inverted_dag.addFunction(node.function_base, children, ""); + chassert(res->result_type == node.result_type); + } + } } } - if (need_inversion) + if (!handled_inversion && need_inversion) res = &inverted_dag.addFunction(FunctionFactory::instance().get("not", context), {res}, ""); + /// Make sure we don't change any data types (e.g. remove LowCardinality). + /// If it turns out that we actually want to change data types sometimes, it's ok to remove this + /// check *and* replace all `addFunction(node.function_base, ...)` calls above with + /// `addFunction(FunctionFactory::instance().get(name, context), ...)` to re-resolve overloads. + if (!node.result_type->equals(*res->result_type)) + throw Exception(ErrorCodes::LOGICAL_ERROR, "KeyCondition inadvertently changed subexpression data type: '{}' -> '{}', column `{}`", node.result_type->getName(), res->result_type->getName(), node.result_name); + to_inverted[&node] = res; return *res; } diff --git a/tests/queries/0_stateless/03215_key_condition_bug.reference b/tests/queries/0_stateless/03215_key_condition_bug.reference new file mode 100644 index 00000000000..84ab67a85e0 --- /dev/null +++ b/tests/queries/0_stateless/03215_key_condition_bug.reference @@ -0,0 +1 @@ +(0) diff --git a/tests/queries/0_stateless/03215_key_condition_bug.sql b/tests/queries/0_stateless/03215_key_condition_bug.sql new file mode 100644 index 00000000000..ef2113e81f8 --- /dev/null +++ b/tests/queries/0_stateless/03215_key_condition_bug.sql @@ -0,0 +1,3 @@ +CREATE TABLE t (x Int8) ENGINE MergeTree ORDER BY x; +INSERT INTO t VALUES (1); +SELECT arrayJoin([tuple((toNullable(10) * toLowCardinality(20)) < materialize(30))]) AS row FROM t WHERE row.1 = 0; \ No newline at end of file From eac2c9fc3d8a88c1033e0f23e048421ecf4db850 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Wed, 31 Jul 2024 17:38:20 +0200 Subject: [PATCH 113/227] Minor change --- src/Databases/DatabaseOnDisk.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Databases/DatabaseOnDisk.cpp b/src/Databases/DatabaseOnDisk.cpp index f419f5811a1..734f354d9a5 100644 --- a/src/Databases/DatabaseOnDisk.cpp +++ b/src/Databases/DatabaseOnDisk.cpp @@ -313,7 +313,7 @@ void DatabaseOnDisk::detachTablePermanently(ContextPtr query_context, const Stri std::lock_guard lock(mutex); if (const auto it = snapshot_detached_tables.find(table_name); it == snapshot_detached_tables.end()) { - throw Exception(ErrorCodes::LOGICAL_ERROR, "Snapshot doesn't contain info about detached table={}", table_name); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Snapshot doesn't contain info about detached table `{}`", table_name); } else { From 69bd306a445a6bc8a55be14bb0080864921f8b69 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Fri, 2 Aug 2024 00:48:25 +0200 Subject: [PATCH 114/227] Fix race condition in system.processes and Settings --- src/Backups/RestoreCoordinationRemote.cpp | 2 +- src/Backups/RestoreCoordinationRemote.h | 2 -- src/Databases/DatabaseLazy.cpp | 2 +- src/Databases/DatabaseLazy.h | 2 +- src/Interpreters/ProcessList.cpp | 2 +- 5 files changed, 4 insertions(+), 6 deletions(-) diff --git a/src/Backups/RestoreCoordinationRemote.cpp b/src/Backups/RestoreCoordinationRemote.cpp index 44214d00be5..0a69bc0eafb 100644 --- a/src/Backups/RestoreCoordinationRemote.cpp +++ b/src/Backups/RestoreCoordinationRemote.cpp @@ -323,7 +323,7 @@ bool RestoreCoordinationRemote::hasConcurrentRestores(const std::atomic return false; bool result = false; - std::string path = zookeeper_path +"/stage"; + std::string path = zookeeper_path + "/stage"; auto holder = with_retries.createRetriesControlHolder("createRootNodes"); holder.retries_ctl.retryLoop( diff --git a/src/Backups/RestoreCoordinationRemote.h b/src/Backups/RestoreCoordinationRemote.h index 9c299865cfa..a3d57e9a4d0 100644 --- a/src/Backups/RestoreCoordinationRemote.h +++ b/src/Backups/RestoreCoordinationRemote.h @@ -61,8 +61,6 @@ private: void createRootNodes(); void removeAllNodes(); - class ReplicatedDatabasesMetadataSync; - /// get_zookeeper will provide a zookeeper client without any fault injection const zkutil::GetZooKeeper get_zookeeper; const String root_zookeeper_path; diff --git a/src/Databases/DatabaseLazy.cpp b/src/Databases/DatabaseLazy.cpp index ca30ee6db15..3fb6d30fcb8 100644 --- a/src/Databases/DatabaseLazy.cpp +++ b/src/Databases/DatabaseLazy.cpp @@ -44,7 +44,7 @@ namespace ErrorCodes DatabaseLazy::DatabaseLazy(const String & name_, const String & metadata_path_, time_t expiration_time_, ContextPtr context_) - : DatabaseOnDisk(name_, metadata_path_, "data/" + escapeForFileName(name_) + "/", "DatabaseLazy (" + name_ + ")", context_) + : DatabaseOnDisk(name_, metadata_path_, std::filesystem::path("data") / escapeForFileName(name_) / "", "DatabaseLazy (" + name_ + ")", context_) , expiration_time(expiration_time_) { } diff --git a/src/Databases/DatabaseLazy.h b/src/Databases/DatabaseLazy.h index 4347649117d..41cfb751141 100644 --- a/src/Databases/DatabaseLazy.h +++ b/src/Databases/DatabaseLazy.h @@ -12,7 +12,7 @@ class DatabaseLazyIterator; class Context; /** Lazy engine of databases. - * Works like DatabaseOrdinary, but stores in memory only cache. + * Works like DatabaseOrdinary, but stores in memory only the cache. * Can be used only with *Log engines. */ class DatabaseLazy final : public DatabaseOnDisk diff --git a/src/Interpreters/ProcessList.cpp b/src/Interpreters/ProcessList.cpp index 271e23a7288..6cb50b310ad 100644 --- a/src/Interpreters/ProcessList.cpp +++ b/src/Interpreters/ProcessList.cpp @@ -657,7 +657,7 @@ QueryStatusInfo QueryStatus::getInfo(bool get_thread_list, bool get_profile_even { if (auto ctx = context.lock()) { - res.query_settings = std::make_shared(ctx->getSettingsRef()); + res.query_settings = std::make_shared(ctx->getSettingsCopy()); res.current_database = ctx->getCurrentDatabase(); } } From 0772ed7f6ac92bb2e016e5db00e85deafeecb127 Mon Sep 17 00:00:00 2001 From: shiyer7474 Date: Fri, 2 Aug 2024 04:02:43 +0000 Subject: [PATCH 115/227] Code style feedback incorporated --- src/Analyzer/Resolve/QueryAnalyzer.cpp | 10 +++++----- ...3209_parameterized_view_with_non_literal_params.sql | 2 ++ 2 files changed, 7 insertions(+), 5 deletions(-) diff --git a/src/Analyzer/Resolve/QueryAnalyzer.cpp b/src/Analyzer/Resolve/QueryAnalyzer.cpp index 6113a38d463..bffdba2f58a 100644 --- a/src/Analyzer/Resolve/QueryAnalyzer.cpp +++ b/src/Analyzer/Resolve/QueryAnalyzer.cpp @@ -4549,11 +4549,11 @@ void QueryAnalyzer::resolveTableFunction(QueryTreeNodePtr & table_function_node, /// Serialize the constant value using datatype specific /// interfaces to match the deserialization in ReplaceQueryParametersVistor. WriteBufferFromOwnString buf; - auto constval = constant->getValue(); - auto realtype = constant->getResultType(); - auto tempcol = realtype->createColumn(); - tempcol->insert(constval); - realtype->getDefaultSerialization()->serializeTextEscaped(*tempcol, 0, buf, {}); + const auto & value = constant->getValue(); + auto real_type = constant->getResultType(); + auto temporary_column = real_type->createColumn(); + temporary_column->insert(value); + real_type->getDefaultSerialization()->serializeTextEscaped(*temporary_column, 0, buf, {}); view_params[identifier_node->getIdentifier().getFullName()] = buf.str(); } } diff --git a/tests/queries/0_stateless/03209_parameterized_view_with_non_literal_params.sql b/tests/queries/0_stateless/03209_parameterized_view_with_non_literal_params.sql index 55795c7a785..f2c61e5cb1d 100644 --- a/tests/queries/0_stateless/03209_parameterized_view_with_non_literal_params.sql +++ b/tests/queries/0_stateless/03209_parameterized_view_with_non_literal_params.sql @@ -90,8 +90,10 @@ select id from ipv4_pv(ipv4param=(select ipaddr from ipv4_table_pv where id=3)); drop view date_pv; drop view date_pv2; +drop view date32_pv; drop view uuid_pv; drop view ipv4_pv; drop table date_table_pv; +drop table date32_table_pv; drop table uuid_table_pv; drop table ipv4_table_pv; From 9c05a0ad5a0269af02ae2234e1d01dc3ce64bce2 Mon Sep 17 00:00:00 2001 From: Michael Stetsyuk Date: Fri, 2 Aug 2024 09:34:32 +0100 Subject: [PATCH 116/227] rm dirs in test_storage_delta --- tests/integration/test_storage_delta/test.py | 28 +++++++------------- 1 file changed, 10 insertions(+), 18 deletions(-) diff --git a/tests/integration/test_storage_delta/test.py b/tests/integration/test_storage_delta/test.py index 384b8296f66..92a870ab360 100644 --- a/tests/integration/test_storage_delta/test.py +++ b/tests/integration/test_storage_delta/test.py @@ -8,6 +8,7 @@ import os import json import time import glob +import shutil import pyspark import delta @@ -52,15 +53,6 @@ def get_spark(): return builder.master("local").getOrCreate() -def remove_local_directory_contents(full_path): - for path in glob.glob(f"{full_path}/**"): - if os.path.isfile(path): - os.unlink(path) - else: - remove_local_directory_contents(path) - os.rmdir(path) - - @pytest.fixture(scope="module") def started_cluster(): try: @@ -179,7 +171,7 @@ def test_single_log_file(started_cluster): ) os.unlink(parquet_data_path) - remove_local_directory_contents(f"/{TABLE_NAME}") + shutil.rmtree(f"/{TABLE_NAME}") def test_partition_by(started_cluster): @@ -203,7 +195,7 @@ def test_partition_by(started_cluster): create_delta_table(instance, TABLE_NAME) assert int(instance.query(f"SELECT count() FROM {TABLE_NAME}")) == 10 - remove_local_directory_contents(f"/{TABLE_NAME}") + shutil.rmtree(f"/{TABLE_NAME}") def test_checkpoint(started_cluster): @@ -280,7 +272,7 @@ def test_checkpoint(started_cluster): ).strip() ) - remove_local_directory_contents(f"/{TABLE_NAME}") + shutil.rmtree(f"/{TABLE_NAME}") spark.sql(f"DROP TABLE {TABLE_NAME}") @@ -321,7 +313,7 @@ def test_multiple_log_files(started_cluster): "SELECT number, toString(number + 1) FROM numbers(200)" ) - remove_local_directory_contents(f"/{TABLE_NAME}") + shutil.rmtree(f"/{TABLE_NAME}") def test_metadata(started_cluster): @@ -357,7 +349,7 @@ def test_metadata(started_cluster): assert int(instance.query(f"SELECT count() FROM {TABLE_NAME}")) == 100 os.unlink(parquet_data_path) - remove_local_directory_contents(f"/{TABLE_NAME}") + shutil.rmtree(f"/{TABLE_NAME}") def test_types(started_cluster): @@ -431,7 +423,7 @@ def test_types(started_cluster): ] ) - remove_local_directory_contents(f"/{result_file}") + shutil.rmtree(f"/{result_file}") spark.sql(f"DROP TABLE {TABLE_NAME}") @@ -496,7 +488,7 @@ def test_restart_broken(started_cluster): assert int(instance.query(f"SELECT count() FROM {TABLE_NAME}")) == 100 os.unlink(parquet_data_path) - remove_local_directory_contents(f"/{TABLE_NAME}") + shutil.rmtree(f"/{TABLE_NAME}") def test_restart_broken_table_function(started_cluster): @@ -553,7 +545,7 @@ def test_restart_broken_table_function(started_cluster): assert int(instance.query(f"SELECT count() FROM {TABLE_NAME}")) == 100 os.unlink(parquet_data_path) - remove_local_directory_contents(f"/{TABLE_NAME}") + shutil.rmtree(f"/{TABLE_NAME}") def test_partition_columns(started_cluster): @@ -753,5 +745,5 @@ SELECT * FROM deltaLake('http://{started_cluster.minio_ip}:{started_cluster.mini == 1 ) - remove_local_directory_contents(f"/{TABLE_NAME}") + shutil.rmtree(f"/{TABLE_NAME}") spark.sql(f"DROP TABLE {TABLE_NAME}") From 27f4e1808e4cf299cd8eaf4a19c3bb979aa4e5bd Mon Sep 17 00:00:00 2001 From: vdimir Date: Fri, 2 Aug 2024 10:43:02 +0200 Subject: [PATCH 117/227] Update tests/queries/0_stateless/02434_cancel_insert_when_client_dies.sh --- .../0_stateless/02434_cancel_insert_when_client_dies.sh | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/tests/queries/0_stateless/02434_cancel_insert_when_client_dies.sh b/tests/queries/0_stateless/02434_cancel_insert_when_client_dies.sh index 45f4194104e..dca8dae22c3 100755 --- a/tests/queries/0_stateless/02434_cancel_insert_when_client_dies.sh +++ b/tests/queries/0_stateless/02434_cancel_insert_when_client_dies.sh @@ -10,10 +10,7 @@ export DATA_FILE="$CLICKHOUSE_TMP/deduptest.tsv" export TEST_MARK="02434_insert_${CLICKHOUSE_DATABASE}_" $CLICKHOUSE_CLIENT -q 'select * from numbers(5000000) format TSV' > $DATA_FILE -$CLICKHOUSE_CLIENT -q "create table dedup_test(A Int64) Engine = MergeTree order by A - settings non_replicated_deduplication_window=1000 - , merge_tree_clear_old_temporary_directories_interval_seconds = 1 - ;" +$CLICKHOUSE_CLIENT -q "create table dedup_test(A Int64) Engine = MergeTree order by A settings non_replicated_deduplication_window=1000, merge_tree_clear_old_temporary_directories_interval_seconds = 1;" $CLICKHOUSE_CLIENT -q "create table dedup_dist(A Int64) Engine = Distributed('test_cluster_one_shard_two_replicas', currentDatabase(), dedup_test)" function insert_data From 6c8f458b0bf9981068c7fecfdd9cef627406419b Mon Sep 17 00:00:00 2001 From: Antonio Andelic Date: Fri, 2 Aug 2024 11:13:41 +0200 Subject: [PATCH 118/227] Fix reloading SQL UDFs with UNION --- .../UserDefinedSQLFunctionFactory.cpp | 8 +++++-- .../UserDefinedSQLObjectsDiskStorage.cpp | 6 ++--- .../UserDefinedSQLObjectsDiskStorage.h | 1 - .../UserDefinedSQLObjectsStorageBase.cpp | 15 +++++++++--- .../UserDefinedSQLObjectsStorageBase.h | 4 ++++ .../UserDefinedSQLObjectsZooKeeperStorage.cpp | 2 +- .../UserDefinedSQLObjectsZooKeeperStorage.h | 2 -- .../NormalizeSelectWithUnionQueryVisitor.h | 2 -- .../test.py | 23 +++++++++++++++++-- .../test.py | 12 ++++++++++ .../03215_udf_with_union.reference | 1 + .../0_stateless/03215_udf_with_union.sql | 14 +++++++++++ 12 files changed, 74 insertions(+), 16 deletions(-) create mode 100644 tests/queries/0_stateless/03215_udf_with_union.reference create mode 100644 tests/queries/0_stateless/03215_udf_with_union.sql diff --git a/src/Functions/UserDefined/UserDefinedSQLFunctionFactory.cpp b/src/Functions/UserDefined/UserDefinedSQLFunctionFactory.cpp index e6796874e50..d0bc812f91d 100644 --- a/src/Functions/UserDefined/UserDefinedSQLFunctionFactory.cpp +++ b/src/Functions/UserDefined/UserDefinedSQLFunctionFactory.cpp @@ -2,6 +2,7 @@ #include #include +#include #include #include #include @@ -9,6 +10,7 @@ #include #include #include +#include #include #include #include @@ -80,13 +82,15 @@ namespace validateFunctionRecursiveness(*function_body, name); } - ASTPtr normalizeCreateFunctionQuery(const IAST & create_function_query) + ASTPtr normalizeCreateFunctionQuery(const IAST & create_function_query, const ContextPtr & context) { auto ptr = create_function_query.clone(); auto & res = typeid_cast(*ptr); res.if_not_exists = false; res.or_replace = false; FunctionNameNormalizer::visit(res.function_core.get()); + NormalizeSelectWithUnionQueryVisitor::Data data{context->getSettingsRef().union_default_mode}; + NormalizeSelectWithUnionQueryVisitor{data}.visit(res.function_core); return ptr; } } @@ -125,7 +129,7 @@ void UserDefinedSQLFunctionFactory::checkCanBeUnregistered(const ContextPtr & co bool UserDefinedSQLFunctionFactory::registerFunction(const ContextMutablePtr & context, const String & function_name, ASTPtr create_function_query, bool throw_if_exists, bool replace_if_exists) { checkCanBeRegistered(context, function_name, *create_function_query); - create_function_query = normalizeCreateFunctionQuery(*create_function_query); + create_function_query = normalizeCreateFunctionQuery(*create_function_query, context); try { diff --git a/src/Functions/UserDefined/UserDefinedSQLObjectsDiskStorage.cpp b/src/Functions/UserDefined/UserDefinedSQLObjectsDiskStorage.cpp index 4c004d2537c..8910b45e79d 100644 --- a/src/Functions/UserDefined/UserDefinedSQLObjectsDiskStorage.cpp +++ b/src/Functions/UserDefined/UserDefinedSQLObjectsDiskStorage.cpp @@ -1,7 +1,7 @@ #include "Functions/UserDefined/UserDefinedSQLObjectsDiskStorage.h" -#include "Functions/UserDefined/UserDefinedSQLFunctionFactory.h" -#include "Functions/UserDefined/UserDefinedSQLObjectType.h" +#include +#include #include #include @@ -54,7 +54,7 @@ namespace } UserDefinedSQLObjectsDiskStorage::UserDefinedSQLObjectsDiskStorage(const ContextPtr & global_context_, const String & dir_path_) - : global_context(global_context_) + : UserDefinedSQLObjectsStorageBase(global_context_) , dir_path{makeDirectoryPathCanonical(dir_path_)} , log{getLogger("UserDefinedSQLObjectsLoaderFromDisk")} { diff --git a/src/Functions/UserDefined/UserDefinedSQLObjectsDiskStorage.h b/src/Functions/UserDefined/UserDefinedSQLObjectsDiskStorage.h index ae0cbd0c589..cafbd140598 100644 --- a/src/Functions/UserDefined/UserDefinedSQLObjectsDiskStorage.h +++ b/src/Functions/UserDefined/UserDefinedSQLObjectsDiskStorage.h @@ -42,7 +42,6 @@ private: ASTPtr tryLoadObject(UserDefinedSQLObjectType object_type, const String & object_name, const String & file_path, bool check_file_exists); String getFilePath(UserDefinedSQLObjectType object_type, const String & object_name) const; - ContextPtr global_context; String dir_path; LoggerPtr log; std::atomic objects_loaded = false; diff --git a/src/Functions/UserDefined/UserDefinedSQLObjectsStorageBase.cpp b/src/Functions/UserDefined/UserDefinedSQLObjectsStorageBase.cpp index f251d11789f..225e919301d 100644 --- a/src/Functions/UserDefined/UserDefinedSQLObjectsStorageBase.cpp +++ b/src/Functions/UserDefined/UserDefinedSQLObjectsStorageBase.cpp @@ -2,7 +2,10 @@ #include +#include +#include #include +#include #include namespace DB @@ -17,18 +20,24 @@ namespace ErrorCodes namespace { -ASTPtr normalizeCreateFunctionQuery(const IAST & create_function_query) +ASTPtr normalizeCreateFunctionQuery(const IAST & create_function_query, const ContextPtr & context) { auto ptr = create_function_query.clone(); auto & res = typeid_cast(*ptr); res.if_not_exists = false; res.or_replace = false; FunctionNameNormalizer::visit(res.function_core.get()); + NormalizeSelectWithUnionQueryVisitor::Data data{context->getSettingsRef().union_default_mode}; + NormalizeSelectWithUnionQueryVisitor{data}.visit(res.function_core); return ptr; } } +UserDefinedSQLObjectsStorageBase::UserDefinedSQLObjectsStorageBase(ContextPtr global_context_) + : global_context(std::move(global_context_)) +{} + ASTPtr UserDefinedSQLObjectsStorageBase::get(const String & object_name) const { std::lock_guard lock(mutex); @@ -148,7 +157,7 @@ void UserDefinedSQLObjectsStorageBase::setAllObjects(const std::vector normalized_functions; for (const auto & [function_name, create_query] : new_objects) - normalized_functions[function_name] = normalizeCreateFunctionQuery(*create_query); + normalized_functions[function_name] = normalizeCreateFunctionQuery(*create_query, global_context); std::lock_guard lock(mutex); object_name_to_create_object_map = std::move(normalized_functions); @@ -166,7 +175,7 @@ std::vector> UserDefinedSQLObjectsStorageBase::getAllO void UserDefinedSQLObjectsStorageBase::setObject(const String & object_name, const IAST & create_object_query) { std::lock_guard lock(mutex); - object_name_to_create_object_map[object_name] = normalizeCreateFunctionQuery(create_object_query); + object_name_to_create_object_map[object_name] = normalizeCreateFunctionQuery(create_object_query, global_context); } void UserDefinedSQLObjectsStorageBase::removeObject(const String & object_name) diff --git a/src/Functions/UserDefined/UserDefinedSQLObjectsStorageBase.h b/src/Functions/UserDefined/UserDefinedSQLObjectsStorageBase.h index cab63a3bfcf..0dbc5586f08 100644 --- a/src/Functions/UserDefined/UserDefinedSQLObjectsStorageBase.h +++ b/src/Functions/UserDefined/UserDefinedSQLObjectsStorageBase.h @@ -4,6 +4,7 @@ #include #include +#include #include @@ -13,6 +14,7 @@ namespace DB class UserDefinedSQLObjectsStorageBase : public IUserDefinedSQLObjectsStorage { public: + explicit UserDefinedSQLObjectsStorageBase(ContextPtr global_context_); ASTPtr get(const String & object_name) const override; ASTPtr tryGet(const String & object_name) const override; @@ -64,6 +66,8 @@ protected: std::unordered_map object_name_to_create_object_map; mutable std::recursive_mutex mutex; + + ContextPtr global_context; }; } diff --git a/src/Functions/UserDefined/UserDefinedSQLObjectsZooKeeperStorage.cpp b/src/Functions/UserDefined/UserDefinedSQLObjectsZooKeeperStorage.cpp index 01e7e3995fa..12c1302a3fe 100644 --- a/src/Functions/UserDefined/UserDefinedSQLObjectsZooKeeperStorage.cpp +++ b/src/Functions/UserDefined/UserDefinedSQLObjectsZooKeeperStorage.cpp @@ -48,7 +48,7 @@ namespace UserDefinedSQLObjectsZooKeeperStorage::UserDefinedSQLObjectsZooKeeperStorage( const ContextPtr & global_context_, const String & zookeeper_path_) - : global_context{global_context_} + : UserDefinedSQLObjectsStorageBase(global_context_) , zookeeper_getter{[global_context_]() { return global_context_->getZooKeeper(); }} , zookeeper_path{zookeeper_path_} , watch_queue{std::make_shared>>(std::numeric_limits::max())} diff --git a/src/Functions/UserDefined/UserDefinedSQLObjectsZooKeeperStorage.h b/src/Functions/UserDefined/UserDefinedSQLObjectsZooKeeperStorage.h index 61002be2bfd..0aa9b198398 100644 --- a/src/Functions/UserDefined/UserDefinedSQLObjectsZooKeeperStorage.h +++ b/src/Functions/UserDefined/UserDefinedSQLObjectsZooKeeperStorage.h @@ -68,8 +68,6 @@ private: void refreshObjects(const zkutil::ZooKeeperPtr & zookeeper, UserDefinedSQLObjectType object_type); void syncObjects(const zkutil::ZooKeeperPtr & zookeeper, UserDefinedSQLObjectType object_type); - ContextPtr global_context; - zkutil::ZooKeeperCachingGetter zookeeper_getter; String zookeeper_path; std::atomic objects_loaded = false; diff --git a/src/Interpreters/NormalizeSelectWithUnionQueryVisitor.h b/src/Interpreters/NormalizeSelectWithUnionQueryVisitor.h index b2f55003da5..b642b5def91 100644 --- a/src/Interpreters/NormalizeSelectWithUnionQueryVisitor.h +++ b/src/Interpreters/NormalizeSelectWithUnionQueryVisitor.h @@ -4,8 +4,6 @@ #include #include -#include - namespace DB { diff --git a/tests/integration/test_replicated_user_defined_functions/test.py b/tests/integration/test_replicated_user_defined_functions/test.py index e5f6683b90b..92d86a8fd2c 100644 --- a/tests/integration/test_replicated_user_defined_functions/test.py +++ b/tests/integration/test_replicated_user_defined_functions/test.py @@ -141,6 +141,9 @@ def test_drop_if_exists(): def test_replication(): node1.query("CREATE FUNCTION f2 AS (x, y) -> x - y") + node1.query( + "CREATE FUNCTION f3 AS () -> (SELECT sum(s) FROM (SELECT 1 as s UNION ALL SELECT 1 as s))" + ) assert ( node1.query("SELECT create_query FROM system.functions WHERE name='f2'") @@ -154,7 +157,11 @@ def test_replication(): assert node1.query("SELECT f2(12,3)") == "9\n" assert node2.query("SELECT f2(12,3)") == "9\n" + assert node1.query("SELECT f3()") == "2\n" + assert node2.query("SELECT f3()") == "2\n" + node1.query("DROP FUNCTION f2") + node1.query("DROP FUNCTION f3") assert ( node1.query("SELECT create_query FROM system.functions WHERE name='f2'") == "" ) @@ -214,7 +221,9 @@ def test_reload_zookeeper(): ) # config reloads, but can still work - node1.query("CREATE FUNCTION f2 AS (x, y) -> x - y") + node1.query( + "CREATE FUNCTION f2 AS () -> (SELECT sum(s) FROM (SELECT 1 as s UNION ALL SELECT 1 as s))" + ) assert_eq_with_retry( node2, "SELECT name FROM system.functions WHERE name IN ['f1', 'f2'] ORDER BY name", @@ -269,7 +278,7 @@ def test_reload_zookeeper(): TSV(["f1", "f2", "f3"]), ) - assert node2.query("SELECT f1(12, 3), f2(12, 3), f3(12, 3)") == TSV([[15, 9, 4]]) + assert node2.query("SELECT f1(12, 3), f2(), f3(12, 3)") == TSV([[15, 2, 4]]) active_zk_connections = get_active_zk_connections() assert ( @@ -307,3 +316,13 @@ def test_start_without_zookeeper(): "CREATE FUNCTION f1 AS (x, y) -> (x + y)\n", ) node1.query("DROP FUNCTION f1") + + +def test_server_restart(): + node1.query( + "CREATE FUNCTION f1 AS () -> (SELECT sum(s) FROM (SELECT 1 as s UNION ALL SELECT 1 as s))" + ) + assert node1.query("SELECT f1()") == "2\n" + node1.restart_clickhouse() + assert node1.query("SELECT f1()") == "2\n" + node1.query("DROP FUNCTION f1") diff --git a/tests/integration/test_user_defined_object_persistence/test.py b/tests/integration/test_user_defined_object_persistence/test.py index 986438a4eed..bd491dfa195 100644 --- a/tests/integration/test_user_defined_object_persistence/test.py +++ b/tests/integration/test_user_defined_object_persistence/test.py @@ -18,20 +18,25 @@ def started_cluster(): def test_persistence(): create_function_query1 = "CREATE FUNCTION MySum1 AS (a, b) -> a + b" create_function_query2 = "CREATE FUNCTION MySum2 AS (a, b) -> MySum1(a, b) + b" + create_function_query3 = "CREATE FUNCTION MyUnion AS () -> (SELECT sum(s) FROM (SELECT 1 as s UNION ALL SELECT 1 as s))" instance.query(create_function_query1) instance.query(create_function_query2) + instance.query(create_function_query3) assert instance.query("SELECT MySum1(1,2)") == "3\n" assert instance.query("SELECT MySum2(1,2)") == "5\n" + assert instance.query("SELECT MyUnion()") == "2\n" instance.restart_clickhouse() assert instance.query("SELECT MySum1(1,2)") == "3\n" assert instance.query("SELECT MySum2(1,2)") == "5\n" + assert instance.query("SELECT MyUnion()") == "2\n" instance.query("DROP FUNCTION MySum2") instance.query("DROP FUNCTION MySum1") + instance.query("DROP FUNCTION MyUnion") instance.restart_clickhouse() @@ -48,3 +53,10 @@ def test_persistence(): or "Function with name 'MySum2' does not exist. In scope SELECT MySum2(1, 2)" in error_message ) + + error_message = instance.query_and_get_error("SELECT MyUnion()") + assert ( + "Unknown function MyUnion" in error_message + or "Function with name 'MyUnion' does not exist. In scope SELECT MyUnion" + in error_message + ) diff --git a/tests/queries/0_stateless/03215_udf_with_union.reference b/tests/queries/0_stateless/03215_udf_with_union.reference new file mode 100644 index 00000000000..0cfbf08886f --- /dev/null +++ b/tests/queries/0_stateless/03215_udf_with_union.reference @@ -0,0 +1 @@ +2 diff --git a/tests/queries/0_stateless/03215_udf_with_union.sql b/tests/queries/0_stateless/03215_udf_with_union.sql new file mode 100644 index 00000000000..00390c5d930 --- /dev/null +++ b/tests/queries/0_stateless/03215_udf_with_union.sql @@ -0,0 +1,14 @@ +DROP FUNCTION IF EXISTS 03215_udf_with_union; +CREATE FUNCTION 03215_udf_with_union AS () -> ( + SELECT sum(s) + FROM + ( + SELECT 1 AS s + UNION ALL + SELECT 1 AS s + ) +); + +SELECT 03215_udf_with_union(); + +DROP FUNCTION 03215_udf_with_union; From af53ed4c02ba52b3f57e97941b37a5931620d447 Mon Sep 17 00:00:00 2001 From: kssenii Date: Fri, 2 Aug 2024 12:08:49 +0200 Subject: [PATCH 119/227] Ping CI From 62f0e09ecbb226ea72b5ee8d812436ef75038e33 Mon Sep 17 00:00:00 2001 From: Antonio Andelic Date: Fri, 2 Aug 2024 12:17:08 +0200 Subject: [PATCH 120/227] Fix setting changes --- src/Core/SettingsChangesHistory.cpp | 264 +--------------------------- 1 file changed, 2 insertions(+), 262 deletions(-) diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index 2438202f6a3..b6ef654438e 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -57,268 +57,6 @@ String ClickHouseVersion::toString() const /// Note: please check if the key already exists to prevent duplicate entries. static std::initializer_list> settings_changes_history_initializer = { - {"24.7", {{"output_format_parquet_write_page_index", false, true, "Add a possibility to write page index into parquet files."}, - {"output_format_binary_encode_types_in_binary_format", false, false, "Added new setting to allow to write type names in binary format in RowBinaryWithNamesAndTypes output format"}, - {"input_format_binary_decode_types_in_binary_format", false, false, "Added new setting to allow to read type names in binary format in RowBinaryWithNamesAndTypes input format"}, - {"output_format_native_encode_types_in_binary_format", false, false, "Added new setting to allow to write type names in binary format in Native output format"}, - {"input_format_native_decode_types_in_binary_format", false, false, "Added new setting to allow to read type names in binary format in Native output format"}, - {"read_in_order_use_buffering", false, true, "Use buffering before merging while reading in order of primary key"}, - {"enable_named_columns_in_function_tuple", false, true, "Generate named tuples in function tuple() when all names are unique and can be treated as unquoted identifiers."}, - {"input_format_json_case_insensitive_column_matching", false, false, "Ignore case when matching JSON keys with CH columns."}, - {"optimize_trivial_insert_select", true, false, "The optimization does not make sense in many cases."}, - {"dictionary_validate_primary_key_type", false, false, "Validate primary key type for dictionaries. By default id type for simple layouts will be implicitly converted to UInt64."}, - {"collect_hash_table_stats_during_joins", false, true, "New setting."}, - {"max_size_to_preallocate_for_joins", 0, 100'000'000, "New setting."}, - {"input_format_orc_read_use_writer_time_zone", false, false, "Whether use the writer's time zone in ORC stripe for ORC row reader, the default ORC row reader's time zone is GMT."}, - {"lightweight_mutation_projection_mode", "throw", "throw", "When lightweight delete happens on a table with projection(s), the possible operations include throw the exception as projection exists, or drop all projection related to this table then do lightweight delete."}, - {"database_replicated_allow_heavy_create", true, false, "Long-running DDL queries (CREATE AS SELECT and POPULATE) for Replicated database engine was forbidden"}, - {"query_plan_merge_filters", false, false, "Allow to merge filters in the query plan"}, - {"azure_sdk_max_retries", 10, 10, "Maximum number of retries in azure sdk"}, - {"azure_sdk_retry_initial_backoff_ms", 10, 10, "Minimal backoff between retries in azure sdk"}, - {"azure_sdk_retry_max_backoff_ms", 1000, 1000, "Maximal backoff between retries in azure sdk"}, - {"merge_tree_min_bytes_per_task_for_remote_reading", 4194304, 2097152, "Value is unified with `filesystem_prefetch_min_bytes_for_single_read_task`"}, - {"ignore_on_cluster_for_replicated_named_collections_queries", false, false, "Ignore ON CLUSTER clause for replicated named collections management queries."}, - {"backup_restore_s3_retry_attempts", 1000,1000, "Setting for Aws::Client::RetryStrategy, Aws::Client does retries itself, 0 means no retries. It takes place only for backup/restore."}, - {"postgresql_connection_attempt_timeout", 2, 2, "Allow to control 'connect_timeout' parameter of PostgreSQL connection."}, - {"postgresql_connection_pool_retries", 2, 2, "Allow to control the number of retries in PostgreSQL connection pool."}, - {"restore_replace_external_table_functions_to_null", false, false, "New setting."}, - {"restore_replace_external_engines_to_null", false, false, "New setting."} - }}, - {"24.6", {{"materialize_skip_indexes_on_insert", true, true, "Added new setting to allow to disable materialization of skip indexes on insert"}, - {"materialize_statistics_on_insert", true, true, "Added new setting to allow to disable materialization of statistics on insert"}, - {"input_format_parquet_use_native_reader", false, false, "When reading Parquet files, to use native reader instead of arrow reader."}, - {"hdfs_throw_on_zero_files_match", false, false, "Allow to throw an error when ListObjects request cannot match any files in HDFS engine instead of empty query result"}, - {"azure_throw_on_zero_files_match", false, false, "Allow to throw an error when ListObjects request cannot match any files in AzureBlobStorage engine instead of empty query result"}, - {"s3_validate_request_settings", true, true, "Allow to disable S3 request settings validation"}, - {"allow_experimental_full_text_index", false, false, "Enable experimental full-text index"}, - {"azure_skip_empty_files", false, false, "Allow to skip empty files in azure table engine"}, - {"hdfs_ignore_file_doesnt_exist", false, false, "Allow to return 0 rows when the requested files don't exist instead of throwing an exception in HDFS table engine"}, - {"azure_ignore_file_doesnt_exist", false, false, "Allow to return 0 rows when the requested files don't exist instead of throwing an exception in AzureBlobStorage table engine"}, - {"s3_ignore_file_doesnt_exist", false, false, "Allow to return 0 rows when the requested files don't exist instead of throwing an exception in S3 table engine"}, - {"s3_max_part_number", 10000, 10000, "Maximum part number number for s3 upload part"}, - {"s3_max_single_operation_copy_size", 32 * 1024 * 1024, 32 * 1024 * 1024, "Maximum size for a single copy operation in s3"}, - {"input_format_parquet_max_block_size", 8192, DEFAULT_BLOCK_SIZE, "Increase block size for parquet reader."}, - {"input_format_parquet_prefer_block_bytes", 0, DEFAULT_BLOCK_SIZE * 256, "Average block bytes output by parquet reader."}, - {"enable_blob_storage_log", true, true, "Write information about blob storage operations to system.blob_storage_log table"}, - {"allow_deprecated_snowflake_conversion_functions", true, false, "Disabled deprecated functions snowflakeToDateTime[64] and dateTime[64]ToSnowflake."}, - {"allow_statistic_optimize", false, false, "Old setting which popped up here being renamed."}, - {"allow_experimental_statistic", false, false, "Old setting which popped up here being renamed."}, - {"allow_statistics_optimize", false, false, "The setting was renamed. The previous name is `allow_statistic_optimize`."}, - {"allow_experimental_statistics", false, false, "The setting was renamed. The previous name is `allow_experimental_statistic`."}, - {"enable_vertical_final", false, true, "Enable vertical final by default again after fixing bug"}, - {"parallel_replicas_custom_key_range_lower", 0, 0, "Add settings to control the range filter when using parallel replicas with dynamic shards"}, - {"parallel_replicas_custom_key_range_upper", 0, 0, "Add settings to control the range filter when using parallel replicas with dynamic shards. A value of 0 disables the upper limit"}, - {"output_format_pretty_display_footer_column_names", 0, 1, "Add a setting to display column names in the footer if there are many rows. Threshold value is controlled by output_format_pretty_display_footer_column_names_min_rows."}, - {"output_format_pretty_display_footer_column_names_min_rows", 0, 50, "Add a setting to control the threshold value for setting output_format_pretty_display_footer_column_names_min_rows. Default 50."}, - {"output_format_csv_serialize_tuple_into_separate_columns", true, true, "A new way of how interpret tuples in CSV format was added."}, - {"input_format_csv_deserialize_separate_columns_into_tuple", true, true, "A new way of how interpret tuples in CSV format was added."}, - {"input_format_csv_try_infer_strings_from_quoted_tuples", true, true, "A new way of how interpret tuples in CSV format was added."}, - }}, - {"24.5", {{"allow_deprecated_error_prone_window_functions", true, false, "Allow usage of deprecated error prone window functions (neighbor, runningAccumulate, runningDifferenceStartingWithFirstValue, runningDifference)"}, - {"allow_experimental_join_condition", false, false, "Support join with inequal conditions which involve columns from both left and right table. e.g. t1.y < t2.y."}, - {"input_format_tsv_crlf_end_of_line", false, false, "Enables reading of CRLF line endings with TSV formats"}, - {"output_format_parquet_use_custom_encoder", false, true, "Enable custom Parquet encoder."}, - {"cross_join_min_rows_to_compress", 0, 10000000, "Minimal count of rows to compress block in CROSS JOIN. Zero value means - disable this threshold. This block is compressed when any of the two thresholds (by rows or by bytes) are reached."}, - {"cross_join_min_bytes_to_compress", 0, 1_GiB, "Minimal size of block to compress in CROSS JOIN. Zero value means - disable this threshold. This block is compressed when any of the two thresholds (by rows or by bytes) are reached."}, - {"http_max_chunk_size", 0, 0, "Internal limitation"}, - {"prefer_external_sort_block_bytes", 0, DEFAULT_BLOCK_SIZE * 256, "Prefer maximum block bytes for external sort, reduce the memory usage during merging."}, - {"input_format_force_null_for_omitted_fields", false, false, "Disable type-defaults for omitted fields when needed"}, - {"cast_string_to_dynamic_use_inference", false, false, "Add setting to allow converting String to Dynamic through parsing"}, - {"allow_experimental_dynamic_type", false, false, "Add new experimental Dynamic type"}, - {"azure_max_blocks_in_multipart_upload", 50000, 50000, "Maximum number of blocks in multipart upload for Azure."}, - {"allow_archive_path_syntax", false, true, "Added new setting to allow disabling archive path syntax."}, - }}, - {"24.4", {{"input_format_json_throw_on_bad_escape_sequence", true, true, "Allow to save JSON strings with bad escape sequences"}, - {"max_parsing_threads", 0, 0, "Add a separate setting to control number of threads in parallel parsing from files"}, - {"ignore_drop_queries_probability", 0, 0, "Allow to ignore drop queries in server with specified probability for testing purposes"}, - {"lightweight_deletes_sync", 2, 2, "The same as 'mutation_sync', but controls only execution of lightweight deletes"}, - {"query_cache_system_table_handling", "save", "throw", "The query cache no longer caches results of queries against system tables"}, - {"input_format_json_ignore_unnecessary_fields", false, true, "Ignore unnecessary fields and not parse them. Enabling this may not throw exceptions on json strings of invalid format or with duplicated fields"}, - {"input_format_hive_text_allow_variable_number_of_columns", false, true, "Ignore extra columns in Hive Text input (if file has more columns than expected) and treat missing fields in Hive Text input as default values."}, - {"allow_experimental_database_replicated", false, true, "Database engine Replicated is now in Beta stage"}, - {"temporary_data_in_cache_reserve_space_wait_lock_timeout_milliseconds", (10 * 60 * 1000), (10 * 60 * 1000), "Wait time to lock cache for sapce reservation in temporary data in filesystem cache"}, - {"optimize_rewrite_sum_if_to_count_if", false, true, "Only available for the analyzer, where it works correctly"}, - {"azure_allow_parallel_part_upload", "true", "true", "Use multiple threads for azure multipart upload."}, - {"max_recursive_cte_evaluation_depth", DBMS_RECURSIVE_CTE_MAX_EVALUATION_DEPTH, DBMS_RECURSIVE_CTE_MAX_EVALUATION_DEPTH, "Maximum limit on recursive CTE evaluation depth"}, - {"query_plan_convert_outer_join_to_inner_join", false, true, "Allow to convert OUTER JOIN to INNER JOIN if filter after JOIN always filters default values"}, - }}, - {"24.3", {{"s3_connect_timeout_ms", 1000, 1000, "Introduce new dedicated setting for s3 connection timeout"}, - {"allow_experimental_shared_merge_tree", false, true, "The setting is obsolete"}, - {"use_page_cache_for_disks_without_file_cache", false, false, "Added userspace page cache"}, - {"read_from_page_cache_if_exists_otherwise_bypass_cache", false, false, "Added userspace page cache"}, - {"page_cache_inject_eviction", false, false, "Added userspace page cache"}, - {"default_table_engine", "None", "MergeTree", "Set default table engine to MergeTree for better usability"}, - {"input_format_json_use_string_type_for_ambiguous_paths_in_named_tuples_inference_from_objects", false, false, "Allow to use String type for ambiguous paths during named tuple inference from JSON objects"}, - {"traverse_shadow_remote_data_paths", false, false, "Traverse shadow directory when query system.remote_data_paths."}, - {"throw_if_deduplication_in_dependent_materialized_views_enabled_with_async_insert", false, true, "Deduplication in dependent materialized view cannot work together with async inserts."}, - {"parallel_replicas_allow_in_with_subquery", false, true, "If true, subquery for IN will be executed on every follower replica"}, - {"log_processors_profiles", false, true, "Enable by default"}, - {"function_locate_has_mysql_compatible_argument_order", false, true, "Increase compatibility with MySQL's locate function."}, - {"allow_suspicious_primary_key", true, false, "Forbid suspicious PRIMARY KEY/ORDER BY for MergeTree (i.e. SimpleAggregateFunction)"}, - {"filesystem_cache_reserve_space_wait_lock_timeout_milliseconds", 1000, 1000, "Wait time to lock cache for sapce reservation in filesystem cache"}, - {"max_parser_backtracks", 0, 1000000, "Limiting the complexity of parsing"}, - {"analyzer_compatibility_join_using_top_level_identifier", false, false, "Force to resolve identifier in JOIN USING from projection"}, - {"distributed_insert_skip_read_only_replicas", false, false, "If true, INSERT into Distributed will skip read-only replicas"}, - {"keeper_max_retries", 10, 10, "Max retries for general keeper operations"}, - {"keeper_retry_initial_backoff_ms", 100, 100, "Initial backoff timeout for general keeper operations"}, - {"keeper_retry_max_backoff_ms", 5000, 5000, "Max backoff timeout for general keeper operations"}, - {"s3queue_allow_experimental_sharded_mode", false, false, "Enable experimental sharded mode of S3Queue table engine. It is experimental because it will be rewritten"}, - {"allow_experimental_analyzer", false, true, "Enable analyzer and planner by default."}, - {"merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability", 0.0, 0.0, "For testing of `PartsSplitter` - split read ranges into intersecting and non intersecting every time you read from MergeTree with the specified probability."}, - {"allow_get_client_http_header", false, false, "Introduced a new function."}, - {"output_format_pretty_row_numbers", false, true, "It is better for usability."}, - {"output_format_pretty_max_value_width_apply_for_single_value", true, false, "Single values in Pretty formats won't be cut."}, - {"output_format_parquet_string_as_string", false, true, "ClickHouse allows arbitrary binary data in the String data type, which is typically UTF-8. Parquet/ORC/Arrow Strings only support UTF-8. That's why you can choose which Arrow's data type to use for the ClickHouse String data type - String or Binary. While Binary would be more correct and compatible, using String by default will correspond to user expectations in most cases."}, - {"output_format_orc_string_as_string", false, true, "ClickHouse allows arbitrary binary data in the String data type, which is typically UTF-8. Parquet/ORC/Arrow Strings only support UTF-8. That's why you can choose which Arrow's data type to use for the ClickHouse String data type - String or Binary. While Binary would be more correct and compatible, using String by default will correspond to user expectations in most cases."}, - {"output_format_arrow_string_as_string", false, true, "ClickHouse allows arbitrary binary data in the String data type, which is typically UTF-8. Parquet/ORC/Arrow Strings only support UTF-8. That's why you can choose which Arrow's data type to use for the ClickHouse String data type - String or Binary. While Binary would be more correct and compatible, using String by default will correspond to user expectations in most cases."}, - {"output_format_parquet_compression_method", "lz4", "zstd", "Parquet/ORC/Arrow support many compression methods, including lz4 and zstd. ClickHouse supports each and every compression method. Some inferior tools, such as 'duckdb', lack support for the faster `lz4` compression method, that's why we set zstd by default."}, - {"output_format_orc_compression_method", "lz4", "zstd", "Parquet/ORC/Arrow support many compression methods, including lz4 and zstd. ClickHouse supports each and every compression method. Some inferior tools, such as 'duckdb', lack support for the faster `lz4` compression method, that's why we set zstd by default."}, - {"output_format_pretty_highlight_digit_groups", false, true, "If enabled and if output is a terminal, highlight every digit corresponding to the number of thousands, millions, etc. with underline."}, - {"geo_distance_returns_float64_on_float64_arguments", false, true, "Increase the default precision."}, - {"azure_max_inflight_parts_for_one_file", 20, 20, "The maximum number of a concurrent loaded parts in multipart upload request. 0 means unlimited."}, - {"azure_strict_upload_part_size", 0, 0, "The exact size of part to upload during multipart upload to Azure blob storage."}, - {"azure_min_upload_part_size", 16*1024*1024, 16*1024*1024, "The minimum size of part to upload during multipart upload to Azure blob storage."}, - {"azure_max_upload_part_size", 5ull*1024*1024*1024, 5ull*1024*1024*1024, "The maximum size of part to upload during multipart upload to Azure blob storage."}, - {"azure_upload_part_size_multiply_factor", 2, 2, "Multiply azure_min_upload_part_size by this factor each time azure_multiply_parts_count_threshold parts were uploaded from a single write to Azure blob storage."}, - {"azure_upload_part_size_multiply_parts_count_threshold", 500, 500, "Each time this number of parts was uploaded to Azure blob storage, azure_min_upload_part_size is multiplied by azure_upload_part_size_multiply_factor."}, - {"output_format_csv_serialize_tuple_into_separate_columns", true, true, "A new way of how interpret tuples in CSV format was added."}, - {"input_format_csv_deserialize_separate_columns_into_tuple", true, true, "A new way of how interpret tuples in CSV format was added."}, - {"input_format_csv_try_infer_strings_from_quoted_tuples", true, true, "A new way of how interpret tuples in CSV format was added."}, - }}, - {"24.2", {{"allow_suspicious_variant_types", true, false, "Don't allow creating Variant type with suspicious variants by default"}, - {"validate_experimental_and_suspicious_types_inside_nested_types", false, true, "Validate usage of experimental and suspicious types inside nested types"}, - {"output_format_values_escape_quote_with_quote", false, false, "If true escape ' with '', otherwise quoted with \\'"}, - {"output_format_pretty_single_large_number_tip_threshold", 0, 1'000'000, "Print a readable number tip on the right side of the table if the block consists of a single number which exceeds this value (except 0)"}, - {"input_format_try_infer_exponent_floats", true, false, "Don't infer floats in exponential notation by default"}, - {"query_plan_optimize_prewhere", true, true, "Allow to push down filter to PREWHERE expression for supported storages"}, - {"async_insert_max_data_size", 1000000, 10485760, "The previous value appeared to be too small."}, - {"async_insert_poll_timeout_ms", 10, 10, "Timeout in milliseconds for polling data from asynchronous insert queue"}, - {"async_insert_use_adaptive_busy_timeout", false, true, "Use adaptive asynchronous insert timeout"}, - {"async_insert_busy_timeout_min_ms", 50, 50, "The minimum value of the asynchronous insert timeout in milliseconds; it also serves as the initial value, which may be increased later by the adaptive algorithm"}, - {"async_insert_busy_timeout_max_ms", 200, 200, "The minimum value of the asynchronous insert timeout in milliseconds; async_insert_busy_timeout_ms is aliased to async_insert_busy_timeout_max_ms"}, - {"async_insert_busy_timeout_increase_rate", 0.2, 0.2, "The exponential growth rate at which the adaptive asynchronous insert timeout increases"}, - {"async_insert_busy_timeout_decrease_rate", 0.2, 0.2, "The exponential growth rate at which the adaptive asynchronous insert timeout decreases"}, - {"format_template_row_format", "", "", "Template row format string can be set directly in query"}, - {"format_template_resultset_format", "", "", "Template result set format string can be set in query"}, - {"split_parts_ranges_into_intersecting_and_non_intersecting_final", true, true, "Allow to split parts ranges into intersecting and non intersecting during FINAL optimization"}, - {"split_intersecting_parts_ranges_into_layers_final", true, true, "Allow to split intersecting parts ranges into layers during FINAL optimization"}, - {"azure_max_single_part_copy_size", 256*1024*1024, 256*1024*1024, "The maximum size of object to copy using single part copy to Azure blob storage."}, - {"min_external_table_block_size_rows", DEFAULT_INSERT_BLOCK_SIZE, DEFAULT_INSERT_BLOCK_SIZE, "Squash blocks passed to external table to specified size in rows, if blocks are not big enough"}, - {"min_external_table_block_size_bytes", DEFAULT_INSERT_BLOCK_SIZE * 256, DEFAULT_INSERT_BLOCK_SIZE * 256, "Squash blocks passed to external table to specified size in bytes, if blocks are not big enough."}, - {"parallel_replicas_prefer_local_join", true, true, "If true, and JOIN can be executed with parallel replicas algorithm, and all storages of right JOIN part are *MergeTree, local JOIN will be used instead of GLOBAL JOIN."}, - {"optimize_time_filter_with_preimage", true, true, "Optimize Date and DateTime predicates by converting functions into equivalent comparisons without conversions (e.g. toYear(col) = 2023 -> col >= '2023-01-01' AND col <= '2023-12-31')"}, - {"extract_key_value_pairs_max_pairs_per_row", 0, 0, "Max number of pairs that can be produced by the `extractKeyValuePairs` function. Used as a safeguard against consuming too much memory."}, - {"default_view_definer", "CURRENT_USER", "CURRENT_USER", "Allows to set default `DEFINER` option while creating a view"}, - {"default_materialized_view_sql_security", "DEFINER", "DEFINER", "Allows to set a default value for SQL SECURITY option when creating a materialized view"}, - {"default_normal_view_sql_security", "INVOKER", "INVOKER", "Allows to set default `SQL SECURITY` option while creating a normal view"}, - {"mysql_map_string_to_text_in_show_columns", false, true, "Reduce the configuration effort to connect ClickHouse with BI tools."}, - {"mysql_map_fixed_string_to_text_in_show_columns", false, true, "Reduce the configuration effort to connect ClickHouse with BI tools."}, - }}, - {"24.1", {{"print_pretty_type_names", false, true, "Better user experience."}, - {"input_format_json_read_bools_as_strings", false, true, "Allow to read bools as strings in JSON formats by default"}, - {"output_format_arrow_use_signed_indexes_for_dictionary", false, true, "Use signed indexes type for Arrow dictionaries by default as it's recommended"}, - {"allow_experimental_variant_type", false, false, "Add new experimental Variant type"}, - {"use_variant_as_common_type", false, false, "Allow to use Variant in if/multiIf if there is no common type"}, - {"output_format_arrow_use_64_bit_indexes_for_dictionary", false, false, "Allow to use 64 bit indexes type in Arrow dictionaries"}, - {"parallel_replicas_mark_segment_size", 128, 128, "Add new setting to control segment size in new parallel replicas coordinator implementation"}, - {"ignore_materialized_views_with_dropped_target_table", false, false, "Add new setting to allow to ignore materialized views with dropped target table"}, - {"output_format_compression_level", 3, 3, "Allow to change compression level in the query output"}, - {"output_format_compression_zstd_window_log", 0, 0, "Allow to change zstd window log in the query output when zstd compression is used"}, - {"enable_zstd_qat_codec", false, false, "Add new ZSTD_QAT codec"}, - {"enable_vertical_final", false, true, "Use vertical final by default"}, - {"output_format_arrow_use_64_bit_indexes_for_dictionary", false, false, "Allow to use 64 bit indexes type in Arrow dictionaries"}, - {"max_rows_in_set_to_optimize_join", 100000, 0, "Disable join optimization as it prevents from read in order optimization"}, - {"output_format_pretty_color", true, "auto", "Setting is changed to allow also for auto value, disabling ANSI escapes if output is not a tty"}, - {"function_visible_width_behavior", 0, 1, "We changed the default behavior of `visibleWidth` to be more precise"}, - {"max_estimated_execution_time", 0, 0, "Separate max_execution_time and max_estimated_execution_time"}, - {"iceberg_engine_ignore_schema_evolution", false, false, "Allow to ignore schema evolution in Iceberg table engine"}, - {"optimize_injective_functions_in_group_by", false, true, "Replace injective functions by it's arguments in GROUP BY section in analyzer"}, - {"update_insert_deduplication_token_in_dependent_materialized_views", false, false, "Allow to update insert deduplication token with table identifier during insert in dependent materialized views"}, - {"azure_max_unexpected_write_error_retries", 4, 4, "The maximum number of retries in case of unexpected errors during Azure blob storage write"}, - {"split_parts_ranges_into_intersecting_and_non_intersecting_final", false, true, "Allow to split parts ranges into intersecting and non intersecting during FINAL optimization"}, - {"split_intersecting_parts_ranges_into_layers_final", true, true, "Allow to split intersecting parts ranges into layers during FINAL optimization"}}}, - {"23.12", {{"allow_suspicious_ttl_expressions", true, false, "It is a new setting, and in previous versions the behavior was equivalent to allowing."}, - {"input_format_parquet_allow_missing_columns", false, true, "Allow missing columns in Parquet files by default"}, - {"input_format_orc_allow_missing_columns", false, true, "Allow missing columns in ORC files by default"}, - {"input_format_arrow_allow_missing_columns", false, true, "Allow missing columns in Arrow files by default"}}}, - {"23.11", {{"parsedatetime_parse_without_leading_zeros", false, true, "Improved compatibility with MySQL DATE_FORMAT/STR_TO_DATE"}}}, - {"23.9", {{"optimize_group_by_constant_keys", false, true, "Optimize group by constant keys by default"}, - {"input_format_json_try_infer_named_tuples_from_objects", false, true, "Try to infer named Tuples from JSON objects by default"}, - {"input_format_json_read_numbers_as_strings", false, true, "Allow to read numbers as strings in JSON formats by default"}, - {"input_format_json_read_arrays_as_strings", false, true, "Allow to read arrays as strings in JSON formats by default"}, - {"input_format_json_infer_incomplete_types_as_strings", false, true, "Allow to infer incomplete types as Strings in JSON formats by default"}, - {"input_format_json_try_infer_numbers_from_strings", true, false, "Don't infer numbers from strings in JSON formats by default to prevent possible parsing errors"}, - {"http_write_exception_in_output_format", false, true, "Output valid JSON/XML on exception in HTTP streaming."}}}, - {"23.8", {{"rewrite_count_distinct_if_with_count_distinct_implementation", false, true, "Rewrite countDistinctIf with count_distinct_implementation configuration"}}}, - {"23.7", {{"function_sleep_max_microseconds_per_block", 0, 3000000, "In previous versions, the maximum sleep time of 3 seconds was applied only for `sleep`, but not for `sleepEachRow` function. In the new version, we introduce this setting. If you set compatibility with the previous versions, we will disable the limit altogether."}}}, - {"23.6", {{"http_send_timeout", 180, 30, "3 minutes seems crazy long. Note that this is timeout for a single network write call, not for the whole upload operation."}, - {"http_receive_timeout", 180, 30, "See http_send_timeout."}}}, - {"23.5", {{"input_format_parquet_preserve_order", true, false, "Allow Parquet reader to reorder rows for better parallelism."}, - {"parallelize_output_from_storages", false, true, "Allow parallelism when executing queries that read from file/url/s3/etc. This may reorder rows."}, - {"use_with_fill_by_sorting_prefix", false, true, "Columns preceding WITH FILL columns in ORDER BY clause form sorting prefix. Rows with different values in sorting prefix are filled independently"}, - {"output_format_parquet_compliant_nested_types", false, true, "Change an internal field name in output Parquet file schema."}}}, - {"23.4", {{"allow_suspicious_indices", true, false, "If true, index can defined with identical expressions"}, - {"allow_nonconst_timezone_arguments", true, false, "Allow non-const timezone arguments in certain time-related functions like toTimeZone(), fromUnixTimestamp*(), snowflakeToDateTime*()."}, - {"connect_timeout_with_failover_ms", 50, 1000, "Increase default connect timeout because of async connect"}, - {"connect_timeout_with_failover_secure_ms", 100, 1000, "Increase default secure connect timeout because of async connect"}, - {"hedged_connection_timeout_ms", 100, 50, "Start new connection in hedged requests after 50 ms instead of 100 to correspond with previous connect timeout"}, - {"formatdatetime_f_prints_single_zero", true, false, "Improved compatibility with MySQL DATE_FORMAT()/STR_TO_DATE()"}, - {"formatdatetime_parsedatetime_m_is_month_name", false, true, "Improved compatibility with MySQL DATE_FORMAT/STR_TO_DATE"}}}, - {"23.3", {{"output_format_parquet_version", "1.0", "2.latest", "Use latest Parquet format version for output format"}, - {"input_format_json_ignore_unknown_keys_in_named_tuple", false, true, "Improve parsing JSON objects as named tuples"}, - {"input_format_native_allow_types_conversion", false, true, "Allow types conversion in Native input forma"}, - {"output_format_arrow_compression_method", "none", "lz4_frame", "Use lz4 compression in Arrow output format by default"}, - {"output_format_parquet_compression_method", "snappy", "lz4", "Use lz4 compression in Parquet output format by default"}, - {"output_format_orc_compression_method", "none", "lz4_frame", "Use lz4 compression in ORC output format by default"}, - {"async_query_sending_for_remote", false, true, "Create connections and send query async across shards"}}}, - {"23.2", {{"output_format_parquet_fixed_string_as_fixed_byte_array", false, true, "Use Parquet FIXED_LENGTH_BYTE_ARRAY type for FixedString by default"}, - {"output_format_arrow_fixed_string_as_fixed_byte_array", false, true, "Use Arrow FIXED_SIZE_BINARY type for FixedString by default"}, - {"query_plan_remove_redundant_distinct", false, true, "Remove redundant Distinct step in query plan"}, - {"optimize_duplicate_order_by_and_distinct", true, false, "Remove duplicate ORDER BY and DISTINCT if it's possible"}, - {"insert_keeper_max_retries", 0, 20, "Enable reconnections to Keeper on INSERT, improve reliability"}}}, - {"23.1", {{"input_format_json_read_objects_as_strings", 0, 1, "Enable reading nested json objects as strings while object type is experimental"}, - {"input_format_json_defaults_for_missing_elements_in_named_tuple", false, true, "Allow missing elements in JSON objects while reading named tuples by default"}, - {"input_format_csv_detect_header", false, true, "Detect header in CSV format by default"}, - {"input_format_tsv_detect_header", false, true, "Detect header in TSV format by default"}, - {"input_format_custom_detect_header", false, true, "Detect header in CustomSeparated format by default"}, - {"query_plan_remove_redundant_sorting", false, true, "Remove redundant sorting in query plan. For example, sorting steps related to ORDER BY clauses in subqueries"}}}, - {"22.12", {{"max_size_to_preallocate_for_aggregation", 10'000'000, 100'000'000, "This optimizes performance"}, - {"query_plan_aggregation_in_order", 0, 1, "Enable some refactoring around query plan"}, - {"format_binary_max_string_size", 0, 1_GiB, "Prevent allocating large amount of memory"}}}, - {"22.11", {{"use_structure_from_insertion_table_in_table_functions", 0, 2, "Improve using structure from insertion table in table functions"}}}, - {"22.9", {{"force_grouping_standard_compatibility", false, true, "Make GROUPING function output the same as in SQL standard and other DBMS"}}}, - {"22.7", {{"cross_to_inner_join_rewrite", 1, 2, "Force rewrite comma join to inner"}, - {"enable_positional_arguments", false, true, "Enable positional arguments feature by default"}, - {"format_csv_allow_single_quotes", true, false, "Most tools don't treat single quote in CSV specially, don't do it by default too"}}}, - {"22.6", {{"output_format_json_named_tuples_as_objects", false, true, "Allow to serialize named tuples as JSON objects in JSON formats by default"}, - {"input_format_skip_unknown_fields", false, true, "Optimize reading subset of columns for some input formats"}}}, - {"22.5", {{"memory_overcommit_ratio_denominator", 0, 1073741824, "Enable memory overcommit feature by default"}, - {"memory_overcommit_ratio_denominator_for_user", 0, 1073741824, "Enable memory overcommit feature by default"}}}, - {"22.4", {{"allow_settings_after_format_in_insert", true, false, "Do not allow SETTINGS after FORMAT for INSERT queries because ClickHouse interpret SETTINGS as some values, which is misleading"}}}, - {"22.3", {{"cast_ipv4_ipv6_default_on_conversion_error", true, false, "Make functions cast(value, 'IPv4') and cast(value, 'IPv6') behave same as toIPv4 and toIPv6 functions"}}}, - {"21.12", {{"stream_like_engine_allow_direct_select", true, false, "Do not allow direct select for Kafka/RabbitMQ/FileLog by default"}}}, - {"21.9", {{"output_format_decimal_trailing_zeros", true, false, "Do not output trailing zeros in text representation of Decimal types by default for better looking output"}, - {"use_hedged_requests", false, true, "Enable Hedged Requests feature by default"}}}, - {"21.7", {{"legacy_column_name_of_tuple_literal", true, false, "Add this setting only for compatibility reasons. It makes sense to set to 'true', while doing rolling update of cluster from version lower than 21.7 to higher"}}}, - {"21.5", {{"async_socket_for_remote", false, true, "Fix all problems and turn on asynchronous reads from socket for remote queries by default again"}}}, - {"21.3", {{"async_socket_for_remote", true, false, "Turn off asynchronous reads from socket for remote queries because of some problems"}, - {"optimize_normalize_count_variants", false, true, "Rewrite aggregate functions that semantically equals to count() as count() by default"}, - {"normalize_function_names", false, true, "Normalize function names to their canonical names, this was needed for projection query routing"}}}, - {"21.2", {{"enable_global_with_statement", false, true, "Propagate WITH statements to UNION queries and all subqueries by default"}}}, - {"21.1", {{"insert_quorum_parallel", false, true, "Use parallel quorum inserts by default. It is significantly more convenient to use than sequential quorum inserts"}, - {"input_format_null_as_default", false, true, "Allow to insert NULL as default for input formats by default"}, - {"optimize_on_insert", false, true, "Enable data optimization on INSERT by default for better user experience"}, - {"use_compact_format_in_distributed_parts_names", false, true, "Use compact format for async INSERT into Distributed tables by default"}}}, - {"20.10", {{"format_regexp_escaping_rule", "Escaped", "Raw", "Use Raw as default escaping rule for Regexp format to male the behaviour more like to what users expect"}}}, - {"20.7", {{"show_table_uuid_in_table_create_query_if_not_nil", true, false, "Stop showing UID of the table in its CREATE query for Engine=Atomic"}}}, - {"20.5", {{"input_format_with_names_use_header", false, true, "Enable using header with names for formats with WithNames/WithNamesAndTypes suffixes"}, - {"allow_suspicious_codecs", true, false, "Don't allow to specify meaningless compression codecs"}}}, - {"20.4", {{"validate_polygons", false, true, "Throw exception if polygon is invalid in function pointInPolygon by default instead of returning possibly wrong results"}}}, - {"19.18", {{"enable_scalar_subquery_optimization", false, true, "Prevent scalar subqueries from (de)serializing large scalar values and possibly avoid running the same subquery more than once"}}}, - {"19.14", {{"any_join_distinct_right_table_keys", true, false, "Disable ANY RIGHT and ANY FULL JOINs by default to avoid inconsistency"}}}, - {"19.12", {{"input_format_defaults_for_omitted_fields", false, true, "Enable calculation of complex default expressions for omitted fields for some input formats, because it should be the expected behaviour"}}}, - {"19.5", {{"max_partitions_per_insert_block", 0, 100, "Add a limit for the number of partitions in one block"}}}, - {"18.12.17", {{"enable_optimize_predicate_expression", 0, 1, "Optimize predicates to subqueries by default"}}}, {"24.12", { } @@ -338,6 +76,7 @@ static std::initializer_list Date: Mon, 29 Jul 2024 12:54:36 +0000 Subject: [PATCH 121/227] Trying to fix test_cache_evicted_by_temporary_data and print debug info --- .../config.d/storage_configuration.xml | 6 +-- .../test_temporary_data_in_cache/test.py | 44 ++++++++++++------- 2 files changed, 31 insertions(+), 19 deletions(-) diff --git a/tests/integration/test_temporary_data_in_cache/configs/config.d/storage_configuration.xml b/tests/integration/test_temporary_data_in_cache/configs/config.d/storage_configuration.xml index 5a087d03266..107864fde0c 100644 --- a/tests/integration/test_temporary_data_in_cache/configs/config.d/storage_configuration.xml +++ b/tests/integration/test_temporary_data_in_cache/configs/config.d/storage_configuration.xml @@ -10,9 +10,9 @@ cache local_disk /tiny_local_cache/ - 10M - 1M - 1M + 12M + 100K + 100K 1 diff --git a/tests/integration/test_temporary_data_in_cache/test.py b/tests/integration/test_temporary_data_in_cache/test.py index cab134dcce2..abdfb5f4064 100644 --- a/tests/integration/test_temporary_data_in_cache/test.py +++ b/tests/integration/test_temporary_data_in_cache/test.py @@ -7,6 +7,9 @@ import fnmatch from helpers.cluster import ClickHouseCluster from helpers.client import QueryRuntimeException + +MB = 1024 * 1024 + cluster = ClickHouseCluster(__file__) node = cluster.add_instance( @@ -36,15 +39,28 @@ def test_cache_evicted_by_temporary_data(start_cluster): q("SELECT sum(size) FROM system.filesystem_cache").strip() ) - assert get_cache_size() == 0 + def dump_debug_info(): + return "\n".join( + [ + ">>> filesystem_cache <<<", + q("SELECT * FROM system.filesystem_cache FORMAT Vertical"), + ">>> remote_data_paths <<<", + q("SELECT * FROM system.remote_data_paths FORMAT Vertical"), + ">>> tiny_local_cache_local_disk <<<", + q( + "SELECT * FROM system.disks WHERE name = 'tiny_local_cache_local_disk' FORMAT Vertical" + ), + ] + ) - assert get_free_space() > 8 * 1024 * 1024 + assert get_cache_size() == 0, dump_debug_info() + assert get_free_space() > 8 * MB, dump_debug_info() # Codec is NONE to make cache size predictable q( - "CREATE TABLE t1 (x UInt64 CODEC(NONE), y UInt64 CODEC(NONE)) ENGINE = MergeTree ORDER BY x SETTINGS storage_policy = 'tiny_local_cache'" + "CREATE TABLE t1 (x UInt64 CODEC(NONE)) ENGINE = MergeTree ORDER BY x SETTINGS storage_policy = 'tiny_local_cache'" ) - q("INSERT INTO t1 SELECT number, number FROM numbers(1024 * 1024)") + q("INSERT INTO t1 SELECT number FROM numbers(1024 * 1024)") # To be sure that nothing is reading the cache and entries for t1 can be evited q("OPTIMIZE TABLE t1 FINAL") @@ -54,11 +70,11 @@ def test_cache_evicted_by_temporary_data(start_cluster): q("SELECT sum(x) FROM t1") cache_size_with_t1 = get_cache_size() - assert cache_size_with_t1 > 8 * 1024 * 1024 + assert cache_size_with_t1 > 8 * MB, dump_debug_info() # Almost all disk space is occupied by t1 cache free_space_with_t1 = get_free_space() - assert free_space_with_t1 < 4 * 1024 * 1024 + assert free_space_with_t1 < 4 * MB, dump_debug_info() # Try to sort the table, but fail because of lack of disk space with pytest.raises(QueryRuntimeException) as exc: @@ -76,31 +92,27 @@ def test_cache_evicted_by_temporary_data(start_cluster): # Some data evicted from cache by temporary data cache_size_after_eviction = get_cache_size() - assert cache_size_after_eviction < cache_size_with_t1 + assert cache_size_after_eviction < cache_size_with_t1, dump_debug_info() # Disk space freed, at least 3 MB, because temporary data tried to write 4 MB - assert get_free_space() > free_space_with_t1 + 3 * 1024 * 1024 + assert get_free_space() > free_space_with_t1 + 3 * MB, dump_debug_info() # Read some data to fill the cache again - q("SELECT avg(y) FROM t1") + q("SELECT avg(x) FROM t1") cache_size_with_t1 = get_cache_size() - assert cache_size_with_t1 > 8 * 1024 * 1024, q( - "SELECT * FROM system.filesystem_cache FORMAT Vertical" - ) + assert cache_size_with_t1 > 8 * MB, dump_debug_info() # Almost all disk space is occupied by t1 cache free_space_with_t1 = get_free_space() - assert free_space_with_t1 < 4 * 1024 * 1024, q( - "SELECT * FROM system.disks WHERE name = 'tiny_local_cache_local_disk' FORMAT Vertical" - ) + assert free_space_with_t1 < 4 * MB, dump_debug_info() node.http_query( "SELECT randomPrintableASCII(1024) FROM numbers(8 * 1024) FORMAT TSV", params={"buffer_size": 0, "wait_end_of_query": 1}, ) - assert get_free_space() > free_space_with_t1 + 3 * 1024 * 1024 + assert get_free_space() > free_space_with_t1 + 3 * MB, dump_debug_info() # not enough space for buffering 32 MB with pytest.raises(Exception) as exc: From 064c0eb9587d9dbd1fa81cdbae8554c22dd11734 Mon Sep 17 00:00:00 2001 From: Andrey Zvonov Date: Fri, 2 Aug 2024 10:35:32 +0000 Subject: [PATCH 122/227] even better healthcheck for ldap --- tests/integration/compose/docker_compose_ldap.yml | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/integration/compose/docker_compose_ldap.yml b/tests/integration/compose/docker_compose_ldap.yml index 1f50b34735d..f49e00400a2 100644 --- a/tests/integration/compose/docker_compose_ldap.yml +++ b/tests/integration/compose/docker_compose_ldap.yml @@ -19,6 +19,7 @@ services: ldapsearch -x -H ldap://localhost:$$LDAP_PORT_NUMBER -D $$LDAP_ADMIN_DN -w $$LDAP_ADMIN_PASSWORD -b $$LDAP_ROOT | grep -c -E "member: cn=j(ohn|ane)doe" | grep 2 >> /dev/null + && cat /run/slapd/slapd.pid interval: 10s retries: 10 timeout: 2s From 02e48436057e45a884a1381e1c9cda9e1fe7de17 Mon Sep 17 00:00:00 2001 From: vdimir Date: Tue, 30 Jul 2024 09:30:39 +0000 Subject: [PATCH 123/227] test_cache_evicted_by_temporary_data drop cache --- .../test_temporary_data_in_cache/test.py | 30 ++++++++++--------- 1 file changed, 16 insertions(+), 14 deletions(-) diff --git a/tests/integration/test_temporary_data_in_cache/test.py b/tests/integration/test_temporary_data_in_cache/test.py index abdfb5f4064..87192a20975 100644 --- a/tests/integration/test_temporary_data_in_cache/test.py +++ b/tests/integration/test_temporary_data_in_cache/test.py @@ -39,19 +39,21 @@ def test_cache_evicted_by_temporary_data(start_cluster): q("SELECT sum(size) FROM system.filesystem_cache").strip() ) - def dump_debug_info(): - return "\n".join( - [ - ">>> filesystem_cache <<<", - q("SELECT * FROM system.filesystem_cache FORMAT Vertical"), - ">>> remote_data_paths <<<", - q("SELECT * FROM system.remote_data_paths FORMAT Vertical"), - ">>> tiny_local_cache_local_disk <<<", - q( - "SELECT * FROM system.disks WHERE name = 'tiny_local_cache_local_disk' FORMAT Vertical" - ), - ] - ) + dump_debug_info = lambda: "\n".join( + [ + ">>> filesystem_cache <<<", + q("SELECT * FROM system.filesystem_cache FORMAT Vertical"), + ">>> remote_data_paths <<<", + q("SELECT * FROM system.remote_data_paths FORMAT Vertical"), + ">>> tiny_local_cache_local_disk <<<", + q( + "SELECT * FROM system.disks WHERE name = 'tiny_local_cache_local_disk' FORMAT Vertical" + ), + ] + ) + + q("SYSTEM DROP FILESYSTEM CACHE") + q("DROP TABLE IF EXISTS t1 SYNC") assert get_cache_size() == 0, dump_debug_info() assert get_free_space() > 8 * MB, dump_debug_info() @@ -124,4 +126,4 @@ def test_cache_evicted_by_temporary_data(start_cluster): str(exc.value), "*Failed to reserve * for temporary file*" ), exc.value - q("DROP TABLE IF EXISTS t1") + q("DROP TABLE IF EXISTS t1 SYNC") From 092c837119a9be11cfcc85b4696e9a9c74d9bbc8 Mon Sep 17 00:00:00 2001 From: Michael Stetsyuk Date: Fri, 2 Aug 2024 12:13:26 +0100 Subject: [PATCH 124/227] randomize table name in test_storage_delta --- tests/integration/test_storage_delta/test.py | 53 +++++++------------- 1 file changed, 17 insertions(+), 36 deletions(-) diff --git a/tests/integration/test_storage_delta/test.py b/tests/integration/test_storage_delta/test.py index 92a870ab360..054b79ff6fe 100644 --- a/tests/integration/test_storage_delta/test.py +++ b/tests/integration/test_storage_delta/test.py @@ -8,7 +8,8 @@ import os import json import time import glob -import shutil +import random +import string import pyspark import delta @@ -53,6 +54,11 @@ def get_spark(): return builder.master("local").getOrCreate() +def randomize_table_name(table_name, random_suffix_length=10): + letters = string.ascii_letters + string.digits + return f"{table_name}{''.join(random.choice(letters) for _ in range(random_suffix_length))}" + + @pytest.fixture(scope="module") def started_cluster(): try: @@ -152,7 +158,7 @@ def test_single_log_file(started_cluster): spark = started_cluster.spark_session minio_client = started_cluster.minio_client bucket = started_cluster.minio_bucket - TABLE_NAME = "test_single_log_file" + TABLE_NAME = randomize_table_name("test_single_log_file") inserted_data = "SELECT number as a, toString(number + 1) as b FROM numbers(100)" parquet_data_path = create_initial_data_file( @@ -170,16 +176,13 @@ def test_single_log_file(started_cluster): inserted_data ) - os.unlink(parquet_data_path) - shutil.rmtree(f"/{TABLE_NAME}") - def test_partition_by(started_cluster): instance = started_cluster.instances["node1"] spark = started_cluster.spark_session minio_client = started_cluster.minio_client bucket = started_cluster.minio_bucket - TABLE_NAME = "test_partition_by" + TABLE_NAME = randomize_table_name("test_partition_by") write_delta_from_df( spark, @@ -195,15 +198,13 @@ def test_partition_by(started_cluster): create_delta_table(instance, TABLE_NAME) assert int(instance.query(f"SELECT count() FROM {TABLE_NAME}")) == 10 - shutil.rmtree(f"/{TABLE_NAME}") - def test_checkpoint(started_cluster): instance = started_cluster.instances["node1"] spark = started_cluster.spark_session minio_client = started_cluster.minio_client bucket = started_cluster.minio_bucket - TABLE_NAME = "test_checkpoint" + TABLE_NAME = randomize_table_name("test_checkpoint") write_delta_from_df( spark, @@ -272,16 +273,13 @@ def test_checkpoint(started_cluster): ).strip() ) - shutil.rmtree(f"/{TABLE_NAME}") - spark.sql(f"DROP TABLE {TABLE_NAME}") - def test_multiple_log_files(started_cluster): instance = started_cluster.instances["node1"] spark = started_cluster.spark_session minio_client = started_cluster.minio_client bucket = started_cluster.minio_bucket - TABLE_NAME = "test_multiple_log_files" + TABLE_NAME = randomize_table_name("test_multiple_log_files") write_delta_from_df( spark, generate_data(spark, 0, 100), f"/{TABLE_NAME}", mode="overwrite" @@ -313,15 +311,13 @@ def test_multiple_log_files(started_cluster): "SELECT number, toString(number + 1) FROM numbers(200)" ) - shutil.rmtree(f"/{TABLE_NAME}") - def test_metadata(started_cluster): instance = started_cluster.instances["node1"] spark = started_cluster.spark_session minio_client = started_cluster.minio_client bucket = started_cluster.minio_bucket - TABLE_NAME = "test_metadata" + TABLE_NAME = randomize_table_name("test_metadata") parquet_data_path = create_initial_data_file( started_cluster, @@ -348,14 +344,11 @@ def test_metadata(started_cluster): create_delta_table(instance, TABLE_NAME) assert int(instance.query(f"SELECT count() FROM {TABLE_NAME}")) == 100 - os.unlink(parquet_data_path) - shutil.rmtree(f"/{TABLE_NAME}") - def test_types(started_cluster): - TABLE_NAME = "test_types" + TABLE_NAME = randomize_table_name("test_types") spark = started_cluster.spark_session - result_file = f"{TABLE_NAME}_result_2" + result_file = randomize_table_name(f"{TABLE_NAME}_result_2") delta_table = ( DeltaTable.create(spark) @@ -423,16 +416,13 @@ def test_types(started_cluster): ] ) - shutil.rmtree(f"/{result_file}") - spark.sql(f"DROP TABLE {TABLE_NAME}") - def test_restart_broken(started_cluster): instance = started_cluster.instances["node1"] spark = started_cluster.spark_session minio_client = started_cluster.minio_client bucket = "broken" - TABLE_NAME = "test_restart_broken" + TABLE_NAME = randomize_table_name("test_restart_broken") if not minio_client.bucket_exists(bucket): minio_client.make_bucket(bucket) @@ -487,16 +477,13 @@ def test_restart_broken(started_cluster): assert int(instance.query(f"SELECT count() FROM {TABLE_NAME}")) == 100 - os.unlink(parquet_data_path) - shutil.rmtree(f"/{TABLE_NAME}") - def test_restart_broken_table_function(started_cluster): instance = started_cluster.instances["node1"] spark = started_cluster.spark_session minio_client = started_cluster.minio_client bucket = "broken2" - TABLE_NAME = "test_restart_broken_table_function" + TABLE_NAME = randomize_table_name("test_restart_broken_table_function") if not minio_client.bucket_exists(bucket): minio_client.make_bucket(bucket) @@ -544,16 +531,13 @@ def test_restart_broken_table_function(started_cluster): assert int(instance.query(f"SELECT count() FROM {TABLE_NAME}")) == 100 - os.unlink(parquet_data_path) - shutil.rmtree(f"/{TABLE_NAME}") - def test_partition_columns(started_cluster): instance = started_cluster.instances["node1"] spark = started_cluster.spark_session minio_client = started_cluster.minio_client bucket = started_cluster.minio_bucket - TABLE_NAME = "test_partition_columns" + TABLE_NAME = randomize_table_name("test_partition_columns") result_file = f"{TABLE_NAME}" partition_columns = ["b", "c", "d", "e"] @@ -744,6 +728,3 @@ SELECT * FROM deltaLake('http://{started_cluster.minio_ip}:{started_cluster.mini ) == 1 ) - - shutil.rmtree(f"/{TABLE_NAME}") - spark.sql(f"DROP TABLE {TABLE_NAME}") From 7d1e958097e716f3ea1e0b7e51d6dfa575229c4c Mon Sep 17 00:00:00 2001 From: Nikita Fomichev Date: Fri, 2 Aug 2024 13:32:59 +0200 Subject: [PATCH 125/227] Integration tests: fix ports clashing problem --- tests/integration/conftest.py | 44 ++++++++++++++++++ tests/integration/helpers/cluster.py | 67 ++++++++++++++++++++++++---- 2 files changed, 103 insertions(+), 8 deletions(-) diff --git a/tests/integration/conftest.py b/tests/integration/conftest.py index f4be31cc532..0a47840ede3 100644 --- a/tests/integration/conftest.py +++ b/tests/integration/conftest.py @@ -2,6 +2,8 @@ import logging import os +import socket +import multiprocessing import pytest # pylint:disable=import-error; for style check from helpers.cluster import run_and_check @@ -11,6 +13,7 @@ from helpers.network import _NetworkManager # # [1]: https://github.com/pytest-dev/pytest/issues/5502 logging.raiseExceptions = False +PORTS_PER_WORKER = 50 @pytest.fixture(scope="session", autouse=True) @@ -111,5 +114,46 @@ def pytest_addoption(parser): ) +def get_n_free_ports(total): + ports = [] + + while len(ports) < total: + with socket.socket() as s: + s.bind(("", 0)) + ports.append(s.getsockname()[1]) + + return ports + + def pytest_configure(config): os.environ["INTEGRATION_TESTS_RUN_ID"] = config.option.run_id + + # When running tests without pytest-xdist, + # the `pytest_xdist_setupnodes` hook is not executed + worker_ports = os.getenv("WORKER_FREE_PORTS", None) + if worker_ports is None: + os.environ["WORKER_FREE_PORTS"] = " ".join( + ([str(p) for p in get_n_free_ports(PORTS_PER_WORKER)]) + ) + + +def pytest_xdist_setupnodes(config, specs): + # Find {PORTS_PER_WORKER} * {number of xdist workers} ports and + # allocate pool of {PORTS_PER_WORKER} ports to each worker + + # Get number of xdist workers + num_workers = 1 + if os.environ.get("PYTEST_XDIST_WORKER", "master") == "master": + num_workers = config.getoption("numprocesses", 1) + if num_workers == "auto": + num_workers = multiprocessing.cpu_count() + + # Get free ports which will be distributed across workers + ports = get_n_free_ports(num_workers * PORTS_PER_WORKER) + + # Iterate over specs of workers and add allocated ports to env variable + for i, spec in enumerate(specs): + start_range = i * PORTS_PER_WORKER + spec.env["WORKER_FREE_PORTS"] = " ".join( + ([str(p) for p in ports[start_range : start_range + PORTS_PER_WORKER]]) + ) diff --git a/tests/integration/helpers/cluster.py b/tests/integration/helpers/cluster.py index 6bc0ece63ca..3480a3089fe 100644 --- a/tests/integration/helpers/cluster.py +++ b/tests/integration/helpers/cluster.py @@ -135,6 +135,52 @@ def get_free_port(): return s.getsockname()[1] +def is_port_free(port: int) -> bool: + try: + with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s: + s.bind(("", port)) + return True + except socket.error: + return False + + +class PortPoolManager: + """ + This class is used for distribution of ports allocated to single pytest-xdist worker + It can be used by multiple ClickHouseCluster instances + """ + + # Shared between instances + all_ports = None + free_ports = None + + def __init__(self): + self.used_ports = [] + + if self.all_ports is None: + worker_ports = os.getenv("WORKER_FREE_PORTS") + ports = [int(p) for p in worker_ports.split(" ")] + + # Static vars + PortPoolManager.all_ports = ports + PortPoolManager.free_ports = ports + + def get_port(self): + for port in self.free_ports: + if is_port_free(port): + self.free_ports.remove(port) + self.used_ports.append(port) + return port + + raise Exception( + f"No free ports: {self.all_ports}", + ) + + def return_used_ports(self): + self.free_ports.extend(self.used_ports) + self.used_ports.clear() + + def retry_exception(num, delay, func, exception=Exception, *args, **kwargs): """ Retry if `func()` throws, `num` times. @@ -716,62 +762,67 @@ class ClickHouseCluster: .stop() ) + self.port_pool = PortPoolManager() + @property def kafka_port(self): if self._kafka_port: return self._kafka_port - self._kafka_port = get_free_port() + self._kafka_port = self.port_pool.get_port() return self._kafka_port @property def schema_registry_port(self): if self._schema_registry_port: return self._schema_registry_port - self._schema_registry_port = get_free_port() + self._schema_registry_port = self.port_pool.get_port() return self._schema_registry_port @property def schema_registry_auth_port(self): if self._schema_registry_auth_port: return self._schema_registry_auth_port - self._schema_registry_auth_port = get_free_port() + self._schema_registry_auth_port = self.port_pool.get_port() return self._schema_registry_auth_port @property def kerberized_kafka_port(self): if self._kerberized_kafka_port: return self._kerberized_kafka_port - self._kerberized_kafka_port = get_free_port() + self._kerberized_kafka_port = self.port_pool.get_port() return self._kerberized_kafka_port @property def azurite_port(self): if self._azurite_port: return self._azurite_port - self._azurite_port = get_free_port() + self._azurite_port = self.port_pool.get_port() return self._azurite_port @property def mongo_port(self): if self._mongo_port: return self._mongo_port - self._mongo_port = get_free_port() + self._mongo_port = self.port_pool.get_port() return self._mongo_port @property def mongo_no_cred_port(self): if self._mongo_no_cred_port: return self._mongo_no_cred_port - self._mongo_no_cred_port = get_free_port() + self._mongo_no_cred_port = self.port_pool.get_port() return self._mongo_no_cred_port @property def redis_port(self): if self._redis_port: return self._redis_port - self._redis_port = get_free_port() + self._redis_port = self.port_pool.get_port() return self._redis_port + def __exit__(self, exc_type, exc_val, exc_tb): + self.port_pool.return_used_ports() + def print_all_docker_pieces(self): res_networks = subprocess.check_output( f"docker network ls --filter name='{self.project_name}*'", From 7d45529fe8d28a6b39deb32d060343bb5d03b64f Mon Sep 17 00:00:00 2001 From: Michael Stetsyuk Date: Fri, 2 Aug 2024 12:35:40 +0100 Subject: [PATCH 126/227] randomize query id in test_checking_s3_blobs_paranoid --- .../test_checking_s3_blobs_paranoid/test.py | 31 ++++++++++++------- 1 file changed, 19 insertions(+), 12 deletions(-) diff --git a/tests/integration/test_checking_s3_blobs_paranoid/test.py b/tests/integration/test_checking_s3_blobs_paranoid/test.py index afe8449b44a..c22142046c7 100644 --- a/tests/integration/test_checking_s3_blobs_paranoid/test.py +++ b/tests/integration/test_checking_s3_blobs_paranoid/test.py @@ -4,6 +4,8 @@ import logging import pytest import os import minio +import random +import string from helpers.cluster import ClickHouseCluster from helpers.mock_servers import start_s3_mock @@ -45,6 +47,11 @@ def cluster(): cluster.shutdown() +def randomize_query_id(query_id, random_suffix_length=10): + letters = string.ascii_letters + string.digits + return f"{query_id}_{''.join(random.choice(letters) for _ in range(random_suffix_length))}" + + @pytest.fixture(scope="module") def init_broken_s3(cluster): yield start_s3_mock(cluster, "broken_s3", "8083") @@ -128,7 +135,7 @@ def test_upload_s3_fail_create_multi_part_upload(cluster, broken_s3, compression broken_s3.setup_at_create_multi_part_upload() - insert_query_id = f"INSERT_INTO_TABLE_FUNCTION_FAIL_CREATE_MPU_{compression}" + insert_query_id = randomize_query_id(f"INSERT_INTO_TABLE_FUNCTION_FAIL_CREATE_MPU_{compression}") error = node.query_and_get_error( f""" INSERT INTO @@ -170,7 +177,7 @@ def test_upload_s3_fail_upload_part_when_multi_part_upload( broken_s3.setup_fake_multpartuploads() broken_s3.setup_at_part_upload(count=1, after=2) - insert_query_id = f"INSERT_INTO_TABLE_FUNCTION_FAIL_UPLOAD_PART_{compression}" + insert_query_id = randomize_query_id(f"INSERT_INTO_TABLE_FUNCTION_FAIL_UPLOAD_PART_{compression}") error = node.query_and_get_error( f""" INSERT INTO @@ -222,7 +229,7 @@ def test_when_error_is_retried(cluster, broken_s3, action_and_message): broken_s3.setup_fake_multpartuploads() broken_s3.setup_at_part_upload(count=3, after=2, action=action) - insert_query_id = f"INSERT_INTO_TABLE_{action}_RETRIED" + insert_query_id = randomize_query_id(f"INSERT_INTO_TABLE_{action}_RETRIED") node.query( f""" INSERT INTO @@ -251,7 +258,7 @@ def test_when_error_is_retried(cluster, broken_s3, action_and_message): assert s3_errors == 3 broken_s3.setup_at_part_upload(count=1000, after=2, action=action) - insert_query_id = f"INSERT_INTO_TABLE_{action}_RETRIED_1" + insert_query_id = randomize_query_id(f"INSERT_INTO_TABLE_{action}_RETRIED_1") error = node.query_and_get_error( f""" INSERT INTO @@ -286,7 +293,7 @@ def test_when_s3_broken_pipe_at_upload_is_retried(cluster, broken_s3): action="broken_pipe", ) - insert_query_id = f"TEST_WHEN_S3_BROKEN_PIPE_AT_UPLOAD" + insert_query_id = randomize_query_id(f"TEST_WHEN_S3_BROKEN_PIPE_AT_UPLOAD") node.query( f""" INSERT INTO @@ -320,7 +327,7 @@ def test_when_s3_broken_pipe_at_upload_is_retried(cluster, broken_s3): after=2, action="broken_pipe", ) - insert_query_id = f"TEST_WHEN_S3_BROKEN_PIPE_AT_UPLOAD_1" + insert_query_id = randomize_query_id(f"TEST_WHEN_S3_BROKEN_PIPE_AT_UPLOAD_1") error = node.query_and_get_error( f""" INSERT INTO @@ -362,7 +369,7 @@ def test_when_s3_connection_reset_by_peer_at_upload_is_retried( action_args=["1"] if send_something else ["0"], ) - insert_query_id = ( + insert_query_id = randomize_query_id( f"TEST_WHEN_S3_CONNECTION_RESET_BY_PEER_AT_UPLOAD_{send_something}" ) node.query( @@ -399,7 +406,7 @@ def test_when_s3_connection_reset_by_peer_at_upload_is_retried( action="connection_reset_by_peer", action_args=["1"] if send_something else ["0"], ) - insert_query_id = ( + insert_query_id = randomize_query_id( f"TEST_WHEN_S3_CONNECTION_RESET_BY_PEER_AT_UPLOAD_{send_something}_1" ) error = node.query_and_get_error( @@ -444,7 +451,7 @@ def test_when_s3_connection_reset_by_peer_at_create_mpu_retried( action_args=["1"] if send_something else ["0"], ) - insert_query_id = ( + insert_query_id = randomize_query_id( f"TEST_WHEN_S3_CONNECTION_RESET_BY_PEER_AT_MULTIPARTUPLOAD_{send_something}" ) node.query( @@ -482,7 +489,7 @@ def test_when_s3_connection_reset_by_peer_at_create_mpu_retried( action_args=["1"] if send_something else ["0"], ) - insert_query_id = ( + insert_query_id = randomize_query_id( f"TEST_WHEN_S3_CONNECTION_RESET_BY_PEER_AT_MULTIPARTUPLOAD_{send_something}_1" ) error = node.query_and_get_error( @@ -522,7 +529,7 @@ def test_query_is_canceled_with_inf_retries(cluster, broken_s3): action="connection_refused", ) - insert_query_id = f"TEST_QUERY_IS_CANCELED_WITH_INF_RETRIES" + insert_query_id = randomize_query_id(f"TEST_QUERY_IS_CANCELED_WITH_INF_RETRIES") request = node.get_query_request( f""" INSERT INTO @@ -580,7 +587,7 @@ def test_adaptive_timeouts(cluster, broken_s3, node_name): count=1000000, ) - insert_query_id = f"TEST_ADAPTIVE_TIMEOUTS_{node_name}" + insert_query_id = randomize_query_id(f"TEST_ADAPTIVE_TIMEOUTS_{node_name}") node.query( f""" INSERT INTO From 34cba1bdda55cdd2409c535be56e4fe6165c894a Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Fri, 2 Aug 2024 11:46:03 +0000 Subject: [PATCH 127/227] Automatic style fix --- tests/integration/test_checking_s3_blobs_paranoid/test.py | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/tests/integration/test_checking_s3_blobs_paranoid/test.py b/tests/integration/test_checking_s3_blobs_paranoid/test.py index c22142046c7..b995b4d6461 100644 --- a/tests/integration/test_checking_s3_blobs_paranoid/test.py +++ b/tests/integration/test_checking_s3_blobs_paranoid/test.py @@ -135,7 +135,9 @@ def test_upload_s3_fail_create_multi_part_upload(cluster, broken_s3, compression broken_s3.setup_at_create_multi_part_upload() - insert_query_id = randomize_query_id(f"INSERT_INTO_TABLE_FUNCTION_FAIL_CREATE_MPU_{compression}") + insert_query_id = randomize_query_id( + f"INSERT_INTO_TABLE_FUNCTION_FAIL_CREATE_MPU_{compression}" + ) error = node.query_and_get_error( f""" INSERT INTO @@ -177,7 +179,9 @@ def test_upload_s3_fail_upload_part_when_multi_part_upload( broken_s3.setup_fake_multpartuploads() broken_s3.setup_at_part_upload(count=1, after=2) - insert_query_id = randomize_query_id(f"INSERT_INTO_TABLE_FUNCTION_FAIL_UPLOAD_PART_{compression}") + insert_query_id = randomize_query_id( + f"INSERT_INTO_TABLE_FUNCTION_FAIL_UPLOAD_PART_{compression}" + ) error = node.query_and_get_error( f""" INSERT INTO From d2d8a16ca6c5c3df31a62894fe2bcfb26d570061 Mon Sep 17 00:00:00 2001 From: Antonio Andelic Date: Fri, 2 Aug 2024 13:51:03 +0200 Subject: [PATCH 128/227] Fix refreshable MVs --- src/Databases/DatabaseOrdinary.cpp | 3 +++ 1 file changed, 3 insertions(+) diff --git a/src/Databases/DatabaseOrdinary.cpp b/src/Databases/DatabaseOrdinary.cpp index 3ab5d3fa697..8808261654f 100644 --- a/src/Databases/DatabaseOrdinary.cpp +++ b/src/Databases/DatabaseOrdinary.cpp @@ -16,6 +16,7 @@ #include #include #include +#include #include #include #include @@ -250,6 +251,8 @@ void DatabaseOrdinary::loadTablesMetadata(ContextPtr local_context, ParsedTables convertMergeTreeToReplicatedIfNeeded(ast, qualified_name, file_name); + NormalizeSelectWithUnionQueryVisitor::Data data{local_context->getSettingsRef().union_default_mode}; + NormalizeSelectWithUnionQueryVisitor{data}.visit(ast); std::lock_guard lock{metadata.mutex}; metadata.parsed_tables[qualified_name] = ParsedTableMetadata{full_path.string(), ast}; metadata.total_dictionaries += create_query->is_dictionary; From cae76458504f46fefe52c7d93594870d0bac4479 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Fri, 2 Aug 2024 14:00:50 +0200 Subject: [PATCH 129/227] Make 02514_null_dictionary_source parallelizable --- tests/queries/0_stateless/02514_null_dictionary_source.sql | 2 -- 1 file changed, 2 deletions(-) diff --git a/tests/queries/0_stateless/02514_null_dictionary_source.sql b/tests/queries/0_stateless/02514_null_dictionary_source.sql index 74fb57707ff..bfd36042f57 100644 --- a/tests/queries/0_stateless/02514_null_dictionary_source.sql +++ b/tests/queries/0_stateless/02514_null_dictionary_source.sql @@ -1,5 +1,3 @@ --- Tags: no-parallel - DROP DICTIONARY IF EXISTS null_dict; CREATE DICTIONARY null_dict ( id UInt64, From 45f54c3633a388075898e3dcabc850dd1347587d Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Fri, 2 Aug 2024 14:01:01 +0200 Subject: [PATCH 130/227] Make 01086_window_view_cleanup parallelizable --- .../0_stateless/01086_window_view_cleanup.sh | 47 ++++++++++--------- 1 file changed, 24 insertions(+), 23 deletions(-) diff --git a/tests/queries/0_stateless/01086_window_view_cleanup.sh b/tests/queries/0_stateless/01086_window_view_cleanup.sh index 113bcffb2af..0bce08523e2 100755 --- a/tests/queries/0_stateless/01086_window_view_cleanup.sh +++ b/tests/queries/0_stateless/01086_window_view_cleanup.sh @@ -1,5 +1,4 @@ #!/usr/bin/env bash -# Tags: no-parallel # Creation of a database with Ordinary engine emits a warning. CLICKHOUSE_CLIENT_SERVER_LOGS_LEVEL=fatal @@ -12,38 +11,40 @@ opts=( "--allow_experimental_analyzer=0" ) -$CLICKHOUSE_CLIENT "${opts[@]}" --allow_deprecated_database_ordinary=1 < Date: Fri, 2 Aug 2024 14:03:27 +0200 Subject: [PATCH 131/227] Integration tests: fix ports clashing problem 2 --- tests/integration/conftest.py | 36 ++++++++++++++--------------------- 1 file changed, 14 insertions(+), 22 deletions(-) diff --git a/tests/integration/conftest.py b/tests/integration/conftest.py index 0a47840ede3..a386ed53009 100644 --- a/tests/integration/conftest.py +++ b/tests/integration/conftest.py @@ -2,11 +2,9 @@ import logging import os -import socket -import multiprocessing import pytest # pylint:disable=import-error; for style check -from helpers.cluster import run_and_check +from helpers.cluster import run_and_check, is_port_free from helpers.network import _NetworkManager # This is a workaround for a problem with logging in pytest [1]. @@ -114,15 +112,16 @@ def pytest_addoption(parser): ) -def get_n_free_ports(total): +def get_unique_free_ports(total): ports = [] + for port in range(30000, 55000): + if is_port_free(port) and port not in ports: + ports.append(port) - while len(ports) < total: - with socket.socket() as s: - s.bind(("", 0)) - ports.append(s.getsockname()[1]) + if len(ports) == total: + return ports - return ports + raise Exception(f"Can't collect {total} ports. Collected: {len(ports)}") def pytest_configure(config): @@ -132,9 +131,8 @@ def pytest_configure(config): # the `pytest_xdist_setupnodes` hook is not executed worker_ports = os.getenv("WORKER_FREE_PORTS", None) if worker_ports is None: - os.environ["WORKER_FREE_PORTS"] = " ".join( - ([str(p) for p in get_n_free_ports(PORTS_PER_WORKER)]) - ) + master_ports = get_unique_free_ports(PORTS_PER_WORKER) + os.environ["WORKER_FREE_PORTS"] = " ".join(([str(p) for p in master_ports])) def pytest_xdist_setupnodes(config, specs): @@ -142,18 +140,12 @@ def pytest_xdist_setupnodes(config, specs): # allocate pool of {PORTS_PER_WORKER} ports to each worker # Get number of xdist workers - num_workers = 1 - if os.environ.get("PYTEST_XDIST_WORKER", "master") == "master": - num_workers = config.getoption("numprocesses", 1) - if num_workers == "auto": - num_workers = multiprocessing.cpu_count() - + num_workers = len(specs) # Get free ports which will be distributed across workers - ports = get_n_free_ports(num_workers * PORTS_PER_WORKER) + ports = get_unique_free_ports(num_workers * PORTS_PER_WORKER) # Iterate over specs of workers and add allocated ports to env variable for i, spec in enumerate(specs): start_range = i * PORTS_PER_WORKER - spec.env["WORKER_FREE_PORTS"] = " ".join( - ([str(p) for p in ports[start_range : start_range + PORTS_PER_WORKER]]) - ) + per_workrer_ports = ports[start_range : start_range + PORTS_PER_WORKER] + spec.env["WORKER_FREE_PORTS"] = " ".join(([str(p) for p in per_workrer_ports])) From 85cecf990ddefc25ff0903b6f97f84cd46dc471a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Fri, 2 Aug 2024 14:06:51 +0200 Subject: [PATCH 132/227] Parallelize 02703_row_policies_for_database_combination --- ...3_row_policies_for_database_combination.sh | 92 +++++++++++++++++++ ..._row_policies_for_database_combination.sql | 88 ------------------ 2 files changed, 92 insertions(+), 88 deletions(-) create mode 100755 tests/queries/0_stateless/02703_row_policies_for_database_combination.sh delete mode 100644 tests/queries/0_stateless/02703_row_policies_for_database_combination.sql diff --git a/tests/queries/0_stateless/02703_row_policies_for_database_combination.sh b/tests/queries/0_stateless/02703_row_policies_for_database_combination.sh new file mode 100755 index 00000000000..f7b7c814d29 --- /dev/null +++ b/tests/queries/0_stateless/02703_row_policies_for_database_combination.sh @@ -0,0 +1,92 @@ +#!/usr/bin/env bash +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +$CLICKHOUSE_CLIENT --multiquery " + +DROP TABLE IF EXISTS 02703_rptable; +DROP TABLE IF EXISTS 02703_rptable_another; +CREATE TABLE 02703_rptable (x UInt8, y UInt8) ENGINE = MergeTree ORDER BY x; + +INSERT INTO 02703_rptable VALUES (1, 10), (2, 20), (3, 30), (4, 40); + +CREATE TABLE 02703_rptable_another ENGINE = MergeTree ORDER BY x AS SELECT * FROM 02703_rptable; + + +DROP ROW POLICY IF EXISTS 02703_filter_1 ON ${CLICKHOUSE_DATABASE}.02703_rptable; +DROP ROW POLICY IF EXISTS 02703_filter_2 ON ${CLICKHOUSE_DATABASE}.*; +DROP ROW POLICY IF EXISTS 02703_filter_3 ON ${CLICKHOUSE_DATABASE}.02703_rptable; +DROP ROW POLICY IF EXISTS 02703_filter_4 ON ${CLICKHOUSE_DATABASE}.02703_rptable; +DROP ROW POLICY IF EXISTS 02703_filter_5 ON ${CLICKHOUSE_DATABASE}.*; + +-- the test assumes users_without_row_policies_can_read_rows is true + +SELECT 'None'; +SELECT * FROM 02703_rptable; + +CREATE ROW POLICY 02703_filter_1 ON ${CLICKHOUSE_DATABASE}.02703_rptable USING x=1 AS permissive TO ALL; +SELECT 'R1: x == 1'; +SELECT * FROM 02703_rptable; + +CREATE ROW POLICY 02703_filter_2 ON ${CLICKHOUSE_DATABASE}.* USING x=2 AS permissive TO ALL; +SELECT 'R1, R2: (x == 1) OR (x == 2)'; +SELECT * FROM 02703_rptable; + +SELECT 'R1, R2: (x == 2) FROM ANOTHER'; +SELECT * FROM 02703_rptable_another; + +CREATE ROW POLICY 02703_filter_3 ON ${CLICKHOUSE_DATABASE}.02703_rptable USING x=3 AS permissive TO ALL; +SELECT 'R1, R2, R3: (x == 1) OR (x == 2) OR (x == 3)'; +SELECT * FROM 02703_rptable; + +CREATE ROW POLICY 02703_filter_4 ON ${CLICKHOUSE_DATABASE}.02703_rptable USING x<=2 AS restrictive TO ALL; +SELECT 'R1, R2, R3, R4: ((x == 1) OR (x == 2) OR (x == 3)) AND (x <= 2)'; +SELECT * FROM 02703_rptable; + +CREATE ROW POLICY 02703_filter_5 ON ${CLICKHOUSE_DATABASE}.* USING y>=20 AS restrictive TO ALL; +SELECT 'R1, R2, R3, R4, R5: ((x == 1) OR (x == 2) OR (x == 3)) AND (x <= 2) AND (y >= 20)'; +SELECT * FROM 02703_rptable; + +CREATE TABLE 02703_after_rp ENGINE = MergeTree ORDER BY x AS SELECT * FROM 02703_rptable; +SELECT * FROM 02703_after_rp; + +-- does not matter if policies or table are created first +SELECT 'R1, R2, R3, R4, R5: (x == 2) AND (y >= 20) FROM AFTER_RP'; +SELECT * FROM 02703_after_rp; + +SELECT 'R1, R2, R3, R4, R5: (x == 2) AND (y >= 20) FROM ANOTHER'; +SELECT * FROM 02703_rptable_another; + +DROP ROW POLICY 02703_filter_1 ON ${CLICKHOUSE_DATABASE}.02703_rptable; +SELECT 'R2, R3, R4, R5: ((x == 2) OR (x == 3)) AND (x <= 2) AND (y >= 20)'; +SELECT * FROM 02703_rptable; + +DROP ROW POLICY 02703_filter_2 ON ${CLICKHOUSE_DATABASE}.*; +SELECT 'R3, R4, R5: (x == 3) AND (x <= 2) AND (y >= 20)'; +SELECT * FROM 02703_rptable; + +DROP ROW POLICY 02703_filter_3 ON ${CLICKHOUSE_DATABASE}.02703_rptable; +SELECT 'R4, R5: (x <= 2) AND (y >= 20)'; +SELECT * FROM 02703_rptable; + +DROP ROW POLICY 02703_filter_4 ON ${CLICKHOUSE_DATABASE}.02703_rptable; +SELECT 'R5: (x >= 2)'; +SELECT * FROM 02703_rptable; + +CREATE TABLE 02703_unexpected_columns (xx UInt8, yy UInt8) ENGINE = MergeTree ORDER BY xx; +SELECT 'Policy not applicable'; +SELECT * FROM 02703_unexpected_columns; -- { serverError UNKNOWN_IDENTIFIER } -- Missing columns: 'x' while processing query + +DROP ROW POLICY 02703_filter_5 ON ${CLICKHOUSE_DATABASE}.*; +SELECT 'None'; +SELECT * FROM 02703_rptable; + +SELECT 'No problematic policy, select works'; +SELECT 'Ok' FROM ${CLICKHOUSE_DATABASE}.02703_unexpected_columns; + +DROP TABLE 02703_rptable; +DROP TABLE 02703_rptable_another; +DROP TABLE 02703_unexpected_columns; + +" diff --git a/tests/queries/0_stateless/02703_row_policies_for_database_combination.sql b/tests/queries/0_stateless/02703_row_policies_for_database_combination.sql deleted file mode 100644 index 8c93fc595ba..00000000000 --- a/tests/queries/0_stateless/02703_row_policies_for_database_combination.sql +++ /dev/null @@ -1,88 +0,0 @@ --- Tags: no-parallel - -DROP DATABASE IF EXISTS 02703_db; -CREATE DATABASE 02703_db; -DROP TABLE IF EXISTS 02703_db.02703_rptable; -DROP TABLE IF EXISTS 02703_db.02703_rptable_another; -CREATE TABLE 02703_db.02703_rptable (x UInt8, y UInt8) ENGINE = MergeTree ORDER BY x; - -INSERT INTO 02703_db.02703_rptable VALUES (1, 10), (2, 20), (3, 30), (4, 40); - -CREATE TABLE 02703_db.02703_rptable_another ENGINE = MergeTree ORDER BY x AS SELECT * FROM 02703_db.02703_rptable; - - -DROP ROW POLICY IF EXISTS 02703_filter_1 ON 02703_db.02703_rptable; -DROP ROW POLICY IF EXISTS 02703_filter_2 ON 02703_db.*; -DROP ROW POLICY IF EXISTS 02703_filter_3 ON 02703_db.02703_rptable; -DROP ROW POLICY IF EXISTS 02703_filter_4 ON 02703_db.02703_rptable; -DROP ROW POLICY IF EXISTS 02703_filter_5 ON 02703_db.*; - --- the test assumes users_without_row_policies_can_read_rows is true - -SELECT 'None'; -SELECT * FROM 02703_db.02703_rptable; - -CREATE ROW POLICY 02703_filter_1 ON 02703_db.02703_rptable USING x=1 AS permissive TO ALL; -SELECT 'R1: x == 1'; -SELECT * FROM 02703_db.02703_rptable; - -CREATE ROW POLICY 02703_filter_2 ON 02703_db.* USING x=2 AS permissive TO ALL; -SELECT 'R1, R2: (x == 1) OR (x == 2)'; -SELECT * FROM 02703_db.02703_rptable; - -SELECT 'R1, R2: (x == 2) FROM ANOTHER'; -SELECT * FROM 02703_db.02703_rptable_another; - -CREATE ROW POLICY 02703_filter_3 ON 02703_db.02703_rptable USING x=3 AS permissive TO ALL; -SELECT 'R1, R2, R3: (x == 1) OR (x == 2) OR (x == 3)'; -SELECT * FROM 02703_db.02703_rptable; - -CREATE ROW POLICY 02703_filter_4 ON 02703_db.02703_rptable USING x<=2 AS restrictive TO ALL; -SELECT 'R1, R2, R3, R4: ((x == 1) OR (x == 2) OR (x == 3)) AND (x <= 2)'; -SELECT * FROM 02703_db.02703_rptable; - -CREATE ROW POLICY 02703_filter_5 ON 02703_db.* USING y>=20 AS restrictive TO ALL; -SELECT 'R1, R2, R3, R4, R5: ((x == 1) OR (x == 2) OR (x == 3)) AND (x <= 2) AND (y >= 20)'; -SELECT * FROM 02703_db.02703_rptable; - -CREATE TABLE 02703_db.02703_after_rp ENGINE = MergeTree ORDER BY x AS SELECT * FROM 02703_db.02703_rptable; -SELECT * FROM 02703_db.02703_after_rp; - --- does not matter if policies or table are created first -SELECT 'R1, R2, R3, R4, R5: (x == 2) AND (y >= 20) FROM AFTER_RP'; -SELECT * FROM 02703_db.02703_after_rp; - -SELECT 'R1, R2, R3, R4, R5: (x == 2) AND (y >= 20) FROM ANOTHER'; -SELECT * FROM 02703_db.02703_rptable_another; - -DROP ROW POLICY 02703_filter_1 ON 02703_db.02703_rptable; -SELECT 'R2, R3, R4, R5: ((x == 2) OR (x == 3)) AND (x <= 2) AND (y >= 20)'; -SELECT * FROM 02703_db.02703_rptable; - -DROP ROW POLICY 02703_filter_2 ON 02703_db.*; -SELECT 'R3, R4, R5: (x == 3) AND (x <= 2) AND (y >= 20)'; -SELECT * FROM 02703_db.02703_rptable; - -DROP ROW POLICY 02703_filter_3 ON 02703_db.02703_rptable; -SELECT 'R4, R5: (x <= 2) AND (y >= 20)'; -SELECT * FROM 02703_db.02703_rptable; - -DROP ROW POLICY 02703_filter_4 ON 02703_db.02703_rptable; -SELECT 'R5: (x >= 2)'; -SELECT * FROM 02703_db.02703_rptable; - -CREATE TABLE 02703_db.02703_unexpected_columns (xx UInt8, yy UInt8) ENGINE = MergeTree ORDER BY xx; -SELECT 'Policy not applicable'; -SELECT * FROM 02703_db.02703_unexpected_columns; -- { serverError UNKNOWN_IDENTIFIER } -- Missing columns: 'x' while processing query - -DROP ROW POLICY 02703_filter_5 ON 02703_db.*; -SELECT 'None'; -SELECT * FROM 02703_db.02703_rptable; - -SELECT 'No problematic policy, select works'; -SELECT 'Ok' FROM 02703_db.02703_unexpected_columns; - -DROP TABLE 02703_db.02703_rptable; -DROP TABLE 02703_db.02703_rptable_another; -DROP TABLE 02703_db.02703_unexpected_columns; -DROP DATABASE 02703_db; From f55784c636c1dcb503dadb2f75bd6b586271bf0d Mon Sep 17 00:00:00 2001 From: vdimir Date: Fri, 2 Aug 2024 12:07:05 +0000 Subject: [PATCH 133/227] Fix 03203_client_benchmark_options --- tests/queries/0_stateless/03203_client_benchmark_options.sh | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/queries/0_stateless/03203_client_benchmark_options.sh b/tests/queries/0_stateless/03203_client_benchmark_options.sh index 37a1f2cd3ac..967db056c0b 100755 --- a/tests/queries/0_stateless/03203_client_benchmark_options.sh +++ b/tests/queries/0_stateless/03203_client_benchmark_options.sh @@ -5,10 +5,10 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CUR_DIR"/../shell_config.sh output=$(${CLICKHOUSE_CLIENT} -t -q "SELECT sleepEachRow(2) FORMAT Null" 2>&1) -{ echo "$output" | grep -q "^2\." && echo "Ok"; } || { echo "Fail"; echo "'$output'"; } +{ number=$(echo "$output" | grep -o "^[0-9]"); [[ -n "$number" && "$number" -ge 2 ]] && echo "Ok"; } || { echo "Fail"; echo "'$output'"; } output=$(${CLICKHOUSE_CLIENT} --time -q "SELECT sleepEachRow(2) FORMAT Null" 2>&1) -{ echo "$output" | grep -q "^2\." && echo "Ok"; } || { echo "Fail"; echo "'$output'"; } +{ number=$(echo "$output" | grep -o "^[0-9]"); [[ -n "$number" && "$number" -ge 2 ]] && echo "Ok"; } || { echo "Fail"; echo "'$output'"; } output=$(${CLICKHOUSE_CLIENT} --memory-usage -q "SELECT sum(number) FROM numbers(10_000) FORMAT Null" 2>&1) { echo "$output" | grep -q "^[0-9]\+$" && echo "Ok"; } || { echo "Fail"; echo "'$output'"; } From cb83274d37f8e90ef9c79451ce01a2ba95363158 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Fri, 2 Aug 2024 14:18:31 +0200 Subject: [PATCH 134/227] Make 02703_row_policy_for_database parallelizable --- .../02703_row_policy_for_database.reference | 22 +++---- .../02703_row_policy_for_database.sh | 59 +++++++++++++++++++ .../02703_row_policy_for_database.sql | 53 ----------------- 3 files changed, 70 insertions(+), 64 deletions(-) create mode 100755 tests/queries/0_stateless/02703_row_policy_for_database.sh delete mode 100644 tests/queries/0_stateless/02703_row_policy_for_database.sql diff --git a/tests/queries/0_stateless/02703_row_policy_for_database.reference b/tests/queries/0_stateless/02703_row_policy_for_database.reference index b67ea69ae72..56b0d8c6f20 100644 --- a/tests/queries/0_stateless/02703_row_policy_for_database.reference +++ b/tests/queries/0_stateless/02703_row_policy_for_database.reference @@ -1,20 +1,20 @@ -- row policies for database - -- SHOW CREATE POLICY db1_02703 ON db1_02703.* -CREATE ROW POLICY db1_02703 ON db1_02703.* FOR SELECT USING 1 TO ALL - -- SHOW CREATE POLICY ON db1_02703.* -CREATE ROW POLICY db1_02703 ON db1_02703.* FOR SELECT USING 1 TO ALL -CREATE ROW POLICY tbl1_02703 ON db1_02703.`table` FOR SELECT USING 1 TO ALL - -- SHOW CREATE POLICY ON db1_02703.`*` + -- SHOW CREATE POLICY default ON default.* +CREATE ROW POLICY default_db_policy ON default.* FOR SELECT USING 1 TO ALL + -- SHOW CREATE POLICY ON default.* +CREATE ROW POLICY default_db_policy ON default.* FOR SELECT USING 1 TO ALL +CREATE ROW POLICY default_tb_policy ON default.`table` FOR SELECT USING 1 TO ALL + -- SHOW CREATE POLICY ON default.`*` R1, R2: (x == 1) OR (x == 2) 1 2 Check system.query_log SELECT \'-- row policies for database\'; [] -SELECT \' -- SHOW CREATE POLICY db1_02703 ON db1_02703.*\'; [] -SELECT \' -- SHOW CREATE POLICY ON db1_02703.*\'; [] -SELECT \' -- SHOW CREATE POLICY ON db1_02703.`*`\'; [] +SELECT \' -- SHOW CREATE POLICY default ON default.*\'; [] +SELECT \' -- SHOW CREATE POLICY ON default.*\'; [] +SELECT \' -- SHOW CREATE POLICY ON default.`*`\'; [] SELECT \'R1, R2: (x == 1) OR (x == 2)\'; [] -SELECT * FROM 02703_rqtable_default; ['`02703_filter_11_db` ON default.*','`02703_filter_11` ON default.`02703_rqtable_default`'] +SELECT * FROM 02703_rqtable_default; ['default_filter_11_db_policy ON default.*','default_filter_11_policy ON default.`02703_rqtable_default`'] SELECT \'Check system.query_log\'; [] -- CREATE DATABASE-LEVEL POLICY IN CURRENT DATABASE -CREATE ROW POLICY db2_02703 ON db1_02703.* TO u1_02703 +CREATE ROW POLICY db2_02703 ON default.* TO user_default diff --git a/tests/queries/0_stateless/02703_row_policy_for_database.sh b/tests/queries/0_stateless/02703_row_policy_for_database.sh new file mode 100755 index 00000000000..e94bc7acd5e --- /dev/null +++ b/tests/queries/0_stateless/02703_row_policy_for_database.sh @@ -0,0 +1,59 @@ +#!/usr/bin/env bash +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +CLICKHOUSE_USER="user_$CLICKHOUSE_DATABASE" + +$CLICKHOUSE_CLIENT --multiquery " + +DROP USER IF EXISTS ${CLICKHOUSE_USER}; +CREATE USER ${CLICKHOUSE_USER}; + +CREATE TABLE ${CLICKHOUSE_DATABASE}.02703_rqtable (x UInt8) ENGINE = MergeTree ORDER BY x; +INSERT INTO ${CLICKHOUSE_DATABASE}.02703_rqtable VALUES (1), (2), (3), (4); + +SELECT '-- row policies for database'; +CREATE ROW POLICY ${CLICKHOUSE_DATABASE}_db_policy ON ${CLICKHOUSE_DATABASE}.* USING 1 AS PERMISSIVE TO ALL; +CREATE ROW POLICY ${CLICKHOUSE_DATABASE}_tb_policy ON ${CLICKHOUSE_DATABASE}.table USING 1 AS PERMISSIVE TO ALL; +SELECT ' -- SHOW CREATE POLICY ${CLICKHOUSE_DATABASE} ON ${CLICKHOUSE_DATABASE}.*'; +SHOW CREATE POLICY ${CLICKHOUSE_DATABASE}_db_policy ON ${CLICKHOUSE_DATABASE}.*; +SELECT ' -- SHOW CREATE POLICY ON ${CLICKHOUSE_DATABASE}.*'; +SHOW CREATE POLICY ON ${CLICKHOUSE_DATABASE}.*; +SELECT ' -- SHOW CREATE POLICY ON ${CLICKHOUSE_DATABASE}.\`*\`'; +SHOW CREATE POLICY ON ${CLICKHOUSE_DATABASE}.\`*\`; +DROP POLICY ${CLICKHOUSE_DATABASE}_db_policy ON ${CLICKHOUSE_DATABASE}.*; +DROP POLICY ${CLICKHOUSE_DATABASE}_tb_policy ON ${CLICKHOUSE_DATABASE}.table; +" + +$CLICKHOUSE_CLIENT --query "CREATE ROW POLICY any_02703 ON *.some_table USING 1 AS PERMISSIVE TO ALL;" 2>&1 | grep -q "SYNTAX_ERROR" + +$CLICKHOUSE_CLIENT --multiquery " +CREATE TABLE 02703_rqtable_default (x UInt8) ENGINE = MergeTree ORDER BY x; + +CREATE ROW POLICY ${CLICKHOUSE_DATABASE}_filter_11_db_policy ON * USING x=1 AS permissive TO ALL; +CREATE ROW POLICY ${CLICKHOUSE_DATABASE}_filter_11_policy ON 02703_rqtable_default USING x=2 AS permissive TO ALL; + +INSERT INTO 02703_rqtable_default VALUES (1), (2), (3), (4); + +SELECT 'R1, R2: (x == 1) OR (x == 2)'; +SELECT * FROM 02703_rqtable_default; + +DROP TABLE 02703_rqtable_default; + +SELECT 'Check system.query_log'; +SYSTEM FLUSH LOGS; +SELECT query, used_row_policies FROM system.query_log WHERE current_database == currentDatabase() AND type == 'QueryStart' AND query_kind == 'Select' ORDER BY event_time_microseconds; + +DROP ROW POLICY ${CLICKHOUSE_DATABASE}_filter_11_db_policy ON *; +DROP ROW POLICY ${CLICKHOUSE_DATABASE}_filter_11_policy ON 02703_rqtable_default; + +USE ${CLICKHOUSE_DATABASE}; +SELECT ' -- CREATE DATABASE-LEVEL POLICY IN CURRENT DATABASE'; +CREATE ROW POLICY db2_02703 ON * TO ${CLICKHOUSE_USER}; +SHOW CREATE POLICY db2_02703 ON *; + +DROP ROW POLICY db2_02703 ON *; + +DROP USER ${CLICKHOUSE_USER}; +" diff --git a/tests/queries/0_stateless/02703_row_policy_for_database.sql b/tests/queries/0_stateless/02703_row_policy_for_database.sql deleted file mode 100644 index 51ce5f4f870..00000000000 --- a/tests/queries/0_stateless/02703_row_policy_for_database.sql +++ /dev/null @@ -1,53 +0,0 @@ --- Tags: no-parallel - -DROP DATABASE IF EXISTS db1_02703; -DROP USER IF EXISTS u1_02703; -CREATE USER u1_02703; - -CREATE DATABASE db1_02703; - -CREATE TABLE db1_02703.02703_rqtable (x UInt8) ENGINE = MergeTree ORDER BY x; -INSERT INTO db1_02703.02703_rqtable VALUES (1), (2), (3), (4); - - -SELECT '-- row policies for database'; -CREATE ROW POLICY db1_02703 ON db1_02703.* USING 1 AS PERMISSIVE TO ALL; -CREATE ROW POLICY tbl1_02703 ON db1_02703.table USING 1 AS PERMISSIVE TO ALL; -SELECT ' -- SHOW CREATE POLICY db1_02703 ON db1_02703.*'; -SHOW CREATE POLICY db1_02703 ON db1_02703.*; -SELECT ' -- SHOW CREATE POLICY ON db1_02703.*'; -SHOW CREATE POLICY ON db1_02703.*; -SELECT ' -- SHOW CREATE POLICY ON db1_02703.`*`'; -SHOW CREATE POLICY ON db1_02703.`*`; -DROP POLICY db1_02703 ON db1_02703.*; -DROP POLICY tbl1_02703 ON db1_02703.table; - -CREATE ROW POLICY any_02703 ON *.some_table USING 1 AS PERMISSIVE TO ALL; -- { clientError SYNTAX_ERROR } - -CREATE TABLE 02703_rqtable_default (x UInt8) ENGINE = MergeTree ORDER BY x; - -CREATE ROW POLICY 02703_filter_11_db ON * USING x=1 AS permissive TO ALL; -CREATE ROW POLICY 02703_filter_11 ON 02703_rqtable_default USING x=2 AS permissive TO ALL; - -INSERT INTO 02703_rqtable_default VALUES (1), (2), (3), (4); - -SELECT 'R1, R2: (x == 1) OR (x == 2)'; -SELECT * FROM 02703_rqtable_default; - -DROP TABLE 02703_rqtable_default; - -SELECT 'Check system.query_log'; -SYSTEM FLUSH LOGS; -SELECT query, used_row_policies FROM system.query_log WHERE current_database == currentDatabase() AND type == 'QueryStart' AND query_kind == 'Select' ORDER BY event_time_microseconds; - -DROP ROW POLICY 02703_filter_11_db ON *; -DROP ROW POLICY 02703_filter_11 ON 02703_rqtable_default; - -USE db1_02703; -SELECT ' -- CREATE DATABASE-LEVEL POLICY IN CURRENT DATABASE'; -CREATE ROW POLICY db2_02703 ON * TO u1_02703; -SHOW CREATE POLICY db2_02703 ON *; - -DROP ROW POLICY db2_02703 ON *; - -DROP USER u1_02703; From 829d07c3a54716bb62cd36d4f2fa24da64ea6a0a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Fri, 2 Aug 2024 14:22:37 +0200 Subject: [PATCH 135/227] Make 02703_row_policies_for_asterisk parallelizable --- .../02703_row_policies_for_asterisk.sh | 14 +++++++++ .../02703_row_policies_for_asterisk.sql | 11 ------- ...3_row_policies_for_database_combination.sh | 30 +++++++++---------- 3 files changed, 29 insertions(+), 26 deletions(-) create mode 100755 tests/queries/0_stateless/02703_row_policies_for_asterisk.sh delete mode 100644 tests/queries/0_stateless/02703_row_policies_for_asterisk.sql diff --git a/tests/queries/0_stateless/02703_row_policies_for_asterisk.sh b/tests/queries/0_stateless/02703_row_policies_for_asterisk.sh new file mode 100755 index 00000000000..f9670e5f6f8 --- /dev/null +++ b/tests/queries/0_stateless/02703_row_policies_for_asterisk.sh @@ -0,0 +1,14 @@ +#!/usr/bin/env bash +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +$CLICKHOUSE_CLIENT --multiquery " + SELECT 'Policy for table \`*\` does not affect other tables in the database'; + CREATE ROW POLICY 02703_asterisk_${CLICKHOUSE_DATABASE}_policy ON ${CLICKHOUSE_DATABASE}.\`*\` USING x=1 AS permissive TO ALL; + CREATE TABLE ${CLICKHOUSE_DATABASE}.\`*\` (x UInt8, y UInt8) ENGINE = MergeTree ORDER BY x AS SELECT 100, 20; + CREATE TABLE ${CLICKHOUSE_DATABASE}.\`other\` (x UInt8, y UInt8) ENGINE = MergeTree ORDER BY x AS SELECT 100, 20; + SELECT 'star', * FROM ${CLICKHOUSE_DATABASE}.\`*\`; + SELECT 'other', * FROM ${CLICKHOUSE_DATABASE}.other; + DROP ROW POLICY 02703_asterisk_${CLICKHOUSE_DATABASE}_policy ON ${CLICKHOUSE_DATABASE}.\`*\`; +" diff --git a/tests/queries/0_stateless/02703_row_policies_for_asterisk.sql b/tests/queries/0_stateless/02703_row_policies_for_asterisk.sql deleted file mode 100644 index 96b1c01a6d6..00000000000 --- a/tests/queries/0_stateless/02703_row_policies_for_asterisk.sql +++ /dev/null @@ -1,11 +0,0 @@ --- Tags: no-parallel - -SELECT 'Policy for table `*` does not affect other tables in the database'; -CREATE DATABASE 02703_db_asterisk; -CREATE ROW POLICY 02703_asterisk ON 02703_db_asterisk.`*` USING x=1 AS permissive TO ALL; -CREATE TABLE 02703_db_asterisk.`*` (x UInt8, y UInt8) ENGINE = MergeTree ORDER BY x AS SELECT 100, 20; -CREATE TABLE 02703_db_asterisk.`other` (x UInt8, y UInt8) ENGINE = MergeTree ORDER BY x AS SELECT 100, 20; -SELECT 'star', * FROM 02703_db_asterisk.`*`; -SELECT 'other', * FROM 02703_db_asterisk.other; -DROP ROW POLICY 02703_asterisk ON 02703_db_asterisk.`*`; -DROP DATABASE 02703_db_asterisk; diff --git a/tests/queries/0_stateless/02703_row_policies_for_database_combination.sh b/tests/queries/0_stateless/02703_row_policies_for_database_combination.sh index f7b7c814d29..35151eed220 100755 --- a/tests/queries/0_stateless/02703_row_policies_for_database_combination.sh +++ b/tests/queries/0_stateless/02703_row_policies_for_database_combination.sh @@ -14,37 +14,37 @@ INSERT INTO 02703_rptable VALUES (1, 10), (2, 20), (3, 30), (4, 40); CREATE TABLE 02703_rptable_another ENGINE = MergeTree ORDER BY x AS SELECT * FROM 02703_rptable; -DROP ROW POLICY IF EXISTS 02703_filter_1 ON ${CLICKHOUSE_DATABASE}.02703_rptable; -DROP ROW POLICY IF EXISTS 02703_filter_2 ON ${CLICKHOUSE_DATABASE}.*; -DROP ROW POLICY IF EXISTS 02703_filter_3 ON ${CLICKHOUSE_DATABASE}.02703_rptable; -DROP ROW POLICY IF EXISTS 02703_filter_4 ON ${CLICKHOUSE_DATABASE}.02703_rptable; -DROP ROW POLICY IF EXISTS 02703_filter_5 ON ${CLICKHOUSE_DATABASE}.*; +DROP ROW POLICY IF EXISTS 02703_filter_policy_${CLICKHOUSE_DATABASE}_1 ON ${CLICKHOUSE_DATABASE}.02703_rptable; +DROP ROW POLICY IF EXISTS 02703_filter_policy_${CLICKHOUSE_DATABASE}_2 ON ${CLICKHOUSE_DATABASE}.*; +DROP ROW POLICY IF EXISTS 02703_filter_policy_${CLICKHOUSE_DATABASE}_3 ON ${CLICKHOUSE_DATABASE}.02703_rptable; +DROP ROW POLICY IF EXISTS 02703_filter_policy_${CLICKHOUSE_DATABASE}_4 ON ${CLICKHOUSE_DATABASE}.02703_rptable; +DROP ROW POLICY IF EXISTS 02703_filter_policy_${CLICKHOUSE_DATABASE}_5 ON ${CLICKHOUSE_DATABASE}.*; -- the test assumes users_without_row_policies_can_read_rows is true SELECT 'None'; SELECT * FROM 02703_rptable; -CREATE ROW POLICY 02703_filter_1 ON ${CLICKHOUSE_DATABASE}.02703_rptable USING x=1 AS permissive TO ALL; +CREATE ROW POLICY 02703_filter_policy_${CLICKHOUSE_DATABASE}_1 ON ${CLICKHOUSE_DATABASE}.02703_rptable USING x=1 AS permissive TO ALL; SELECT 'R1: x == 1'; SELECT * FROM 02703_rptable; -CREATE ROW POLICY 02703_filter_2 ON ${CLICKHOUSE_DATABASE}.* USING x=2 AS permissive TO ALL; +CREATE ROW POLICY 02703_filter_policy_${CLICKHOUSE_DATABASE}_2 ON ${CLICKHOUSE_DATABASE}.* USING x=2 AS permissive TO ALL; SELECT 'R1, R2: (x == 1) OR (x == 2)'; SELECT * FROM 02703_rptable; SELECT 'R1, R2: (x == 2) FROM ANOTHER'; SELECT * FROM 02703_rptable_another; -CREATE ROW POLICY 02703_filter_3 ON ${CLICKHOUSE_DATABASE}.02703_rptable USING x=3 AS permissive TO ALL; +CREATE ROW POLICY 02703_filter_policy_${CLICKHOUSE_DATABASE}_3 ON ${CLICKHOUSE_DATABASE}.02703_rptable USING x=3 AS permissive TO ALL; SELECT 'R1, R2, R3: (x == 1) OR (x == 2) OR (x == 3)'; SELECT * FROM 02703_rptable; -CREATE ROW POLICY 02703_filter_4 ON ${CLICKHOUSE_DATABASE}.02703_rptable USING x<=2 AS restrictive TO ALL; +CREATE ROW POLICY 02703_filter_policy_${CLICKHOUSE_DATABASE}_4 ON ${CLICKHOUSE_DATABASE}.02703_rptable USING x<=2 AS restrictive TO ALL; SELECT 'R1, R2, R3, R4: ((x == 1) OR (x == 2) OR (x == 3)) AND (x <= 2)'; SELECT * FROM 02703_rptable; -CREATE ROW POLICY 02703_filter_5 ON ${CLICKHOUSE_DATABASE}.* USING y>=20 AS restrictive TO ALL; +CREATE ROW POLICY 02703_filter_policy_${CLICKHOUSE_DATABASE}_5 ON ${CLICKHOUSE_DATABASE}.* USING y>=20 AS restrictive TO ALL; SELECT 'R1, R2, R3, R4, R5: ((x == 1) OR (x == 2) OR (x == 3)) AND (x <= 2) AND (y >= 20)'; SELECT * FROM 02703_rptable; @@ -58,19 +58,19 @@ SELECT * FROM 02703_after_rp; SELECT 'R1, R2, R3, R4, R5: (x == 2) AND (y >= 20) FROM ANOTHER'; SELECT * FROM 02703_rptable_another; -DROP ROW POLICY 02703_filter_1 ON ${CLICKHOUSE_DATABASE}.02703_rptable; +DROP ROW POLICY 02703_filter_policy_${CLICKHOUSE_DATABASE}_1 ON ${CLICKHOUSE_DATABASE}.02703_rptable; SELECT 'R2, R3, R4, R5: ((x == 2) OR (x == 3)) AND (x <= 2) AND (y >= 20)'; SELECT * FROM 02703_rptable; -DROP ROW POLICY 02703_filter_2 ON ${CLICKHOUSE_DATABASE}.*; +DROP ROW POLICY 02703_filter_policy_${CLICKHOUSE_DATABASE}_2 ON ${CLICKHOUSE_DATABASE}.*; SELECT 'R3, R4, R5: (x == 3) AND (x <= 2) AND (y >= 20)'; SELECT * FROM 02703_rptable; -DROP ROW POLICY 02703_filter_3 ON ${CLICKHOUSE_DATABASE}.02703_rptable; +DROP ROW POLICY 02703_filter_policy_${CLICKHOUSE_DATABASE}_3 ON ${CLICKHOUSE_DATABASE}.02703_rptable; SELECT 'R4, R5: (x <= 2) AND (y >= 20)'; SELECT * FROM 02703_rptable; -DROP ROW POLICY 02703_filter_4 ON ${CLICKHOUSE_DATABASE}.02703_rptable; +DROP ROW POLICY 02703_filter_policy_${CLICKHOUSE_DATABASE}_4 ON ${CLICKHOUSE_DATABASE}.02703_rptable; SELECT 'R5: (x >= 2)'; SELECT * FROM 02703_rptable; @@ -78,7 +78,7 @@ CREATE TABLE 02703_unexpected_columns (xx UInt8, yy UInt8) ENGINE = MergeTree OR SELECT 'Policy not applicable'; SELECT * FROM 02703_unexpected_columns; -- { serverError UNKNOWN_IDENTIFIER } -- Missing columns: 'x' while processing query -DROP ROW POLICY 02703_filter_5 ON ${CLICKHOUSE_DATABASE}.*; +DROP ROW POLICY 02703_filter_policy_${CLICKHOUSE_DATABASE}_5 ON ${CLICKHOUSE_DATABASE}.*; SELECT 'None'; SELECT * FROM 02703_rptable; From 5256e8e6d08b5076be49b56fdebadb668892771c Mon Sep 17 00:00:00 2001 From: Nikita Fomichev Date: Fri, 2 Aug 2024 14:36:33 +0200 Subject: [PATCH 136/227] Integration tests: fix ports clashing problem 3 --- tests/integration/conftest.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/integration/conftest.py b/tests/integration/conftest.py index a386ed53009..b4c86a1cd2f 100644 --- a/tests/integration/conftest.py +++ b/tests/integration/conftest.py @@ -132,7 +132,7 @@ def pytest_configure(config): worker_ports = os.getenv("WORKER_FREE_PORTS", None) if worker_ports is None: master_ports = get_unique_free_ports(PORTS_PER_WORKER) - os.environ["WORKER_FREE_PORTS"] = " ".join(([str(p) for p in master_ports])) + os.environ["WORKER_FREE_PORTS"] = " ".join([str(p) for p in master_ports]) def pytest_xdist_setupnodes(config, specs): @@ -148,4 +148,4 @@ def pytest_xdist_setupnodes(config, specs): for i, spec in enumerate(specs): start_range = i * PORTS_PER_WORKER per_workrer_ports = ports[start_range : start_range + PORTS_PER_WORKER] - spec.env["WORKER_FREE_PORTS"] = " ".join(([str(p) for p in per_workrer_ports])) + spec.env["WORKER_FREE_PORTS"] = " ".join([str(p) for p in per_workrer_ports]) From db0bce33526abf16e705b9e56d178d6e2c45a36b Mon Sep 17 00:00:00 2001 From: kssenii Date: Fri, 2 Aug 2024 15:01:15 +0200 Subject: [PATCH 137/227] Try make the code more understandable --- src/Interpreters/Cache/FileCache.cpp | 164 ++++++++++++------------- src/Interpreters/Cache/FileCache.h | 12 +- tests/config/config.d/storage_conf.xml | 2 +- 3 files changed, 80 insertions(+), 98 deletions(-) diff --git a/src/Interpreters/Cache/FileCache.cpp b/src/Interpreters/Cache/FileCache.cpp index 0a03f5dcc7d..4c17afb79be 100644 --- a/src/Interpreters/Cache/FileCache.cpp +++ b/src/Interpreters/Cache/FileCache.cpp @@ -318,7 +318,29 @@ FileSegments FileCache::getImpl(const LockedKey & locked_key, const FileSegment: std::vector FileCache::splitRange(size_t offset, size_t size, size_t aligned_size) { - assert(size > 0); + chassert(size > 0); + chassert(size <= aligned_size); + + /// Consider this example to understand why we need to account here for both `size` and `aligned_size`. + /// [________________]__________________] <-- requested range + /// ^ ^ + /// right offset aligned_right_offset + /// [_________] <-- last cached file segment, e.g. we have uncovered suffix of the requested range + /// [________________] + /// size + /// [____________________________________] + /// aligned_size + /// + /// So it is possible that we split this hole range into sub-segments by `max_file_segment_size` + /// and get something like this: + /// + /// [________________________] + /// ^ ^ + /// right_offset right_offset + max_file_segment_size + /// e.g. there is no need to create sub-segment for range (right_offset + max_file_segment_size, aligned_right_offset]. + /// Because its left offset would be bigger than right_offset. + /// Therefore, we set end_pos_non_included as offset+size, but remaining_size as aligned_size. + std::vector ranges; size_t current_pos = offset; @@ -339,42 +361,23 @@ std::vector FileCache::splitRange(size_t offset, size_t size return ranges; } -FileSegments FileCache::splitRangeIntoFileSegments( +FileSegments FileCache::createFileSegmentsFromRanges( LockedKey & locked_key, - size_t offset, - size_t size, - size_t aligned_size, - FileSegment::State state, + const std::vector & ranges, + size_t & file_segments_count, size_t file_segments_limit, const CreateFileSegmentSettings & create_settings) { - chassert(size > 0); - chassert(size <= aligned_size); - /// We take `size` as a soft limit and `aligned_size` as a hard limit. - - auto current_pos = offset; - auto end_pos_non_included = offset + size; - - size_t current_file_segment_size; - size_t remaining_size = aligned_size; - - FileSegments file_segments; - const size_t max_size = max_file_segment_size.load(); - while (current_pos < end_pos_non_included && (!file_segments_limit || file_segments.size() < file_segments_limit)) + FileSegments result; + for (const auto & r : ranges) { - current_file_segment_size = std::min(remaining_size, max_size); - remaining_size -= current_file_segment_size; - - auto file_segment_metadata_it = addFileSegment( - locked_key, current_pos, current_file_segment_size, state, create_settings, nullptr); - file_segments.push_back(file_segment_metadata_it->second->file_segment); - - current_pos += current_file_segment_size; + if (file_segments_limit && file_segments_count >= file_segments_limit) + break; + auto metadata_it = addFileSegment(locked_key, r.left, r.size(), FileSegment::State::EMPTY, create_settings, nullptr); + result.push_back(metadata_it->second->file_segment); + ++file_segments_count; } - - chassert(file_segments.size() == file_segments_limit || file_segments.back()->range().contains(offset + size - 1), - fmt::format("Offset: {}, size: {}, file segments: {}", offset, size, toString(file_segments))); - return file_segments; + return result; } void FileCache::fillHolesWithEmptyFileSegments( @@ -448,18 +451,9 @@ void FileCache::fillHolesWithEmptyFileSegments( } else { - auto ranges = splitRange(current_pos, hole_size, hole_size); - FileSegments hole; - for (const auto & r : ranges) - { - auto metadata_it = addFileSegment(locked_key, r.left, r.size(), FileSegment::State::EMPTY, create_settings, nullptr); - hole.push_back(metadata_it->second->file_segment); - ++processed_count; - - if (is_limit_reached()) - break; - } - file_segments.splice(it, std::move(hole)); + const auto ranges = splitRange(current_pos, hole_size, hole_size); + auto hole_segments = createFileSegmentsFromRanges(locked_key, ranges, processed_count, file_segments_limit, create_settings); + file_segments.splice(it, std::move(hole_segments)); } if (is_limit_reached()) @@ -493,29 +487,20 @@ void FileCache::fillHolesWithEmptyFileSegments( /// segmentN auto hole_size = range.right - current_pos + 1; - auto non_aligned_size = non_aligned_right_offset - current_pos + 1; + auto non_aligned_hole_size = non_aligned_right_offset - current_pos + 1; if (fill_with_detached_file_segments) { auto file_segment = std::make_shared( - locked_key.getKey(), current_pos, hole_size, FileSegment::State::DETACHED, create_settings); + locked_key.getKey(), current_pos, non_aligned_hole_size, FileSegment::State::DETACHED, create_settings); file_segments.insert(file_segments.end(), file_segment); } else { - auto ranges = splitRange(current_pos, non_aligned_size, hole_size); - FileSegments hole; - for (const auto & r : ranges) - { - auto metadata_it = addFileSegment(locked_key, r.left, r.size(), FileSegment::State::EMPTY, create_settings, nullptr); - hole.push_back(metadata_it->second->file_segment); - ++processed_count; - - if (is_limit_reached()) - break; - } - file_segments.splice(it, std::move(hole)); + const auto ranges = splitRange(current_pos, non_aligned_hole_size, hole_size); + auto hole_segments = createFileSegmentsFromRanges(locked_key, ranges, processed_count, file_segments_limit, create_settings); + file_segments.splice(it, std::move(hole_segments)); if (is_limit_reached()) erase_unprocessed(); @@ -548,8 +533,9 @@ FileSegmentsHolderPtr FileCache::set( } else { - file_segments = splitRangeIntoFileSegments( - *locked_key, offset, size, size, FileSegment::State::EMPTY, /* file_segments_limit */0, create_settings); + const auto ranges = splitRange(offset, size, size); + size_t file_segments_count = 0; + file_segments = createFileSegmentsFromRanges(*locked_key, ranges, file_segments_count, /* file_segments_limit */0, create_settings); } return std::make_unique(std::move(file_segments)); @@ -569,23 +555,27 @@ FileCache::getOrSet( assertInitialized(); - FileSegment::Range range(offset, offset + size - 1); + FileSegment::Range initial_range(offset, offset + size - 1); + /// result_range is initial range, which will be adjusted according to + /// 1. aligned offset, alighed_end_offset + /// 2. max_file_segments_limit + FileSegment::Range result_range = initial_range; - const auto aligned_offset = roundDownToMultiple(range.left, boundary_alignment); - auto aligned_end_offset = std::min(roundUpToMultiple(offset + size, boundary_alignment), file_size) - 1; + const auto aligned_offset = roundDownToMultiple(initial_range.left, boundary_alignment); + auto aligned_end_offset = std::min(roundUpToMultiple(initial_range.right + 1, boundary_alignment), file_size) - 1; - chassert(aligned_offset <= range.left); - chassert(aligned_end_offset >= range.right); + chassert(aligned_offset <= initial_range.left); + chassert(aligned_end_offset >= initial_range.right); auto locked_key = metadata.lockKeyMetadata(key, CacheMetadata::KeyNotFoundPolicy::CREATE_EMPTY, user); /// Get all segments which intersect with the given range. - auto file_segments = getImpl(*locked_key, range, file_segments_limit); + auto file_segments = getImpl(*locked_key, initial_range, file_segments_limit); if (file_segments_limit) { chassert(file_segments.size() <= file_segments_limit); if (file_segments.size() == file_segments_limit) - range.right = aligned_end_offset = file_segments.back()->range().right; + result_range.right = aligned_end_offset = file_segments.back()->range().right; } /// Check case if we have uncovered prefix, e.g. @@ -597,11 +587,11 @@ FileCache::getOrSet( /// [ ] /// ^----^ /// uncovered prefix. - const bool has_uncovered_prefix = file_segments.empty() || range.left < file_segments.front()->range().left; + const bool has_uncovered_prefix = file_segments.empty() || result_range.left < file_segments.front()->range().left; - if (aligned_offset < range.left && has_uncovered_prefix) + if (aligned_offset < result_range.left && has_uncovered_prefix) { - auto prefix_range = FileSegment::Range(aligned_offset, file_segments.empty() ? range.left - 1 : file_segments.front()->range().left - 1); + auto prefix_range = FileSegment::Range(aligned_offset, file_segments.empty() ? result_range.left - 1 : file_segments.front()->range().left - 1); auto prefix_file_segments = getImpl(*locked_key, prefix_range, /* file_segments_limit */0); if (prefix_file_segments.empty()) @@ -610,7 +600,7 @@ FileCache::getOrSet( /// ^ ^ ^ /// aligned_offset range.left range.right /// [___] [__________] <-- current cache (example) - range.left = aligned_offset; + result_range.left = aligned_offset; } else { @@ -621,10 +611,10 @@ FileCache::getOrSet( /// ^ /// prefix_file_segments.back().right - chassert(prefix_file_segments.back()->range().right < range.left); + chassert(prefix_file_segments.back()->range().right < result_range.left); chassert(prefix_file_segments.back()->range().right >= aligned_offset); - range.left = prefix_file_segments.back()->range().right + 1; + result_range.left = prefix_file_segments.back()->range().right + 1; } } @@ -637,11 +627,11 @@ FileCache::getOrSet( /// [___] /// ^---^ /// uncovered_suffix - const bool has_uncovered_suffix = file_segments.empty() || file_segments.back()->range().right < range.right; + const bool has_uncovered_suffix = file_segments.empty() || file_segments.back()->range().right < result_range.right; - if (range.right < aligned_end_offset && has_uncovered_suffix) + if (result_range.right < aligned_end_offset && has_uncovered_suffix) { - auto suffix_range = FileSegment::Range(range.right, aligned_end_offset); + auto suffix_range = FileSegment::Range(result_range.right, aligned_end_offset); /// We need to get 1 file segment, so file_segments_limit = 1 here. auto suffix_file_segments = getImpl(*locked_key, suffix_range, /* file_segments_limit */1); @@ -652,7 +642,7 @@ FileCache::getOrSet( /// range.left range.right aligned_end_offset /// [___] [___] <-- current cache (example) - range.right = aligned_end_offset; + result_range.right = aligned_end_offset; } else { @@ -662,35 +652,33 @@ FileCache::getOrSet( /// [___] [___] [_________] <-- current cache (example) /// ^ /// suffix_file_segments.front().left - range.right = suffix_file_segments.front()->range().left - 1; + result_range.right = suffix_file_segments.front()->range().left - 1; } } - chassert(range.left >= aligned_offset); - if (file_segments.empty()) { - file_segments = splitRangeIntoFileSegments( - *locked_key, range.left, /* size */offset + size - range.left, /* aligned_size */range.size(), - FileSegment::State::EMPTY, file_segments_limit, create_settings); + auto ranges = splitRange(result_range.left, initial_range.size() + (initial_range.left - result_range.left), result_range.size()); + size_t file_segments_count = file_segments.size(); + file_segments.splice(file_segments.end(), createFileSegmentsFromRanges(*locked_key, ranges, file_segments_count, file_segments_limit, create_settings)); } else { - chassert(file_segments.front()->range().right >= range.left); - chassert(file_segments.back()->range().left <= range.right); + chassert(file_segments.front()->range().right >= result_range.left); + chassert(file_segments.back()->range().left <= result_range.right); fillHolesWithEmptyFileSegments( - *locked_key, file_segments, range, offset + size - 1, file_segments_limit, /* fill_with_detached */false, create_settings); + *locked_key, file_segments, result_range, offset + size - 1, file_segments_limit, /* fill_with_detached */false, create_settings); - if (!file_segments.front()->range().contains(range.left)) + if (!file_segments.front()->range().contains(result_range.left)) { throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected {} to include {} " "(end offset: {}, aligned offset: {}, aligned end offset: {})", - file_segments.front()->range().toString(), offset, range.right, aligned_offset, aligned_end_offset); + file_segments.front()->range().toString(), offset, result_range.right, aligned_offset, aligned_end_offset); } } - chassert(file_segments_limit ? file_segments.back()->range().left <= range.right : file_segments.back()->range().contains(range.right)); + chassert(file_segments_limit ? file_segments.back()->range().left <= result_range.right : file_segments.back()->range().contains(result_range.right)); chassert(!file_segments_limit || file_segments.size() <= file_segments_limit); return std::make_unique(std::move(file_segments)); diff --git a/src/Interpreters/Cache/FileCache.h b/src/Interpreters/Cache/FileCache.h index 3f7eec73b56..07be802a940 100644 --- a/src/Interpreters/Cache/FileCache.h +++ b/src/Interpreters/Cache/FileCache.h @@ -265,16 +265,10 @@ private: /// each subrange size must be less or equal to max_file_segment_size. std::vector splitRange(size_t offset, size_t size, size_t aligned_size); - /// Split range into subranges by max_file_segment_size (same as in splitRange()) - /// and create a new file segment for each subrange. - /// If `file_segments_limit` > 0, create no more than first file_segments_limit - /// file segments. - FileSegments splitRangeIntoFileSegments( + FileSegments createFileSegmentsFromRanges( LockedKey & locked_key, - size_t offset, - size_t size, - size_t aligned_size, - FileSegment::State state, + const std::vector & ranges, + size_t & file_segments_count, size_t file_segments_limit, const CreateFileSegmentSettings & create_settings); diff --git a/tests/config/config.d/storage_conf.xml b/tests/config/config.d/storage_conf.xml index 4daa64b520d..e106e3a0e6b 100644 --- a/tests/config/config.d/storage_conf.xml +++ b/tests/config/config.d/storage_conf.xml @@ -19,7 +19,7 @@ cache s3_disk s3_cache/ - 100Mi + 104857600 5Mi 1 100 From 8d979680060f10e6bcec3fc83fb3bdbaa7bb3deb Mon Sep 17 00:00:00 2001 From: Lennard Eijsackers Date: Fri, 2 Aug 2024 15:13:37 +0200 Subject: [PATCH 138/227] Use FunctionArgumentDescriptors to check bitSlice function + add test case Signed-off-by: Lennard Eijsackers --- src/Functions/bitSlice.cpp | 30 ++++++++----------- ...214_bitslice_argument_evaluation.reference | 0 .../03214_bitslice_argument_evaluation.sql | 10 +++++++ 3 files changed, 22 insertions(+), 18 deletions(-) create mode 100644 tests/queries/0_stateless/03214_bitslice_argument_evaluation.reference create mode 100644 tests/queries/0_stateless/03214_bitslice_argument_evaluation.sql diff --git a/src/Functions/bitSlice.cpp b/src/Functions/bitSlice.cpp index e2b455846d8..f1d3bb57221 100644 --- a/src/Functions/bitSlice.cpp +++ b/src/Functions/bitSlice.cpp @@ -40,28 +40,22 @@ public: bool useDefaultImplementationForConstants() const override { return true; } - DataTypePtr getReturnTypeImpl(const DataTypes & arguments) const override + DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override { - const size_t number_of_arguments = arguments.size(); + FunctionArgumentDescriptors mandatory_args{ + {"s", static_cast(&isStringOrFixedString), nullptr, "String"}, + {"offset", static_cast(&isNativeNumber), nullptr, "(U)Int8, (U)Int16, (U)Int32, (U)Int64 or Float"}, + }; - if (number_of_arguments < 2 || number_of_arguments > 3) - throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, - "Number of arguments for function {} doesn't match: passed {}, should be 2 or 3", - getName(), number_of_arguments); + FunctionArgumentDescriptors optional_args{ + {"length", static_cast(&isNativeNumber), nullptr, "(U)Int8, (U)Int16, (U)Int32, (U)Int64 or Float"}, + }; - if (!isString(arguments[0]) && !isStringOrFixedString(arguments[0])) - throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of argument of function {}", - arguments[0]->getName(), getName()); - if (arguments[0]->onlyNull()) - return arguments[0]; + validateFunctionArguments(*this, arguments, mandatory_args, optional_args); - if (!isNativeNumber(arguments[1])) - throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of second argument of function {}", - arguments[1]->getName(), getName()); - - if (number_of_arguments == 3 && !isNativeNumber(arguments[2])) - throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of second argument of function {}", - arguments[2]->getName(), getName()); + const auto & type = arguments[0].type; + if (type->onlyNull()) + return type; return std::make_shared(); } diff --git a/tests/queries/0_stateless/03214_bitslice_argument_evaluation.reference b/tests/queries/0_stateless/03214_bitslice_argument_evaluation.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03214_bitslice_argument_evaluation.sql b/tests/queries/0_stateless/03214_bitslice_argument_evaluation.sql new file mode 100644 index 00000000000..b8488600fcb --- /dev/null +++ b/tests/queries/0_stateless/03214_bitslice_argument_evaluation.sql @@ -0,0 +1,10 @@ +-- No arguments passed +SELECT bitSlice(); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +-- Invalid 1st argument passed +SELECT bitSlice(1, 1); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +-- Valid 1st argument, invalid 2nd argument passed +SELECT bitSlice('Hello', 'World'); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +-- Valid 1st argument & 2nd argument, invalid 3rd argument passed +SELECT bitSlice('Hello', 1, 'World'); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +-- More arguments then expected +SELECT bitSlice('Hello', 1, 1, 'World'); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } From 9c7464e0653782af385dbc884dd3acecfc69c6cc Mon Sep 17 00:00:00 2001 From: Nikita Fomichev Date: Fri, 2 Aug 2024 16:04:11 +0200 Subject: [PATCH 139/227] Stateless tests: reduce pure_http_client timeout to get reasons of timed out tests --- tests/queries/0_stateless/helpers/pure_http_client.py | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tests/queries/0_stateless/helpers/pure_http_client.py b/tests/queries/0_stateless/helpers/pure_http_client.py index 0e7a4d27f4f..7a8efec36bb 100644 --- a/tests/queries/0_stateless/helpers/pure_http_client.py +++ b/tests/queries/0_stateless/helpers/pure_http_client.py @@ -18,7 +18,7 @@ class ClickHouseClient: self.host = host def query( - self, query, connection_timeout=1500, settings=dict(), binary_result=False + self, query, connection_timeout=500, settings=dict(), binary_result=False ): NUMBER_OF_TRIES = 30 DELAY = 10 @@ -47,12 +47,12 @@ class ClickHouseClient: else: raise ValueError(r.text) - def query_return_df(self, query, connection_timeout=1500): + def query_return_df(self, query, connection_timeout=500): data = self.query(query, connection_timeout) df = pd.read_csv(io.StringIO(data), sep="\t") return df - def query_with_data(self, query, data, connection_timeout=1500, settings=dict()): + def query_with_data(self, query, data, connection_timeout=500, settings=dict()): params = { "query": query, "timeout_before_checking_execution_speed": 120, From 2c9cef38e56c65ec9bbe7f3af21d4865662f6e9a Mon Sep 17 00:00:00 2001 From: Nikita Fomichev Date: Fri, 2 Aug 2024 16:05:39 +0200 Subject: [PATCH 140/227] Stateless tests: fix hanging tests `02473_multistep_prewhere*` `00411_long_accurate_number_comparison*` --- ...411_long_accurate_number_comparison.python | 19 ++++++------------- .../02473_multistep_prewhere.python | 4 ++-- .../02473_multistep_split_prewhere.python | 4 ++-- .../0_stateless/helpers/pure_http_client.py | 17 ++++++++++++++++- 4 files changed, 26 insertions(+), 18 deletions(-) diff --git a/tests/queries/0_stateless/00411_long_accurate_number_comparison.python b/tests/queries/0_stateless/00411_long_accurate_number_comparison.python index 045de9ee7ee..38b108a696f 100644 --- a/tests/queries/0_stateless/00411_long_accurate_number_comparison.python +++ b/tests/queries/0_stateless/00411_long_accurate_number_comparison.python @@ -2,23 +2,16 @@ import os, itertools, urllib.request, urllib.parse, urllib.error, urllib.request, urllib.error, urllib.parse, sys +CURDIR = os.path.dirname(os.path.realpath(__file__)) +sys.path.insert(0, os.path.join(CURDIR, "helpers")) -def get_ch_answer(query): - return ( - urllib.request.urlopen( - os.environ.get( - "CLICKHOUSE_URL", - "http://localhost:" + os.environ.get("CLICKHOUSE_PORT_HTTP", "8123"), - ), - data=query.encode(), - ) - .read() - .decode() - ) +from pure_http_client import ClickHouseClient + +client = ClickHouseClient() def check_answers(query, answer): - ch_answer = get_ch_answer(query) + ch_answer = client.query(query) if ch_answer.strip() != answer.strip(): print("FAIL on query:", query) print("Expected answer:", answer) diff --git a/tests/queries/0_stateless/02473_multistep_prewhere.python b/tests/queries/0_stateless/02473_multistep_prewhere.python index 11095202039..09326b6365d 100644 --- a/tests/queries/0_stateless/02473_multistep_prewhere.python +++ b/tests/queries/0_stateless/02473_multistep_prewhere.python @@ -6,7 +6,7 @@ import sys CURDIR = os.path.dirname(os.path.realpath(__file__)) sys.path.insert(0, os.path.join(CURDIR, "helpers")) -from pure_http_client import ClickHouseClient +from pure_http_client import ClickHouseClient, requests_session_with_retries class Tester: @@ -195,7 +195,7 @@ def main(): default_index_granularity = 10 total_rows = 7 * default_index_granularity step = default_index_granularity - session = requests.Session() + session = requests_session_with_retries() for index_granularity in [ default_index_granularity - 1, default_index_granularity, diff --git a/tests/queries/0_stateless/02473_multistep_split_prewhere.python b/tests/queries/0_stateless/02473_multistep_split_prewhere.python index 19444994fd2..10e94059171 100644 --- a/tests/queries/0_stateless/02473_multistep_split_prewhere.python +++ b/tests/queries/0_stateless/02473_multistep_split_prewhere.python @@ -6,7 +6,7 @@ import sys CURDIR = os.path.dirname(os.path.realpath(__file__)) sys.path.insert(0, os.path.join(CURDIR, "helpers")) -from pure_http_client import ClickHouseClient +from pure_http_client import ClickHouseClient, requests_session_with_retries class Tester: @@ -161,7 +161,7 @@ def main(): default_index_granularity = 10 total_rows = 8 * default_index_granularity step = default_index_granularity - session = requests.Session() + session = requests_session_with_retries() for index_granularity in [default_index_granularity - 1, default_index_granularity]: tester = Tester(session, url, index_granularity, total_rows) # Test combinations of ranges of columns c and d diff --git a/tests/queries/0_stateless/helpers/pure_http_client.py b/tests/queries/0_stateless/helpers/pure_http_client.py index 7a8efec36bb..a31a91e0550 100644 --- a/tests/queries/0_stateless/helpers/pure_http_client.py +++ b/tests/queries/0_stateless/helpers/pure_http_client.py @@ -1,7 +1,8 @@ import os import io -import sys import requests +from requests.adapters import HTTPAdapter +from requests.packages.urllib3.util.retry import Retry import time import pandas as pd @@ -77,3 +78,17 @@ class ClickHouseClient: return result else: raise ValueError(r.text) + + +def requests_session_with_retries(retries=3, timeout=180): + session = requests.Session() + retry = Retry( + total=retries, + read=retries, + connect=retries, + ) + adapter = HTTPAdapter(max_retries=retry) + session.mount("http://", adapter) + session.mount("https://", adapter) + session.timeout = timeout + return session From aefed7cdd62e874f7507afe69d803c9164a283ea Mon Sep 17 00:00:00 2001 From: alesapin Date: Fri, 2 Aug 2024 16:06:53 +0200 Subject: [PATCH 141/227] Update tests/queries/0_stateless/03164_s3_settings_for_queries_and_merges.sql --- .../0_stateless/03164_s3_settings_for_queries_and_merges.sql | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/queries/0_stateless/03164_s3_settings_for_queries_and_merges.sql b/tests/queries/0_stateless/03164_s3_settings_for_queries_and_merges.sql index 001ef382850..a6932e0536c 100644 --- a/tests/queries/0_stateless/03164_s3_settings_for_queries_and_merges.sql +++ b/tests/queries/0_stateless/03164_s3_settings_for_queries_and_merges.sql @@ -21,6 +21,7 @@ SYSTEM DROP MARK CACHE; SELECT count() FROM t_compact_bytes_s3 WHERE NOT ignore(c2, c4); SYSTEM FLUSH LOGS; +-- Errors in S3 requests will be automatically retried, however ProfileEvents can be wrong. That is why we subtract errors. SELECT ProfileEvents['S3ReadRequestsCount'] - ProfileEvents['S3ReadRequestsErrors'], ProfileEvents['ReadBufferFromS3Bytes'] < ProfileEvents['ReadCompressedBytes'] * 1.1 From 664e131f4f2e46fc216305c440e840a5a5784328 Mon Sep 17 00:00:00 2001 From: Nikita Fomichev Date: Fri, 2 Aug 2024 16:09:48 +0200 Subject: [PATCH 142/227] Integration tests: fix ports clashing problem 4 --- tests/integration/conftest.py | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tests/integration/conftest.py b/tests/integration/conftest.py index b4c86a1cd2f..aa235118aed 100644 --- a/tests/integration/conftest.py +++ b/tests/integration/conftest.py @@ -1,4 +1,6 @@ #!/usr/bin/env python3 +# pylint: disable=unused-argument +# pylint: disable=broad-exception-raised import logging import os From b3e2ce695514d4d314ed8ac1ecdb111c5f94ac7d Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Wed, 31 Jul 2024 17:38:20 +0200 Subject: [PATCH 143/227] Minor change --- src/Databases/DatabaseOnDisk.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Databases/DatabaseOnDisk.cpp b/src/Databases/DatabaseOnDisk.cpp index f419f5811a1..734f354d9a5 100644 --- a/src/Databases/DatabaseOnDisk.cpp +++ b/src/Databases/DatabaseOnDisk.cpp @@ -313,7 +313,7 @@ void DatabaseOnDisk::detachTablePermanently(ContextPtr query_context, const Stri std::lock_guard lock(mutex); if (const auto it = snapshot_detached_tables.find(table_name); it == snapshot_detached_tables.end()) { - throw Exception(ErrorCodes::LOGICAL_ERROR, "Snapshot doesn't contain info about detached table={}", table_name); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Snapshot doesn't contain info about detached table `{}`", table_name); } else { From 2110b765d19ce4e68d0c23eab219e199aec0ea96 Mon Sep 17 00:00:00 2001 From: Lennard Eijsackers Date: Fri, 2 Aug 2024 17:53:12 +0200 Subject: [PATCH 144/227] Style check fix + adding debug info to query output Signed-off-by: Lennard Eijsackers --- src/Functions/bitSlice.cpp | 2 -- .../03214_bitslice_argument_evaluation.reference | 11 +++++++++++ .../03214_bitslice_argument_evaluation.sql | 1 + 3 files changed, 12 insertions(+), 2 deletions(-) diff --git a/src/Functions/bitSlice.cpp b/src/Functions/bitSlice.cpp index f1d3bb57221..f24473351ae 100644 --- a/src/Functions/bitSlice.cpp +++ b/src/Functions/bitSlice.cpp @@ -18,9 +18,7 @@ using namespace GatherUtils; namespace ErrorCodes { extern const int ILLEGAL_COLUMN; - extern const int ILLEGAL_TYPE_OF_ARGUMENT; extern const int ZERO_ARRAY_OR_TUPLE_INDEX; - extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; } class FunctionBitSlice : public IFunction diff --git a/tests/queries/0_stateless/03214_bitslice_argument_evaluation.reference b/tests/queries/0_stateless/03214_bitslice_argument_evaluation.reference index e69de29bb2d..1731dfa0d79 100644 --- a/tests/queries/0_stateless/03214_bitslice_argument_evaluation.reference +++ b/tests/queries/0_stateless/03214_bitslice_argument_evaluation.reference @@ -0,0 +1,11 @@ +-- { echo } +-- No arguments passed +SELECT bitSlice(); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +-- Invalid 1st argument passed +SELECT bitSlice(1, 1); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +-- Valid 1st argument, invalid 2nd argument passed +SELECT bitSlice('Hello', 'World'); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +-- Valid 1st argument & 2nd argument, invalid 3rd argument passed +SELECT bitSlice('Hello', 1, 'World'); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +-- More arguments then expected +SELECT bitSlice('Hello', 1, 1, 'World'); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } diff --git a/tests/queries/0_stateless/03214_bitslice_argument_evaluation.sql b/tests/queries/0_stateless/03214_bitslice_argument_evaluation.sql index b8488600fcb..1731dfa0d79 100644 --- a/tests/queries/0_stateless/03214_bitslice_argument_evaluation.sql +++ b/tests/queries/0_stateless/03214_bitslice_argument_evaluation.sql @@ -1,3 +1,4 @@ +-- { echo } -- No arguments passed SELECT bitSlice(); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } -- Invalid 1st argument passed From 364622f567028ffc70785b681fc246d7151eef04 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Wed, 31 Jul 2024 17:38:20 +0200 Subject: [PATCH 145/227] Minor change --- src/Databases/DatabaseOnDisk.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Databases/DatabaseOnDisk.cpp b/src/Databases/DatabaseOnDisk.cpp index f419f5811a1..734f354d9a5 100644 --- a/src/Databases/DatabaseOnDisk.cpp +++ b/src/Databases/DatabaseOnDisk.cpp @@ -313,7 +313,7 @@ void DatabaseOnDisk::detachTablePermanently(ContextPtr query_context, const Stri std::lock_guard lock(mutex); if (const auto it = snapshot_detached_tables.find(table_name); it == snapshot_detached_tables.end()) { - throw Exception(ErrorCodes::LOGICAL_ERROR, "Snapshot doesn't contain info about detached table={}", table_name); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Snapshot doesn't contain info about detached table `{}`", table_name); } else { From fcb0ce7361f74dd8d97a3007f77248f293b2ce5f Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Fri, 2 Aug 2024 18:17:32 +0200 Subject: [PATCH 146/227] Fix docs build --- docker/docs/builder/run.sh | 1 - 1 file changed, 1 deletion(-) diff --git a/docker/docs/builder/run.sh b/docker/docs/builder/run.sh index 01c15cb4b0f..d73adb5d279 100755 --- a/docker/docs/builder/run.sh +++ b/docker/docs/builder/run.sh @@ -26,7 +26,6 @@ sed -i '/onBrokenMarkdownLinks:/ s/ignore/error/g' docusaurus.config.js if [[ $# -lt 1 ]] || [[ "$1" == "--"* ]]; then export CI=true - yarn install exec yarn build "$@" fi From 89c47df559ba23d988f8af3c342e0c8d5531f4b8 Mon Sep 17 00:00:00 2001 From: avogar Date: Fri, 2 Aug 2024 16:52:46 +0000 Subject: [PATCH 147/227] Fix variant as common type in if function with Tuples and Maps --- src/Functions/if.cpp | 10 ++++++++++ .../03215_varian_as_common_type_tuple_map.reference | 10 ++++++++++ .../03215_varian_as_common_type_tuple_map.sql | 7 +++++++ 3 files changed, 27 insertions(+) create mode 100644 tests/queries/0_stateless/03215_varian_as_common_type_tuple_map.reference create mode 100644 tests/queries/0_stateless/03215_varian_as_common_type_tuple_map.sql diff --git a/src/Functions/if.cpp b/src/Functions/if.cpp index 07dbee27a9d..64da6e95a43 100644 --- a/src/Functions/if.cpp +++ b/src/Functions/if.cpp @@ -632,6 +632,11 @@ private: ColumnPtr executeTuple(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const { + /// For different Tuples the result type can be Variant with this Tuples if use_variant_as_common_type=1. + /// In this case we should use generic implementation. + if (!isTuple(result_type)) + return nullptr; + /// Calculate function for each corresponding elements of tuples. const ColumnWithTypeAndName & arg1 = arguments[1]; @@ -677,6 +682,11 @@ private: ColumnPtr executeMap(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const { + /// For different Maps the result type can be Variant with this Maps if use_variant_as_common_type=1. + /// In this case we should use generic implementation. + if (!isMap(result_type)) + return nullptr; + auto extract_kv_from_map = [](const ColumnMap * map) { const ColumnTuple & tuple = map->getNestedData(); diff --git a/tests/queries/0_stateless/03215_varian_as_common_type_tuple_map.reference b/tests/queries/0_stateless/03215_varian_as_common_type_tuple_map.reference new file mode 100644 index 00000000000..7dabd5388f4 --- /dev/null +++ b/tests/queries/0_stateless/03215_varian_as_common_type_tuple_map.reference @@ -0,0 +1,10 @@ +('0') Variant(Tuple(String), Tuple(\n number UInt64)) +(1) Variant(Tuple(String), Tuple(\n number UInt64)) +('2') Variant(Tuple(String), Tuple(\n number UInt64)) +(3) Variant(Tuple(String), Tuple(\n number UInt64)) +('4') Variant(Tuple(String), Tuple(\n number UInt64)) +{'0':'0'} Variant(Map(String, String), Map(UInt64, UInt64)) +{1:1} Variant(Map(String, String), Map(UInt64, UInt64)) +{'2':'2'} Variant(Map(String, String), Map(UInt64, UInt64)) +{3:3} Variant(Map(String, String), Map(UInt64, UInt64)) +{'4':'4'} Variant(Map(String, String), Map(UInt64, UInt64)) diff --git a/tests/queries/0_stateless/03215_varian_as_common_type_tuple_map.sql b/tests/queries/0_stateless/03215_varian_as_common_type_tuple_map.sql new file mode 100644 index 00000000000..4a9a788ab18 --- /dev/null +++ b/tests/queries/0_stateless/03215_varian_as_common_type_tuple_map.sql @@ -0,0 +1,7 @@ +set use_variant_as_common_type = 1; +set allow_experimental_variant_type = 1; + +SELECT if(number % 2, tuple(number), tuple(toString(number))) as res, toTypeName(res) FROM numbers(5); +SELECT if(number % 2, map(number, number), map(toString(number), toString(number))) as res, toTypeName(res) FROM numbers(5); + + From b38c46a87d6eacbc7805562deb07ce586fd7e0fb Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Wed, 31 Jul 2024 17:38:20 +0200 Subject: [PATCH 148/227] Minor change --- src/Databases/DatabaseOnDisk.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Databases/DatabaseOnDisk.cpp b/src/Databases/DatabaseOnDisk.cpp index f419f5811a1..734f354d9a5 100644 --- a/src/Databases/DatabaseOnDisk.cpp +++ b/src/Databases/DatabaseOnDisk.cpp @@ -313,7 +313,7 @@ void DatabaseOnDisk::detachTablePermanently(ContextPtr query_context, const Stri std::lock_guard lock(mutex); if (const auto it = snapshot_detached_tables.find(table_name); it == snapshot_detached_tables.end()) { - throw Exception(ErrorCodes::LOGICAL_ERROR, "Snapshot doesn't contain info about detached table={}", table_name); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Snapshot doesn't contain info about detached table `{}`", table_name); } else { From e3239c6ee11eb5bf0466fb750c58125868885ec8 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Fri, 2 Aug 2024 19:09:16 +0200 Subject: [PATCH 149/227] Fix bad log message in JIT for sorting --- src/Core/SortDescription.cpp | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/src/Core/SortDescription.cpp b/src/Core/SortDescription.cpp index 9edc79a1ff1..1b3f81f8547 100644 --- a/src/Core/SortDescription.cpp +++ b/src/Core/SortDescription.cpp @@ -103,7 +103,15 @@ static std::string getSortDescriptionDump(const SortDescription & description, c WriteBufferFromOwnString buffer; for (size_t i = 0; i < description.size(); ++i) - buffer << header_types[i]->getName() << ' ' << description[i].direction << ' ' << description[i].nulls_direction; + { + if (i != 0) + buffer << ", "; + + buffer << "(type: " << header_types[i]->getName() + << ", direction: " << description[i].direction + << ", nulls_direction: " << description[i].nulls_direction + << ")"; + } return buffer.str(); } From 3c4389ec4d78db55ce742e5d5a3b0ed050c9c9e6 Mon Sep 17 00:00:00 2001 From: Denny Crane Date: Fri, 2 Aug 2024 14:57:19 -0300 Subject: [PATCH 150/227] doc/fix max_partitions_to_read description --- .../operations/settings/merge-tree-settings.md | 7 ++----- .../en/operations/settings/query-complexity.md | 18 ++++++++++++++++-- 2 files changed, 18 insertions(+), 7 deletions(-) diff --git a/docs/en/operations/settings/merge-tree-settings.md b/docs/en/operations/settings/merge-tree-settings.md index 7278b91f90d..67fa45c20cd 100644 --- a/docs/en/operations/settings/merge-tree-settings.md +++ b/docs/en/operations/settings/merge-tree-settings.md @@ -119,11 +119,6 @@ Minimum size of blocks of uncompressed data required for compression when writin You can also specify this setting in the global settings (see [min_compress_block_size](/docs/en/operations/settings/settings.md/#min-compress-block-size) setting). The value specified when table is created overrides the global value for this setting. -## max_partitions_to_read - -Limits the maximum number of partitions that can be accessed in one query. -You can also specify setting [max_partitions_to_read](/docs/en/operations/settings/merge-tree-settings.md/#max-partitions-to-read) in the global setting. - ## max_suspicious_broken_parts If the number of broken parts in a single partition exceeds the `max_suspicious_broken_parts` value, automatic deletion is denied. @@ -691,6 +686,8 @@ Possible values: Default value: -1 (unlimited). +You can also specify a query complexity setting [max_partitions_to_read](query-complexity#max-partitions-to-read) at a query / session / profile level. + ## min_age_to_force_merge_seconds {#min_age_to_force_merge_seconds} Merge parts if every part in the range is older than the value of `min_age_to_force_merge_seconds`. diff --git a/docs/en/operations/settings/query-complexity.md b/docs/en/operations/settings/query-complexity.md index 2a20e74e20f..14ccb1167f9 100644 --- a/docs/en/operations/settings/query-complexity.md +++ b/docs/en/operations/settings/query-complexity.md @@ -188,7 +188,7 @@ If you set `timeout_before_checking_execution_speed `to 0, ClickHouse will use c What to do if the query is run longer than `max_execution_time` or the estimated running time is longer than `max_estimated_execution_time`: `throw` or `break`. By default, `throw`. -# max_execution_time_leaf +## max_execution_time_leaf Similar semantic to `max_execution_time` but only apply on leaf node for distributed or remote queries. @@ -204,7 +204,7 @@ We can use `max_execution_time_leaf` as the query settings: SELECT count() FROM cluster(cluster, view(SELECT * FROM t)) SETTINGS max_execution_time_leaf = 10; ``` -# timeout_overflow_mode_leaf +## timeout_overflow_mode_leaf What to do when the query in leaf node run longer than `max_execution_time_leaf`: `throw` or `break`. By default, `throw`. @@ -426,3 +426,17 @@ Example: ``` Default value: 0 (Infinite count of simultaneous sessions). + +## max_partitions_to_read {#max-partitions-to-read} + +Limits the maximum number of partitions that can be accessed in one query. + +The setting value specified when the table is created can be overridden via query-level setting. + +Possible values: + +- Any positive integer. + +Default value: -1 (unlimited). + +You can also specify a MergeTree setting [max_partitions_to_read](merge-tree-settings#max-partitions-to-read) in tables' setting. From ce39957983af8bdd7d97e4a3729e2f97d3e0cb85 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Fri, 2 Aug 2024 20:11:25 +0200 Subject: [PATCH 151/227] Remove capitalization in test reports --- tests/ci/report.py | 10 +++------- 1 file changed, 3 insertions(+), 7 deletions(-) diff --git a/tests/ci/report.py b/tests/ci/report.py index f50ed4c1f85..3f0fc596824 100644 --- a/tests/ci/report.py +++ b/tests/ci/report.py @@ -293,9 +293,9 @@ class JobReport: start_time: str duration: float additional_files: Union[Sequence[str], Sequence[Path]] - # clickhouse version, build job only + # ClickHouse version, build job only version: str = "" - # checkname to set in commit status, set if differs from jjob name + # check_name to be set in commit status, set it if it differs from the job name check_name: str = "" # directory with artifacts to upload on s3 build_dir_for_upload: Union[Path, str] = "" @@ -667,11 +667,7 @@ ColorTheme = Tuple[str, str, str] def _format_header( header: str, branch_name: str, branch_url: Optional[str] = None ) -> str: - # Following line does not lower CI->Ci and SQLancer->Sqlancer. It only - # capitalizes the first letter and doesn't touch the rest of the word - result = " ".join([w[0].upper() + w[1:] for w in header.split(" ") if w]) - result = result.replace("Clickhouse", "ClickHouse") - result = result.replace("clickhouse", "ClickHouse") + result = header if "ClickHouse" not in result: result = f"ClickHouse {result}" if branch_url: From 2c9b61d047c1afe22b0fa0a967a87db8bd4cf62f Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Fri, 2 Aug 2024 20:16:44 +0200 Subject: [PATCH 152/227] Miscellaneous --- tests/ci/ci.py | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/tests/ci/ci.py b/tests/ci/ci.py index 935fe472e50..6ca84a346e2 100644 --- a/tests/ci/ci.py +++ b/tests/ci/ci.py @@ -1019,7 +1019,7 @@ def _get_ext_check_name(check_name: str) -> str: return check_name_with_group -def _cancel_pr_wf(s3: S3Helper, pr_number: int, cancel_sync: bool = False) -> None: +def _cancel_pr_workflow(s3: S3Helper, pr_number: int, cancel_sync: bool = False) -> None: wf_data = CiMetadata(s3, pr_number).fetch_meta() if not cancel_sync: if not wf_data.run_id: @@ -1368,12 +1368,12 @@ def main() -> int: assert indata, "Run config must be provided via --infile" _update_gh_statuses_action(indata=indata, s3=s3) - ### CANCEL PREVIOUS WORKFLOW RUN + ### CANCEL THE PREVIOUS WORKFLOW RUN elif args.cancel_previous_run: if pr_info.is_merge_queue: - _cancel_pr_wf(s3, pr_info.merged_pr) + _cancel_pr_workflow(s3, pr_info.merged_pr) elif pr_info.is_pr: - _cancel_pr_wf(s3, pr_info.number, cancel_sync=True) + _cancel_pr_workflow(s3, pr_info.number, cancel_sync=True) else: assert False, "BUG! Not supported scenario" From bd3606dac4954c673ec6c38dd6fbdb70bc7b53cc Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Fri, 2 Aug 2024 20:17:07 +0200 Subject: [PATCH 153/227] Fix typos --- tests/ci/commit_status_helper.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/ci/commit_status_helper.py b/tests/ci/commit_status_helper.py index fdc9c002b66..908ac4a7dca 100644 --- a/tests/ci/commit_status_helper.py +++ b/tests/ci/commit_status_helper.py @@ -301,7 +301,7 @@ def get_worst_state(statuses: CommitStatuses) -> StatusType: def create_ci_report(pr_info: PRInfo, statuses: CommitStatuses) -> str: - """The function converst the statuses to TestResults and uploads the report + """The function converts the statuses to TestResults and uploads the report to S3 tests bucket. Then it returns the URL""" test_results = [] # type: TestResults for status in statuses: From 675afda17210ca7e8e71e0899a5ed14d7227fb55 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Fri, 2 Aug 2024 20:22:08 +0200 Subject: [PATCH 154/227] Fix check names in the CI Logs database --- tests/ci/clickhouse_helper.py | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/tests/ci/clickhouse_helper.py b/tests/ci/clickhouse_helper.py index 287970cce9a..0725f7100d1 100644 --- a/tests/ci/clickhouse_helper.py +++ b/tests/ci/clickhouse_helper.py @@ -3,6 +3,7 @@ import fileinput import json import logging import time +import os from pathlib import Path from typing import Any, Dict, List, Optional @@ -298,6 +299,11 @@ class CiLogsCredentials: def get_docker_arguments( self, pr_info: PRInfo, check_start_time: str, check_name: str ) -> str: + run_by_hash_total = int(os.getenv("RUN_BY_HASH_TOTAL", "0")) + if run_by_hash_total > 1: + run_by_hash_num = int(os.getenv("RUN_BY_HASH_NUM", "0")) + check_name = f"{check_name} [{run_by_hash_num + 1}/{run_by_hash_total}]" + self.create_ci_logs_credentials() if not self.config_path.exists(): logging.info("Do not use external logs pushing") From a45ba44dbaa2ed43eb63e49fe609a01be978eac9 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Fri, 2 Aug 2024 18:28:38 +0000 Subject: [PATCH 155/227] Automatic style fix --- tests/ci/ci.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tests/ci/ci.py b/tests/ci/ci.py index 6ca84a346e2..805296d2bb2 100644 --- a/tests/ci/ci.py +++ b/tests/ci/ci.py @@ -1019,7 +1019,9 @@ def _get_ext_check_name(check_name: str) -> str: return check_name_with_group -def _cancel_pr_workflow(s3: S3Helper, pr_number: int, cancel_sync: bool = False) -> None: +def _cancel_pr_workflow( + s3: S3Helper, pr_number: int, cancel_sync: bool = False +) -> None: wf_data = CiMetadata(s3, pr_number).fetch_meta() if not cancel_sync: if not wf_data.run_id: From a37eeb0f211c0c7b6251a9108d6b939e73c9a66e Mon Sep 17 00:00:00 2001 From: Michael Kolupaev Date: Fri, 2 Aug 2024 20:13:27 +0000 Subject: [PATCH 156/227] Allow types to change, re-resolve overloads --- src/Storages/MergeTree/KeyCondition.cpp | 19 +++++++++---------- 1 file changed, 9 insertions(+), 10 deletions(-) diff --git a/src/Storages/MergeTree/KeyCondition.cpp b/src/Storages/MergeTree/KeyCondition.cpp index eaf9f0af623..91f054c3a71 100644 --- a/src/Storages/MergeTree/KeyCondition.cpp +++ b/src/Storages/MergeTree/KeyCondition.cpp @@ -627,7 +627,8 @@ static const ActionsDAG::Node & cloneASTWithInversionPushDown( } } - res = &inverted_dag.addFunction(node.function_base, children, ""); + auto function_builder = FunctionFactory::instance().get(name, context); + res = &inverted_dag.addFunction(function_builder, children, ""); handled_inversion = true; } else if (need_inversion && (name == "and" || name == "or")) @@ -668,8 +669,13 @@ static const ActionsDAG::Node & cloneASTWithInversionPushDown( } else { - res = &inverted_dag.addFunction(node.function_base, children, ""); - chassert(res->result_type == node.result_type); + /// Can't just addFunction(node.function_base) because argument types may have + /// changed slightly because of our transformations, e.g. maybe some subexpression + /// changed constness, which caused some function return value to change LowCardinality-ness. + /// (I don't have a specific counterexample, but it seems likely that it exists. + /// One was fixed in the past: https://github.com/ClickHouse/ClickHouse/issues/65143 ) + auto function_builder = FunctionFactory::instance().get(name, context); + res = &inverted_dag.addFunction(function_builder, children, ""); } } } @@ -678,13 +684,6 @@ static const ActionsDAG::Node & cloneASTWithInversionPushDown( if (!handled_inversion && need_inversion) res = &inverted_dag.addFunction(FunctionFactory::instance().get("not", context), {res}, ""); - /// Make sure we don't change any data types (e.g. remove LowCardinality). - /// If it turns out that we actually want to change data types sometimes, it's ok to remove this - /// check *and* replace all `addFunction(node.function_base, ...)` calls above with - /// `addFunction(FunctionFactory::instance().get(name, context), ...)` to re-resolve overloads. - if (!node.result_type->equals(*res->result_type)) - throw Exception(ErrorCodes::LOGICAL_ERROR, "KeyCondition inadvertently changed subexpression data type: '{}' -> '{}', column `{}`", node.result_type->getName(), res->result_type->getName(), node.result_name); - to_inverted[&node] = res; return *res; } From eb0e12099a8fa45d79b2cff96e02ee273e879efa Mon Sep 17 00:00:00 2001 From: Michael Kolupaev Date: Fri, 2 Aug 2024 21:18:35 +0000 Subject: [PATCH 157/227] Another attempt --- src/Storages/MergeTree/KeyCondition.cpp | 27 ++++++++++++++++--------- 1 file changed, 18 insertions(+), 9 deletions(-) diff --git a/src/Storages/MergeTree/KeyCondition.cpp b/src/Storages/MergeTree/KeyCondition.cpp index 91f054c3a71..9115cb5608e 100644 --- a/src/Storages/MergeTree/KeyCondition.cpp +++ b/src/Storages/MergeTree/KeyCondition.cpp @@ -627,8 +627,7 @@ static const ActionsDAG::Node & cloneASTWithInversionPushDown( } } - auto function_builder = FunctionFactory::instance().get(name, context); - res = &inverted_dag.addFunction(function_builder, children, ""); + res = &inverted_dag.addFunction(node.function_base, children, ""); handled_inversion = true; } else if (need_inversion && (name == "and" || name == "or")) @@ -669,13 +668,23 @@ static const ActionsDAG::Node & cloneASTWithInversionPushDown( } else { - /// Can't just addFunction(node.function_base) because argument types may have - /// changed slightly because of our transformations, e.g. maybe some subexpression - /// changed constness, which caused some function return value to change LowCardinality-ness. - /// (I don't have a specific counterexample, but it seems likely that it exists. - /// One was fixed in the past: https://github.com/ClickHouse/ClickHouse/issues/65143 ) - auto function_builder = FunctionFactory::instance().get(name, context); - res = &inverted_dag.addFunction(function_builder, children, ""); + /// Make sure we don't change types of function arguments (e.g. remove LowCardinality). + /// Otherwise the function may crash when passed columns of unexpected types. + /// * Why not check this for all subexperessions rather than function arguments? + /// Because types may change, e.g. in `NOT (u64 AND u64)` -> `(NOT u64 OR NOT u64)` + /// the AND's args were UInt64, but OR's args are UInt8. + /// * Why not re-resolve function overload, using FunctionFactory::instance().get(name, context)? + /// Because some functions can't be found through FunctionFactory, e.g. FunctionCapture. + /// (But maybe we could re-resolve only if argument types changed.) + for (size_t i = 0; i < children.size(); ++i) + { + if (!node.children[i]->result_type->equals(*children[i]->result_type)) + throw Exception( + ErrorCodes::LOGICAL_ERROR, "KeyCondition inadvertently changed subexpression data type: '{}' -> '{}', column `{}`", + node.children[i]->result_type->getName(), children[i]->result_type->getName(), node.children[i]->result_name); + } + + res = &inverted_dag.addFunction(node.function_base, children, ""); } } } From dd0ae04f90314ce6d5dbe748605e66f1a6d9024f Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Wed, 31 Jul 2024 17:38:20 +0200 Subject: [PATCH 158/227] Minor change --- src/Databases/DatabaseOnDisk.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Databases/DatabaseOnDisk.cpp b/src/Databases/DatabaseOnDisk.cpp index f419f5811a1..734f354d9a5 100644 --- a/src/Databases/DatabaseOnDisk.cpp +++ b/src/Databases/DatabaseOnDisk.cpp @@ -313,7 +313,7 @@ void DatabaseOnDisk::detachTablePermanently(ContextPtr query_context, const Stri std::lock_guard lock(mutex); if (const auto it = snapshot_detached_tables.find(table_name); it == snapshot_detached_tables.end()) { - throw Exception(ErrorCodes::LOGICAL_ERROR, "Snapshot doesn't contain info about detached table={}", table_name); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Snapshot doesn't contain info about detached table `{}`", table_name); } else { From a431ab3e4b6f925924a81d99997e6c028ae7950f Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 3 Aug 2024 01:31:02 +0200 Subject: [PATCH 159/227] Improve dashboard --- programs/server/dashboard.html | 66 +++++++++++++++++++++++++++------- 1 file changed, 54 insertions(+), 12 deletions(-) diff --git a/programs/server/dashboard.html b/programs/server/dashboard.html index 45f988f7b1e..71880b9e228 100644 --- a/programs/server/dashboard.html +++ b/programs/server/dashboard.html @@ -17,7 +17,7 @@ --input-shadow-color: rgba(0, 255, 0, 1); --error-color: red; --global-error-color: white; - --legend-background: rgba(255, 255, 255, 0.75); + --legend-background: rgba(255, 255, 0, 0.75); --title-color: #666; --text-color: black; --edit-title-background: #FEE; @@ -41,7 +41,7 @@ --moving-shadow-color: rgba(255, 255, 255, 0.25); --input-shadow-color: rgba(255, 128, 0, 0.25); --error-color: #F66; - --legend-background: rgba(255, 255, 255, 0.25); + --legend-background: rgba(0, 96, 128, 0.75); --title-color: white; --text-color: white; --edit-title-background: #364f69; @@ -1004,14 +1004,14 @@ function legendAsTooltipPlugin({ className, style = { background: "var(--legend- className && legendEl.classList.add(className); uPlot.assign(legendEl.style, { - textAlign: "left", + textAlign: "right", pointerEvents: "none", display: "none", position: "absolute", left: 0, top: 0, - zIndex: 100, - boxShadow: "2px 2px 10px rgba(0,0,0,0.1)", + zIndex: 200, + boxShadow: "2px 2px 10px rgba(0, 0, 0, 0.1)", ...style }); @@ -1051,8 +1051,10 @@ function legendAsTooltipPlugin({ className, style = { background: "var(--legend- function update(u) { let { left, top } = u.cursor; - left -= legendEl.clientWidth / 2; - top -= legendEl.clientHeight / 2; + /// This will make the balloon to the right of the cursor when the cursor is on the left side, and vise-versa, + /// avoiding the borders of the chart. + left -= legendEl.clientWidth * (left / u.width); + top -= legendEl.clientHeight; legendEl.style.transform = "translate(" + left + "px, " + top + "px)"; if (multiline) { @@ -1229,14 +1231,53 @@ async function draw(idx, chart, url_params, query) { let sync = uPlot.sync("sync"); - let axis = { + function formatDateTime(t) { + return (new Date(t * 1000)).toISOString().replace('T', '\n').replace('.000Z', ''); + } + + function formatDateTimes(self, ticks) { + return ticks.map((t, idx) => { + let res = formatDateTime(t); + if (idx == 0 || res.substring(0, 10) != formatDateTime(ticks[idx - 1]).substring(0, 10)) { + return res; + } else { + return res.substring(11); + } + }); + } + + function formatValue(v) { + const a = Math.abs(v); + if (a >= 1000000000000000) { return (v / 1000000000000000) + 'P'; } + if (a >= 1000000000000) { return (v / 1000000000000) + 'T'; } + if (a >= 1000000000) { return (v / 1000000000) + 'G'; } + if (a >= 1000000) { return (v / 1000000) + 'M'; } + if (a >= 1000) { return (v / 1000) + 'K'; } + if (a > 0 && a < 0.001) { return (v * 1000000) + "μ"; } + return v; + } + + let axis_x = { stroke: axes_color, grid: { width: 1 / devicePixelRatio, stroke: grid_color }, - ticks: { width: 1 / devicePixelRatio, stroke: grid_color } + ticks: { width: 1 / devicePixelRatio, stroke: grid_color }, + values: formatDateTimes, + space: 80, + incrs: [1, 5, 10, 15, 30, + 60, 60 * 5, 60 * 10, 60 * 15, 60 * 30, + 3600, 3600 * 2, 3600 * 3, 3600 * 4, 3600 * 6, 3600 * 12, + 3600 * 24], }; - let axes = [axis, axis]; - let series = [{ label: "x" }]; + let axis_y = { + stroke: axes_color, + grid: { width: 1 / devicePixelRatio, stroke: grid_color }, + ticks: { width: 1 / devicePixelRatio, stroke: grid_color }, + values: (self, ticks) => ticks.map(formatValue) + }; + + let axes = [axis_x, axis_y]; + let series = [{ label: "time", value: (self, t) => formatDateTime(t) }]; let data = [reply.data[reply.meta[0].name]]; // Treat every column as series @@ -1254,9 +1295,10 @@ async function draw(idx, chart, url_params, query) { const opts = { width: chart.clientWidth, height: chart.clientHeight, + scales: { x: { time: false } }, /// Because we want to split and format time on our own. axes, series, - padding: [ null, null, null, (Math.round(max_value * 100) / 100).toString().length * 6 - 10 ], + padding: [ null, null, null, 3 ], plugins: [ legendAsTooltipPlugin() ], cursor: { sync: { From a6f9dd4447cbb475cbf77b07de35b40fbcad50b1 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 3 Aug 2024 01:35:10 +0200 Subject: [PATCH 160/227] Improve dashboard --- programs/server/dashboard.html | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/programs/server/dashboard.html b/programs/server/dashboard.html index 71880b9e228..c69acec7858 100644 --- a/programs/server/dashboard.html +++ b/programs/server/dashboard.html @@ -1010,7 +1010,7 @@ function legendAsTooltipPlugin({ className, style = { background: "var(--legend- position: "absolute", left: 0, top: 0, - zIndex: 200, + zIndex: 100, boxShadow: "2px 2px 10px rgba(0, 0, 0, 0.1)", ...style }); From 95659de26573bdb17ab2b5649e6dad96fb75c479 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 3 Aug 2024 01:45:41 +0200 Subject: [PATCH 161/227] Fix invalid detection of an empty result --- programs/server/dashboard.html | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/programs/server/dashboard.html b/programs/server/dashboard.html index c69acec7858..238254f4ef8 100644 --- a/programs/server/dashboard.html +++ b/programs/server/dashboard.html @@ -1141,7 +1141,7 @@ async function draw(idx, chart, url_params, query) { let {reply, error} = await doFetch(query, url_params); if (!error) { - if (reply.rows.length == 0) { + if (reply.rows == 0) { error = "Query returned empty result."; } else if (reply.meta.length < 2) { error = "Query should return at least two columns: unix timestamp and value."; From a99f9bb603f78437fba8d3ebb031c2f41d00cd58 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 3 Aug 2024 01:50:27 +0200 Subject: [PATCH 162/227] Focus on the mass editor --- programs/server/dashboard.html | 1 + 1 file changed, 1 insertion(+) diff --git a/programs/server/dashboard.html b/programs/server/dashboard.html index 238254f4ef8..8fb07d5da3b 100644 --- a/programs/server/dashboard.html +++ b/programs/server/dashboard.html @@ -945,6 +945,7 @@ function showMassEditor() { let editor = document.getElementById('mass-editor-textarea'); editor.value = JSON.stringify({params: params, queries: queries}, null, 2); + editor.focus(); mass_editor_active = true; } From eeb8c1caac9e8e2ba2f3a1a86f5603281e161610 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 3 Aug 2024 02:06:53 +0200 Subject: [PATCH 163/227] Improve margins when there are many parameters --- programs/server/dashboard.html | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/programs/server/dashboard.html b/programs/server/dashboard.html index 8fb07d5da3b..344de779065 100644 --- a/programs/server/dashboard.html +++ b/programs/server/dashboard.html @@ -256,6 +256,7 @@ font-weight: bold; user-select: none; cursor: pointer; + margin-bottom: 1rem; } #run:hover { @@ -309,7 +310,7 @@ color: var(--param-text-color); display: inline-block; box-shadow: 1px 1px 0 var(--shadow-color); - margin-bottom: 1rem; + margin-bottom: 0.5rem; } input:focus { From 090fb59194462324507d75f032aa803303c3e041 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 3 Aug 2024 02:20:58 +0200 Subject: [PATCH 164/227] Automatic field width of chart parameters --- programs/server/dashboard.html | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/programs/server/dashboard.html b/programs/server/dashboard.html index 344de779065..0b099b15536 100644 --- a/programs/server/dashboard.html +++ b/programs/server/dashboard.html @@ -218,6 +218,7 @@ #chart-params .param { width: 6%; + font-family: monospace; } input { @@ -658,6 +659,10 @@ function insertParam(name, value) { param_value.value = value; param_value.spellcheck = false; + let setWidth = e => { e.style.width = (e.value.length + 1) + 'ch' }; + if (value) { setWidth(param_value); } + param_value.addEventListener('input', e => setWidth(e.target)); + param_wrapper.appendChild(param_name); param_wrapper.appendChild(param_value); document.getElementById('chart-params').appendChild(param_wrapper); From 9a017528a4685fc4ed7eec7ba37f9e9804972c3b Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Wed, 31 Jul 2024 17:38:20 +0200 Subject: [PATCH 165/227] Minor change --- src/Databases/DatabaseOnDisk.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Databases/DatabaseOnDisk.cpp b/src/Databases/DatabaseOnDisk.cpp index f419f5811a1..734f354d9a5 100644 --- a/src/Databases/DatabaseOnDisk.cpp +++ b/src/Databases/DatabaseOnDisk.cpp @@ -313,7 +313,7 @@ void DatabaseOnDisk::detachTablePermanently(ContextPtr query_context, const Stri std::lock_guard lock(mutex); if (const auto it = snapshot_detached_tables.find(table_name); it == snapshot_detached_tables.end()) { - throw Exception(ErrorCodes::LOGICAL_ERROR, "Snapshot doesn't contain info about detached table={}", table_name); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Snapshot doesn't contain info about detached table `{}`", table_name); } else { From dfeb1991164bd6c8b0efc8bdcfe9dcd5b8906928 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 3 Aug 2024 03:06:37 +0200 Subject: [PATCH 166/227] Fix locking inside TimerDescriptor --- src/Common/TimerDescriptor.cpp | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/src/Common/TimerDescriptor.cpp b/src/Common/TimerDescriptor.cpp index 9a171ae9487..ce290a1cb31 100644 --- a/src/Common/TimerDescriptor.cpp +++ b/src/Common/TimerDescriptor.cpp @@ -2,6 +2,7 @@ #include #include +#include #include #include @@ -75,10 +76,22 @@ void TimerDescriptor::drain() const /// or since the last successful read(2), then the buffer given to read(2) returns an unsigned 8-byte integer (uint64_t) /// containing the number of expirations that have occurred. /// (The returned value is in host byte order—that is, the native byte order for integers on the host machine.) + + /// Due to a bug in Linux Kernel, reading from timerfd in non-blocking mode can be still blocking. + /// Avoid it with polling. + Epoll epoll; + epoll.add(timer_fd); + epoll_event event; + event.data.fd = -1; + size_t ready_count = epoll.getManyReady(1, &event, 0); + if (!ready_count) + return; + uint64_t buf; while (true) { ssize_t res = ::read(timer_fd, &buf, sizeof(buf)); + if (res < 0) { /// man timerfd_create: From e491c51cdc9a94018fc6918f74232cfff747436a Mon Sep 17 00:00:00 2001 From: Michael Kolupaev Date: Sat, 3 Aug 2024 01:49:53 +0000 Subject: [PATCH 167/227] e --- src/Storages/MergeTree/KeyCondition.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Storages/MergeTree/KeyCondition.cpp b/src/Storages/MergeTree/KeyCondition.cpp index 9115cb5608e..2b89344d3d9 100644 --- a/src/Storages/MergeTree/KeyCondition.cpp +++ b/src/Storages/MergeTree/KeyCondition.cpp @@ -670,7 +670,7 @@ static const ActionsDAG::Node & cloneASTWithInversionPushDown( { /// Make sure we don't change types of function arguments (e.g. remove LowCardinality). /// Otherwise the function may crash when passed columns of unexpected types. - /// * Why not check this for all subexperessions rather than function arguments? + /// * Why not check this for all subexpressions rather than function arguments? /// Because types may change, e.g. in `NOT (u64 AND u64)` -> `(NOT u64 OR NOT u64)` /// the AND's args were UInt64, but OR's args are UInt8. /// * Why not re-resolve function overload, using FunctionFactory::instance().get(name, context)? From 8cb2e308f41638ebb6ba7fddbd4f0bf89d4d612e Mon Sep 17 00:00:00 2001 From: shiyer7474 Date: Sat, 3 Aug 2024 01:55:24 +0000 Subject: [PATCH 168/227] Only new analyzer --- .../03209_parameterized_view_with_non_literal_params.sql | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/03209_parameterized_view_with_non_literal_params.sql b/tests/queries/0_stateless/03209_parameterized_view_with_non_literal_params.sql index f2c61e5cb1d..7fe84929910 100644 --- a/tests/queries/0_stateless/03209_parameterized_view_with_non_literal_params.sql +++ b/tests/queries/0_stateless/03209_parameterized_view_with_non_literal_params.sql @@ -1,4 +1,4 @@ - +SET allow_experimental_analyzer = 1; select 'Test with Date parameter'; drop table if exists date_table_pv; From 96e826d154e9b55cf035a3ce025ac81194455ebc Mon Sep 17 00:00:00 2001 From: Michael Kolupaev Date: Sat, 3 Aug 2024 02:57:33 +0000 Subject: [PATCH 169/227] Change tactics again --- src/Storages/MergeTree/KeyCondition.cpp | 31 +++++++++++++++---------- 1 file changed, 19 insertions(+), 12 deletions(-) diff --git a/src/Storages/MergeTree/KeyCondition.cpp b/src/Storages/MergeTree/KeyCondition.cpp index 2b89344d3d9..dfb43c4e75d 100644 --- a/src/Storages/MergeTree/KeyCondition.cpp +++ b/src/Storages/MergeTree/KeyCondition.cpp @@ -668,23 +668,30 @@ static const ActionsDAG::Node & cloneASTWithInversionPushDown( } else { - /// Make sure we don't change types of function arguments (e.g. remove LowCardinality). - /// Otherwise the function may crash when passed columns of unexpected types. - /// * Why not check this for all subexpressions rather than function arguments? - /// Because types may change, e.g. in `NOT (u64 AND u64)` -> `(NOT u64 OR NOT u64)` - /// the AND's args were UInt64, but OR's args are UInt8. - /// * Why not re-resolve function overload, using FunctionFactory::instance().get(name, context)? - /// Because some functions can't be found through FunctionFactory, e.g. FunctionCapture. - /// (But maybe we could re-resolve only if argument types changed.) + /// Argument types could change slightly because of our transformations, e.g. + /// LowCardinality can be added because some subexpressions became constant + /// (in particular, sets). If that happens, re-run function overload resolver. + /// Otherwise don't re-run it because some functions may not be available + /// through FunctionFactory::get(), e.g. FunctionCapture. + bool types_changed = false; for (size_t i = 0; i < children.size(); ++i) { if (!node.children[i]->result_type->equals(*children[i]->result_type)) - throw Exception( - ErrorCodes::LOGICAL_ERROR, "KeyCondition inadvertently changed subexpression data type: '{}' -> '{}', column `{}`", - node.children[i]->result_type->getName(), children[i]->result_type->getName(), node.children[i]->result_name); + { + types_changed = true; + break; + } } - res = &inverted_dag.addFunction(node.function_base, children, ""); + if (types_changed) + { + auto function_builder = FunctionFactory::instance().get(name, context); + res = &inverted_dag.addFunction(function_builder, children, ""); + } + else + { + res = &inverted_dag.addFunction(node.function_base, children, ""); + } } } } From f97abf69949f8822d70f4b1251e1945f279dd0ec Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Sat, 3 Aug 2024 09:35:37 +0200 Subject: [PATCH 170/227] tests: avoid endless wait in 01042_system_reload_dictionary_reloads_completely Signed-off-by: Azat Khuzhin --- ...em_reload_dictionary_reloads_completely.sh | 21 ++++++++++++++----- 1 file changed, 16 insertions(+), 5 deletions(-) diff --git a/tests/queries/0_stateless/01042_system_reload_dictionary_reloads_completely.sh b/tests/queries/0_stateless/01042_system_reload_dictionary_reloads_completely.sh index 03dd376f802..ebc4110332f 100755 --- a/tests/queries/0_stateless/01042_system_reload_dictionary_reloads_completely.sh +++ b/tests/queries/0_stateless/01042_system_reload_dictionary_reloads_completely.sh @@ -8,6 +8,18 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) set -e -o pipefail +# Wait when the dictionary will update the value for 13 on its own: +function wait_for_dict_upate() +{ + for ((i = 0; i < 100; ++i)); do + if [ "$(${CLICKHOUSE_CLIENT} --query "SELECT dictGetInt64('${CLICKHOUSE_DATABASE}.dict', 'y', toUInt64(13))")" != -1 ]; then + return 0 + fi + sleep 0.5 + done + return 1 +} + $CLICKHOUSE_CLIENT < ', dictGetInt64('${CLICKHOUSE_DATABASE $CLICKHOUSE_CLIENT --query "INSERT INTO ${CLICKHOUSE_DATABASE}.table VALUES (13, 103, now())" $CLICKHOUSE_CLIENT --query "INSERT INTO ${CLICKHOUSE_DATABASE}.table VALUES (14, 104, now() - INTERVAL 1 DAY)" -# Wait when the dictionary will update the value for 13 on its own: -while [ "$(${CLICKHOUSE_CLIENT} --query "SELECT dictGetInt64('${CLICKHOUSE_DATABASE}.dict', 'y', toUInt64(13))")" = -1 ] -do - sleep 0.5 -done +if ! wait_for_dict_upate; then + echo "Dictionary had not been reloaded" >&2 + exit 1 +fi $CLICKHOUSE_CLIENT --query "SELECT '13 -> ', dictGetInt64('${CLICKHOUSE_DATABASE}.dict', 'y', toUInt64(13))" From 40cd5467c18d65a6624d273ac1a8fd9cc9257d8c Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Sat, 3 Aug 2024 09:39:33 +0200 Subject: [PATCH 171/227] tests: fix 01042_system_reload_dictionary_reloads_completely flakiness (increase lag) The test fails in case of INSERT takes > 1 sec: 2024.08.02 13:06:07.746869 [ 45445 ] {c9b55378-6bc5-46d5-80c1-5385a880f88b} executeQuery: (from [::1]:37208) (comment: 01042_system_reload_dictionary_reloads_completely.sh) CREATE DICTIONARY test_m4lx2bit.dict ( x Int64 DEFAULT -1, y Int64 DEFAULT -1, insert_time DateTime ) PRIMARY KEY x SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'table' DB 'test_m4lx2bit' UPDATE_FIELD 'insert_time')) LAYOUT(FLAT()) LIFETIME(1); (stage: Complete) ... 2024.08.02 13:06:08.263832 [ 59987 ] {744da223-67b9-4e32-b12a-eb2550a92fdb} DictionaryFactory: Created dictionary source 'ClickHouse: test_m4lx2bit.table' for dictionary '5b2b98a9-9372-47c9-bda3-830794cb96e7' 2024.08.02 13:06:08.268118 [ 59987 ] {744da223-67b9-4e32-b12a-eb2550a92fdb} executeQuery: (internal) SELECT `x`, `y`, `insert_time` FROM `test_m4lx2bit`.`table`; (stage: Complete) ... 2024.08.02 13:06:09.193190 [ 45445 ] {b6033498-4666-452f-bcf9-02ecf257ba7f} executeQuery: (from [::1]:37262) (comment: 01042_system_reload_dictionary_reloads_completely.sh) INSERT INTO test_m4lx2bit.table VALUES (stage: Complete) ... 2024.08.02 13:06:11.342119 [ 50962 ] {} executeQuery: (internal) SELECT `x`, `y`, `insert_time` FROM `test_m4lx2bit`.`table` WHERE insert_time >= '2024-08-02 13:06:07'; (stage: Complete) ... 2024.08.02 13:06:11.832158 [ 45445 ] {b6033498-4666-452f-bcf9-02ecf257ba7f} TCPHandler: Processed in 2.642106236 sec. ... 2024.08.02 13:06:16.357448 [ 41632 ] {} executeQuery: (internal) SELECT `x`, `y`, `insert_time` FROM `test_m4lx2bit`.`table` WHERE insert_time >= '2024-08-02 13:06:10'; (stage: Complete) Signed-off-by: Azat Khuzhin --- .../01042_system_reload_dictionary_reloads_completely.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/01042_system_reload_dictionary_reloads_completely.sh b/tests/queries/0_stateless/01042_system_reload_dictionary_reloads_completely.sh index ebc4110332f..453e1bb8f0a 100755 --- a/tests/queries/0_stateless/01042_system_reload_dictionary_reloads_completely.sh +++ b/tests/queries/0_stateless/01042_system_reload_dictionary_reloads_completely.sh @@ -31,7 +31,7 @@ CREATE DICTIONARY ${CLICKHOUSE_DATABASE}.dict insert_time DateTime ) PRIMARY KEY x -SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'table' DB '${CLICKHOUSE_DATABASE}' UPDATE_FIELD 'insert_time')) +SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'table' DB '${CLICKHOUSE_DATABASE}' UPDATE_FIELD 'insert_time' UPDATE_LAG 60)) LAYOUT(FLAT()) LIFETIME(1); EOF From 6ce6af0647590f4b58a6ab87ee5f29b8487e8c2f Mon Sep 17 00:00:00 2001 From: Vitaly Baranov Date: Sat, 3 Aug 2024 14:16:24 +0200 Subject: [PATCH 172/227] Fix completion RESTORE ON CLUSTER. --- src/Backups/RestorerFromBackup.cpp | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-) diff --git a/src/Backups/RestorerFromBackup.cpp b/src/Backups/RestorerFromBackup.cpp index 3056f9fe421..278af9d4eb3 100644 --- a/src/Backups/RestorerFromBackup.cpp +++ b/src/Backups/RestorerFromBackup.cpp @@ -222,10 +222,19 @@ void RestorerFromBackup::setStage(const String & new_stage, const String & messa if (restore_coordination) { restore_coordination->setStage(new_stage, message); - if (new_stage == Stage::FINDING_TABLES_IN_BACKUP) - restore_coordination->waitForStage(new_stage, on_cluster_first_sync_timeout); - else - restore_coordination->waitForStage(new_stage); + + /// The initiator of a RESTORE ON CLUSTER query waits for other hosts to complete their work (see waitForStage(Stage::COMPLETED) in BackupsWorker::doRestore), + /// but other hosts shouldn't wait for each others' completion. (That's simply unnecessary and also + /// the initiator may start cleaning up (e.g. removing restore-coordination ZooKeeper nodes) once all other hosts are in Stage::COMPLETED.) + bool need_wait = (new_stage != Stage::COMPLETED); + + if (need_wait) + { + if (new_stage == Stage::FINDING_TABLES_IN_BACKUP) + restore_coordination->waitForStage(new_stage, on_cluster_first_sync_timeout); + else + restore_coordination->waitForStage(new_stage); + } } } From 73080d25a93aa715c4126ed82b89e3ada89d7a3f Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 3 Aug 2024 15:10:05 +0200 Subject: [PATCH 173/227] Fix test `00002_log_and_exception_messages_formatting` --- src/Databases/DatabaseHDFS.cpp | 4 ++-- src/Storages/ObjectStorage/HDFS/Configuration.cpp | 6 +++--- .../00002_log_and_exception_messages_formatting.sql | 5 ++++- 3 files changed, 9 insertions(+), 6 deletions(-) diff --git a/src/Databases/DatabaseHDFS.cpp b/src/Databases/DatabaseHDFS.cpp index eccaae5f22e..f58f1b76e71 100644 --- a/src/Databases/DatabaseHDFS.cpp +++ b/src/Databases/DatabaseHDFS.cpp @@ -75,8 +75,8 @@ std::string DatabaseHDFS::getTablePath(const std::string & table_name) const return table_name; if (source.empty()) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Bad hdfs url: {}. " - "It should have structure 'hdfs://:/path'", table_name); + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Bad HDFS URL: {}. " + "It should have the following structure 'hdfs://:/path'", table_name); return fs::path(source) / table_name; } diff --git a/src/Storages/ObjectStorage/HDFS/Configuration.cpp b/src/Storages/ObjectStorage/HDFS/Configuration.cpp index e8071be6f02..85eb29a3868 100644 --- a/src/Storages/ObjectStorage/HDFS/Configuration.cpp +++ b/src/Storages/ObjectStorage/HDFS/Configuration.cpp @@ -142,11 +142,11 @@ void StorageHDFSConfiguration::setURL(const std::string & url_) { auto pos = url_.find("//"); if (pos == std::string::npos) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Bad hdfs url: {}", url_); + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Bad HDFS URL: {}. It should have the following structure 'hdfs://:/path'", url_); pos = url_.find('/', pos + 2); if (pos == std::string::npos) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Bad hdfs url: {}", url_); + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Bad HDFS URL: {}. It should have the following structure 'hdfs://:/path'", url_); path = url_.substr(pos + 1); if (!path.starts_with('/')) @@ -155,7 +155,7 @@ void StorageHDFSConfiguration::setURL(const std::string & url_) url = url_.substr(0, pos); paths = {path}; - LOG_TRACE(getLogger("StorageHDFSConfiguration"), "Using url: {}, path: {}", url, path); + LOG_TRACE(getLogger("StorageHDFSConfiguration"), "Using URL: {}, path: {}", url, path); } void StorageHDFSConfiguration::addStructureAndFormatToArgs( diff --git a/tests/queries/0_stateless/00002_log_and_exception_messages_formatting.sql b/tests/queries/0_stateless/00002_log_and_exception_messages_formatting.sql index 07c42d6d039..e916fdfc1ff 100644 --- a/tests/queries/0_stateless/00002_log_and_exception_messages_formatting.sql +++ b/tests/queries/0_stateless/00002_log_and_exception_messages_formatting.sql @@ -163,7 +163,10 @@ create temporary table known_short_messages (s String) as select * from (select '{} -> {}', '{} {}', '{}%', - '{}: {}' + '{}: {}', + 'Unknown data type family: {}', + 'Cannot load time zone {}', + 'Unknown table engine {}' ] as arr) array join arr; -- Check that we don't have too many short meaningless message patterns. From a749223251b7e580f5d7bbcb4fc59aa6b5fffbe2 Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Sat, 3 Aug 2024 22:43:12 +0800 Subject: [PATCH 174/227] change as request --- src/Functions/printf.cpp | 39 ++++++++++++++----- .../0_stateless/03203_function_printf.sql | 7 +++- 2 files changed, 36 insertions(+), 10 deletions(-) diff --git a/src/Functions/printf.cpp b/src/Functions/printf.cpp index 3efe854a53b..3cf3efaf534 100644 --- a/src/Functions/printf.cpp +++ b/src/Functions/printf.cpp @@ -6,11 +6,10 @@ #include #include #include +#include #include #include -#include -#include #include #include #include @@ -22,6 +21,7 @@ namespace ErrorCodes extern const int ILLEGAL_COLUMN; extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; extern const int ILLEGAL_TYPE_OF_ARGUMENT; +extern const int BAD_ARGUMENTS; } namespace @@ -52,10 +52,9 @@ private: [[maybe_unused]] String toString() const { - std::ostringstream oss; // STYLE_CHECK_ALLOW_STD_STRING_STREAM - oss << "format:" << format << ", rows:" << rows << ", is_literal:" << is_literal << ", input:" << input.dumpStructure() - << std::endl; - return oss.str(); + WriteBufferFromOwnString buf; + buf << "format:" << format << ", rows:" << rows << ", is_literal:" << is_literal << ", input:" << input.dumpStructure() << "\n"; + return buf.str(); } private: @@ -229,9 +228,31 @@ public: ColumnsWithTypeAndName concat_args(instructions.size()); for (size_t i = 0; i < instructions.size(); ++i) { - // std::cout << "instruction[" << i << "]:" << instructions[i].toString() << std::endl; - concat_args[i] = instructions[i].execute(); - // std::cout << "concat_args[" << i << "]:" << concat_args[i].dumpStructure() << std::endl; + const auto & instruction = instructions[i]; + try + { + // std::cout << "instruction[" << i << "]:" << instructions[i].toString() << std::endl; + concat_args[i] = instruction.execute(); + // std::cout << "concat_args[" << i << "]:" << concat_args[i].dumpStructure() << std::endl; + } + catch (const fmt::v9::format_error & e) + { + if (instruction.is_literal) + throw Exception( + ErrorCodes::BAD_ARGUMENTS, + "Bad format {} in function {} without input argument, reason: {}", + instruction.format, + getName(), + e.what()); + else + throw Exception( + ErrorCodes::BAD_ARGUMENTS, + "Bad format {} in function {} with {} as input argument, reason: {}", + instructions[i].format, + getName(), + instruction.input.dumpStructure(), + e.what()); + } } auto res = function_concat->build(concat_args)->execute(concat_args, std::make_shared(), input_rows_count); diff --git a/tests/queries/0_stateless/03203_function_printf.sql b/tests/queries/0_stateless/03203_function_printf.sql index c41cbf0b5e9..6ff4699c8a7 100644 --- a/tests/queries/0_stateless/03203_function_printf.sql +++ b/tests/queries/0_stateless/03203_function_printf.sql @@ -31,4 +31,9 @@ select printf('%%.2e: %.2e', 123.456) = '%.2e: 1.23e+02'; select printf('%%.2g: %.2g', 123.456) = '%.2g: 1.2e+02'; -- Testing character formats with precision -select printf('%%.2s: %.2s', 'abc') = '%.2s: ab'; \ No newline at end of file +select printf('%%.2s: %.2s', 'abc') = '%.2s: ab'; + +select printf('%%X: %X', 123.123); -- { serverError BAD_ARGUMENTS } +select printf('%%A: %A', 'abc'); -- { serverError BAD_ARGUMENTS } +select printf('%%s: %s', 100); -- { serverError BAD_ARGUMENTS } +select printf('%%n: %n', 100); -- { serverError BAD_ARGUMENTS } From fc651cc0c61feb37e9cf104612cc0ac0cd7448e9 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 3 Aug 2024 17:57:48 +0200 Subject: [PATCH 175/227] Fix strange code in HostResolvePool --- src/Common/HostResolvePool.cpp | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) diff --git a/src/Common/HostResolvePool.cpp b/src/Common/HostResolvePool.cpp index cad64ee7204..e8a05a269bc 100644 --- a/src/Common/HostResolvePool.cpp +++ b/src/Common/HostResolvePool.cpp @@ -253,18 +253,18 @@ void HostResolver::updateImpl(Poco::Timestamp now, std::vector Date: Sat, 3 Aug 2024 18:30:33 +0200 Subject: [PATCH 176/227] Fix typo --- programs/server/Server.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/programs/server/Server.cpp b/programs/server/Server.cpp index 20db4c2773c..7800ee9ff00 100644 --- a/programs/server/Server.cpp +++ b/programs/server/Server.cpp @@ -849,7 +849,7 @@ try #endif #if defined(SANITIZER) - LOG_INFO(log, "Query Profiler disabled because they cannot work under sanitizers" + LOG_INFO(log, "Query Profiler is disabled because it cannot work under sanitizers" " when two different stack unwinding methods will interfere with each other."); #endif From 60648e5240fecb92344ff029d2b280f542c3a86e Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 3 Aug 2024 18:41:17 +0200 Subject: [PATCH 177/227] Revert "Add replication lag and recovery time metrics" --- src/Databases/DatabaseReplicated.cpp | 57 ++++------------- src/Databases/DatabaseReplicated.h | 10 +-- src/Databases/DatabaseReplicatedWorker.cpp | 21 ------- src/Databases/DatabaseReplicatedWorker.h | 5 -- src/Storages/System/StorageSystemClusters.cpp | 37 ++++------- src/Storages/System/StorageSystemClusters.h | 4 +- .../test_recovery_time_metric/__init__.py | 0 .../configs/config.xml | 41 ------------- .../test_recovery_time_metric/test.py | 61 ------------------- .../02117_show_create_table_system.reference | 2 - .../03206_replication_lag_metric.reference | 4 -- .../03206_replication_lag_metric.sql | 11 ---- 12 files changed, 27 insertions(+), 226 deletions(-) delete mode 100644 tests/integration/test_recovery_time_metric/__init__.py delete mode 100644 tests/integration/test_recovery_time_metric/configs/config.xml delete mode 100644 tests/integration/test_recovery_time_metric/test.py delete mode 100644 tests/queries/0_stateless/03206_replication_lag_metric.reference delete mode 100644 tests/queries/0_stateless/03206_replication_lag_metric.sql diff --git a/src/Databases/DatabaseReplicated.cpp b/src/Databases/DatabaseReplicated.cpp index d2dee9b5994..f127ccbc224 100644 --- a/src/Databases/DatabaseReplicated.cpp +++ b/src/Databases/DatabaseReplicated.cpp @@ -12,7 +12,6 @@ #include #include #include -#include #include #include #include @@ -339,12 +338,9 @@ ClusterPtr DatabaseReplicated::getClusterImpl(bool all_groups) const return std::make_shared(getContext()->getSettingsRef(), shards, params); } -ReplicasInfo DatabaseReplicated::tryGetReplicasInfo(const ClusterPtr & cluster_) const +std::vector DatabaseReplicated::tryGetAreReplicasActive(const ClusterPtr & cluster_) const { - Strings paths_get, paths_exists; - - paths_get.emplace_back(fs::path(zookeeper_path) / "max_log_ptr"); - + Strings paths; const auto & addresses_with_failover = cluster_->getShardsAddresses(); const auto & shards_info = cluster_->getShardsInfo(); for (size_t shard_index = 0; shard_index < shards_info.size(); ++shard_index) @@ -352,59 +348,32 @@ ReplicasInfo DatabaseReplicated::tryGetReplicasInfo(const ClusterPtr & cluster_) for (const auto & replica : addresses_with_failover[shard_index]) { String full_name = getFullReplicaName(replica.database_shard_name, replica.database_replica_name); - paths_exists.emplace_back(fs::path(zookeeper_path) / "replicas" / full_name / "active"); - paths_get.emplace_back(fs::path(zookeeper_path) / "replicas" / full_name / "log_ptr"); + paths.emplace_back(fs::path(zookeeper_path) / "replicas" / full_name / "active"); } } try { auto current_zookeeper = getZooKeeper(); - auto get_res = current_zookeeper->get(paths_get); - auto exist_res = current_zookeeper->exists(paths_exists); - chassert(get_res.size() == exist_res.size() + 1); + auto res = current_zookeeper->exists(paths); - auto max_log_ptr_zk = get_res[0]; - if (max_log_ptr_zk.error != Coordination::Error::ZOK) - throw Coordination::Exception(max_log_ptr_zk.error); + std::vector statuses; + statuses.resize(paths.size()); - UInt32 max_log_ptr = parse(max_log_ptr_zk.data); + for (size_t i = 0; i < res.size(); ++i) + if (res[i].error == Coordination::Error::ZOK) + statuses[i] = 1; - ReplicasInfo replicas_info; - replicas_info.resize(exist_res.size()); - - size_t global_replica_index = 0; - for (size_t shard_index = 0; shard_index < shards_info.size(); ++shard_index) - { - for (const auto & replica : addresses_with_failover[shard_index]) - { - auto replica_active = exist_res[global_replica_index]; - auto replica_log_ptr = get_res[global_replica_index + 1]; - - if (replica_active.error != Coordination::Error::ZOK && replica_active.error != Coordination::Error::ZNONODE) - throw Coordination::Exception(replica_active.error); - - if (replica_log_ptr.error != Coordination::Error::ZOK) - throw Coordination::Exception(replica_log_ptr.error); - - replicas_info[global_replica_index] = ReplicaInfo{ - .is_active = replica_active.error == Coordination::Error::ZOK, - .replication_lag = max_log_ptr - parse(replica_log_ptr.data), - .recovery_time = replica.is_local ? ddl_worker->getCurrentInitializationDurationMs() : 0, - }; - - ++global_replica_index; - } - } - - return replicas_info; - } catch (...) + return statuses; + } + catch (...) { tryLogCurrentException(log); return {}; } } + void DatabaseReplicated::fillClusterAuthInfo(String collection_name, const Poco::Util::AbstractConfiguration & config_ref) { const auto & config_prefix = fmt::format("named_collections.{}", collection_name); diff --git a/src/Databases/DatabaseReplicated.h b/src/Databases/DatabaseReplicated.h index 5a1570ae2e2..27ab262d1f1 100644 --- a/src/Databases/DatabaseReplicated.h +++ b/src/Databases/DatabaseReplicated.h @@ -17,14 +17,6 @@ using ZooKeeperPtr = std::shared_ptr; class Cluster; using ClusterPtr = std::shared_ptr; -struct ReplicaInfo -{ - bool is_active; - UInt32 replication_lag; - UInt64 recovery_time; -}; -using ReplicasInfo = std::vector; - class DatabaseReplicated : public DatabaseAtomic { public: @@ -92,7 +84,7 @@ public: static void dropReplica(DatabaseReplicated * database, const String & database_zookeeper_path, const String & shard, const String & replica, bool throw_if_noop); - ReplicasInfo tryGetReplicasInfo(const ClusterPtr & cluster_) const; + std::vector tryGetAreReplicasActive(const ClusterPtr & cluster_) const; void renameDatabase(ContextPtr query_context, const String & new_name) override; diff --git a/src/Databases/DatabaseReplicatedWorker.cpp b/src/Databases/DatabaseReplicatedWorker.cpp index 4e7408aa96e..1ef88dc03bc 100644 --- a/src/Databases/DatabaseReplicatedWorker.cpp +++ b/src/Databases/DatabaseReplicatedWorker.cpp @@ -32,12 +32,6 @@ DatabaseReplicatedDDLWorker::DatabaseReplicatedDDLWorker(DatabaseReplicated * db bool DatabaseReplicatedDDLWorker::initializeMainThread() { - { - std::lock_guard lock(initialization_duration_timer_mutex); - initialization_duration_timer.emplace(); - initialization_duration_timer->start(); - } - while (!stop_flag) { try @@ -75,10 +69,6 @@ bool DatabaseReplicatedDDLWorker::initializeMainThread() initializeReplication(); initialized = true; - { - std::lock_guard lock(initialization_duration_timer_mutex); - initialization_duration_timer.reset(); - } return true; } catch (...) @@ -88,11 +78,6 @@ bool DatabaseReplicatedDDLWorker::initializeMainThread() } } - { - std::lock_guard lock(initialization_duration_timer_mutex); - initialization_duration_timer.reset(); - } - return false; } @@ -474,10 +459,4 @@ UInt32 DatabaseReplicatedDDLWorker::getLogPointer() const return max_id.load(); } -UInt64 DatabaseReplicatedDDLWorker::getCurrentInitializationDurationMs() const -{ - std::lock_guard lock(initialization_duration_timer_mutex); - return initialization_duration_timer ? initialization_duration_timer->elapsedMilliseconds() : 0; -} - } diff --git a/src/Databases/DatabaseReplicatedWorker.h b/src/Databases/DatabaseReplicatedWorker.h index 2309c831839..41edf2221b8 100644 --- a/src/Databases/DatabaseReplicatedWorker.h +++ b/src/Databases/DatabaseReplicatedWorker.h @@ -36,8 +36,6 @@ public: DatabaseReplicated * const database, bool committed = false); /// NOLINT UInt32 getLogPointer() const; - - UInt64 getCurrentInitializationDurationMs() const; private: bool initializeMainThread() override; void initializeReplication(); @@ -58,9 +56,6 @@ private: ZooKeeperPtr active_node_holder_zookeeper; /// It will remove "active" node when database is detached zkutil::EphemeralNodeHolderPtr active_node_holder; - - std::optional initialization_duration_timer; - mutable std::mutex initialization_duration_timer_mutex; }; } diff --git a/src/Storages/System/StorageSystemClusters.cpp b/src/Storages/System/StorageSystemClusters.cpp index d03b600b6ef..160c8d6270e 100644 --- a/src/Storages/System/StorageSystemClusters.cpp +++ b/src/Storages/System/StorageSystemClusters.cpp @@ -31,8 +31,6 @@ ColumnsDescription StorageSystemClusters::getColumnsDescription() {"database_shard_name", std::make_shared(), "The name of the `Replicated` database shard (for clusters that belong to a `Replicated` database)."}, {"database_replica_name", std::make_shared(), "The name of the `Replicated` database replica (for clusters that belong to a `Replicated` database)."}, {"is_active", std::make_shared(std::make_shared()), "The status of the Replicated database replica (for clusters that belong to a Replicated database): 1 means 'replica is online', 0 means 'replica is offline', NULL means 'unknown'."}, - {"replication_lag", std::make_shared(std::make_shared()), "The replication lag of the `Replicated` database replica (for clusters that belong to a Replicated database)."}, - {"recovery_time", std::make_shared(std::make_shared()), "The recovery time of the `Replicated` database replica (for clusters that belong to a Replicated database), in milliseconds."}, }; description.setAliases({ @@ -48,30 +46,31 @@ void StorageSystemClusters::fillData(MutableColumns & res_columns, ContextPtr co writeCluster(res_columns, name_and_cluster, {}); const auto databases = DatabaseCatalog::instance().getDatabases(); - for (const auto & [database_name, database] : databases) + for (const auto & name_and_database : databases) { - if (const auto * replicated = typeid_cast(database.get())) + if (const auto * replicated = typeid_cast(name_and_database.second.get())) { + if (auto database_cluster = replicated->tryGetCluster()) - writeCluster(res_columns, {database_name, database_cluster}, - replicated->tryGetReplicasInfo(database_cluster)); + writeCluster(res_columns, {name_and_database.first, database_cluster}, + replicated->tryGetAreReplicasActive(database_cluster)); if (auto database_cluster = replicated->tryGetAllGroupsCluster()) - writeCluster(res_columns, {DatabaseReplicated::ALL_GROUPS_CLUSTER_PREFIX + database_name, database_cluster}, - replicated->tryGetReplicasInfo(database_cluster)); + writeCluster(res_columns, {DatabaseReplicated::ALL_GROUPS_CLUSTER_PREFIX + name_and_database.first, database_cluster}, + replicated->tryGetAreReplicasActive(database_cluster)); } } } void StorageSystemClusters::writeCluster(MutableColumns & res_columns, const NameAndCluster & name_and_cluster, - const ReplicasInfo & replicas_info) + const std::vector & is_active) { const String & cluster_name = name_and_cluster.first; const ClusterPtr & cluster = name_and_cluster.second; const auto & shards_info = cluster->getShardsInfo(); const auto & addresses_with_failover = cluster->getShardsAddresses(); - size_t global_replica_idx = 0; + size_t replica_idx = 0; for (size_t shard_index = 0; shard_index < shards_info.size(); ++shard_index) { const auto & shard_info = shards_info[shard_index]; @@ -100,24 +99,10 @@ void StorageSystemClusters::writeCluster(MutableColumns & res_columns, const Nam res_columns[i++]->insert(pool_status[replica_index].estimated_recovery_time.count()); res_columns[i++]->insert(address.database_shard_name); res_columns[i++]->insert(address.database_replica_name); - if (replicas_info.empty()) - { + if (is_active.empty()) res_columns[i++]->insertDefault(); - res_columns[i++]->insertDefault(); - res_columns[i++]->insertDefault(); - } else - { - const auto & replica_info = replicas_info[global_replica_idx]; - res_columns[i++]->insert(replica_info.is_active); - res_columns[i++]->insert(replica_info.replication_lag); - if (replica_info.recovery_time != 0) - res_columns[i++]->insert(replica_info.recovery_time); - else - res_columns[i++]->insertDefault(); - } - - ++global_replica_idx; + res_columns[i++]->insert(is_active[replica_idx++]); } } } diff --git a/src/Storages/System/StorageSystemClusters.h b/src/Storages/System/StorageSystemClusters.h index f6e08734896..0f7c792261d 100644 --- a/src/Storages/System/StorageSystemClusters.h +++ b/src/Storages/System/StorageSystemClusters.h @@ -1,10 +1,10 @@ #pragma once -#include #include #include #include + namespace DB { @@ -27,7 +27,7 @@ protected: using NameAndCluster = std::pair>; void fillData(MutableColumns & res_columns, ContextPtr context, const ActionsDAG::Node *, std::vector) const override; - static void writeCluster(MutableColumns & res_columns, const NameAndCluster & name_and_cluster, const ReplicasInfo & replicas_info); + static void writeCluster(MutableColumns & res_columns, const NameAndCluster & name_and_cluster, const std::vector & is_active); }; } diff --git a/tests/integration/test_recovery_time_metric/__init__.py b/tests/integration/test_recovery_time_metric/__init__.py deleted file mode 100644 index e69de29bb2d..00000000000 diff --git a/tests/integration/test_recovery_time_metric/configs/config.xml b/tests/integration/test_recovery_time_metric/configs/config.xml deleted file mode 100644 index bad9b1fa9ea..00000000000 --- a/tests/integration/test_recovery_time_metric/configs/config.xml +++ /dev/null @@ -1,41 +0,0 @@ - - 9000 - - - - - - - - - default - - - - - - 2181 - 1 - /var/lib/clickhouse/coordination/log - /var/lib/clickhouse/coordination/snapshots - - 20000 - - - - 1 - localhost - 9444 - - - - - - - localhost - 2181 - - 20000 - - - diff --git a/tests/integration/test_recovery_time_metric/test.py b/tests/integration/test_recovery_time_metric/test.py deleted file mode 100644 index 6fcf2fad423..00000000000 --- a/tests/integration/test_recovery_time_metric/test.py +++ /dev/null @@ -1,61 +0,0 @@ -import pytest -from helpers.cluster import ClickHouseCluster - -cluster = ClickHouseCluster(__file__) -node = cluster.add_instance( - "node", - main_configs=["configs/config.xml"], - stay_alive=True, -) - - -@pytest.fixture(scope="module") -def start_cluster(): - try: - cluster.start() - yield cluster - finally: - cluster.shutdown() - - -def test_recovery_time_metric(start_cluster): - node.query( - """ - DROP DATABASE IF EXISTS rdb; - CREATE DATABASE rdb - ENGINE = Replicated('/test/test_recovery_time_metric', 'shard1', 'replica1') - """ - ) - - node.query( - """ - DROP TABLE IF EXISTS rdb.t; - CREATE TABLE rdb.t - ( - `x` UInt32 - ) - ENGINE = MergeTree - ORDER BY x - """ - ) - - node.exec_in_container(["bash", "-c", "rm /var/lib/clickhouse/metadata/rdb/t.sql"]) - - node.restart_clickhouse() - - ret = int( - node.query( - """ - SELECT recovery_time - FROM system.clusters - WHERE cluster = 'rdb' - """ - ).strip() - ) - assert ret > 0 - - node.query( - """ - DROP DATABASE rdb - """ - ) diff --git a/tests/queries/0_stateless/02117_show_create_table_system.reference b/tests/queries/0_stateless/02117_show_create_table_system.reference index 32e8b2f4312..cfae4fee6c2 100644 --- a/tests/queries/0_stateless/02117_show_create_table_system.reference +++ b/tests/queries/0_stateless/02117_show_create_table_system.reference @@ -52,8 +52,6 @@ CREATE TABLE system.clusters `database_shard_name` String, `database_replica_name` String, `is_active` Nullable(UInt8), - `replication_lag` Nullable(UInt32), - `recovery_time` Nullable(UInt64), `name` String ALIAS cluster ) ENGINE = SystemClusters diff --git a/tests/queries/0_stateless/03206_replication_lag_metric.reference b/tests/queries/0_stateless/03206_replication_lag_metric.reference deleted file mode 100644 index 02f4a7264b1..00000000000 --- a/tests/queries/0_stateless/03206_replication_lag_metric.reference +++ /dev/null @@ -1,4 +0,0 @@ -0 -2 -0 -2 diff --git a/tests/queries/0_stateless/03206_replication_lag_metric.sql b/tests/queries/0_stateless/03206_replication_lag_metric.sql deleted file mode 100644 index 998c332a11c..00000000000 --- a/tests/queries/0_stateless/03206_replication_lag_metric.sql +++ /dev/null @@ -1,11 +0,0 @@ --- Tags: no-parallel - -CREATE DATABASE rdb1 ENGINE = Replicated('/test/test_replication_lag_metric', 'shard1', 'replica1'); -CREATE DATABASE rdb2 ENGINE = Replicated('/test/test_replication_lag_metric', 'shard1', 'replica2'); - -SET distributed_ddl_task_timeout = 0; -CREATE TABLE rdb1.t (id UInt32) ENGINE = ReplicatedMergeTree ORDER BY id; -SELECT replication_lag FROM system.clusters WHERE cluster IN ('rdb1', 'rdb2') ORDER BY cluster ASC, replica_num ASC; - -DROP DATABASE rdb1; -DROP DATABASE rdb2; From a19750234153e760907f3c7bc040f949100534df Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Sat, 3 Aug 2024 23:11:20 +0200 Subject: [PATCH 178/227] Fix test retries Should fix issues like: - 02494_zero_copy_projection_cancel_fetch - https://s3.amazonaws.com/clickhouse-test-reports/67719/40cd5467c18d65a6624d273ac1a8fd9cc9257d8c/stateless_tests__tsan__s3_storage__[4_4].html Signed-off-by: Azat Khuzhin --- tests/clickhouse-test | 1 - 1 file changed, 1 deletion(-) diff --git a/tests/clickhouse-test b/tests/clickhouse-test index a29c786e998..877548e577e 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -2218,7 +2218,6 @@ def run_tests_array(all_tests_with_params: Tuple[List[str], int, TestSuite, bool args, test_suite, client_options, server_logs_level ) test_result = test_case.process_result(test_result, MESSAGES) - break except TimeoutError: break finally: From f06ae2f5518ff8cb610b337d4900fd6f0088190f Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Sat, 3 Aug 2024 23:27:19 +0200 Subject: [PATCH 179/227] Fill only selected columns from system.clusters Some of them pretty heavy, i.e. is_active for ReplicatedDatabase This should fix 02903_rmt_retriable_merge_exception flakiness [1]. [1]: https://s3.amazonaws.com/clickhouse-test-reports/67687/89c47df559ba23d988f8af3c342e0c8d5531f4b8/fast_test.html Signed-off-by: Azat Khuzhin --- src/Storages/System/StorageSystemClusters.cpp | 82 ++++++++++++------- src/Storages/System/StorageSystemClusters.h | 6 +- 2 files changed, 56 insertions(+), 32 deletions(-) diff --git a/src/Storages/System/StorageSystemClusters.cpp b/src/Storages/System/StorageSystemClusters.cpp index 160c8d6270e..9c5c07ae49f 100644 --- a/src/Storages/System/StorageSystemClusters.cpp +++ b/src/Storages/System/StorageSystemClusters.cpp @@ -40,10 +40,10 @@ ColumnsDescription StorageSystemClusters::getColumnsDescription() return description; } -void StorageSystemClusters::fillData(MutableColumns & res_columns, ContextPtr context, const ActionsDAG::Node *, std::vector) const +void StorageSystemClusters::fillData(MutableColumns & res_columns, ContextPtr context, const ActionsDAG::Node *, std::vector columns_mask) const { for (const auto & name_and_cluster : context->getClusters()) - writeCluster(res_columns, name_and_cluster, {}); + writeCluster(res_columns, columns_mask, name_and_cluster, /* replicated= */ nullptr); const auto databases = DatabaseCatalog::instance().getDatabases(); for (const auto & name_and_database : databases) @@ -52,18 +52,15 @@ void StorageSystemClusters::fillData(MutableColumns & res_columns, ContextPtr co { if (auto database_cluster = replicated->tryGetCluster()) - writeCluster(res_columns, {name_and_database.first, database_cluster}, - replicated->tryGetAreReplicasActive(database_cluster)); + writeCluster(res_columns, columns_mask, {name_and_database.first, database_cluster}, replicated); if (auto database_cluster = replicated->tryGetAllGroupsCluster()) - writeCluster(res_columns, {DatabaseReplicated::ALL_GROUPS_CLUSTER_PREFIX + name_and_database.first, database_cluster}, - replicated->tryGetAreReplicasActive(database_cluster)); + writeCluster(res_columns, columns_mask, {DatabaseReplicated::ALL_GROUPS_CLUSTER_PREFIX + name_and_database.first, database_cluster}, replicated); } } } -void StorageSystemClusters::writeCluster(MutableColumns & res_columns, const NameAndCluster & name_and_cluster, - const std::vector & is_active) +void StorageSystemClusters::writeCluster(MutableColumns & res_columns, const std::vector & columns_mask, const NameAndCluster & name_and_cluster, const DatabaseReplicated * replicated) { const String & cluster_name = name_and_cluster.first; const ClusterPtr & cluster = name_and_cluster.second; @@ -79,30 +76,55 @@ void StorageSystemClusters::writeCluster(MutableColumns & res_columns, const Nam for (size_t replica_index = 0; replica_index < shard_addresses.size(); ++replica_index) { - size_t i = 0; + size_t src_index = 0, res_index = 0; const auto & address = shard_addresses[replica_index]; - res_columns[i++]->insert(cluster_name); - res_columns[i++]->insert(shard_info.shard_num); - res_columns[i++]->insert(shard_info.weight); - res_columns[i++]->insert(shard_info.has_internal_replication); - res_columns[i++]->insert(replica_index + 1); - res_columns[i++]->insert(address.host_name); - auto resolved = address.getResolvedAddress(); - res_columns[i++]->insert(resolved ? resolved->host().toString() : String()); - res_columns[i++]->insert(address.port); - res_columns[i++]->insert(address.is_local); - res_columns[i++]->insert(address.user); - res_columns[i++]->insert(address.default_database); - res_columns[i++]->insert(pool_status[replica_index].error_count); - res_columns[i++]->insert(pool_status[replica_index].slowdown_count); - res_columns[i++]->insert(pool_status[replica_index].estimated_recovery_time.count()); - res_columns[i++]->insert(address.database_shard_name); - res_columns[i++]->insert(address.database_replica_name); - if (is_active.empty()) - res_columns[i++]->insertDefault(); - else - res_columns[i++]->insert(is_active[replica_idx++]); + if (columns_mask[src_index++]) + res_columns[res_index++]->insert(cluster_name); + if (columns_mask[src_index++]) + res_columns[res_index++]->insert(shard_info.shard_num); + if (columns_mask[src_index++]) + res_columns[res_index++]->insert(shard_info.weight); + if (columns_mask[src_index++]) + res_columns[res_index++]->insert(shard_info.has_internal_replication); + if (columns_mask[src_index++]) + res_columns[res_index++]->insert(replica_index + 1); + if (columns_mask[src_index++]) + res_columns[res_index++]->insert(address.host_name); + if (columns_mask[src_index++]) + { + auto resolved = address.getResolvedAddress(); + res_columns[res_index++]->insert(resolved ? resolved->host().toString() : String()); + } + if (columns_mask[src_index++]) + res_columns[res_index++]->insert(address.port); + if (columns_mask[src_index++]) + res_columns[res_index++]->insert(address.is_local); + if (columns_mask[src_index++]) + res_columns[res_index++]->insert(address.user); + if (columns_mask[src_index++]) + res_columns[res_index++]->insert(address.default_database); + if (columns_mask[src_index++]) + res_columns[res_index++]->insert(pool_status[replica_index].error_count); + if (columns_mask[src_index++]) + res_columns[res_index++]->insert(pool_status[replica_index].slowdown_count); + if (columns_mask[src_index++]) + res_columns[res_index++]->insert(pool_status[replica_index].estimated_recovery_time.count()); + if (columns_mask[src_index++]) + res_columns[res_index++]->insert(address.database_shard_name); + if (columns_mask[src_index++]) + res_columns[res_index++]->insert(address.database_replica_name); + if (columns_mask[src_index++]) + { + std::vector is_active; + if (replicated) + is_active = replicated->tryGetAreReplicasActive(name_and_cluster.second); + + if (is_active.empty()) + res_columns[res_index++]->insertDefault(); + else + res_columns[res_index++]->insert(is_active[replica_idx++]); + } } } } diff --git a/src/Storages/System/StorageSystemClusters.h b/src/Storages/System/StorageSystemClusters.h index 0f7c792261d..f6adb902f43 100644 --- a/src/Storages/System/StorageSystemClusters.h +++ b/src/Storages/System/StorageSystemClusters.h @@ -10,6 +10,7 @@ namespace DB class Context; class Cluster; +class DatabaseReplicated; /** Implements system table 'clusters' * that allows to obtain information about available clusters @@ -26,8 +27,9 @@ protected: using IStorageSystemOneBlock::IStorageSystemOneBlock; using NameAndCluster = std::pair>; - void fillData(MutableColumns & res_columns, ContextPtr context, const ActionsDAG::Node *, std::vector) const override; - static void writeCluster(MutableColumns & res_columns, const NameAndCluster & name_and_cluster, const std::vector & is_active); + void fillData(MutableColumns & res_columns, ContextPtr context, const ActionsDAG::Node *, std::vector columns_mask) const override; + static void writeCluster(MutableColumns & res_columns, const std::vector & columns_mask, const NameAndCluster & name_and_cluster, const DatabaseReplicated * replicated); + bool supportsColumnsMask() const override { return true; } }; } From 9d0e066cda8d0ccb6bd4f9e07fee36a2bfae707a Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Sat, 3 Aug 2024 23:41:10 +0200 Subject: [PATCH 180/227] Bump NuRaft (to properly catch thread exceptions) Refs: https://github.com/ClickHouse/NuRaft/pull/75 Refs: https://github.com/eBay/NuRaft/pull/525 Signed-off-by: Azat Khuzhin --- contrib/NuRaft | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/contrib/NuRaft b/contrib/NuRaft index cb5dc3c906e..c2b0811f164 160000 --- a/contrib/NuRaft +++ b/contrib/NuRaft @@ -1 +1 @@ -Subproject commit cb5dc3c906e80f253e9ce9535807caef827cc2e0 +Subproject commit c2b0811f164a7948208489562dab4f186eb305ce From 8562a6106c286882f26383086e52e399106893be Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 3 Aug 2024 23:45:18 +0200 Subject: [PATCH 181/227] Better safety thresholds in `arrayWithConstant` --- src/Functions/array/arrayWithConstant.cpp | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-) diff --git a/src/Functions/array/arrayWithConstant.cpp b/src/Functions/array/arrayWithConstant.cpp index 48262870553..4cbc6404b9b 100644 --- a/src/Functions/array/arrayWithConstant.cpp +++ b/src/Functions/array/arrayWithConstant.cpp @@ -1,9 +1,9 @@ #include -#include #include #include #include #include +#include namespace DB @@ -15,7 +15,8 @@ namespace ErrorCodes extern const int TOO_LARGE_ARRAY_SIZE; } -/// Reasonable threshold. +/// Reasonable thresholds. +static constexpr Int64 max_array_size_in_columns_bytes = 1000000000; static constexpr size_t max_arrays_size_in_columns = 1000000000; @@ -63,12 +64,19 @@ public: auto array_size = col_num->getInt(i); if (unlikely(array_size < 0)) - throw Exception(ErrorCodes::TOO_LARGE_ARRAY_SIZE, "Array size cannot be negative: while executing function {}", getName()); + throw Exception(ErrorCodes::TOO_LARGE_ARRAY_SIZE, "Array size {} cannot be negative: while executing function {}", array_size, getName()); + + Int64 estimated_size = 0; + if (unlikely(common::mulOverflow(array_size, col_value->byteSize(), estimated_size))) + throw Exception(ErrorCodes::TOO_LARGE_ARRAY_SIZE, "Array size {} with element size {} bytes is too large: while executing function {}", array_size, col_value->byteSize(), getName()); + + if (unlikely(estimated_size > max_array_size_in_columns_bytes)) + throw Exception(ErrorCodes::TOO_LARGE_ARRAY_SIZE, "Array size {} with element size {} bytes is too large: while executing function {}", array_size, col_value->byteSize(), getName()); offset += array_size; if (unlikely(offset > max_arrays_size_in_columns)) - throw Exception(ErrorCodes::TOO_LARGE_ARRAY_SIZE, "Too large array size while executing function {}", getName()); + throw Exception(ErrorCodes::TOO_LARGE_ARRAY_SIZE, "Too large array size {} (will generate at least {} elements) while executing function {}", array_size, offset, getName()); offsets.push_back(offset); } From 7a066a6505108b14bd49da8766c1bc473a978b1f Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 3 Aug 2024 23:48:51 +0200 Subject: [PATCH 182/227] Add a test --- .../0_stateless/03216_arrayWithConstant_limits.reference | 1 + tests/queries/0_stateless/03216_arrayWithConstant_limits.sql | 3 +++ 2 files changed, 4 insertions(+) create mode 100644 tests/queries/0_stateless/03216_arrayWithConstant_limits.reference create mode 100644 tests/queries/0_stateless/03216_arrayWithConstant_limits.sql diff --git a/tests/queries/0_stateless/03216_arrayWithConstant_limits.reference b/tests/queries/0_stateless/03216_arrayWithConstant_limits.reference new file mode 100644 index 00000000000..825319e1c5b --- /dev/null +++ b/tests/queries/0_stateless/03216_arrayWithConstant_limits.reference @@ -0,0 +1 @@ +10000000 diff --git a/tests/queries/0_stateless/03216_arrayWithConstant_limits.sql b/tests/queries/0_stateless/03216_arrayWithConstant_limits.sql new file mode 100644 index 00000000000..c46524c50e6 --- /dev/null +++ b/tests/queries/0_stateless/03216_arrayWithConstant_limits.sql @@ -0,0 +1,3 @@ +SELECT arrayWithConstant(96142475, ['qMUF']); -- { serverError TOO_LARGE_ARRAY_SIZE } +SELECT arrayWithConstant(100000000, materialize([[[[[[[[[['Hello, world!']]]]]]]]]])); -- { serverError TOO_LARGE_ARRAY_SIZE } +SELECT length(arrayWithConstant(10000000, materialize([[[[[[[[[['Hello world']]]]]]]]]]))); From 6013e4b81d29f2b46bec64859132a4e2ff1bbdc9 Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Wed, 17 Jul 2024 20:52:48 +0200 Subject: [PATCH 183/227] poco/MongoDB: Support broader UUID types (generated by python uuid.UUID) Signed-off-by: Azat Khuzhin --- base/poco/MongoDB/src/Binary.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/base/poco/MongoDB/src/Binary.cpp b/base/poco/MongoDB/src/Binary.cpp index ea814d6969f..47f3453630f 100644 --- a/base/poco/MongoDB/src/Binary.cpp +++ b/base/poco/MongoDB/src/Binary.cpp @@ -76,7 +76,7 @@ std::string Binary::toString(int indent) const UUID Binary::uuid() const { - if (_subtype == 0x04 && _buffer.size() == 16) + if ((_subtype == 0x04 || _subtype == 0x03) && _buffer.size() == 16) { UUID uuid; uuid.copyFrom((const char*) _buffer.begin()); From 2f00c962711e13ca00af324366421fe4593b4ce6 Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Wed, 17 Jul 2024 20:34:15 +0200 Subject: [PATCH 184/227] Support true UUID type for MongoDB engine Signed-off-by: Azat Khuzhin --- base/poco/Foundation/include/Poco/UUID.h | 7 +++ base/poco/MongoDB/src/Binary.cpp | 2 +- src/Processors/Sources/MongoDBSource.cpp | 46 +++++++++++++++++-- .../integration/test_storage_mongodb/test.py | 33 +++++++++++++ 4 files changed, 82 insertions(+), 6 deletions(-) diff --git a/base/poco/Foundation/include/Poco/UUID.h b/base/poco/Foundation/include/Poco/UUID.h index df67ef73e4b..6466d226b2e 100644 --- a/base/poco/Foundation/include/Poco/UUID.h +++ b/base/poco/Foundation/include/Poco/UUID.h @@ -19,6 +19,7 @@ #include "Poco/Foundation.h" +#include namespace Poco @@ -135,6 +136,12 @@ public: static const UUID & x500(); /// Returns the namespace identifier for the X500 namespace. + UInt32 getTimeLow() const { return _timeLow; } + UInt16 getTimeMid() const { return _timeMid; } + UInt16 getTimeHiAndVersion() const { return _timeHiAndVersion; } + UInt16 getClockSeq() const { return _clockSeq; } + std::array getNode() const { return std::array{_node[0], _node[1], _node[2], _node[3], _node[4], _node[5]}; } + protected: UUID(UInt32 timeLow, UInt32 timeMid, UInt32 timeHiAndVersion, UInt16 clockSeq, UInt8 node[]); UUID(const char * bytes, Version version); diff --git a/base/poco/MongoDB/src/Binary.cpp b/base/poco/MongoDB/src/Binary.cpp index 47f3453630f..8b0e6baeccb 100644 --- a/base/poco/MongoDB/src/Binary.cpp +++ b/base/poco/MongoDB/src/Binary.cpp @@ -82,7 +82,7 @@ UUID Binary::uuid() const uuid.copyFrom((const char*) _buffer.begin()); return uuid; } - throw BadCastException("Invalid subtype"); + throw BadCastException("Invalid subtype: " + std::to_string(_subtype) + ", size: " + std::to_string(_buffer.size())); } diff --git a/src/Processors/Sources/MongoDBSource.cpp b/src/Processors/Sources/MongoDBSource.cpp index 0d583cf6be5..e00a541b300 100644 --- a/src/Processors/Sources/MongoDBSource.cpp +++ b/src/Processors/Sources/MongoDBSource.cpp @@ -4,6 +4,7 @@ #include #include +#include #include #include #include @@ -17,6 +18,7 @@ #include #include #include +#include "base/types.h" #include #include @@ -45,8 +47,28 @@ namespace using ValueType = ExternalResultDescription::ValueType; using ObjectId = Poco::MongoDB::ObjectId; using MongoArray = Poco::MongoDB::Array; + using MongoUUID = Poco::MongoDB::Binary::Ptr; + UUID parsePocoUUID(const Poco::UUID & src) + { + UUID uuid; + + std::array src_node = src.getNode(); + UInt64 node = 0; + node |= UInt64(src_node[0]) << 40; + node |= UInt64(src_node[1]) << 32; + node |= UInt64(src_node[2]) << 24; + node |= UInt64(src_node[3]) << 16; + node |= UInt64(src_node[4]) << 8; + node |= src_node[5]; + + UUIDHelpers::getHighBytes(uuid) = UInt64(src.getTimeLow()) << 32 | UInt32(src.getTimeMid() << 16 | src.getTimeHiAndVersion()); + UUIDHelpers::getLowBytes(uuid) = UInt64(src.getClockSeq()) << 48 | node; + + return uuid; + } + template Field getNumber(const Poco::MongoDB::Element & value, const std::string & name) { @@ -149,12 +171,20 @@ namespace else if (which.isUUID()) parser = [](const Poco::MongoDB::Element & value, const std::string & name) -> Field { - if (value.type() != Poco::MongoDB::ElementTraits::TypeId) - throw Exception(ErrorCodes::TYPE_MISMATCH, "Type mismatch, expected String (UUID), got type id = {} for column {}", + if (value.type() == Poco::MongoDB::ElementTraits::TypeId) + { + String string = static_cast &>(value).value(); + return parse(string); + } + else if (value.type() == Poco::MongoDB::ElementTraits::TypeId) + { + const Poco::UUID & poco_uuid = static_cast &>(value).value()->uuid(); + return parsePocoUUID(poco_uuid); + } + else + throw Exception(ErrorCodes::TYPE_MISMATCH, "Type mismatch, expected String/UUID, got type id = {} for column {}", toString(value.type()), name); - String string = static_cast &>(value).value(); - return parse(string); }; else throw Exception(ErrorCodes::BAD_ARGUMENTS, "Type conversion to {} is not supported", nested->getName()); @@ -286,8 +316,14 @@ namespace String string = static_cast &>(value).value(); assert_cast(column).getData().push_back(parse(string)); } + else if (value.type() == Poco::MongoDB::ElementTraits::TypeId) + { + const Poco::UUID & poco_uuid = static_cast &>(value).value()->uuid(); + UUID uuid = parsePocoUUID(poco_uuid); + assert_cast(column).getData().push_back(uuid); + } else - throw Exception(ErrorCodes::TYPE_MISMATCH, "Type mismatch, expected String (UUID), got type id = {} for column {}", + throw Exception(ErrorCodes::TYPE_MISMATCH, "Type mismatch, expected String/UUID, got type id = {} for column {}", toString(value.type()), name); break; } diff --git a/tests/integration/test_storage_mongodb/test.py b/tests/integration/test_storage_mongodb/test.py index 3957afe8b29..1a1a790e8e8 100644 --- a/tests/integration/test_storage_mongodb/test.py +++ b/tests/integration/test_storage_mongodb/test.py @@ -1,4 +1,5 @@ import pymongo +from uuid import UUID import pytest from helpers.client import QueryRuntimeException @@ -72,6 +73,28 @@ def test_simple_select(started_cluster): simple_mongo_table.drop() +@pytest.mark.parametrize("started_cluster", [False], indirect=["started_cluster"]) +def test_uuid(started_cluster): + mongo_connection = get_mongo_connection(started_cluster) + db = mongo_connection["test"] + db.add_user("root", "clickhouse") + mongo_table = db["uuid_table"] + mongo_table.insert({"key": 0, "data": UUID("f0e77736-91d1-48ce-8f01-15123ca1c7ed")}) + + node = started_cluster.instances["node"] + node.query( + "CREATE TABLE uuid_mongo_table(key UInt64, data UUID) ENGINE = MongoDB('mongo1:27017', 'test', 'uuid_table', 'root', 'clickhouse')" + ) + + assert node.query("SELECT COUNT() FROM uuid_mongo_table") == "1\n" + assert ( + node.query("SELECT data from uuid_mongo_table where key = 0") + == "f0e77736-91d1-48ce-8f01-15123ca1c7ed\n" + ) + node.query("DROP TABLE uuid_mongo_table") + mongo_table.drop() + + @pytest.mark.parametrize("started_cluster", [False], indirect=["started_cluster"]) def test_simple_select_from_view(started_cluster): mongo_connection = get_mongo_connection(started_cluster) @@ -140,6 +163,10 @@ def test_arrays(started_cluster): "f0e77736-91d1-48ce-8f01-15123ca1c7ed", "93376a07-c044-4281-a76e-ad27cf6973c5", ], + "arr_mongo_uuid": [ + UUID("f0e77736-91d1-48ce-8f01-15123ca1c7ed"), + UUID("93376a07-c044-4281-a76e-ad27cf6973c5"), + ], "arr_arr_bool": [ [True, False, True], [True], @@ -174,6 +201,7 @@ def test_arrays(started_cluster): "arr_datetime Array(DateTime)," "arr_string Array(String)," "arr_uuid Array(UUID)," + "arr_mongo_uuid Array(UUID)," "arr_arr_bool Array(Array(Bool))," "arr_empty Array(UInt64)," "arr_null Array(UInt64)," @@ -222,6 +250,11 @@ def test_arrays(started_cluster): == "['f0e77736-91d1-48ce-8f01-15123ca1c7ed','93376a07-c044-4281-a76e-ad27cf6973c5']\n" ) + assert ( + node.query(f"SELECT arr_mongo_uuid FROM arrays_mongo_table WHERE key = 42") + == "['f0e77736-91d1-48ce-8f01-15123ca1c7ed','93376a07-c044-4281-a76e-ad27cf6973c5']\n" + ) + assert ( node.query(f"SELECT arr_arr_bool FROM arrays_mongo_table WHERE key = 42") == "[[true,false,true],[true],[],[],[false],[false]]\n" From a051eb2a5d1c29a61d3c2d9e2f3f0841bbb2817c Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sun, 4 Aug 2024 00:24:17 +0200 Subject: [PATCH 185/227] Fix tests --- src/Databases/DatabaseHDFS.cpp | 2 +- src/Storages/ObjectStorage/HDFS/HDFSCommon.cpp | 2 +- tests/integration/test_storage_hdfs/test.py | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/src/Databases/DatabaseHDFS.cpp b/src/Databases/DatabaseHDFS.cpp index f58f1b76e71..7fa67a5678e 100644 --- a/src/Databases/DatabaseHDFS.cpp +++ b/src/Databases/DatabaseHDFS.cpp @@ -51,7 +51,7 @@ DatabaseHDFS::DatabaseHDFS(const String & name_, const String & source_url, Cont if (!source.empty()) { if (!re2::RE2::FullMatch(source, std::string(HDFS_HOST_REGEXP))) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Bad hdfs host: {}. " + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Bad HDFS host: {}. " "It should have structure 'hdfs://:'", source); context_->getGlobalContext()->getRemoteHostFilter().checkURL(Poco::URI(source)); diff --git a/src/Storages/ObjectStorage/HDFS/HDFSCommon.cpp b/src/Storages/ObjectStorage/HDFS/HDFSCommon.cpp index 365828bcc40..7f8727eea1c 100644 --- a/src/Storages/ObjectStorage/HDFS/HDFSCommon.cpp +++ b/src/Storages/ObjectStorage/HDFS/HDFSCommon.cpp @@ -192,7 +192,7 @@ String getNameNodeCluster(const String &hdfs_url) void checkHDFSURL(const String & url) { if (!re2::RE2::FullMatch(url, std::string(HDFS_URL_REGEXP))) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Bad hdfs url: {}. It should have structure 'hdfs://:/'", url); + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Bad HDFS URL: {}. It should have structure 'hdfs://:/'", url); } } diff --git a/tests/integration/test_storage_hdfs/test.py b/tests/integration/test_storage_hdfs/test.py index 47d8f44c0b7..ccd2c7eaf11 100644 --- a/tests/integration/test_storage_hdfs/test.py +++ b/tests/integration/test_storage_hdfs/test.py @@ -157,7 +157,7 @@ def test_bad_hdfs_uri(started_cluster): ) except Exception as ex: print(ex) - assert "Bad hdfs url" in str(ex) + assert "Bad HDFS URL" in str(ex) try: node1.query( "create table BadStorage2 (id UInt32, name String, weight Float64) ENGINE = HDFS('hdfs://hdfs100500:9000/other_storage', 'TSV')" From 185b6a54da8f24a97f130bdebe7bb1ec2bd266c0 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sun, 4 Aug 2024 00:34:58 +0200 Subject: [PATCH 186/227] Merge with master --- src/Core/SettingsChangesHistory.cpp | 261 +--------------------------- 1 file changed, 1 insertion(+), 260 deletions(-) diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index 3f07bfdb933..5b94391bade 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -57,266 +57,6 @@ String ClickHouseVersion::toString() const /// Note: please check if the key already exists to prevent duplicate entries. static std::initializer_list> settings_changes_history_initializer = { - {"24.8", {{"input_format_json_max_depth", 1000000, 1000, "It was unlimited in previous versions, but that was unsafe."}}}, - {"24.7", {{"output_format_parquet_write_page_index", false, true, "Add a possibility to write page index into parquet files."}, - {"output_format_binary_encode_types_in_binary_format", false, false, "Added new setting to allow to write type names in binary format in RowBinaryWithNamesAndTypes output format"}, - {"input_format_binary_decode_types_in_binary_format", false, false, "Added new setting to allow to read type names in binary format in RowBinaryWithNamesAndTypes input format"}, - {"output_format_native_encode_types_in_binary_format", false, false, "Added new setting to allow to write type names in binary format in Native output format"}, - {"input_format_native_decode_types_in_binary_format", false, false, "Added new setting to allow to read type names in binary format in Native output format"}, - {"read_in_order_use_buffering", false, true, "Use buffering before merging while reading in order of primary key"}, - {"enable_named_columns_in_function_tuple", false, true, "Generate named tuples in function tuple() when all names are unique and can be treated as unquoted identifiers."}, - {"input_format_json_case_insensitive_column_matching", false, false, "Ignore case when matching JSON keys with CH columns."}, - {"optimize_trivial_insert_select", true, false, "The optimization does not make sense in many cases."}, - {"dictionary_validate_primary_key_type", false, false, "Validate primary key type for dictionaries. By default id type for simple layouts will be implicitly converted to UInt64."}, - {"collect_hash_table_stats_during_joins", false, true, "New setting."}, - {"max_size_to_preallocate_for_joins", 0, 100'000'000, "New setting."}, - {"input_format_orc_reader_time_zone_name", "GMT", "GMT", "The time zone name for ORC row reader, the default ORC row reader's time zone is GMT."}, - {"lightweight_mutation_projection_mode", "throw", "throw", "When lightweight delete happens on a table with projection(s), the possible operations include throw the exception as projection exists, or drop all projection related to this table then do lightweight delete."}, - {"database_replicated_allow_heavy_create", true, false, "Long-running DDL queries (CREATE AS SELECT and POPULATE) for Replicated database engine was forbidden"}, - {"query_plan_merge_filters", false, false, "Allow to merge filters in the query plan"}, - {"azure_sdk_max_retries", 10, 10, "Maximum number of retries in azure sdk"}, - {"azure_sdk_retry_initial_backoff_ms", 10, 10, "Minimal backoff between retries in azure sdk"}, - {"azure_sdk_retry_max_backoff_ms", 1000, 1000, "Maximal backoff between retries in azure sdk"}, - {"merge_tree_min_bytes_per_task_for_remote_reading", 4194304, 2097152, "Value is unified with `filesystem_prefetch_min_bytes_for_single_read_task`"}, - {"ignore_on_cluster_for_replicated_named_collections_queries", false, false, "Ignore ON CLUSTER clause for replicated named collections management queries."}, - {"backup_restore_s3_retry_attempts", 1000,1000, "Setting for Aws::Client::RetryStrategy, Aws::Client does retries itself, 0 means no retries. It takes place only for backup/restore."}, - {"postgresql_connection_attempt_timeout", 2, 2, "Allow to control 'connect_timeout' parameter of PostgreSQL connection."}, - {"postgresql_connection_pool_retries", 2, 2, "Allow to control the number of retries in PostgreSQL connection pool."} - }}, - {"24.6", {{"materialize_skip_indexes_on_insert", true, true, "Added new setting to allow to disable materialization of skip indexes on insert"}, - {"materialize_statistics_on_insert", true, true, "Added new setting to allow to disable materialization of statistics on insert"}, - {"input_format_parquet_use_native_reader", false, false, "When reading Parquet files, to use native reader instead of arrow reader."}, - {"hdfs_throw_on_zero_files_match", false, false, "Allow to throw an error when ListObjects request cannot match any files in HDFS engine instead of empty query result"}, - {"azure_throw_on_zero_files_match", false, false, "Allow to throw an error when ListObjects request cannot match any files in AzureBlobStorage engine instead of empty query result"}, - {"s3_validate_request_settings", true, true, "Allow to disable S3 request settings validation"}, - {"allow_experimental_full_text_index", false, false, "Enable experimental full-text index"}, - {"azure_skip_empty_files", false, false, "Allow to skip empty files in azure table engine"}, - {"hdfs_ignore_file_doesnt_exist", false, false, "Allow to return 0 rows when the requested files don't exist instead of throwing an exception in HDFS table engine"}, - {"azure_ignore_file_doesnt_exist", false, false, "Allow to return 0 rows when the requested files don't exist instead of throwing an exception in AzureBlobStorage table engine"}, - {"s3_ignore_file_doesnt_exist", false, false, "Allow to return 0 rows when the requested files don't exist instead of throwing an exception in S3 table engine"}, - {"s3_max_part_number", 10000, 10000, "Maximum part number number for s3 upload part"}, - {"s3_max_single_operation_copy_size", 32 * 1024 * 1024, 32 * 1024 * 1024, "Maximum size for a single copy operation in s3"}, - {"input_format_parquet_max_block_size", 8192, DEFAULT_BLOCK_SIZE, "Increase block size for parquet reader."}, - {"input_format_parquet_prefer_block_bytes", 0, DEFAULT_BLOCK_SIZE * 256, "Average block bytes output by parquet reader."}, - {"enable_blob_storage_log", true, true, "Write information about blob storage operations to system.blob_storage_log table"}, - {"allow_deprecated_snowflake_conversion_functions", true, false, "Disabled deprecated functions snowflakeToDateTime[64] and dateTime[64]ToSnowflake."}, - {"allow_statistic_optimize", false, false, "Old setting which popped up here being renamed."}, - {"allow_experimental_statistic", false, false, "Old setting which popped up here being renamed."}, - {"allow_statistics_optimize", false, false, "The setting was renamed. The previous name is `allow_statistic_optimize`."}, - {"allow_experimental_statistics", false, false, "The setting was renamed. The previous name is `allow_experimental_statistic`."}, - {"enable_vertical_final", false, true, "Enable vertical final by default again after fixing bug"}, - {"parallel_replicas_custom_key_range_lower", 0, 0, "Add settings to control the range filter when using parallel replicas with dynamic shards"}, - {"parallel_replicas_custom_key_range_upper", 0, 0, "Add settings to control the range filter when using parallel replicas with dynamic shards. A value of 0 disables the upper limit"}, - {"output_format_pretty_display_footer_column_names", 0, 1, "Add a setting to display column names in the footer if there are many rows. Threshold value is controlled by output_format_pretty_display_footer_column_names_min_rows."}, - {"output_format_pretty_display_footer_column_names_min_rows", 0, 50, "Add a setting to control the threshold value for setting output_format_pretty_display_footer_column_names_min_rows. Default 50."}, - {"output_format_csv_serialize_tuple_into_separate_columns", true, true, "A new way of how interpret tuples in CSV format was added."}, - {"input_format_csv_deserialize_separate_columns_into_tuple", true, true, "A new way of how interpret tuples in CSV format was added."}, - {"input_format_csv_try_infer_strings_from_quoted_tuples", true, true, "A new way of how interpret tuples in CSV format was added."}, - }}, - {"24.5", {{"allow_deprecated_error_prone_window_functions", true, false, "Allow usage of deprecated error prone window functions (neighbor, runningAccumulate, runningDifferenceStartingWithFirstValue, runningDifference)"}, - {"allow_experimental_join_condition", false, false, "Support join with inequal conditions which involve columns from both left and right table. e.g. t1.y < t2.y."}, - {"input_format_tsv_crlf_end_of_line", false, false, "Enables reading of CRLF line endings with TSV formats"}, - {"output_format_parquet_use_custom_encoder", false, true, "Enable custom Parquet encoder."}, - {"cross_join_min_rows_to_compress", 0, 10000000, "Minimal count of rows to compress block in CROSS JOIN. Zero value means - disable this threshold. This block is compressed when any of the two thresholds (by rows or by bytes) are reached."}, - {"cross_join_min_bytes_to_compress", 0, 1_GiB, "Minimal size of block to compress in CROSS JOIN. Zero value means - disable this threshold. This block is compressed when any of the two thresholds (by rows or by bytes) are reached."}, - {"http_max_chunk_size", 0, 0, "Internal limitation"}, - {"prefer_external_sort_block_bytes", 0, DEFAULT_BLOCK_SIZE * 256, "Prefer maximum block bytes for external sort, reduce the memory usage during merging."}, - {"input_format_force_null_for_omitted_fields", false, false, "Disable type-defaults for omitted fields when needed"}, - {"cast_string_to_dynamic_use_inference", false, false, "Add setting to allow converting String to Dynamic through parsing"}, - {"allow_experimental_dynamic_type", false, false, "Add new experimental Dynamic type"}, - {"azure_max_blocks_in_multipart_upload", 50000, 50000, "Maximum number of blocks in multipart upload for Azure."}, - }}, - {"24.4", {{"input_format_json_throw_on_bad_escape_sequence", true, true, "Allow to save JSON strings with bad escape sequences"}, - {"max_parsing_threads", 0, 0, "Add a separate setting to control number of threads in parallel parsing from files"}, - {"ignore_drop_queries_probability", 0, 0, "Allow to ignore drop queries in server with specified probability for testing purposes"}, - {"lightweight_deletes_sync", 2, 2, "The same as 'mutation_sync', but controls only execution of lightweight deletes"}, - {"query_cache_system_table_handling", "save", "throw", "The query cache no longer caches results of queries against system tables"}, - {"input_format_json_ignore_unnecessary_fields", false, true, "Ignore unnecessary fields and not parse them. Enabling this may not throw exceptions on json strings of invalid format or with duplicated fields"}, - {"input_format_hive_text_allow_variable_number_of_columns", false, true, "Ignore extra columns in Hive Text input (if file has more columns than expected) and treat missing fields in Hive Text input as default values."}, - {"allow_experimental_database_replicated", false, true, "Database engine Replicated is now in Beta stage"}, - {"temporary_data_in_cache_reserve_space_wait_lock_timeout_milliseconds", (10 * 60 * 1000), (10 * 60 * 1000), "Wait time to lock cache for sapce reservation in temporary data in filesystem cache"}, - {"optimize_rewrite_sum_if_to_count_if", false, true, "Only available for the analyzer, where it works correctly"}, - {"azure_allow_parallel_part_upload", "true", "true", "Use multiple threads for azure multipart upload."}, - {"max_recursive_cte_evaluation_depth", DBMS_RECURSIVE_CTE_MAX_EVALUATION_DEPTH, DBMS_RECURSIVE_CTE_MAX_EVALUATION_DEPTH, "Maximum limit on recursive CTE evaluation depth"}, - {"query_plan_convert_outer_join_to_inner_join", false, true, "Allow to convert OUTER JOIN to INNER JOIN if filter after JOIN always filters default values"}, - }}, - {"24.3", {{"s3_connect_timeout_ms", 1000, 1000, "Introduce new dedicated setting for s3 connection timeout"}, - {"allow_experimental_shared_merge_tree", false, true, "The setting is obsolete"}, - {"use_page_cache_for_disks_without_file_cache", false, false, "Added userspace page cache"}, - {"read_from_page_cache_if_exists_otherwise_bypass_cache", false, false, "Added userspace page cache"}, - {"page_cache_inject_eviction", false, false, "Added userspace page cache"}, - {"default_table_engine", "None", "MergeTree", "Set default table engine to MergeTree for better usability"}, - {"input_format_json_use_string_type_for_ambiguous_paths_in_named_tuples_inference_from_objects", false, false, "Allow to use String type for ambiguous paths during named tuple inference from JSON objects"}, - {"traverse_shadow_remote_data_paths", false, false, "Traverse shadow directory when query system.remote_data_paths."}, - {"throw_if_deduplication_in_dependent_materialized_views_enabled_with_async_insert", false, true, "Deduplication in dependent materialized view cannot work together with async inserts."}, - {"parallel_replicas_allow_in_with_subquery", false, true, "If true, subquery for IN will be executed on every follower replica"}, - {"log_processors_profiles", false, true, "Enable by default"}, - {"function_locate_has_mysql_compatible_argument_order", false, true, "Increase compatibility with MySQL's locate function."}, - {"allow_suspicious_primary_key", true, false, "Forbid suspicious PRIMARY KEY/ORDER BY for MergeTree (i.e. SimpleAggregateFunction)"}, - {"filesystem_cache_reserve_space_wait_lock_timeout_milliseconds", 1000, 1000, "Wait time to lock cache for sapce reservation in filesystem cache"}, - {"max_parser_backtracks", 0, 1000000, "Limiting the complexity of parsing"}, - {"analyzer_compatibility_join_using_top_level_identifier", false, false, "Force to resolve identifier in JOIN USING from projection"}, - {"distributed_insert_skip_read_only_replicas", false, false, "If true, INSERT into Distributed will skip read-only replicas"}, - {"keeper_max_retries", 10, 10, "Max retries for general keeper operations"}, - {"keeper_retry_initial_backoff_ms", 100, 100, "Initial backoff timeout for general keeper operations"}, - {"keeper_retry_max_backoff_ms", 5000, 5000, "Max backoff timeout for general keeper operations"}, - {"s3queue_allow_experimental_sharded_mode", false, false, "Enable experimental sharded mode of S3Queue table engine. It is experimental because it will be rewritten"}, - {"allow_experimental_analyzer", false, true, "Enable analyzer and planner by default."}, - {"merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability", 0.0, 0.0, "For testing of `PartsSplitter` - split read ranges into intersecting and non intersecting every time you read from MergeTree with the specified probability."}, - {"allow_get_client_http_header", false, false, "Introduced a new function."}, - {"output_format_pretty_row_numbers", false, true, "It is better for usability."}, - {"output_format_pretty_max_value_width_apply_for_single_value", true, false, "Single values in Pretty formats won't be cut."}, - {"output_format_parquet_string_as_string", false, true, "ClickHouse allows arbitrary binary data in the String data type, which is typically UTF-8. Parquet/ORC/Arrow Strings only support UTF-8. That's why you can choose which Arrow's data type to use for the ClickHouse String data type - String or Binary. While Binary would be more correct and compatible, using String by default will correspond to user expectations in most cases."}, - {"output_format_orc_string_as_string", false, true, "ClickHouse allows arbitrary binary data in the String data type, which is typically UTF-8. Parquet/ORC/Arrow Strings only support UTF-8. That's why you can choose which Arrow's data type to use for the ClickHouse String data type - String or Binary. While Binary would be more correct and compatible, using String by default will correspond to user expectations in most cases."}, - {"output_format_arrow_string_as_string", false, true, "ClickHouse allows arbitrary binary data in the String data type, which is typically UTF-8. Parquet/ORC/Arrow Strings only support UTF-8. That's why you can choose which Arrow's data type to use for the ClickHouse String data type - String or Binary. While Binary would be more correct and compatible, using String by default will correspond to user expectations in most cases."}, - {"output_format_parquet_compression_method", "lz4", "zstd", "Parquet/ORC/Arrow support many compression methods, including lz4 and zstd. ClickHouse supports each and every compression method. Some inferior tools, such as 'duckdb', lack support for the faster `lz4` compression method, that's why we set zstd by default."}, - {"output_format_orc_compression_method", "lz4", "zstd", "Parquet/ORC/Arrow support many compression methods, including lz4 and zstd. ClickHouse supports each and every compression method. Some inferior tools, such as 'duckdb', lack support for the faster `lz4` compression method, that's why we set zstd by default."}, - {"output_format_pretty_highlight_digit_groups", false, true, "If enabled and if output is a terminal, highlight every digit corresponding to the number of thousands, millions, etc. with underline."}, - {"geo_distance_returns_float64_on_float64_arguments", false, true, "Increase the default precision."}, - {"azure_max_inflight_parts_for_one_file", 20, 20, "The maximum number of a concurrent loaded parts in multipart upload request. 0 means unlimited."}, - {"azure_strict_upload_part_size", 0, 0, "The exact size of part to upload during multipart upload to Azure blob storage."}, - {"azure_min_upload_part_size", 16*1024*1024, 16*1024*1024, "The minimum size of part to upload during multipart upload to Azure blob storage."}, - {"azure_max_upload_part_size", 5ull*1024*1024*1024, 5ull*1024*1024*1024, "The maximum size of part to upload during multipart upload to Azure blob storage."}, - {"azure_upload_part_size_multiply_factor", 2, 2, "Multiply azure_min_upload_part_size by this factor each time azure_multiply_parts_count_threshold parts were uploaded from a single write to Azure blob storage."}, - {"azure_upload_part_size_multiply_parts_count_threshold", 500, 500, "Each time this number of parts was uploaded to Azure blob storage, azure_min_upload_part_size is multiplied by azure_upload_part_size_multiply_factor."}, - {"output_format_csv_serialize_tuple_into_separate_columns", true, true, "A new way of how interpret tuples in CSV format was added."}, - {"input_format_csv_deserialize_separate_columns_into_tuple", true, true, "A new way of how interpret tuples in CSV format was added."}, - {"input_format_csv_try_infer_strings_from_quoted_tuples", true, true, "A new way of how interpret tuples in CSV format was added."}, - }}, - {"24.2", {{"allow_suspicious_variant_types", true, false, "Don't allow creating Variant type with suspicious variants by default"}, - {"validate_experimental_and_suspicious_types_inside_nested_types", false, true, "Validate usage of experimental and suspicious types inside nested types"}, - {"output_format_values_escape_quote_with_quote", false, false, "If true escape ' with '', otherwise quoted with \\'"}, - {"output_format_pretty_single_large_number_tip_threshold", 0, 1'000'000, "Print a readable number tip on the right side of the table if the block consists of a single number which exceeds this value (except 0)"}, - {"input_format_try_infer_exponent_floats", true, false, "Don't infer floats in exponential notation by default"}, - {"query_plan_optimize_prewhere", true, true, "Allow to push down filter to PREWHERE expression for supported storages"}, - {"async_insert_max_data_size", 1000000, 10485760, "The previous value appeared to be too small."}, - {"async_insert_poll_timeout_ms", 10, 10, "Timeout in milliseconds for polling data from asynchronous insert queue"}, - {"async_insert_use_adaptive_busy_timeout", false, true, "Use adaptive asynchronous insert timeout"}, - {"async_insert_busy_timeout_min_ms", 50, 50, "The minimum value of the asynchronous insert timeout in milliseconds; it also serves as the initial value, which may be increased later by the adaptive algorithm"}, - {"async_insert_busy_timeout_max_ms", 200, 200, "The minimum value of the asynchronous insert timeout in milliseconds; async_insert_busy_timeout_ms is aliased to async_insert_busy_timeout_max_ms"}, - {"async_insert_busy_timeout_increase_rate", 0.2, 0.2, "The exponential growth rate at which the adaptive asynchronous insert timeout increases"}, - {"async_insert_busy_timeout_decrease_rate", 0.2, 0.2, "The exponential growth rate at which the adaptive asynchronous insert timeout decreases"}, - {"format_template_row_format", "", "", "Template row format string can be set directly in query"}, - {"format_template_resultset_format", "", "", "Template result set format string can be set in query"}, - {"split_parts_ranges_into_intersecting_and_non_intersecting_final", true, true, "Allow to split parts ranges into intersecting and non intersecting during FINAL optimization"}, - {"split_intersecting_parts_ranges_into_layers_final", true, true, "Allow to split intersecting parts ranges into layers during FINAL optimization"}, - {"azure_max_single_part_copy_size", 256*1024*1024, 256*1024*1024, "The maximum size of object to copy using single part copy to Azure blob storage."}, - {"min_external_table_block_size_rows", DEFAULT_INSERT_BLOCK_SIZE, DEFAULT_INSERT_BLOCK_SIZE, "Squash blocks passed to external table to specified size in rows, if blocks are not big enough"}, - {"min_external_table_block_size_bytes", DEFAULT_INSERT_BLOCK_SIZE * 256, DEFAULT_INSERT_BLOCK_SIZE * 256, "Squash blocks passed to external table to specified size in bytes, if blocks are not big enough."}, - {"parallel_replicas_prefer_local_join", true, true, "If true, and JOIN can be executed with parallel replicas algorithm, and all storages of right JOIN part are *MergeTree, local JOIN will be used instead of GLOBAL JOIN."}, - {"optimize_time_filter_with_preimage", true, true, "Optimize Date and DateTime predicates by converting functions into equivalent comparisons without conversions (e.g. toYear(col) = 2023 -> col >= '2023-01-01' AND col <= '2023-12-31')"}, - {"extract_key_value_pairs_max_pairs_per_row", 0, 0, "Max number of pairs that can be produced by the `extractKeyValuePairs` function. Used as a safeguard against consuming too much memory."}, - {"default_view_definer", "CURRENT_USER", "CURRENT_USER", "Allows to set default `DEFINER` option while creating a view"}, - {"default_materialized_view_sql_security", "DEFINER", "DEFINER", "Allows to set a default value for SQL SECURITY option when creating a materialized view"}, - {"default_normal_view_sql_security", "INVOKER", "INVOKER", "Allows to set default `SQL SECURITY` option while creating a normal view"}, - {"mysql_map_string_to_text_in_show_columns", false, true, "Reduce the configuration effort to connect ClickHouse with BI tools."}, - {"mysql_map_fixed_string_to_text_in_show_columns", false, true, "Reduce the configuration effort to connect ClickHouse with BI tools."}, - }}, - {"24.1", {{"print_pretty_type_names", false, true, "Better user experience."}, - {"input_format_json_read_bools_as_strings", false, true, "Allow to read bools as strings in JSON formats by default"}, - {"output_format_arrow_use_signed_indexes_for_dictionary", false, true, "Use signed indexes type for Arrow dictionaries by default as it's recommended"}, - {"allow_experimental_variant_type", false, false, "Add new experimental Variant type"}, - {"use_variant_as_common_type", false, false, "Allow to use Variant in if/multiIf if there is no common type"}, - {"output_format_arrow_use_64_bit_indexes_for_dictionary", false, false, "Allow to use 64 bit indexes type in Arrow dictionaries"}, - {"parallel_replicas_mark_segment_size", 128, 128, "Add new setting to control segment size in new parallel replicas coordinator implementation"}, - {"ignore_materialized_views_with_dropped_target_table", false, false, "Add new setting to allow to ignore materialized views with dropped target table"}, - {"output_format_compression_level", 3, 3, "Allow to change compression level in the query output"}, - {"output_format_compression_zstd_window_log", 0, 0, "Allow to change zstd window log in the query output when zstd compression is used"}, - {"enable_zstd_qat_codec", false, false, "Add new ZSTD_QAT codec"}, - {"enable_vertical_final", false, true, "Use vertical final by default"}, - {"output_format_arrow_use_64_bit_indexes_for_dictionary", false, false, "Allow to use 64 bit indexes type in Arrow dictionaries"}, - {"max_rows_in_set_to_optimize_join", 100000, 0, "Disable join optimization as it prevents from read in order optimization"}, - {"output_format_pretty_color", true, "auto", "Setting is changed to allow also for auto value, disabling ANSI escapes if output is not a tty"}, - {"function_visible_width_behavior", 0, 1, "We changed the default behavior of `visibleWidth` to be more precise"}, - {"max_estimated_execution_time", 0, 0, "Separate max_execution_time and max_estimated_execution_time"}, - {"iceberg_engine_ignore_schema_evolution", false, false, "Allow to ignore schema evolution in Iceberg table engine"}, - {"optimize_injective_functions_in_group_by", false, true, "Replace injective functions by it's arguments in GROUP BY section in analyzer"}, - {"update_insert_deduplication_token_in_dependent_materialized_views", false, false, "Allow to update insert deduplication token with table identifier during insert in dependent materialized views"}, - {"azure_max_unexpected_write_error_retries", 4, 4, "The maximum number of retries in case of unexpected errors during Azure blob storage write"}, - {"split_parts_ranges_into_intersecting_and_non_intersecting_final", false, true, "Allow to split parts ranges into intersecting and non intersecting during FINAL optimization"}, - {"split_intersecting_parts_ranges_into_layers_final", true, true, "Allow to split intersecting parts ranges into layers during FINAL optimization"}}}, - {"23.12", {{"allow_suspicious_ttl_expressions", true, false, "It is a new setting, and in previous versions the behavior was equivalent to allowing."}, - {"input_format_parquet_allow_missing_columns", false, true, "Allow missing columns in Parquet files by default"}, - {"input_format_orc_allow_missing_columns", false, true, "Allow missing columns in ORC files by default"}, - {"input_format_arrow_allow_missing_columns", false, true, "Allow missing columns in Arrow files by default"}}}, - {"23.11", {{"parsedatetime_parse_without_leading_zeros", false, true, "Improved compatibility with MySQL DATE_FORMAT/STR_TO_DATE"}}}, - {"23.9", {{"optimize_group_by_constant_keys", false, true, "Optimize group by constant keys by default"}, - {"input_format_json_try_infer_named_tuples_from_objects", false, true, "Try to infer named Tuples from JSON objects by default"}, - {"input_format_json_read_numbers_as_strings", false, true, "Allow to read numbers as strings in JSON formats by default"}, - {"input_format_json_read_arrays_as_strings", false, true, "Allow to read arrays as strings in JSON formats by default"}, - {"input_format_json_infer_incomplete_types_as_strings", false, true, "Allow to infer incomplete types as Strings in JSON formats by default"}, - {"input_format_json_try_infer_numbers_from_strings", true, false, "Don't infer numbers from strings in JSON formats by default to prevent possible parsing errors"}, - {"http_write_exception_in_output_format", false, true, "Output valid JSON/XML on exception in HTTP streaming."}}}, - {"23.8", {{"rewrite_count_distinct_if_with_count_distinct_implementation", false, true, "Rewrite countDistinctIf with count_distinct_implementation configuration"}}}, - {"23.7", {{"function_sleep_max_microseconds_per_block", 0, 3000000, "In previous versions, the maximum sleep time of 3 seconds was applied only for `sleep`, but not for `sleepEachRow` function. In the new version, we introduce this setting. If you set compatibility with the previous versions, we will disable the limit altogether."}}}, - {"23.6", {{"http_send_timeout", 180, 30, "3 minutes seems crazy long. Note that this is timeout for a single network write call, not for the whole upload operation."}, - {"http_receive_timeout", 180, 30, "See http_send_timeout."}}}, - {"23.5", {{"input_format_parquet_preserve_order", true, false, "Allow Parquet reader to reorder rows for better parallelism."}, - {"parallelize_output_from_storages", false, true, "Allow parallelism when executing queries that read from file/url/s3/etc. This may reorder rows."}, - {"use_with_fill_by_sorting_prefix", false, true, "Columns preceding WITH FILL columns in ORDER BY clause form sorting prefix. Rows with different values in sorting prefix are filled independently"}, - {"output_format_parquet_compliant_nested_types", false, true, "Change an internal field name in output Parquet file schema."}}}, - {"23.4", {{"allow_suspicious_indices", true, false, "If true, index can defined with identical expressions"}, - {"allow_nonconst_timezone_arguments", true, false, "Allow non-const timezone arguments in certain time-related functions like toTimeZone(), fromUnixTimestamp*(), snowflakeToDateTime*()."}, - {"connect_timeout_with_failover_ms", 50, 1000, "Increase default connect timeout because of async connect"}, - {"connect_timeout_with_failover_secure_ms", 100, 1000, "Increase default secure connect timeout because of async connect"}, - {"hedged_connection_timeout_ms", 100, 50, "Start new connection in hedged requests after 50 ms instead of 100 to correspond with previous connect timeout"}, - {"formatdatetime_f_prints_single_zero", true, false, "Improved compatibility with MySQL DATE_FORMAT()/STR_TO_DATE()"}, - {"formatdatetime_parsedatetime_m_is_month_name", false, true, "Improved compatibility with MySQL DATE_FORMAT/STR_TO_DATE"}}}, - {"23.3", {{"output_format_parquet_version", "1.0", "2.latest", "Use latest Parquet format version for output format"}, - {"input_format_json_ignore_unknown_keys_in_named_tuple", false, true, "Improve parsing JSON objects as named tuples"}, - {"input_format_native_allow_types_conversion", false, true, "Allow types conversion in Native input forma"}, - {"output_format_arrow_compression_method", "none", "lz4_frame", "Use lz4 compression in Arrow output format by default"}, - {"output_format_parquet_compression_method", "snappy", "lz4", "Use lz4 compression in Parquet output format by default"}, - {"output_format_orc_compression_method", "none", "lz4_frame", "Use lz4 compression in ORC output format by default"}, - {"async_query_sending_for_remote", false, true, "Create connections and send query async across shards"}}}, - {"23.2", {{"output_format_parquet_fixed_string_as_fixed_byte_array", false, true, "Use Parquet FIXED_LENGTH_BYTE_ARRAY type for FixedString by default"}, - {"output_format_arrow_fixed_string_as_fixed_byte_array", false, true, "Use Arrow FIXED_SIZE_BINARY type for FixedString by default"}, - {"query_plan_remove_redundant_distinct", false, true, "Remove redundant Distinct step in query plan"}, - {"optimize_duplicate_order_by_and_distinct", true, false, "Remove duplicate ORDER BY and DISTINCT if it's possible"}, - {"insert_keeper_max_retries", 0, 20, "Enable reconnections to Keeper on INSERT, improve reliability"}}}, - {"23.1", {{"input_format_json_read_objects_as_strings", 0, 1, "Enable reading nested json objects as strings while object type is experimental"}, - {"input_format_json_defaults_for_missing_elements_in_named_tuple", false, true, "Allow missing elements in JSON objects while reading named tuples by default"}, - {"input_format_csv_detect_header", false, true, "Detect header in CSV format by default"}, - {"input_format_tsv_detect_header", false, true, "Detect header in TSV format by default"}, - {"input_format_custom_detect_header", false, true, "Detect header in CustomSeparated format by default"}, - {"query_plan_remove_redundant_sorting", false, true, "Remove redundant sorting in query plan. For example, sorting steps related to ORDER BY clauses in subqueries"}}}, - {"22.12", {{"max_size_to_preallocate_for_aggregation", 10'000'000, 100'000'000, "This optimizes performance"}, - {"query_plan_aggregation_in_order", 0, 1, "Enable some refactoring around query plan"}, - {"format_binary_max_string_size", 0, 1_GiB, "Prevent allocating large amount of memory"}}}, - {"22.11", {{"use_structure_from_insertion_table_in_table_functions", 0, 2, "Improve using structure from insertion table in table functions"}}}, - {"22.9", {{"force_grouping_standard_compatibility", false, true, "Make GROUPING function output the same as in SQL standard and other DBMS"}}}, - {"22.7", {{"cross_to_inner_join_rewrite", 1, 2, "Force rewrite comma join to inner"}, - {"enable_positional_arguments", false, true, "Enable positional arguments feature by default"}, - {"format_csv_allow_single_quotes", true, false, "Most tools don't treat single quote in CSV specially, don't do it by default too"}}}, - {"22.6", {{"output_format_json_named_tuples_as_objects", false, true, "Allow to serialize named tuples as JSON objects in JSON formats by default"}, - {"input_format_skip_unknown_fields", false, true, "Optimize reading subset of columns for some input formats"}}}, - {"22.5", {{"memory_overcommit_ratio_denominator", 0, 1073741824, "Enable memory overcommit feature by default"}, - {"memory_overcommit_ratio_denominator_for_user", 0, 1073741824, "Enable memory overcommit feature by default"}}}, - {"22.4", {{"allow_settings_after_format_in_insert", true, false, "Do not allow SETTINGS after FORMAT for INSERT queries because ClickHouse interpret SETTINGS as some values, which is misleading"}}}, - {"22.3", {{"cast_ipv4_ipv6_default_on_conversion_error", true, false, "Make functions cast(value, 'IPv4') and cast(value, 'IPv6') behave same as toIPv4 and toIPv6 functions"}}}, - {"21.12", {{"stream_like_engine_allow_direct_select", true, false, "Do not allow direct select for Kafka/RabbitMQ/FileLog by default"}}}, - {"21.9", {{"output_format_decimal_trailing_zeros", true, false, "Do not output trailing zeros in text representation of Decimal types by default for better looking output"}, - {"use_hedged_requests", false, true, "Enable Hedged Requests feature by default"}}}, - {"21.7", {{"legacy_column_name_of_tuple_literal", true, false, "Add this setting only for compatibility reasons. It makes sense to set to 'true', while doing rolling update of cluster from version lower than 21.7 to higher"}}}, - {"21.5", {{"async_socket_for_remote", false, true, "Fix all problems and turn on asynchronous reads from socket for remote queries by default again"}}}, - {"21.3", {{"async_socket_for_remote", true, false, "Turn off asynchronous reads from socket for remote queries because of some problems"}, - {"optimize_normalize_count_variants", false, true, "Rewrite aggregate functions that semantically equals to count() as count() by default"}, - {"normalize_function_names", false, true, "Normalize function names to their canonical names, this was needed for projection query routing"}}}, - {"21.2", {{"enable_global_with_statement", false, true, "Propagate WITH statements to UNION queries and all subqueries by default"}}}, - {"21.1", {{"insert_quorum_parallel", false, true, "Use parallel quorum inserts by default. It is significantly more convenient to use than sequential quorum inserts"}, - {"input_format_null_as_default", false, true, "Allow to insert NULL as default for input formats by default"}, - {"optimize_on_insert", false, true, "Enable data optimization on INSERT by default for better user experience"}, - {"use_compact_format_in_distributed_parts_names", false, true, "Use compact format for async INSERT into Distributed tables by default"}}}, - {"20.10", {{"format_regexp_escaping_rule", "Escaped", "Raw", "Use Raw as default escaping rule for Regexp format to male the behaviour more like to what users expect"}}}, - {"20.7", {{"show_table_uuid_in_table_create_query_if_not_nil", true, false, "Stop showing UID of the table in its CREATE query for Engine=Atomic"}}}, - {"20.5", {{"input_format_with_names_use_header", false, true, "Enable using header with names for formats with WithNames/WithNamesAndTypes suffixes"}, - {"allow_suspicious_codecs", true, false, "Don't allow to specify meaningless compression codecs"}}}, - {"20.4", {{"validate_polygons", false, true, "Throw exception if polygon is invalid in function pointInPolygon by default instead of returning possibly wrong results"}}}, - {"19.18", {{"enable_scalar_subquery_optimization", false, true, "Prevent scalar subqueries from (de)serializing large scalar values and possibly avoid running the same subquery more than once"}}}, - {"19.14", {{"any_join_distinct_right_table_keys", true, false, "Disable ANY RIGHT and ANY FULL JOINs by default to avoid inconsistency"}}}, - {"19.12", {{"input_format_defaults_for_omitted_fields", false, true, "Enable calculation of complex default expressions for omitted fields for some input formats, because it should be the expected behaviour"}}}, - {"19.5", {{"max_partitions_per_insert_block", 0, 100, "Add a limit for the number of partitions in one block"}}}, - {"18.12.17", {{"enable_optimize_predicate_expression", 0, 1, "Optimize predicates to subqueries by default"}}}, {"24.12", { } @@ -335,6 +75,7 @@ static std::initializer_list Date: Sun, 4 Aug 2024 09:02:19 +0200 Subject: [PATCH 187/227] Use RabbitMQ without management in tests (attempt to improve startup) Sometimes startup can take ~90 seconds [1]: 2024-08-03 23:11:38.756067+00:00 [info] <0.9.0> Time to start RabbitMQ: 94651980 us Unlike normally ~10 [2]: 2024-08-04 00:33:07.016137+00:00 [info] <0.9.0> Time to start RabbitMQ: 10082489 us [1]: https://s3.amazonaws.com/clickhouse-test-reports/67737/b4e3bbcb82158bea4f5db1d9f5c28cfb741d1d51/integration_tests__asan__old_analyzer__[4_6].html [2]: https://s3.amazonaws.com/clickhouse-test-reports/66671/2f00c962711e13ca00af324366421fe4593b4ce6/integration_tests__asan__old_analyzer__[4_6].html I've tried locally, the difference is very small, 3135665 us (+management) vs (2740747 us), but still something, and who knows how it works under pressure. Signed-off-by: Azat Khuzhin --- tests/integration/compose/docker_compose_rabbitmq.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/integration/compose/docker_compose_rabbitmq.yml b/tests/integration/compose/docker_compose_rabbitmq.yml index 61b21e0e3d9..94c7f0111c4 100644 --- a/tests/integration/compose/docker_compose_rabbitmq.yml +++ b/tests/integration/compose/docker_compose_rabbitmq.yml @@ -2,7 +2,7 @@ version: '2.3' services: rabbitmq1: - image: rabbitmq:3.12.6-management-alpine + image: rabbitmq:3.12.6-alpine hostname: rabbitmq1 expose: - ${RABBITMQ_PORT:-5672} From 264be9c598b42d91ee0a19f718c4d9a4291c7bc4 Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Sun, 4 Aug 2024 09:06:50 +0200 Subject: [PATCH 188/227] tests: increase timeout for RabbitMQ startup Signed-off-by: Azat Khuzhin --- tests/integration/helpers/cluster.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/integration/helpers/cluster.py b/tests/integration/helpers/cluster.py index 9259c720ff0..a1bdee33d57 100644 --- a/tests/integration/helpers/cluster.py +++ b/tests/integration/helpers/cluster.py @@ -2371,7 +2371,7 @@ class ClickHouseCluster: time.sleep(0.5) raise Exception("Cannot wait PostgreSQL Java Client container") - def wait_rabbitmq_to_start(self, timeout=30): + def wait_rabbitmq_to_start(self, timeout=60): self.print_all_docker_pieces() self.rabbitmq_ip = self.get_instance_ip(self.rabbitmq_host) From dc527b6fd1dfdffb1d177237bffc69cd110cd2a7 Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Sun, 4 Aug 2024 09:07:05 +0200 Subject: [PATCH 189/227] tests: detailed errors for RabbitMQ startup Signed-off-by: Azat Khuzhin --- tests/integration/helpers/cluster.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/integration/helpers/cluster.py b/tests/integration/helpers/cluster.py index a1bdee33d57..56d111629c9 100644 --- a/tests/integration/helpers/cluster.py +++ b/tests/integration/helpers/cluster.py @@ -2399,7 +2399,7 @@ class ClickHouseCluster: ) rabbitmq_debuginfo(self.rabbitmq_docker_id, self.rabbitmq_cookie) except Exception as e: - logging.debug("Unable to get logs from docker.") + logging.debug(f"Unable to get logs from docker: {e}.") raise Exception("Cannot wait RabbitMQ container") def wait_nats_is_available(self, max_retries=5): From 27db715761ec66b0dfc915269270697e6cf67909 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Sun, 4 Aug 2024 09:52:36 +0000 Subject: [PATCH 190/227] Incorporate review feedback --- src/Functions/CountSubstringsImpl.h | 16 +++++++++----- src/Functions/MatchImpl.h | 34 ++++++++++++++--------------- 2 files changed, 28 insertions(+), 22 deletions(-) diff --git a/src/Functions/CountSubstringsImpl.h b/src/Functions/CountSubstringsImpl.h index 8e91bc3aeb4..b1cefae6f1d 100644 --- a/src/Functions/CountSubstringsImpl.h +++ b/src/Functions/CountSubstringsImpl.h @@ -38,10 +38,10 @@ struct CountSubstringsImpl const ColumnPtr & start_pos, PaddedPODArray & res, [[maybe_unused]] ColumnUInt8 * res_null, - size_t /*input_rows_count*/) + size_t input_rows_count) { /// `res_null` serves as an output parameter for implementing an XYZOrNull variant. - assert(!res_null); + chassert(!res_null); const UInt8 * const begin = haystack_data.data(); const UInt8 * const end = haystack_data.data() + haystack_data.size(); @@ -81,6 +81,8 @@ struct CountSubstringsImpl } pos = begin + haystack_offsets[i]; ++i; + + chassert(i < input_rows_count); } } @@ -116,7 +118,7 @@ struct CountSubstringsImpl [[maybe_unused]] ColumnUInt8 * res_null) { /// `res_null` serves as an output parameter for implementing an XYZOrNull variant. - assert(!res_null); + chassert(!res_null); Impl::toLowerIfNeed(haystack); Impl::toLowerIfNeed(needle); @@ -154,8 +156,10 @@ struct CountSubstringsImpl [[maybe_unused]] ColumnUInt8 * res_null, size_t input_rows_count) { + chassert(input_rows_count == haystack_offsets.size()); + /// `res_null` serves as an output parameter for implementing an XYZOrNull variant. - assert(!res_null); + chassert(!res_null); ColumnString::Offset prev_haystack_offset = 0; ColumnString::Offset prev_needle_offset = 0; @@ -210,8 +214,10 @@ struct CountSubstringsImpl [[maybe_unused]] ColumnUInt8 * res_null, size_t input_rows_count) { + chassert(input_rows_count == needle_offsets.size()); + /// `res_null` serves as an output parameter for implementing an XYZOrNull variant. - assert(!res_null); + chassert(!res_null); /// NOTE You could use haystack indexing. But this is a rare case. ColumnString::Offset prev_needle_offset = 0; diff --git a/src/Functions/MatchImpl.h b/src/Functions/MatchImpl.h index ceac753fe79..7dc93ba79e0 100644 --- a/src/Functions/MatchImpl.h +++ b/src/Functions/MatchImpl.h @@ -131,11 +131,11 @@ struct MatchImpl size_t input_rows_count) { /// `res_null` serves as an output parameter for implementing an XYZOrNull variant. - assert(!res_null); + chassert(!res_null); - assert(res.size() == haystack_offsets.size()); - assert(res.size() == input_rows_count); - assert(start_pos_ == nullptr); + chassert(res.size() == haystack_offsets.size()); + chassert(res.size() == input_rows_count); + chassert(start_pos_ == nullptr); if (input_rows_count == 0) return; @@ -295,10 +295,10 @@ struct MatchImpl size_t input_rows_count) { /// `res_null` serves as an output parameter for implementing an XYZOrNull variant. - assert(!res_null); + chassert(!res_null); - assert(res.size() == haystack.size() / N); - assert(res.size() == input_rows_count); + chassert(res.size() == haystack.size() / N); + chassert(res.size() == input_rows_count); if (input_rows_count == 0) return; @@ -468,12 +468,12 @@ struct MatchImpl size_t input_rows_count) { /// `res_null` serves as an output parameter for implementing an XYZOrNull variant. - assert(!res_null); + chassert(!res_null); - assert(haystack_offsets.size() == needle_offset.size()); - assert(res.size() == haystack_offsets.size()); - assert(res.size() == input_rows_count); - assert(start_pos_ == nullptr); + chassert(haystack_offsets.size() == needle_offset.size()); + chassert(res.size() == haystack_offsets.size()); + chassert(res.size() == input_rows_count); + chassert(start_pos_ == nullptr); if (input_rows_count == 0) return; @@ -577,12 +577,12 @@ struct MatchImpl size_t input_rows_count) { /// `res_null` serves as an output parameter for implementing an XYZOrNull variant. - assert(!res_null); + chassert(!res_null); - assert(res.size() == input_rows_count); - assert(res.size() == haystack.size() / N); - assert(res.size() == needle_offset.size()); - assert(start_pos_ == nullptr); + chassert(res.size() == input_rows_count); + chassert(res.size() == haystack.size() / N); + chassert(res.size() == needle_offset.size()); + chassert(start_pos_ == nullptr); if (haystack.empty()) return; From f0aaac3bd19c21a796f54ba080fd67f92959131a Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Sun, 4 Aug 2024 14:14:28 +0200 Subject: [PATCH 191/227] tests: remove useless retries from test_ttl_move::test_alter_with_merge_work Signed-off-by: Azat Khuzhin --- tests/integration/test_ttl_move/test.py | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tests/integration/test_ttl_move/test.py b/tests/integration/test_ttl_move/test.py index 3b79ea7916d..48a6224347d 100644 --- a/tests/integration/test_ttl_move/test.py +++ b/tests/integration/test_ttl_move/test.py @@ -76,7 +76,7 @@ def get_used_disks_for_table(node, table_name, partition=None): ) -def check_used_disks_with_retry(node, table_name, expected_disks, retries): +def check_used_disks_with_retry(node, table_name, expected_disks, retries=1): for _ in range(retries): used_disks = get_used_disks_for_table(node, table_name) if set(used_disks).issubset(expected_disks): @@ -1635,9 +1635,9 @@ def test_alter_with_merge_work(started_cluster, name, engine, positive): optimize_table(20) if positive: - assert check_used_disks_with_retry(node1, name, set(["external"]), 100) + assert check_used_disks_with_retry(node1, name, set(["external"])) else: - assert check_used_disks_with_retry(node1, name, set(["jbod1", "jbod2"]), 50) + assert check_used_disks_with_retry(node1, name, set(["jbod1", "jbod2"])) time.sleep(5) From 47dbc5e05b12213a08c25ade9536603a3fd2b175 Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Sun, 4 Aug 2024 14:16:17 +0200 Subject: [PATCH 192/227] tests: add debug info into test_ttl_move::test_alter_with_merge_work Signed-off-by: Azat Khuzhin --- tests/integration/test_ttl_move/test.py | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/tests/integration/test_ttl_move/test.py b/tests/integration/test_ttl_move/test.py index 48a6224347d..4ebe9a30699 100644 --- a/tests/integration/test_ttl_move/test.py +++ b/tests/integration/test_ttl_move/test.py @@ -1635,9 +1635,17 @@ def test_alter_with_merge_work(started_cluster, name, engine, positive): optimize_table(20) if positive: - assert check_used_disks_with_retry(node1, name, set(["external"])) + assert check_used_disks_with_retry( + node1, name, set(["external"]) + ), "Parts: " + node1.query( + f"SELECT disk_name, name FROM system.parts WHERE table = '{name}' AND active = 1" + ) else: - assert check_used_disks_with_retry(node1, name, set(["jbod1", "jbod2"])) + assert check_used_disks_with_retry( + node1, name, set(["jbod1", "jbod2"]) + ), "Parts: " + node1.query( + f"SELECT disk_name, name FROM system.parts WHERE table = '{name}' AND active = 1" + ) time.sleep(5) From 062490e1b40a8df8d63fca567b11e7dd26cf52ee Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Sun, 4 Aug 2024 14:22:24 +0200 Subject: [PATCH 193/227] tests: fix test_ttl_move::test_alter_with_merge_work flakiness MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Increase timeout for TTL DELETE, since otherwise if other routines will take too long, the part will be removed when it should be still be on "external" disk: 2024.08.04 03:48:53.803032 [ 622 ] {} default.mt_test_alter_with_merge_work_1722743323 (9dc6904a-f082-4f06-be7a-efe4733e811c): Will drop empty part all_1_3_4_4 And this is how part_log looks like: SELECT event_time, event_type, rows, part_name, error, database, disk_name FROM system.part_log WHERE `table` = 'mt_test_alter_with_merge_work_1722743323' ORDER BY event_time ASC Query id: a118b3cd-e4fe-45a5-b675-d73bdd887d79 ┌──────────event_time─┬─event_type─┬─rows─┬─part_name───┬─error─┬─database─┬─disk_name─┐ 1. │ 2024-08-04 03:48:44 │ NewPart │ 2 │ all_1_1_0 │ 0 │ default │ jbod1 │ 2. │ 2024-08-04 03:48:44 │ NewPart │ 2 │ all_2_2_0 │ 0 │ default │ jbod2 │ 3. │ 2024-08-04 03:48:45 │ NewPart │ 2 │ all_3_3_0 │ 0 │ default │ jbod1 │ 4. │ 2024-08-04 03:48:46 │ MutatePart │ 2 │ all_1_1_0_4 │ 0 │ default │ jbod1 │ 5. │ 2024-08-04 03:48:46 │ MutatePart │ 2 │ all_2_2_0_4 │ 0 │ default │ jbod2 │ 6. │ 2024-08-04 03:48:46 │ MutatePart │ 2 │ all_3_3_0_4 │ 0 │ default │ jbod1 │ 7. │ 2024-08-04 03:48:47 │ MovePart │ 2 │ all_1_1_0_4 │ 0 │ default │ external │ 8. │ 2024-08-04 03:48:47 │ MovePart │ 2 │ all_3_3_0_4 │ 0 │ default │ jbod2 │ 9. │ 2024-08-04 03:48:47 │ MergeParts │ 6 │ all_1_3_1_4 │ 0 │ default │ jbod2 │ 10. │ 2024-08-04 03:48:48 │ MovePart │ 6 │ all_1_3_1_4 │ 0 │ default │ external │ 11. │ 2024-08-04 03:48:52 │ MergeParts │ 4 │ all_1_3_2_4 │ 0 │ default │ external │ 12. │ 2024-08-04 03:48:53 │ MergeParts │ 0 │ all_1_3_3_4 │ 0 │ default │ external │ # rows==0 13. │ 2024-08-04 03:48:53 │ MergeParts │ 0 │ all_1_3_4_4 │ 0 │ default │ external │ └─────────────────────┴────────────┴──────┴─────────────┴───────┴──────────┴───────────┘ CI: https://s3.amazonaws.com/clickhouse-test-reports/66671/2f00c962711e13ca00af324366421fe4593b4ce6/integration_tests__tsan__[5_6].html Signed-off-by: Azat Khuzhin --- tests/integration/test_ttl_move/test.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/integration/test_ttl_move/test.py b/tests/integration/test_ttl_move/test.py index 4ebe9a30699..925bdf9baaa 100644 --- a/tests/integration/test_ttl_move/test.py +++ b/tests/integration/test_ttl_move/test.py @@ -1613,7 +1613,7 @@ def test_alter_with_merge_work(started_cluster, name, engine, positive): ALTER TABLE {name} MODIFY TTL d1 + INTERVAL 0 SECOND TO DISK 'jbod2', d1 + INTERVAL 5 SECOND TO VOLUME 'external', - d1 + INTERVAL 10 SECOND DELETE + d1 + INTERVAL 30 SECOND DELETE """.format( name=name ) @@ -1647,7 +1647,7 @@ def test_alter_with_merge_work(started_cluster, name, engine, positive): f"SELECT disk_name, name FROM system.parts WHERE table = '{name}' AND active = 1" ) - time.sleep(5) + time.sleep(25) optimize_table(20) From 9f31488e502c2b2c02e3058f9794829aac14f8b9 Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Sun, 4 Aug 2024 15:43:18 +0200 Subject: [PATCH 194/227] Fix dictionary hang in case of CANNOT_SCHEDULE_TASK while loading On CI you can find that 01747_executable_pool_dictionary_implicit_key can hang [1], it is possible due to after CANNOT_SCHEDULE_TASK the async loading will hang: 2024.07.18 03:56:32.365226 [ 6138 ] {6206a18f-668c-4a5c-a5ad-07f577220762} ExternalDictionariesLoader: Will load the object 'executable_pool_simple_implicit_key' in background, force = false, loading_id = 2 2024.07.18 03:56:32.368005 [ 6138 ] {6206a18f-668c-4a5c-a5ad-07f577220762} executeQuery: Code: 439. DB::Exception: Cannot schedule a task: fault injected (threads=766, jobs=746): In scope SELECT dictGet('executable_pool_simple_implicit_key', 'a', toUInt64(1)). (CANNOT_SCHEDULE_TASK) (version 24.7.1.2241) (from [::1]:56446) (comment: 01747_executable_pool_dictionary_implicit_key.sql) (in query: SELECT dictGet('executable_pool_simple_implicit_key', 'a', toUInt64(1));), Stack trace (when copying this message, always include the lines below): 0. /build/contrib/llvm-project/libcxx/include/exception:141: Poco::Exception::Exception(String const&, int) @ 0x0000000015f8a292 1. /build/src/Common/Exception.cpp:110: DB::Exception::Exception(DB::Exception::MessageMasked&&, int, bool) @ 0x000000000c3df6b9 2. /build/contrib/llvm-project/libcxx/include/string:1499: DB::Exception::Exception(PreformattedMessage&&, int) @ 0x0000000006de714c 3. /build/contrib/llvm-project/libcxx/include/vector:438: DB::Exception::Exception(int, FormatStringHelperImpl::type, std::type_identity::type, std::type_identity::type>, String const&, unsigned long&&, unsigned long&) @ 0x000000000c4838eb 4. /build/src/Common/ThreadPool.cpp:0: void ThreadPoolImpl::scheduleImpl(std::function, Priority, std::optional, bool)::'lambda'(String const&)::operator()(String const&) const @ 0x000000000c4832d3 5. /build/src/Common/ThreadPool.cpp:186: void ThreadPoolImpl::scheduleImpl(std::function, Priority, std::optional, bool) @ 0x000000000c47e7db 6. /build/contrib/llvm-project/libcxx/include/__functional/function.h:818: ? @ 0x000000000c47ec8d 7. /build/contrib/llvm-project/libcxx/include/__functional/function.h:818: ? @ 0x000000001114b16e 8. /build/contrib/llvm-project/libcxx/include/__memory/shared_ptr.h:701: DB::ExternalLoader::LoadingDispatcher::startLoading(DB::ExternalLoader::LoadingDispatcher::Info&, bool, unsigned long) @ 0x0000000011147733 9. /build/src/Interpreters/ExternalLoader.cpp:837: DB::ExternalLoader::LoadingDispatcher::loadImpl(String const&, std::chrono::duration>, bool, std::unique_lock&)::'lambda'()::operator()() const @ 0x0000000011158bf9 10. /build/contrib/llvm-project/libcxx/include/__mutex_base:397: DB::ExternalLoader::LoadingDispatcher::loadImpl(String const&, std::chrono::duration>, bool, std::unique_lock&) @ 0x00000000111588bc 11. /build/src/Interpreters/ExternalLoader.cpp:604: DB::ExternalLoader::LoadResult DB::ExternalLoader::LoadingDispatcher::tryLoad(String const&, std::chrono::duration>) @ 0x00000000111440bf 12. /build/src/Interpreters/ExternalLoader.cpp:1381: std::shared_ptr DB::ExternalLoader::load, void>(String const&) const @ 0x00000000111442f5 13. /build/contrib/llvm-project/libcxx/include/__memory/shared_ptr.h:587: DB::ExternalDictionariesLoader::getDictionary(String const&, std::shared_ptr) const @ 0x0000000011141028 14. /build/src/Functions/FunctionsExternalDictionaries.h:76: DB::FunctionDictHelper::getDictionary(String const&) @ 0x00000000071d28ec ... 2024.07.18 03:58:29.000900 [ 48468 ] {8cf63d7e-dcbf-4af6-bd7c-0e1789ddce3b} executeQuery: (from [::1]:40410) (comment: 01747_executable_pool_dictionary_implicit_key.sql) SELECT dictGet('executable_pool_simple_implicit_key', 'a', toUInt64(1)); (stage: Complete) # and no more rows for 8cf63d7e-dcbf-4af6-bd7c-0e1789ddce3b [1]: https://s3.amazonaws.com/clickhouse-test-reports/66495/bc029ed8207ac75e96e9cb48cb79d27a9ffa4e2f/stress_test__debug_.html The problem that it should be properly cancelled, otherwise it will not be loaded in loadImpl(), but will be waited. Signed-off-by: Azat Khuzhin --- src/Interpreters/ExternalLoader.cpp | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/src/Interpreters/ExternalLoader.cpp b/src/Interpreters/ExternalLoader.cpp index 96405f35f3f..511300be2e0 100644 --- a/src/Interpreters/ExternalLoader.cpp +++ b/src/Interpreters/ExternalLoader.cpp @@ -922,7 +922,16 @@ private: if (enable_async_loading) { /// Put a job to the thread pool for the loading. - auto thread = ThreadFromGlobalPool{&LoadingDispatcher::doLoading, this, info.name, loading_id, forced_to_reload, min_id_to_finish_loading_dependencies_, true, CurrentThread::getGroup()}; + ThreadFromGlobalPool thread; + try + { + thread = ThreadFromGlobalPool{&LoadingDispatcher::doLoading, this, info.name, loading_id, forced_to_reload, min_id_to_finish_loading_dependencies_, true, CurrentThread::getGroup()}; + } + catch (...) + { + cancelLoading(info); + throw; + } loading_threads.try_emplace(loading_id, std::move(thread)); } else From dea95e6c332392a87ec3175bb32a50770a57d65b Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Sun, 14 Jul 2024 16:06:07 +0200 Subject: [PATCH 195/227] Ensure that llvm-symbolizer is used for symbolizing sanitizer reports Since you don't want to fallback to addr2line: # addr2line $ time /bin/test set_flag_impl: Success set_flag_if: Success /usr/bin/addr2line: DWARF error: invalid or unhandled FORM value: 0x23 ================== WARNING: ThreadSanitizer: data race (pid=18) ... real 3m8.580s user 0m21.967s sys 0m40.628s # llvm-symbolizer $ time ./test set_flag_impl: Success set_flag_if: Success ================== WARNING: ThreadSanitizer: data race (pid=24884) real 0m0.028s user 0m0.003s sys 0m0.006s Signed-off-by: Azat Khuzhin --- docker/test/base/Dockerfile | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/docker/test/base/Dockerfile b/docker/test/base/Dockerfile index a81826ed6b5..e7ac62604dc 100644 --- a/docker/test/base/Dockerfile +++ b/docker/test/base/Dockerfile @@ -35,7 +35,9 @@ ENV UBSAN_OPTIONS='print_stacktrace=1 max_allocation_size_mb=32768' ENV MSAN_OPTIONS='abort_on_error=1 poison_in_dtor=1 max_allocation_size_mb=32768' ENV LSAN_OPTIONS='max_allocation_size_mb=32768' -# for external_symbolizer_path +# for external_symbolizer_path, and also ensure that llvm-symbolizer really +# exists (since you don't want to fallback to addr2line, it is very slow) +RUN test -f /usr/bin/llvm-symbolizer-${LLVM_VERSION} RUN ln -s /usr/bin/llvm-symbolizer-${LLVM_VERSION} /usr/bin/llvm-symbolizer RUN echo "en_US.UTF-8 UTF-8" > /etc/locale.gen && locale-gen en_US.UTF-8 From 1f2bb6676f39259ec20dd0b077381d251b591d3a Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Sun, 14 Jul 2024 16:08:09 +0200 Subject: [PATCH 196/227] Remove outdated comments about workaround for #64086 Signed-off-by: Azat Khuzhin --- docker/test/util/Dockerfile | 8 -------- 1 file changed, 8 deletions(-) diff --git a/docker/test/util/Dockerfile b/docker/test/util/Dockerfile index babddbd228c..dc928ba7195 100644 --- a/docker/test/util/Dockerfile +++ b/docker/test/util/Dockerfile @@ -5,14 +5,6 @@ FROM ubuntu:22.04 ARG apt_archive="http://archive.ubuntu.com" RUN sed -i "s|http://archive.ubuntu.com|$apt_archive|g" /etc/apt/sources.list -# FIXME: rebuild for clang 18.1.3, that contains a workaround [1] for -# sanitizers issue [2]: -# -# $ git tag --contains c2a57034eff048cd36c563c8e0051db3a70991b3 | tail -1 -# llvmorg-18.1.3 -# -# [1]: https://github.com/llvm/llvm-project/commit/c2a57034eff048cd36c563c8e0051db3a70991b3 -# [2]: https://github.com/ClickHouse/ClickHouse/issues/64086 ENV DEBIAN_FRONTEND=noninteractive LLVM_VERSION=18 RUN apt-get update \ From eed2edd7db5f2424d85a66bfee63281cba5a5c94 Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Tue, 2 Jul 2024 12:23:10 +0200 Subject: [PATCH 197/227] Comment certificateFile/privateKeyFile/dhParamsFile in keeper config Otherwise you will get annoying messages at startup: 2024.07.02 10:03:38.331593 [ 1 ] {} CertificateReloader: Cannot obtain modification time for certificate file /etc/clickhouse-keeper/server.crt, skipping update. errno: 2, strerror: 0 2024.07.02 10:03:38.331658 [ 1 ] {} CertificateReloader: Cannot obtain modification time for key file /etc/clickhouse-keeper/server.key, skipping update. errno: 2, strerror: 0 2024.07.02 10:03:38.341085 [ 1 ] {} CertificateReloader: Poco::Exception. Code: 1000, e.code() = 0, SSL context exception: Error loading private key from file /etc/clickhouse-keeper/server.key: error:80000002:system library::No such file or directory Signed-off-by: Azat Khuzhin --- programs/keeper/keeper_config.xml | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/programs/keeper/keeper_config.xml b/programs/keeper/keeper_config.xml index 4cf84cffc86..efd0010d184 100644 --- a/programs/keeper/keeper_config.xml +++ b/programs/keeper/keeper_config.xml @@ -66,14 +66,14 @@ - /etc/clickhouse-keeper/server.crt - /etc/clickhouse-keeper/server.key + + - /etc/clickhouse-keeper/dhparam.pem + none true true From a75c4b9f9fe519cb74a47d88efb55216da22a64b Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sun, 4 Aug 2024 18:07:40 +0200 Subject: [PATCH 198/227] Introduce `no-flaky-check` tag --- tests/clickhouse-test | 3 +++ .../00002_log_and_exception_messages_formatting.sql | 2 +- 2 files changed, 4 insertions(+), 1 deletion(-) diff --git a/tests/clickhouse-test b/tests/clickhouse-test index 893fded8d23..b70dd61a25a 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -1232,6 +1232,9 @@ class TestCase: ): return FailureReason.SKIP + elif "no-flaky-check" in tags and (args.test_runs > 1): + return FailureReason.SKIP + elif tags: for build_flag in args.build_flags: if "no-" + build_flag in tags: diff --git a/tests/queries/0_stateless/00002_log_and_exception_messages_formatting.sql b/tests/queries/0_stateless/00002_log_and_exception_messages_formatting.sql index 07c42d6d039..32db77cd8dd 100644 --- a/tests/queries/0_stateless/00002_log_and_exception_messages_formatting.sql +++ b/tests/queries/0_stateless/00002_log_and_exception_messages_formatting.sql @@ -1,4 +1,4 @@ --- Tags: no-parallel, no-fasttest, no-ubsan, no-batch +-- Tags: no-parallel, no-fasttest, no-ubsan, no-batch, no-flaky-check -- no-parallel because we want to run this test when most of the other tests already passed -- If this test fails, see the "Top patterns of log messages" diagnostics in the end of run.log From 87e0cf6b7635b7249d72c6ac0e7aad75c863d01d Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sun, 4 Aug 2024 18:25:36 +0200 Subject: [PATCH 199/227] Fix stack overflow in JSONMergePatch --- src/Common/JSONParsers/RapidJSONParser.h | 12 ++++++--- src/Functions/jsonMergePatch.cpp | 34 +++++++++++++++--------- 2 files changed, 29 insertions(+), 17 deletions(-) diff --git a/src/Common/JSONParsers/RapidJSONParser.h b/src/Common/JSONParsers/RapidJSONParser.h index 6c5ea938bfe..ad7a4cbf53a 100644 --- a/src/Common/JSONParsers/RapidJSONParser.h +++ b/src/Common/JSONParsers/RapidJSONParser.h @@ -3,10 +3,14 @@ #include "config.h" #if USE_RAPIDJSON -# include -# include -# include -# include "ElementTypes.h" + +/// Prevent stack overflow: +#define RAPIDJSON_PARSE_DEFAULT_FLAGS (kParseIterativeFlag) + +#include +#include +#include +#include "ElementTypes.h" namespace DB { diff --git a/src/Functions/jsonMergePatch.cpp b/src/Functions/jsonMergePatch.cpp index a83daacdbf6..3bde415aabf 100644 --- a/src/Functions/jsonMergePatch.cpp +++ b/src/Functions/jsonMergePatch.cpp @@ -10,12 +10,14 @@ #if USE_RAPIDJSON -#include "rapidjson/document.h" -#include "rapidjson/writer.h" -#include "rapidjson/stringbuffer.h" -#include "rapidjson/filewritestream.h" -#include "rapidjson/prettywriter.h" -#include "rapidjson/filereadstream.h" +/// Prevent stack overflow: +#define RAPIDJSON_PARSE_DEFAULT_FLAGS (kParseIterativeFlag) + +#include +#include +#include +#include +#include namespace DB @@ -31,17 +33,17 @@ namespace ErrorCodes namespace { - // select jsonMergePatch('{"a":1}','{"name": "joey"}','{"name": "tom"}','{"name": "zoey"}'); + // select JSONMergePatch('{"a":1}','{"name": "joey"}','{"name": "tom"}','{"name": "zoey"}'); // || // \/ // ┌───────────────────────┐ // │ {"a":1,"name":"zoey"} │ // └───────────────────────┘ - class FunctionjsonMergePatch : public IFunction + class FunctionJSONMergePatch : public IFunction { public: - static constexpr auto name = "jsonMergePatch"; - static FunctionPtr create(ContextPtr) { return std::make_shared(); } + static constexpr auto name = "JSONMergePatch"; + static FunctionPtr create(ContextPtr) { return std::make_shared(); } String getName() const override { return name; } bool isVariadic() const override { return true; } @@ -98,7 +100,11 @@ namespace const char * json = str_ref.data; document.Parse(json); - if (document.HasParseError() || !document.IsObject()) + + if (document.HasParseError()) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Wrong JSON string to merge: {}", rapidjson::GetParseError_En(document.GetParseError())); + + if (!document.IsObject()) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Wrong JSON string to merge. Expected JSON object"); }; @@ -162,10 +168,12 @@ namespace } -REGISTER_FUNCTION(jsonMergePatch) +REGISTER_FUNCTION(JSONMergePatch) { - factory.registerFunction(FunctionDocumentation{ + factory.registerFunction(FunctionDocumentation{ .description="Returns the merged JSON object string, which is formed by merging multiple JSON objects."}); + + factory.registerAlias("jsonMergePatch", "JSONMergePatch"); } } From 37da0b3c307d249a078b94cb7bc53d8acdd22a98 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sun, 4 Aug 2024 18:27:37 +0200 Subject: [PATCH 200/227] Add a test --- .../03217_json_merge_patch_stack_overflow.reference | 0 .../03217_json_merge_patch_stack_overflow.sql | 9 +++++++++ 2 files changed, 9 insertions(+) create mode 100644 tests/queries/0_stateless/03217_json_merge_patch_stack_overflow.reference create mode 100644 tests/queries/0_stateless/03217_json_merge_patch_stack_overflow.sql diff --git a/tests/queries/0_stateless/03217_json_merge_patch_stack_overflow.reference b/tests/queries/0_stateless/03217_json_merge_patch_stack_overflow.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03217_json_merge_patch_stack_overflow.sql b/tests/queries/0_stateless/03217_json_merge_patch_stack_overflow.sql new file mode 100644 index 00000000000..4b366b08c6b --- /dev/null +++ b/tests/queries/0_stateless/03217_json_merge_patch_stack_overflow.sql @@ -0,0 +1,9 @@ +-- Tags: no-fasttest +-- Needs rapidjson library +SELECT JSONMergePatch(REPEAT('{"c":', 1000000)); -- { serverError BAD_ARGUMENTS } +SELECT JSONMergePatch(REPEAT('{"c":', 100000)); -- { serverError BAD_ARGUMENTS } +SELECT JSONMergePatch(REPEAT('{"c":', 10000)); -- { serverError BAD_ARGUMENTS } +SELECT JSONMergePatch(REPEAT('{"c":', 1000)); -- { serverError BAD_ARGUMENTS } +SELECT JSONMergePatch(REPEAT('{"c":', 100)); -- { serverError BAD_ARGUMENTS } +SELECT JSONMergePatch(REPEAT('{"c":', 10)); -- { serverError BAD_ARGUMENTS } +SELECT JSONMergePatch(REPEAT('{"c":', 1)); -- { serverError BAD_ARGUMENTS } From a433115434ebe4c1d69f2ed200005fa93c7adcb7 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sun, 4 Aug 2024 19:18:00 +0200 Subject: [PATCH 201/227] Fix typo --- docker/test/util/process_functional_tests_result.py | 6 +++--- docs/changelogs/v20.5.1.3833-prestable.md | 2 +- docs/changelogs/v21.11.1.8636-prestable.md | 6 +++--- docs/changelogs/v21.12.1.9017-prestable.md | 10 +++++----- docs/changelogs/v21.5.1.6601-prestable.md | 2 +- docs/changelogs/v21.6.9.7-stable.md | 2 +- docs/changelogs/v21.7.9.7-stable.md | 2 +- docs/changelogs/v21.8.1.7409-prestable.md | 2 +- docs/changelogs/v21.8.5.7-lts.md | 2 +- docs/changelogs/v21.9.1.8000-prestable.md | 2 +- docs/changelogs/v22.1.1.2542-prestable.md | 4 ++-- docs/changelogs/v22.4.1.2305-prestable.md | 2 +- docs/changelogs/v23.4.1.1943-stable.md | 3 +-- docs/changelogs/v23.6.1.1524-stable.md | 2 +- docs/zh/changelog/index.md | 2 +- src/Common/SystemLogBase.cpp | 2 +- 16 files changed, 25 insertions(+), 26 deletions(-) diff --git a/docker/test/util/process_functional_tests_result.py b/docker/test/util/process_functional_tests_result.py index 3da1a8f3674..aa2ea686c46 100755 --- a/docker/test/util/process_functional_tests_result.py +++ b/docker/test/util/process_functional_tests_result.py @@ -161,11 +161,11 @@ def process_result(result_path, broken_tests): retries, test_results, ) = process_test_log(result_path, broken_tests) - is_flacky_check = 1 < int(os.environ.get("NUM_TRIES", 1)) - logging.info("Is flaky check: %s", is_flacky_check) + is_flaky_check = 1 < int(os.environ.get("NUM_TRIES", 1)) + logging.info("Is flaky check: %s", is_flaky_check) # If no tests were run (success == 0) it indicates an error (e.g. server did not start or crashed immediately) # But it's Ok for "flaky checks" - they can contain just one test for check which is marked as skipped. - if failed != 0 or unknown != 0 or (success == 0 and (not is_flacky_check)): + if failed != 0 or unknown != 0 or (success == 0 and (not is_flaky_check)): state = "failure" if hung: diff --git a/docs/changelogs/v20.5.1.3833-prestable.md b/docs/changelogs/v20.5.1.3833-prestable.md index 79d61fb12f9..3de263f5e37 100644 --- a/docs/changelogs/v20.5.1.3833-prestable.md +++ b/docs/changelogs/v20.5.1.3833-prestable.md @@ -331,7 +331,7 @@ * Fix several non significant errors in unit tests. [#11262](https://github.com/ClickHouse/ClickHouse/pull/11262) ([alesapin](https://github.com/alesapin)). * Add a test for Join table engine from @donmikel. This closes [#9158](https://github.com/ClickHouse/ClickHouse/issues/9158). [#11265](https://github.com/ClickHouse/ClickHouse/pull/11265) ([Alexey Milovidov](https://github.com/alexey-milovidov)). * Repeat test in CI if `curl` invocation was timed out. It is possible due to system hangups for 10+ seconds that are typical in our CI infrastructure. This fixes [#11267](https://github.com/ClickHouse/ClickHouse/issues/11267). [#11268](https://github.com/ClickHouse/ClickHouse/pull/11268) ([Alexey Milovidov](https://github.com/alexey-milovidov)). -* Fix potentially flacky test `00731_long_merge_tree_select_opened_files.sh`. It does not fail frequently but we have discovered potential race condition in this test while experimenting with ThreadFuzzer: [#9814](https://github.com/ClickHouse/ClickHouse/issues/9814) See [link](https://clickhouse-test-reports.s3.yandex.net/9814/40e3023e215df22985d275bf85f4d2290897b76b/functional_stateless_tests_(unbundled).html#fail1) for the example. [#11270](https://github.com/ClickHouse/ClickHouse/pull/11270) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Fix potentially flaky test `00731_long_merge_tree_select_opened_files.sh`. It does not fail frequently but we have discovered potential race condition in this test while experimenting with ThreadFuzzer: [#9814](https://github.com/ClickHouse/ClickHouse/issues/9814) See [link](https://clickhouse-test-reports.s3.yandex.net/9814/40e3023e215df22985d275bf85f4d2290897b76b/functional_stateless_tests_(unbundled).html#fail1) for the example. [#11270](https://github.com/ClickHouse/ClickHouse/pull/11270) ([Alexey Milovidov](https://github.com/alexey-milovidov)). * Now clickhouse-test check the server aliveness before tests run. [#11285](https://github.com/ClickHouse/ClickHouse/pull/11285) ([alesapin](https://github.com/alesapin)). * Emit a warning if server was build in debug or with sanitizers. [#11304](https://github.com/ClickHouse/ClickHouse/pull/11304) ([Alexey Milovidov](https://github.com/alexey-milovidov)). * Better check for hung queries in clickhouse-test. [#11321](https://github.com/ClickHouse/ClickHouse/pull/11321) ([Alexey Milovidov](https://github.com/alexey-milovidov)). diff --git a/docs/changelogs/v21.11.1.8636-prestable.md b/docs/changelogs/v21.11.1.8636-prestable.md index d6a435dd3ce..95c8580d591 100644 --- a/docs/changelogs/v21.11.1.8636-prestable.md +++ b/docs/changelogs/v21.11.1.8636-prestable.md @@ -280,7 +280,7 @@ sidebar_label: 2022 * Cleanup unbundled image [#29689](https://github.com/ClickHouse/ClickHouse/pull/29689) ([Azat Khuzhin](https://github.com/azat)). * Fix memory tracking for merges and mutations [#29691](https://github.com/ClickHouse/ClickHouse/pull/29691) ([Azat Khuzhin](https://github.com/azat)). * Fix data-race in WriteIndirectBuffer (used in DiskMemory) [#29692](https://github.com/ClickHouse/ClickHouse/pull/29692) ([Azat Khuzhin](https://github.com/azat)). -* Fix flacky test [#29706](https://github.com/ClickHouse/ClickHouse/pull/29706) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Fix flaky test [#29706](https://github.com/ClickHouse/ClickHouse/pull/29706) ([Kseniia Sumarokova](https://github.com/kssenii)). * BorrowedObjectPool condition variable notify fix [#29722](https://github.com/ClickHouse/ClickHouse/pull/29722) ([Maksim Kita](https://github.com/kitaisreal)). * Better exception message for local interactive [#29737](https://github.com/ClickHouse/ClickHouse/pull/29737) ([Kseniia Sumarokova](https://github.com/kssenii)). * Fix --stage for clickhouse-local [#29745](https://github.com/ClickHouse/ClickHouse/pull/29745) ([Azat Khuzhin](https://github.com/azat)). @@ -308,7 +308,7 @@ sidebar_label: 2022 * Fix client [#29864](https://github.com/ClickHouse/ClickHouse/pull/29864) ([Kseniia Sumarokova](https://github.com/kssenii)). * Remove some more streams. [#29898](https://github.com/ClickHouse/ClickHouse/pull/29898) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). * Add logging in ZooKeeper client [#29901](https://github.com/ClickHouse/ClickHouse/pull/29901) ([Alexander Tokmakov](https://github.com/tavplubix)). -* Fix some flacky tests [#29902](https://github.com/ClickHouse/ClickHouse/pull/29902) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Fix some flaky tests [#29902](https://github.com/ClickHouse/ClickHouse/pull/29902) ([Kseniia Sumarokova](https://github.com/kssenii)). * Grep server log even if it contains binary data [#29903](https://github.com/ClickHouse/ClickHouse/pull/29903) ([Alexander Tokmakov](https://github.com/tavplubix)). * Cosmetic refactoring of server constants. [#29913](https://github.com/ClickHouse/ClickHouse/pull/29913) ([Amos Bird](https://github.com/amosbird)). * Format improvement of AlterQuery [#29916](https://github.com/ClickHouse/ClickHouse/pull/29916) ([flynn](https://github.com/ucasfl)). @@ -465,7 +465,7 @@ sidebar_label: 2022 * Fix docs release [#30933](https://github.com/ClickHouse/ClickHouse/pull/30933) ([alesapin](https://github.com/alesapin)). * Fix style check [#30937](https://github.com/ClickHouse/ClickHouse/pull/30937) ([alesapin](https://github.com/alesapin)). * Fix file progress for clickhouse-local [#30938](https://github.com/ClickHouse/ClickHouse/pull/30938) ([Kseniia Sumarokova](https://github.com/kssenii)). -* Fix flacky test [#30940](https://github.com/ClickHouse/ClickHouse/pull/30940) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Fix flaky test [#30940](https://github.com/ClickHouse/ClickHouse/pull/30940) ([Kseniia Sumarokova](https://github.com/kssenii)). * Fix reading from TinyLog [#30941](https://github.com/ClickHouse/ClickHouse/pull/30941) ([Vitaly Baranov](https://github.com/vitlibar)). * Add github to known hosts in docs release [#30947](https://github.com/ClickHouse/ClickHouse/pull/30947) ([alesapin](https://github.com/alesapin)). * Parse json from response in ci checks [#30948](https://github.com/ClickHouse/ClickHouse/pull/30948) ([alesapin](https://github.com/alesapin)). diff --git a/docs/changelogs/v21.12.1.9017-prestable.md b/docs/changelogs/v21.12.1.9017-prestable.md index bd84873e67a..f5416664d35 100644 --- a/docs/changelogs/v21.12.1.9017-prestable.md +++ b/docs/changelogs/v21.12.1.9017-prestable.md @@ -220,7 +220,7 @@ sidebar_label: 2022 * Fix test_backward_compatibility [#30950](https://github.com/ClickHouse/ClickHouse/pull/30950) ([Ilya Yatsishin](https://github.com/qoega)). * Add stress test to github actions [#30952](https://github.com/ClickHouse/ClickHouse/pull/30952) ([alesapin](https://github.com/alesapin)). * Try smaller blacklist of non parallel integration tests [#30963](https://github.com/ClickHouse/ClickHouse/pull/30963) ([Ilya Yatsishin](https://github.com/qoega)). -* Fix flacky test [#30967](https://github.com/ClickHouse/ClickHouse/pull/30967) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Fix flaky test [#30967](https://github.com/ClickHouse/ClickHouse/pull/30967) ([Kseniia Sumarokova](https://github.com/kssenii)). * Move access-rights source code [#30973](https://github.com/ClickHouse/ClickHouse/pull/30973) ([Vitaly Baranov](https://github.com/vitlibar)). * Set output_format_avro_rows_in_file default to 1 [#30990](https://github.com/ClickHouse/ClickHouse/pull/30990) ([Kruglov Pavel](https://github.com/Avogar)). * Remove remaining usages of Y_IGNORE [#30993](https://github.com/ClickHouse/ClickHouse/pull/30993) ([Yuriy Chernyshov](https://github.com/georgthegreat)). @@ -353,7 +353,7 @@ sidebar_label: 2022 * Support toUInt8/toInt8 for if constant condition optimization. [#31866](https://github.com/ClickHouse/ClickHouse/pull/31866) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). * Added -no-sanitize=unsigned-integer-overflow build flag [#31881](https://github.com/ClickHouse/ClickHouse/pull/31881) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). * Fix typos [#31886](https://github.com/ClickHouse/ClickHouse/pull/31886) ([Anton Popov](https://github.com/CurtizJ)). -* Try to fix flacky test. [#31889](https://github.com/ClickHouse/ClickHouse/pull/31889) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Try to fix flaky test. [#31889](https://github.com/ClickHouse/ClickHouse/pull/31889) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). * Reduce the files that depend on parser headers [#31896](https://github.com/ClickHouse/ClickHouse/pull/31896) ([Raúl Marín](https://github.com/Algunenano)). * Fix magic_enum for debug helpers (fixes build w/ USE_DEBUG_HELPERS) [#31922](https://github.com/ClickHouse/ClickHouse/pull/31922) ([Azat Khuzhin](https://github.com/azat)). * Remove some trash from build [#31923](https://github.com/ClickHouse/ClickHouse/pull/31923) ([Alexey Milovidov](https://github.com/alexey-milovidov)). @@ -387,7 +387,7 @@ sidebar_label: 2022 * make looping in H3 funcs uniform [#32110](https://github.com/ClickHouse/ClickHouse/pull/32110) ([Bharat Nallan](https://github.com/bharatnc)). * Remove PVS check from master [#32114](https://github.com/ClickHouse/ClickHouse/pull/32114) ([alesapin](https://github.com/alesapin)). * Fix flaky keeper whitelist test [#32115](https://github.com/ClickHouse/ClickHouse/pull/32115) ([alesapin](https://github.com/alesapin)). -* Fix flacky test test_executable_storage_input [#32118](https://github.com/ClickHouse/ClickHouse/pull/32118) ([Maksim Kita](https://github.com/kitaisreal)). +* Fix flaky test test_executable_storage_input [#32118](https://github.com/ClickHouse/ClickHouse/pull/32118) ([Maksim Kita](https://github.com/kitaisreal)). * Fix data race in `removePartAndEnqueueFetch(...)` [#32119](https://github.com/ClickHouse/ClickHouse/pull/32119) ([Alexander Tokmakov](https://github.com/tavplubix)). * Move fuzzers and unit tests to another group [#32120](https://github.com/ClickHouse/ClickHouse/pull/32120) ([alesapin](https://github.com/alesapin)). * Add a test with 20000 mutations in one query [#32122](https://github.com/ClickHouse/ClickHouse/pull/32122) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). @@ -411,11 +411,11 @@ sidebar_label: 2022 * Add test for [#32186](https://github.com/ClickHouse/ClickHouse/issues/32186) [#32203](https://github.com/ClickHouse/ClickHouse/pull/32203) ([Raúl Marín](https://github.com/Algunenano)). * Fix uncaught exception in DatabaseLazy [#32206](https://github.com/ClickHouse/ClickHouse/pull/32206) ([Alexander Tokmakov](https://github.com/tavplubix)). * Update ASTCreateQuery.cpp [#32208](https://github.com/ClickHouse/ClickHouse/pull/32208) ([Kseniia Sumarokova](https://github.com/kssenii)). -* Fix flacky fileLog test (probably) [#32209](https://github.com/ClickHouse/ClickHouse/pull/32209) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Fix flaky fileLog test (probably) [#32209](https://github.com/ClickHouse/ClickHouse/pull/32209) ([Kseniia Sumarokova](https://github.com/kssenii)). * Fix jemalloc under osx [#32219](https://github.com/ClickHouse/ClickHouse/pull/32219) ([Azat Khuzhin](https://github.com/azat)). * Add missing timezones to some tests [#32222](https://github.com/ClickHouse/ClickHouse/pull/32222) ([Kseniia Sumarokova](https://github.com/kssenii)). * Fix versioning of aggregate functions (fixes performance tests) [#32236](https://github.com/ClickHouse/ClickHouse/pull/32236) ([Azat Khuzhin](https://github.com/azat)). -* Disable window view tests temporarily because still flacky [#32257](https://github.com/ClickHouse/ClickHouse/pull/32257) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Disable window view tests temporarily because still flaky [#32257](https://github.com/ClickHouse/ClickHouse/pull/32257) ([Kseniia Sumarokova](https://github.com/kssenii)). * Fix typo in tupleToNameValuePairs doc [#32262](https://github.com/ClickHouse/ClickHouse/pull/32262) ([Vladimir C](https://github.com/vdimir)). * Fix possible Pipeline stuck in case of StrictResize processor. [#32270](https://github.com/ClickHouse/ClickHouse/pull/32270) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). * Fix possible crash in DataTypeAggregateFunction [#32287](https://github.com/ClickHouse/ClickHouse/pull/32287) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). diff --git a/docs/changelogs/v21.5.1.6601-prestable.md b/docs/changelogs/v21.5.1.6601-prestable.md index b7dd8ae87c0..69ea9cb8d0e 100644 --- a/docs/changelogs/v21.5.1.6601-prestable.md +++ b/docs/changelogs/v21.5.1.6601-prestable.md @@ -158,7 +158,7 @@ sidebar_label: 2022 * MemoryStorage sync comments and code [#22721](https://github.com/ClickHouse/ClickHouse/pull/22721) ([Maksim Kita](https://github.com/kitaisreal)). * Fix potential segfault on Keeper startup [#22743](https://github.com/ClickHouse/ClickHouse/pull/22743) ([alesapin](https://github.com/alesapin)). * Avoid using harmful function rand() [#22744](https://github.com/ClickHouse/ClickHouse/pull/22744) ([Amos Bird](https://github.com/amosbird)). -* Fix flacky hedged tests [#22746](https://github.com/ClickHouse/ClickHouse/pull/22746) ([Kruglov Pavel](https://github.com/Avogar)). +* Fix flaky hedged tests [#22746](https://github.com/ClickHouse/ClickHouse/pull/22746) ([Kruglov Pavel](https://github.com/Avogar)). * add more messages when flushing the logs [#22761](https://github.com/ClickHouse/ClickHouse/pull/22761) ([Alexander Kuzmenkov](https://github.com/akuzm)). * Moved BorrowedObjectPool to common [#22764](https://github.com/ClickHouse/ClickHouse/pull/22764) ([Maksim Kita](https://github.com/kitaisreal)). * Functions ExternalDictionaries standardize exception throw [#22821](https://github.com/ClickHouse/ClickHouse/pull/22821) ([Maksim Kita](https://github.com/kitaisreal)). diff --git a/docs/changelogs/v21.6.9.7-stable.md b/docs/changelogs/v21.6.9.7-stable.md index 0a989e4d6b7..533c58badac 100644 --- a/docs/changelogs/v21.6.9.7-stable.md +++ b/docs/changelogs/v21.6.9.7-stable.md @@ -55,7 +55,7 @@ sidebar_label: 2022 * Try fix rabbitmq tests [#26826](https://github.com/ClickHouse/ClickHouse/pull/26826) ([Kseniia Sumarokova](https://github.com/kssenii)). * One more library bridge fix [#26873](https://github.com/ClickHouse/ClickHouse/pull/26873) ([Kseniia Sumarokova](https://github.com/kssenii)). * Update PVS checksum [#27317](https://github.com/ClickHouse/ClickHouse/pull/27317) ([Alexander Tokmakov](https://github.com/tavplubix)). -* Fix flacky test [#27383](https://github.com/ClickHouse/ClickHouse/pull/27383) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Fix flaky test [#27383](https://github.com/ClickHouse/ClickHouse/pull/27383) ([Kseniia Sumarokova](https://github.com/kssenii)). * Fix throw without exception in MySQL source. [#28027](https://github.com/ClickHouse/ClickHouse/pull/28027) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). * Fix race between REPLACE PARTITION and MOVE PARTITION [#28035](https://github.com/ClickHouse/ClickHouse/pull/28035) ([Alexander Tokmakov](https://github.com/tavplubix)). * Follow-up to [#28016](https://github.com/ClickHouse/ClickHouse/issues/28016) [#28036](https://github.com/ClickHouse/ClickHouse/pull/28036) ([Alexander Tokmakov](https://github.com/tavplubix)). diff --git a/docs/changelogs/v21.7.9.7-stable.md b/docs/changelogs/v21.7.9.7-stable.md index 7aaab54af6b..684d0e8995e 100644 --- a/docs/changelogs/v21.7.9.7-stable.md +++ b/docs/changelogs/v21.7.9.7-stable.md @@ -35,7 +35,7 @@ sidebar_label: 2022 #### NOT FOR CHANGELOG / INSIGNIFICANT * Fix prometheus metric name [#26140](https://github.com/ClickHouse/ClickHouse/pull/26140) ([Vladimir C](https://github.com/vdimir)). -* Fix flacky test [#27383](https://github.com/ClickHouse/ClickHouse/pull/27383) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Fix flaky test [#27383](https://github.com/ClickHouse/ClickHouse/pull/27383) ([Kseniia Sumarokova](https://github.com/kssenii)). * Fix throw without exception in MySQL source. [#28027](https://github.com/ClickHouse/ClickHouse/pull/28027) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). * Fix race between REPLACE PARTITION and MOVE PARTITION [#28035](https://github.com/ClickHouse/ClickHouse/pull/28035) ([Alexander Tokmakov](https://github.com/tavplubix)). * Follow-up to [#28016](https://github.com/ClickHouse/ClickHouse/issues/28016) [#28036](https://github.com/ClickHouse/ClickHouse/pull/28036) ([Alexander Tokmakov](https://github.com/tavplubix)). diff --git a/docs/changelogs/v21.8.1.7409-prestable.md b/docs/changelogs/v21.8.1.7409-prestable.md index cb6ab82b30f..6ef2f1b50d2 100644 --- a/docs/changelogs/v21.8.1.7409-prestable.md +++ b/docs/changelogs/v21.8.1.7409-prestable.md @@ -101,7 +101,7 @@ sidebar_label: 2022 * Separate log files for separate runs in stress test [#25741](https://github.com/ClickHouse/ClickHouse/pull/25741) ([Alexey Milovidov](https://github.com/alexey-milovidov)). * Fix slow performance test [#25742](https://github.com/ClickHouse/ClickHouse/pull/25742) ([Alexey Milovidov](https://github.com/alexey-milovidov)). * DatabaseAtomic EXCHANGE DICTIONARIES fix test [#25753](https://github.com/ClickHouse/ClickHouse/pull/25753) ([Maksim Kita](https://github.com/kitaisreal)). -* Try fix flacky rabbitmq test [#25756](https://github.com/ClickHouse/ClickHouse/pull/25756) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Try fix flaky rabbitmq test [#25756](https://github.com/ClickHouse/ClickHouse/pull/25756) ([Kseniia Sumarokova](https://github.com/kssenii)). * Add a test for [#13993](https://github.com/ClickHouse/ClickHouse/issues/13993) [#25758](https://github.com/ClickHouse/ClickHouse/pull/25758) ([Alexey Milovidov](https://github.com/alexey-milovidov)). * Set follow-fork-mode child for gdb in stress/fasttest/fuzzer [#25769](https://github.com/ClickHouse/ClickHouse/pull/25769) ([Azat Khuzhin](https://github.com/azat)). * Ignore TOO_DEEP_RECURSION server exception during fuzzing [#25770](https://github.com/ClickHouse/ClickHouse/pull/25770) ([Azat Khuzhin](https://github.com/azat)). diff --git a/docs/changelogs/v21.8.5.7-lts.md b/docs/changelogs/v21.8.5.7-lts.md index fa459e093f7..4d0727e362c 100644 --- a/docs/changelogs/v21.8.5.7-lts.md +++ b/docs/changelogs/v21.8.5.7-lts.md @@ -40,7 +40,7 @@ sidebar_label: 2022 * Fix several bugs in ZooKeeper snapshots deserialization [#26127](https://github.com/ClickHouse/ClickHouse/pull/26127) ([alesapin](https://github.com/alesapin)). * Fix prometheus metric name [#26140](https://github.com/ClickHouse/ClickHouse/pull/26140) ([Vladimir C](https://github.com/vdimir)). -* Fix flacky test [#27383](https://github.com/ClickHouse/ClickHouse/pull/27383) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Fix flaky test [#27383](https://github.com/ClickHouse/ClickHouse/pull/27383) ([Kseniia Sumarokova](https://github.com/kssenii)). * Fix throw without exception in MySQL source. [#28027](https://github.com/ClickHouse/ClickHouse/pull/28027) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). * Fix race between REPLACE PARTITION and MOVE PARTITION [#28035](https://github.com/ClickHouse/ClickHouse/pull/28035) ([Alexander Tokmakov](https://github.com/tavplubix)). * Follow-up to [#28016](https://github.com/ClickHouse/ClickHouse/issues/28016) [#28036](https://github.com/ClickHouse/ClickHouse/pull/28036) ([Alexander Tokmakov](https://github.com/tavplubix)). diff --git a/docs/changelogs/v21.9.1.8000-prestable.md b/docs/changelogs/v21.9.1.8000-prestable.md index bc921a68693..70ae3697e92 100644 --- a/docs/changelogs/v21.9.1.8000-prestable.md +++ b/docs/changelogs/v21.9.1.8000-prestable.md @@ -346,7 +346,7 @@ sidebar_label: 2022 * Update PVS checksum [#27317](https://github.com/ClickHouse/ClickHouse/pull/27317) ([Alexander Tokmakov](https://github.com/tavplubix)). * Fix 01300_client_save_history_when_terminated_long [#27324](https://github.com/ClickHouse/ClickHouse/pull/27324) ([Raúl Marín](https://github.com/Algunenano)). * Try update contrib/zlib-ng [#27327](https://github.com/ClickHouse/ClickHouse/pull/27327) ([Ilya Yatsishin](https://github.com/qoega)). -* Fix flacky test [#27383](https://github.com/ClickHouse/ClickHouse/pull/27383) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Fix flaky test [#27383](https://github.com/ClickHouse/ClickHouse/pull/27383) ([Kseniia Sumarokova](https://github.com/kssenii)). * Add and check system.mutations for database filter [#27384](https://github.com/ClickHouse/ClickHouse/pull/27384) ([Azat Khuzhin](https://github.com/azat)). * Correct the key data type used in mapContains [#27423](https://github.com/ClickHouse/ClickHouse/pull/27423) ([Fuwang Hu](https://github.com/fuwhu)). * Fix tests for WithMergeableStateAfterAggregationAndLimit [#27424](https://github.com/ClickHouse/ClickHouse/pull/27424) ([Azat Khuzhin](https://github.com/azat)). diff --git a/docs/changelogs/v22.1.1.2542-prestable.md b/docs/changelogs/v22.1.1.2542-prestable.md index cacd13c1e12..3b0422abb11 100644 --- a/docs/changelogs/v22.1.1.2542-prestable.md +++ b/docs/changelogs/v22.1.1.2542-prestable.md @@ -398,7 +398,7 @@ sidebar_label: 2022 * test for [#24410](https://github.com/ClickHouse/ClickHouse/issues/24410) [#33265](https://github.com/ClickHouse/ClickHouse/pull/33265) ([Denny Crane](https://github.com/den-crane)). * Wait for RabbitMQ container to actually start when it was restarted in test on purpose [#33266](https://github.com/ClickHouse/ClickHouse/pull/33266) ([Kseniia Sumarokova](https://github.com/kssenii)). * Mark max_alter_threads as obsolete [#33268](https://github.com/ClickHouse/ClickHouse/pull/33268) ([Denny Crane](https://github.com/den-crane)). -* Fix azure tests flackyness because of azure server closing connection [#33269](https://github.com/ClickHouse/ClickHouse/pull/33269) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Fix azure tests flakyness because of azure server closing connection [#33269](https://github.com/ClickHouse/ClickHouse/pull/33269) ([Kseniia Sumarokova](https://github.com/kssenii)). * Test for [#26920](https://github.com/ClickHouse/ClickHouse/issues/26920) [#33272](https://github.com/ClickHouse/ClickHouse/pull/33272) ([Denny Crane](https://github.com/den-crane)). * Fix test_storage_kafka failures by adjusting retention.ms [#33278](https://github.com/ClickHouse/ClickHouse/pull/33278) ([Azat Khuzhin](https://github.com/azat)). * Disable FunctionConvertFromString::canBeExecutedOnDefaultArguments [#33286](https://github.com/ClickHouse/ClickHouse/pull/33286) ([Vladimir C](https://github.com/vdimir)). @@ -447,7 +447,7 @@ sidebar_label: 2022 * Update mongodb.md [#33585](https://github.com/ClickHouse/ClickHouse/pull/33585) ([Kseniia Sumarokova](https://github.com/kssenii)). * Restore existing static builds links [#33597](https://github.com/ClickHouse/ClickHouse/pull/33597) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). * Fix pylint for run_check.py [#33600](https://github.com/ClickHouse/ClickHouse/pull/33600) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). -* Fix flacky test_dictionaries_postgresql/ [#33601](https://github.com/ClickHouse/ClickHouse/pull/33601) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Fix flaky test_dictionaries_postgresql/ [#33601](https://github.com/ClickHouse/ClickHouse/pull/33601) ([Kseniia Sumarokova](https://github.com/kssenii)). * Make ZooKeeper client better interpret keeper server connection reject [#33602](https://github.com/ClickHouse/ClickHouse/pull/33602) ([alesapin](https://github.com/alesapin)). * Fix broken workflow dependencies [#33608](https://github.com/ClickHouse/ClickHouse/pull/33608) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). * Force rebuild images in CI [#33609](https://github.com/ClickHouse/ClickHouse/pull/33609) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). diff --git a/docs/changelogs/v22.4.1.2305-prestable.md b/docs/changelogs/v22.4.1.2305-prestable.md index b277137ca7e..e8304e6505f 100644 --- a/docs/changelogs/v22.4.1.2305-prestable.md +++ b/docs/changelogs/v22.4.1.2305-prestable.md @@ -410,7 +410,7 @@ sidebar_label: 2022 * Fix mongodb test with new cert [#36161](https://github.com/ClickHouse/ClickHouse/pull/36161) ([alesapin](https://github.com/alesapin)). * Some fixes for ReplicatedMergeTree [#36163](https://github.com/ClickHouse/ClickHouse/pull/36163) ([Alexander Tokmakov](https://github.com/tavplubix)). * clickhouse-client: properly cancel query in case of error during formatting data [#36164](https://github.com/ClickHouse/ClickHouse/pull/36164) ([Azat Khuzhin](https://github.com/azat)). -* Fix flacky test 01161_all_system_tables under s3 storage [#36175](https://github.com/ClickHouse/ClickHouse/pull/36175) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Fix flaky test 01161_all_system_tables under s3 storage [#36175](https://github.com/ClickHouse/ClickHouse/pull/36175) ([Kseniia Sumarokova](https://github.com/kssenii)). * Revert "Fix possible mutation stuck due to race with DROP_RANGE" [#36190](https://github.com/ClickHouse/ClickHouse/pull/36190) ([Azat Khuzhin](https://github.com/azat)). * Use atomic instead of mutex + condvar in ParallelReadBuffer [#36192](https://github.com/ClickHouse/ClickHouse/pull/36192) ([Kruglov Pavel](https://github.com/Avogar)). * Follow-up to [#36138](https://github.com/ClickHouse/ClickHouse/issues/36138) [#36194](https://github.com/ClickHouse/ClickHouse/pull/36194) ([Alexander Tokmakov](https://github.com/tavplubix)). diff --git a/docs/changelogs/v23.4.1.1943-stable.md b/docs/changelogs/v23.4.1.1943-stable.md index ea16f5856be..34590ba9d37 100644 --- a/docs/changelogs/v23.4.1.1943-stable.md +++ b/docs/changelogs/v23.4.1.1943-stable.md @@ -321,7 +321,7 @@ sidebar_label: 2023 * Add a test for [#38128](https://github.com/ClickHouse/ClickHouse/issues/38128) [#48817](https://github.com/ClickHouse/ClickHouse/pull/48817) ([Alexey Milovidov](https://github.com/alexey-milovidov)). * Remove excessive logging [#48826](https://github.com/ClickHouse/ClickHouse/pull/48826) ([Alexey Milovidov](https://github.com/alexey-milovidov)). * remove duplicate indentwith in clang-format [#48834](https://github.com/ClickHouse/ClickHouse/pull/48834) ([cluster](https://github.com/infdahai)). -* Try fix flacky test_concurrent_alter_move_and_drop [#48843](https://github.com/ClickHouse/ClickHouse/pull/48843) ([Sergei Trifonov](https://github.com/serxa)). +* Try fix flaky test_concurrent_alter_move_and_drop [#48843](https://github.com/ClickHouse/ClickHouse/pull/48843) ([Sergei Trifonov](https://github.com/serxa)). * fix the race wait loading parts [#48844](https://github.com/ClickHouse/ClickHouse/pull/48844) ([Sema Checherinda](https://github.com/CheSema)). * suppress assert of progress for test_system_replicated_fetches [#48856](https://github.com/ClickHouse/ClickHouse/pull/48856) ([Han Fei](https://github.com/hanfei1991)). * Fix: do not run test_store_cleanup_disk_s3 in parallel [#48863](https://github.com/ClickHouse/ClickHouse/pull/48863) ([Igor Nikonov](https://github.com/devcrafter)). @@ -372,4 +372,3 @@ sidebar_label: 2023 * suppress two timeout tests [#49175](https://github.com/ClickHouse/ClickHouse/pull/49175) ([Han Fei](https://github.com/hanfei1991)). * Document makeDateTime() and its variants [#49183](https://github.com/ClickHouse/ClickHouse/pull/49183) ([Robert Schulze](https://github.com/rschu1ze)). * Fix after [#49110](https://github.com/ClickHouse/ClickHouse/issues/49110) [#49206](https://github.com/ClickHouse/ClickHouse/pull/49206) ([Kseniia Sumarokova](https://github.com/kssenii)). - diff --git a/docs/changelogs/v23.6.1.1524-stable.md b/docs/changelogs/v23.6.1.1524-stable.md index b91c5340789..0de9ab37653 100644 --- a/docs/changelogs/v23.6.1.1524-stable.md +++ b/docs/changelogs/v23.6.1.1524-stable.md @@ -263,7 +263,7 @@ sidebar_label: 2023 * Fix broken labeling for `manual approve` [#51405](https://github.com/ClickHouse/ClickHouse/pull/51405) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). * Fix parts lifetime in `MergeTreeTransaction` [#51407](https://github.com/ClickHouse/ClickHouse/pull/51407) ([Alexander Tokmakov](https://github.com/tavplubix)). * Fix flaky test test_skip_empty_files [#51409](https://github.com/ClickHouse/ClickHouse/pull/51409) ([Kruglov Pavel](https://github.com/Avogar)). -* fix flacky test test_profile_events_s3 [#51412](https://github.com/ClickHouse/ClickHouse/pull/51412) ([Sema Checherinda](https://github.com/CheSema)). +* fix flaky test test_profile_events_s3 [#51412](https://github.com/ClickHouse/ClickHouse/pull/51412) ([Sema Checherinda](https://github.com/CheSema)). * Update README.md [#51413](https://github.com/ClickHouse/ClickHouse/pull/51413) ([Tyler Hannan](https://github.com/tylerhannan)). * Replace try/catch logic in hasTokenOrNull() by something more lightweight [#51425](https://github.com/ClickHouse/ClickHouse/pull/51425) ([Robert Schulze](https://github.com/rschu1ze)). * Add retries to `tlsv1_3` tests [#51434](https://github.com/ClickHouse/ClickHouse/pull/51434) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). diff --git a/docs/zh/changelog/index.md b/docs/zh/changelog/index.md index cd77a8c03cf..fb50dfcee85 100644 --- a/docs/zh/changelog/index.md +++ b/docs/zh/changelog/index.md @@ -252,7 +252,7 @@ sidebar_label: "\u53D8\u66F4\u65E5\u5FD7" - 抑制MSan下的一些测试失败。 [#8780](https://github.com/ClickHouse/ClickHouse/pull/8780) ([Alexander Kuzmenkov](https://github.com/akuzm)) - 加速 “exception while insert” 测试 此测试通常在具有复盖率的调试版本中超时。 [#8711](https://github.com/ClickHouse/ClickHouse/pull/8711) ([阿列克谢-米洛维多夫](https://github.com/alexey-milovidov)) - 更新 `libcxx` 和 `libcxxabi` 为了主人 在准备 [#9304](https://github.com/ClickHouse/ClickHouse/issues/9304) [#9308](https://github.com/ClickHouse/ClickHouse/pull/9308) ([阿列克谢-米洛维多夫](https://github.com/alexey-milovidov)) -- 修复flacky测试 `00910_zookeeper_test_alter_compression_codecs`. [#9525](https://github.com/ClickHouse/ClickHouse/pull/9525) ([阿列克谢-米洛维多夫](https://github.com/alexey-milovidov)) +- 修复flaky测试 `00910_zookeeper_test_alter_compression_codecs`. [#9525](https://github.com/ClickHouse/ClickHouse/pull/9525) ([阿列克谢-米洛维多夫](https://github.com/alexey-milovidov)) - 清理重复的链接器标志。 确保链接器不会查找意想不到的符号。 [#9433](https://github.com/ClickHouse/ClickHouse/pull/9433) ([阿莫斯鸟](https://github.com/amosbird)) - 添加 `clickhouse-odbc` 驱动程序进入测试图像。 这允许通过自己的ODBC驱动程序测试ClickHouse与ClickHouse的交互。 [#9348](https://github.com/ClickHouse/ClickHouse/pull/9348) ([filimonov](https://github.com/filimonov)) - 修复单元测试中的几个错误。 [#9047](https://github.com/ClickHouse/ClickHouse/pull/9047) ([阿利沙平](https://github.com/alesapin)) diff --git a/src/Common/SystemLogBase.cpp b/src/Common/SystemLogBase.cpp index a9307c3be99..7d2c15714e2 100644 --- a/src/Common/SystemLogBase.cpp +++ b/src/Common/SystemLogBase.cpp @@ -65,7 +65,7 @@ void SystemLogQueue::push(LogElement&& element) /// Memory can be allocated while resizing on queue.push_back. /// The size of allocation can be in order of a few megabytes. /// But this should not be accounted for query memory usage. - /// Otherwise the tests like 01017_uniqCombined_memory_usage.sql will be flacky. + /// Otherwise the tests like 01017_uniqCombined_memory_usage.sql will be flaky. MemoryTrackerBlockerInThread temporarily_disable_memory_tracker; /// Should not log messages under mutex. From bd5022432909bcb633f31617a764ca223ad585d5 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sun, 4 Aug 2024 20:57:22 +0200 Subject: [PATCH 202/227] Whitespaces --- docs/changelogs/v23.4.1.1943-stable.md | 1 - src/Processors/Chunk.h | 2 +- tests/queries/0_stateless/02099_tsv_raw_format.sh | 7 +++---- 3 files changed, 4 insertions(+), 6 deletions(-) diff --git a/docs/changelogs/v23.4.1.1943-stable.md b/docs/changelogs/v23.4.1.1943-stable.md index ea16f5856be..6ab6d8b457a 100644 --- a/docs/changelogs/v23.4.1.1943-stable.md +++ b/docs/changelogs/v23.4.1.1943-stable.md @@ -372,4 +372,3 @@ sidebar_label: 2023 * suppress two timeout tests [#49175](https://github.com/ClickHouse/ClickHouse/pull/49175) ([Han Fei](https://github.com/hanfei1991)). * Document makeDateTime() and its variants [#49183](https://github.com/ClickHouse/ClickHouse/pull/49183) ([Robert Schulze](https://github.com/rschu1ze)). * Fix after [#49110](https://github.com/ClickHouse/ClickHouse/issues/49110) [#49206](https://github.com/ClickHouse/ClickHouse/pull/49206) ([Kseniia Sumarokova](https://github.com/kssenii)). - diff --git a/src/Processors/Chunk.h b/src/Processors/Chunk.h index 1348966c0d3..f45e2c4619e 100644 --- a/src/Processors/Chunk.h +++ b/src/Processors/Chunk.h @@ -22,7 +22,7 @@ public: }; -template +template class ChunkInfoCloneable : public ChunkInfo { public: diff --git a/tests/queries/0_stateless/02099_tsv_raw_format.sh b/tests/queries/0_stateless/02099_tsv_raw_format.sh index 16b695e4037..026607ac6d5 100755 --- a/tests/queries/0_stateless/02099_tsv_raw_format.sh +++ b/tests/queries/0_stateless/02099_tsv_raw_format.sh @@ -14,7 +14,7 @@ do echo $format $CLICKHOUSE_CLIENT -q "INSERT INTO test_02099 SELECT number, toString(number), toDate(number) FROM numbers(3)" $CLICKHOUSE_CLIENT -q "SELECT * FROM test_02099 FORMAT $format" - + $CLICKHOUSE_CLIENT -q "SELECT * FROM test_02099 FORMAT $format" | $CLICKHOUSE_CLIENT -q "INSERT INTO test_02099 FORMAT $format" $CLICKHOUSE_CLIENT -q "SELECT * FROM test_02099" @@ -49,13 +49,12 @@ $CLICKHOUSE_CLIENT -q "DROP TABLE test_nullable_string_02099" $CLICKHOUSE_CLIENT -q "DROP TABLE IF EXISTS test_parallel_parsing_02099" $CLICKHOUSE_CLIENT -q "CREATE TABLE test_parallel_parsing_02099 (x UInt64, a Array(UInt64), s String) ENGINE=Memory()"; -$CLICKHOUSE_CLIENT -q "SELECT number AS x, range(number % 50) AS a, toString(a) AS s FROM numbers(1000000) FORMAT TSVRaw" | $CLICKHOUSE_CLIENT --input_format_parallel_parsing=0 -q "INSERT INTO test_parallel_parsing_02099 FORMAT TSVRaw" +$CLICKHOUSE_CLIENT -q "SELECT number AS x, range(number % 50) AS a, toString(a) AS s FROM numbers(1000000) FORMAT TSVRaw" | $CLICKHOUSE_CLIENT --input_format_parallel_parsing=0 -q "INSERT INTO test_parallel_parsing_02099 FORMAT TSVRaw" $CLICKHOUSE_CLIENT -q "SELECT * FROM test_parallel_parsing_02099 ORDER BY x" | md5sum $CLICKHOUSE_CLIENT -q "TRUNCATE TABLE test_parallel_parsing_02099" -$CLICKHOUSE_CLIENT -q "SELECT number AS x, range(number % 50) AS a, toString(a) AS s FROM numbers(1000000) FORMAT TSVRaw" | $CLICKHOUSE_CLIENT --input_format_parallel_parsing=1 -q "INSERT INTO test_parallel_parsing_02099 FORMAT TSVRaw" +$CLICKHOUSE_CLIENT -q "SELECT number AS x, range(number % 50) AS a, toString(a) AS s FROM numbers(1000000) FORMAT TSVRaw" | $CLICKHOUSE_CLIENT --input_format_parallel_parsing=1 -q "INSERT INTO test_parallel_parsing_02099 FORMAT TSVRaw" $CLICKHOUSE_CLIENT -q "SELECT * FROM test_parallel_parsing_02099 ORDER BY x" | md5sum $CLICKHOUSE_CLIENT -q "DROP TABLE test_parallel_parsing_02099" - From a49924b3834d577827279f6278e5df21e3065035 Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Sun, 4 Aug 2024 13:27:06 +0200 Subject: [PATCH 203/227] tests: avoid leaving leftovers after test_storage_mongodb (fixes flaky chec) CI: https://s3.amazonaws.com/clickhouse-test-reports/66671/2f00c962711e13ca00af324366421fe4593b4ce6/integration_tests_flaky_check__asan_.html Signed-off-by: Azat Khuzhin --- tests/integration/test_storage_mongodb/test.py | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/tests/integration/test_storage_mongodb/test.py b/tests/integration/test_storage_mongodb/test.py index 1a1a790e8e8..6e2d15b03be 100644 --- a/tests/integration/test_storage_mongodb/test.py +++ b/tests/integration/test_storage_mongodb/test.py @@ -410,6 +410,7 @@ def test_no_credentials(started_cluster): simple_mongo_table.insert_many(data) node = started_cluster.instances["node"] + node.query("drop table if exists simple_mongo_table_2") node.query( "create table simple_mongo_table_2(key UInt64, data String) engine = MongoDB('mongo2:27017', 'test', 'simple_table', '', '')" ) @@ -439,10 +440,13 @@ def test_auth_source(started_cluster): simple_mongo_table.insert_many(data) node = started_cluster.instances["node"] + node.query("drop table if exists simple_mongo_table_fail") node.query( "create table simple_mongo_table_fail(key UInt64, data String) engine = MongoDB('mongo2:27017', 'test', 'simple_table', 'root', 'clickhouse')" ) node.query_and_get_error("SELECT count() FROM simple_mongo_table_fail") + + node.query("drop table if exists simple_mongo_table_ok") node.query( "create table simple_mongo_table_ok(key UInt64, data String) engine = MongoDB('mongo2:27017', 'test', 'simple_table', 'root', 'clickhouse', 'authSource=admin')" ) From c200f437746ddbd03e5ce7c7ad9d9613015ce929 Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Sun, 4 Aug 2024 16:00:46 +0200 Subject: [PATCH 204/227] ci: fix stateless runner to correctly gather artifacts when server failed Right now it simply does not work due to "set -e", with it you cannot use "foo=$(false)" since bash will break execution after, rewrite it to a plain shell, with "if". Also use ZSTD everywhere (ugh) Signed-off-by: Azat Khuzhin --- docker/test/stateless/run.sh | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/docker/test/stateless/run.sh b/docker/test/stateless/run.sh index da17b82d91b..5c15c05652b 100755 --- a/docker/test/stateless/run.sh +++ b/docker/test/stateless/run.sh @@ -299,22 +299,22 @@ stop_logs_replication failed_to_save_logs=0 for table in query_log zookeeper_log trace_log transactions_info_log metric_log blob_storage_log error_log do - err=$(clickhouse-client -q "select * from system.$table into outfile '/test_output/$table.tsv.gz' format TSVWithNamesAndTypes") - echo "$err" - [[ "0" != "${#err}" ]] && failed_to_save_logs=1 + if ! clickhouse-client -q "select * from system.$table into outfile '/test_output/$table.tsv.zst' format TSVWithNamesAndTypes"; then + failed_to_save_logs=1 + fi if [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then - err=$( { clickhouse-client --port 19000 -q "select * from system.$table format TSVWithNamesAndTypes" | zstd --threads=0 > /test_output/$table.1.tsv.zst; } 2>&1 ) - echo "$err" - [[ "0" != "${#err}" ]] && failed_to_save_logs=1 - err=$( { clickhouse-client --port 29000 -q "select * from system.$table format TSVWithNamesAndTypes" | zstd --threads=0 > /test_output/$table.2.tsv.zst; } 2>&1 ) - echo "$err" - [[ "0" != "${#err}" ]] && failed_to_save_logs=1 + if ! clickhouse-client --port 19000 -q "select * from system.$table into outfile '/test_output/$table.1.tsv.zst' format TSVWithNamesAndTypes"; then + failed_to_save_logs=1 + fi + if ! clickhouse-client --port 29000 -q "select * from system.$table into outfile '/test_output/$table.2.tsv.zst' format TSVWithNamesAndTypes"; then + failed_to_save_logs=1 + fi fi if [[ "$USE_SHARED_CATALOG" -eq 1 ]]; then - err=$( { clickhouse-client --port 19000 -q "select * from system.$table format TSVWithNamesAndTypes" | zstd --threads=0 > /test_output/$table.1.tsv.zst; } 2>&1 ) - echo "$err" - [[ "0" != "${#err}" ]] && failed_to_save_logs=1 + if ! clickhouse-client --port 29000 -q "select * from system.$table into outfile '/test_output/$table.2.tsv.zst' format TSVWithNamesAndTypes"; then + failed_to_save_logs=1 + fi fi done From 43cf85ef2841190a5d12a592a65a5181bb110661 Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Sun, 4 Aug 2024 16:09:48 +0200 Subject: [PATCH 205/227] ci: collect basic issues in stateless tests (dmesg, fatal and similar) Signed-off-by: Azat Khuzhin --- docker/test/stateless/run.sh | 9 +++++++++ docker/test/stateless/stress_tests.lib | 2 +- 2 files changed, 10 insertions(+), 1 deletion(-) diff --git a/docker/test/stateless/run.sh b/docker/test/stateless/run.sh index 5c15c05652b..b352539cc1a 100755 --- a/docker/test/stateless/run.sh +++ b/docker/test/stateless/run.sh @@ -3,6 +3,12 @@ # shellcheck disable=SC1091 source /setup_export_logs.sh +# shellcheck source=../stateless/stress_tests.lib +source /stress_tests.lib + +# Avoid overlaps with previous runs +dmesg --clear + # fail on errors, verbose and export all env variables set -e -x -a @@ -420,4 +426,7 @@ if [[ "$USE_SHARED_CATALOG" -eq 1 ]]; then tar -chf /test_output/coordination1.tar /var/lib/clickhouse1/coordination ||: fi +# Grep logs for sanitizer asserts, crashes and other critical errors +check_logs_for_critical_errors + collect_core_dumps diff --git a/docker/test/stateless/stress_tests.lib b/docker/test/stateless/stress_tests.lib index 682da1df837..36782101fa7 100644 --- a/docker/test/stateless/stress_tests.lib +++ b/docker/test/stateless/stress_tests.lib @@ -242,7 +242,7 @@ function check_server_start() function check_logs_for_critical_errors() { # Sanitizer asserts - sed -n '/WARNING:.*anitizer/,/^$/p' /var/log/clickhouse-server/stderr.log >> /test_output/tmp + sed -n '/WARNING:.*anitizer/,/^$/p' /var/log/clickhouse-server/stderr*.log >> /test_output/tmp rg -Fav -e "ASan doesn't fully support makecontext/swapcontext functions" -e "DB::Exception" /test_output/tmp > /dev/null \ && echo -e "Sanitizer assert (in stderr.log)$FAIL$(head_escaped /test_output/tmp)" >> /test_output/test_results.tsv \ || echo -e "No sanitizer asserts$OK" >> /test_output/test_results.tsv From 83be27cdf59ec5886abc23441f5cf92740b04c57 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sun, 4 Aug 2024 23:34:01 +0200 Subject: [PATCH 206/227] Fix test `02833_concurrent_sessions`, Fix test `02835_drop_user_during_session` --- .../0_stateless/02833_concurrent_sessions.sh | 18 ++++++++++++- .../02835_drop_user_during_session.sh | 26 +++++++++++++++---- 2 files changed, 38 insertions(+), 6 deletions(-) diff --git a/tests/queries/0_stateless/02833_concurrent_sessions.sh b/tests/queries/0_stateless/02833_concurrent_sessions.sh index 846661cfeed..feaff3a38a3 100755 --- a/tests/queries/0_stateless/02833_concurrent_sessions.sh +++ b/tests/queries/0_stateless/02833_concurrent_sessions.sh @@ -137,7 +137,23 @@ ${CLICKHOUSE_CLIENT} -q "SELECT count(*) FROM system.session_log WHERE user IN ( for user in "${ALL_USERS[@]}"; do ${CLICKHOUSE_CLIENT} -q "DROP USER ${user}" echo "Corresponding LoginSuccess/Logout" - ${CLICKHOUSE_CLIENT} -q "SELECT COUNT(*) FROM (SELECT ${SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = '${user}' AND type = 'LoginSuccess' INTERSECT SELECT ${SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = '${user}' AND type = 'Logout')" + + # The client can exit sooner than the server records its disconnection and closes the session. + # When the client disconnects, two processes happen at the same time and are in the race condition: + # - the client application exits and returns control to the shell; + # - the server closes the session and records the logout event to the session log. + # We cannot expect that after the control is returned to the shell, the server records the logout event. + while true + do + [[ 3 -eq $(${CLICKHOUSE_CLIENT} -q " + SELECT COUNT(*) FROM ( + SELECT ${SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = '${user}' AND type = 'LoginSuccess' + INTERSECT + SELECT ${SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = '${user}' AND type = 'Logout' + )") ]] && echo 3 && break; + sleep 0.1 + done + echo "LoginFailure" ${CLICKHOUSE_CLIENT} -q "SELECT COUNT(*) FROM system.session_log WHERE user = '${user}' AND type = 'LoginFailure'" done diff --git a/tests/queries/0_stateless/02835_drop_user_during_session.sh b/tests/queries/0_stateless/02835_drop_user_during_session.sh index 347ebd22f96..c32003a2a11 100755 --- a/tests/queries/0_stateless/02835_drop_user_during_session.sh +++ b/tests/queries/0_stateless/02835_drop_user_during_session.sh @@ -24,7 +24,7 @@ function http_session() ${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}&user=${user}&password=pass" -d "SELECT COUNT(*) FROM system.numbers" } -function http_with_session_id_session() +function http_with_session_id_session() { local user=$1 ${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}&user=${user}&password=pass" -d "SELECT COUNT(*) FROM system.numbers" @@ -104,11 +104,27 @@ wait ${CLICKHOUSE_CLIENT} -q "SYSTEM FLUSH LOGS" -echo "port_0_sessions:" +echo "port_0_sessions:" ${CLICKHOUSE_CLIENT} -q "SELECT count(*) FROM system.session_log WHERE user = '${TEST_USER}' AND client_port = 0" echo "address_0_sessions:" ${CLICKHOUSE_CLIENT} -q "SELECT count(*) FROM system.session_log WHERE user = '${TEST_USER}' AND client_address = toIPv6('::')" -echo "Corresponding LoginSuccess/Logout" -${CLICKHOUSE_CLIENT} -q "SELECT COUNT(*) FROM (SELECT ${SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = '${TEST_USER}' AND type = 'LoginSuccess' INTERSECT SELECT ${SESSION_LOG_MATCHING_FIELDS}, FROM system.session_log WHERE user = '${TEST_USER}' AND type = 'Logout')" +echo "Corresponding LoginSuccess/Logout" + +# The client can exit sooner than the server records its disconnection and closes the session. +# When the client disconnects, two processes happen at the same time and are in the race condition: +# - the client application exits and returns control to the shell; +# - the server closes the session and records the logout event to the session log. +# We cannot expect that after the control is returned to the shell, the server records the logout event. +while true +do + [[ 9 -eq $(${CLICKHOUSE_CLIENT} -q " + SELECT COUNT(*) FROM ( + SELECT ${SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = '${TEST_USER}' AND type = 'LoginSuccess' + INTERSECT + SELECT ${SESSION_LOG_MATCHING_FIELDS}, FROM system.session_log WHERE user = '${TEST_USER}' AND type = 'Logout' + )") ]] && echo 9 && break; + sleep 0.1 +done + echo "LoginFailure" -${CLICKHOUSE_CLIENT} -q "SELECT COUNT(*) FROM system.session_log WHERE user = '${TEST_USER}' AND type = 'LoginFailure'" +${CLICKHOUSE_CLIENT} -q "SELECT COUNT(*) FROM system.session_log WHERE user = '${TEST_USER}' AND type = 'LoginFailure'" From 0bdaa57023ef69e49ab5cf9d54ed2e52c1fd2dae Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sun, 4 Aug 2024 23:53:54 +0200 Subject: [PATCH 207/227] Fix diagnostics in the test script --- docker/test/stateless/run.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docker/test/stateless/run.sh b/docker/test/stateless/run.sh index da17b82d91b..bec8b9cc4d1 100755 --- a/docker/test/stateless/run.sh +++ b/docker/test/stateless/run.sh @@ -260,7 +260,7 @@ function run_tests() | ts '%Y-%m-%d %H:%M:%S' \ | tee -a test_output/test_result.txt set -e - DURATION=$((START_TIME - SECONDS)) + DURATION=$((SECONDS - START_TIME)) echo "Elapsed ${DURATION} seconds." if [[ $DURATION -ge $TIMEOUT ]] From a573b2926e13ea31c7947b0429b1b5723c7fb938 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sun, 4 Aug 2024 23:59:01 +0200 Subject: [PATCH 208/227] Fixes for the script --- docker/test/stateless/run.sh | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/docker/test/stateless/run.sh b/docker/test/stateless/run.sh index da17b82d91b..8b9e729970c 100755 --- a/docker/test/stateless/run.sh +++ b/docker/test/stateless/run.sh @@ -72,8 +72,12 @@ if [[ -n "$BUGFIX_VALIDATE_CHECK" ]] && [[ "$BUGFIX_VALIDATE_CHECK" -eq 1 ]]; th remove_keeper_config "latest_logs_cache_size_threshold" "[[:digit:]]\+" fi +export IS_FLAKY_CHECK=0 + # For flaky check we also enable thread fuzzer if [ "$NUM_TRIES" -gt "1" ]; then + export IS_FLAKY_CHECK=1 + export THREAD_FUZZER_CPU_TIME_PERIOD_US=1000 export THREAD_FUZZER_SLEEP_PROBABILITY=0.1 export THREAD_FUZZER_SLEEP_TIME_US_MAX=100000 From 538761b43dbf704d0700548e61b4034ef66c5766 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Mon, 5 Aug 2024 00:02:32 +0200 Subject: [PATCH 209/227] Fix flaky check --- tests/clickhouse-test | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/clickhouse-test b/tests/clickhouse-test index b70dd61a25a..38b0e99760e 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -1232,7 +1232,7 @@ class TestCase: ): return FailureReason.SKIP - elif "no-flaky-check" in tags and (args.test_runs > 1): + elif "no-flaky-check" in tags and (1 == int(os.environ.get("IS_FLAKY_CHECK", 0))): return FailureReason.SKIP elif tags: From 270dddc00fd5533ba24914e7d9ae0aebc50a3fd0 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Mon, 5 Aug 2024 00:22:48 +0200 Subject: [PATCH 210/227] Fix test `02231_bloom_filter_sizing` --- tests/queries/0_stateless/02231_bloom_filter_sizing.reference | 4 ++-- tests/queries/0_stateless/02231_bloom_filter_sizing.sql | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/tests/queries/0_stateless/02231_bloom_filter_sizing.reference b/tests/queries/0_stateless/02231_bloom_filter_sizing.reference index bdba311c092..aabadfc92fd 100644 --- a/tests/queries/0_stateless/02231_bloom_filter_sizing.reference +++ b/tests/queries/0_stateless/02231_bloom_filter_sizing.reference @@ -1,6 +1,6 @@ Bloom filter on sort key -10000 +1000 0 Bloom filter on non-sort key -10000 +1000 0 diff --git a/tests/queries/0_stateless/02231_bloom_filter_sizing.sql b/tests/queries/0_stateless/02231_bloom_filter_sizing.sql index 233e3111067..ee896675d64 100644 --- a/tests/queries/0_stateless/02231_bloom_filter_sizing.sql +++ b/tests/queries/0_stateless/02231_bloom_filter_sizing.sql @@ -12,7 +12,7 @@ INSERT INTO bloom_filter_sizing_pk SELECT number % 100 as key, -- 100 unique keys number as value -- whatever -FROM numbers(1000 * 1000); +FROM numbers(100_000); -- -- Merge everything into a single part @@ -40,7 +40,7 @@ SELECT number % 100 as key1, -- 100 unique keys rand() % 100 as key2, -- 100 unique keys number as value -- whatever -FROM numbers(1000 * 1000); +FROM numbers(100_000); -- -- Merge everything into a single part From 5a860fcc3aaf4e77025cd9286da10d0587134c3f Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Mon, 5 Aug 2024 00:25:56 +0200 Subject: [PATCH 211/227] Update 02099_tsv_raw_format.sh --- tests/queries/0_stateless/02099_tsv_raw_format.sh | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/queries/0_stateless/02099_tsv_raw_format.sh b/tests/queries/0_stateless/02099_tsv_raw_format.sh index 026607ac6d5..a69c96ab613 100755 --- a/tests/queries/0_stateless/02099_tsv_raw_format.sh +++ b/tests/queries/0_stateless/02099_tsv_raw_format.sh @@ -1,4 +1,5 @@ #!/usr/bin/env bash +# Tags: long CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh From 7adeaf9c28016b29453d6be2bca1aa1ef866ecdf Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Sun, 4 Aug 2024 22:35:07 +0000 Subject: [PATCH 212/227] Automatic style fix --- tests/clickhouse-test | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tests/clickhouse-test b/tests/clickhouse-test index 38b0e99760e..907d773337a 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -1232,7 +1232,9 @@ class TestCase: ): return FailureReason.SKIP - elif "no-flaky-check" in tags and (1 == int(os.environ.get("IS_FLAKY_CHECK", 0))): + elif "no-flaky-check" in tags and ( + 1 == int(os.environ.get("IS_FLAKY_CHECK", 0)) + ): return FailureReason.SKIP elif tags: From 6b1e184e12bed759487d89f54f5ac4f269dffda2 Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Sun, 4 Aug 2024 22:50:46 +0200 Subject: [PATCH 213/227] Print original query for AST formatting check on CI It may be tricky to understand the root cause of the AST formatting issue in case of syntax error, so add one knob to control this - debug_ast_formatting_print_original_query. And CI contains core dumps anyway, so let's enable for CI. P.S. There was concern from @al13n321 that printing original query even in debug build is not a good idea [1], hence a knob for this. [1]: https://github.com/ClickHouse/ClickHouse/pull/63357/files#r1674809348 Signed-off-by: Azat Khuzhin --- src/Interpreters/executeQuery.cpp | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/src/Interpreters/executeQuery.cpp b/src/Interpreters/executeQuery.cpp index ba6fc0f14a0..ce58f7f922c 100644 --- a/src/Interpreters/executeQuery.cpp +++ b/src/Interpreters/executeQuery.cpp @@ -796,10 +796,9 @@ static std::tuple executeQueryImpl( catch (const Exception & e) { if (e.code() == ErrorCodes::SYNTAX_ERROR) - /// Don't print the original query text because it may contain sensitive data. throw Exception(ErrorCodes::LOGICAL_ERROR, - "Inconsistent AST formatting: the query:\n{}\ncannot parse.", - formatted1); + "Inconsistent AST formatting: the query:\n{}\ncannot parse query back from {}", + formatted1, std::string_view(begin, end-begin)); else throw; } From 8bca80f4dd08d8ad05db0325b96365d82e6c4076 Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Sun, 4 Aug 2024 22:35:21 +0200 Subject: [PATCH 214/227] Fix REPLACE modifier formatting (forbid omitting brackets) It is too tricky to verify does brackets required or not, i.e. "SELECT * REPLACE(1/3/3 AS dummy)" will be formatted to "SELECT * REPLACE (1/3)/3 AS dummy" which is already invalid query. So let's simply always print them. Signed-off-by: Azat Khuzhin v2: move the fix into correct place ASTColumnsReplaceTransformer::formatImpl() instead of ASTColumnsReplaceTransformer::Replacement::formatImpl() --- src/Parsers/ASTColumnsTransformers.cpp | 8 ++------ .../03220_replace_formatting.reference | 16 ++++++++++++++++ .../0_stateless/03220_replace_formatting.sh | 14 ++++++++++++++ 3 files changed, 32 insertions(+), 6 deletions(-) create mode 100644 tests/queries/0_stateless/03220_replace_formatting.reference create mode 100755 tests/queries/0_stateless/03220_replace_formatting.sh diff --git a/src/Parsers/ASTColumnsTransformers.cpp b/src/Parsers/ASTColumnsTransformers.cpp index 2a61892f8cc..332ebca3bdb 100644 --- a/src/Parsers/ASTColumnsTransformers.cpp +++ b/src/Parsers/ASTColumnsTransformers.cpp @@ -323,9 +323,7 @@ void ASTColumnsReplaceTransformer::formatImpl(const FormatSettings & settings, F { settings.ostr << (settings.hilite ? hilite_keyword : "") << "REPLACE" << (is_strict ? " STRICT " : " ") << (settings.hilite ? hilite_none : ""); - if (children.size() > 1) - settings.ostr << "("; - + settings.ostr << "("; for (ASTs::const_iterator it = children.begin(); it != children.end(); ++it) { if (it != children.begin()) @@ -333,9 +331,7 @@ void ASTColumnsReplaceTransformer::formatImpl(const FormatSettings & settings, F (*it)->formatImpl(settings, state, frame); } - - if (children.size() > 1) - settings.ostr << ")"; + settings.ostr << ")"; } void ASTColumnsReplaceTransformer::appendColumnName(WriteBuffer & ostr) const diff --git a/tests/queries/0_stateless/03220_replace_formatting.reference b/tests/queries/0_stateless/03220_replace_formatting.reference new file mode 100644 index 00000000000..cbcd63839b1 --- /dev/null +++ b/tests/queries/0_stateless/03220_replace_formatting.reference @@ -0,0 +1,16 @@ +SELECT * REPLACE ((1 / 3) / 3 AS dummy) +SELECT * REPLACE ((1 / 3) / 3 AS dummy) +SELECT * REPLACE STRICT (1 AS id, 2 AS value) +FROM +( + SELECT + 0 AS id, + 1 AS value +) +SELECT * REPLACE STRICT (1 AS id, 2 AS value) +FROM +( + SELECT + 0 AS id, + 1 AS value +) diff --git a/tests/queries/0_stateless/03220_replace_formatting.sh b/tests/queries/0_stateless/03220_replace_formatting.sh new file mode 100755 index 00000000000..1c11ed6da8d --- /dev/null +++ b/tests/queries/0_stateless/03220_replace_formatting.sh @@ -0,0 +1,14 @@ +#!/usr/bin/env bash + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +q=$($CLICKHOUSE_FORMAT <<<"SELECT * REPLACE(1/3/3 AS dummy)") +echo "$q" +$CLICKHOUSE_FORMAT <<<"$q" + +# multiple columns +q=$($CLICKHOUSE_FORMAT <<<"SELECT * REPLACE STRICT (1 AS id, 2 AS value) FROM (SELECT 0 id, 1 value)") +echo "$q" +$CLICKHOUSE_FORMAT <<<"$q" From 2a7ad3a1f979708fd152e364296c505db8926aba Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Sun, 4 Aug 2024 23:21:19 +0200 Subject: [PATCH 215/227] Update test references for new REPLACE modifier syntax Signed-off-by: Azat Khuzhin --- .../01913_fix_column_transformer_replace_format.reference | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/01913_fix_column_transformer_replace_format.reference b/tests/queries/0_stateless/01913_fix_column_transformer_replace_format.reference index 33be11c07d5..6fabd33c804 100644 --- a/tests/queries/0_stateless/01913_fix_column_transformer_replace_format.reference +++ b/tests/queries/0_stateless/01913_fix_column_transformer_replace_format.reference @@ -1 +1 @@ -CREATE VIEW default.my_view\n(\n `Id` UInt32,\n `Object.Key` Array(UInt16),\n `Object.Value` Array(String)\n)\nAS SELECT * REPLACE arrayMap(x -> (x + 1), `Object.Key`) AS `Object.Key`\nFROM default.my_table +CREATE VIEW default.my_view\n(\n `Id` UInt32,\n `Object.Key` Array(UInt16),\n `Object.Value` Array(String)\n)\nAS SELECT * REPLACE (arrayMap(x -> (x + 1), `Object.Key`) AS `Object.Key`)\nFROM default.my_table From b47f8a733f926c0f52a5837414bcd6ecfece9089 Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Mon, 5 Aug 2024 08:40:35 +0200 Subject: [PATCH 216/227] ci: fix basic errors collecting after stateless tests Signed-off-by: Azat Khuzhin --- docker/test/stateless/run.sh | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/docker/test/stateless/run.sh b/docker/test/stateless/run.sh index b352539cc1a..c582d3a982b 100755 --- a/docker/test/stateless/run.sh +++ b/docker/test/stateless/run.sh @@ -393,6 +393,8 @@ do | zstd --threads=0 > "/test_output/trace-log-$trace_type-flamegraph.tsv.zst" ||: done +# Grep logs for sanitizer asserts, crashes and other critical errors +check_logs_for_critical_errors # Compressed (FIXME: remove once only github actions will be left) rm /var/log/clickhouse-server/clickhouse-server.log @@ -426,7 +428,4 @@ if [[ "$USE_SHARED_CATALOG" -eq 1 ]]; then tar -chf /test_output/coordination1.tar /var/lib/clickhouse1/coordination ||: fi -# Grep logs for sanitizer asserts, crashes and other critical errors -check_logs_for_critical_errors - collect_core_dumps From cdbc4f357324ad0b41d46b6e54475ac2cebdc630 Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Mon, 5 Aug 2024 08:45:20 +0200 Subject: [PATCH 217/227] ci: fail the test if the entrypoint script failed Signed-off-by: Azat Khuzhin --- tests/ci/functional_test_check.py | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/tests/ci/functional_test_check.py b/tests/ci/functional_test_check.py index 66db082677f..52970404d2d 100644 --- a/tests/ci/functional_test_check.py +++ b/tests/ci/functional_test_check.py @@ -166,6 +166,7 @@ def _get_statless_tests_to_run(pr_info: PRInfo) -> List[str]: def process_results( + ret_code: int, result_directory: Path, server_log_path: Path, ) -> Tuple[StatusType, str, TestResults, List[Path]]: @@ -192,6 +193,9 @@ def process_results( logging.info("Files in result folder %s", os.listdir(result_directory)) return ERROR, "Invalid check_status.tsv", test_results, additional_files state, description = status[0][0], status[0][1] + if ret_code != 0: + state = ERROR + description += " (but script exited with an error)" try: results_path = result_directory / "test_results.tsv" @@ -339,7 +343,7 @@ def main(): ci_logs_credentials.clean_ci_logs_from_credentials(run_log_path) state, description, test_results, additional_logs = process_results( - result_path, server_log_path + retcode, result_path, server_log_path ) else: print( From 9ce55b69b49ac8426ed0b3db16b95964e3c4db4d Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Mon, 5 Aug 2024 11:22:55 +0200 Subject: [PATCH 218/227] Fix possible CANNOT_READ_ALL_DATA during server startup in performance tests CI [1]: 2024.08.04 22:09:11.646800 [ 1052 ] {} Application: Code: 33. DB::Exception: Cannot read all data. Bytes read: 0. Bytes expected: 4.: While checking access for disk backups. (CANNOT_READ_ALL_DATA), Stack trace (when copying this message, always include the lines below): [1]: https://s3.amazonaws.com/clickhouse-test-reports/64955/6702acf6f2e4a0ee9697066e38006631fc7f69df/performance_comparison__aarch64__[2_4].html Signed-off-by: Azat Khuzhin --- tests/performance/scripts/compare.sh | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tests/performance/scripts/compare.sh b/tests/performance/scripts/compare.sh index cb56ab6c5bf..da7bbf77a28 100755 --- a/tests/performance/scripts/compare.sh +++ b/tests/performance/scripts/compare.sh @@ -71,6 +71,8 @@ function configure { # Use the new config for both servers, so that we can change it in a PR. rm right/config/config.d/text_log.xml ||: + # backups disk uses absolute path, and this overlaps between servers, that could lead to errors + rm right/config/config.d/backups.xml ||: cp -rv right/config left ||: # Start a temporary server to rename the tables From a499cd25c7e12c05f2f8fa3fe546715c751ad88d Mon Sep 17 00:00:00 2001 From: avogar Date: Mon, 5 Aug 2024 09:31:41 +0000 Subject: [PATCH 219/227] Fix for integers --- src/Functions/if.cpp | 16 ++++++---------- ...3215_varian_as_common_type_integers.reference | 8 ++++++++ .../03215_varian_as_common_type_integers.sql | 8 ++++++++ 3 files changed, 22 insertions(+), 10 deletions(-) create mode 100644 tests/queries/0_stateless/03215_varian_as_common_type_integers.reference create mode 100644 tests/queries/0_stateless/03215_varian_as_common_type_integers.sql diff --git a/src/Functions/if.cpp b/src/Functions/if.cpp index 64da6e95a43..8829b3c4ff1 100644 --- a/src/Functions/if.cpp +++ b/src/Functions/if.cpp @@ -632,11 +632,6 @@ private: ColumnPtr executeTuple(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const { - /// For different Tuples the result type can be Variant with this Tuples if use_variant_as_common_type=1. - /// In this case we should use generic implementation. - if (!isTuple(result_type)) - return nullptr; - /// Calculate function for each corresponding elements of tuples. const ColumnWithTypeAndName & arg1 = arguments[1]; @@ -682,11 +677,6 @@ private: ColumnPtr executeMap(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const { - /// For different Maps the result type can be Variant with this Maps if use_variant_as_common_type=1. - /// In this case we should use generic implementation. - if (!isMap(result_type)) - return nullptr; - auto extract_kv_from_map = [](const ColumnMap * map) { const ColumnTuple & tuple = map->getNestedData(); @@ -1243,6 +1233,12 @@ public: throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}. " "Must be ColumnUInt8 or ColumnConstUInt8.", arg_cond.column->getName(), getName()); + /// If result is Variant, always use generic implementation. + /// Using typed implementations may lead to incorrect result column type when + /// resulting Variant is created by use_variant_when_no_common_type. + if (isVariant(result_type)) + return executeGeneric(cond_col, arguments, input_rows_count, use_variant_when_no_common_type); + auto call = [&](const auto & types) -> bool { using Types = std::decay_t; diff --git a/tests/queries/0_stateless/03215_varian_as_common_type_integers.reference b/tests/queries/0_stateless/03215_varian_as_common_type_integers.reference new file mode 100644 index 00000000000..c5edc9e9963 --- /dev/null +++ b/tests/queries/0_stateless/03215_varian_as_common_type_integers.reference @@ -0,0 +1,8 @@ +0 Variant(Int64, UInt64) +1 Variant(Int64, UInt64) +0 Variant(Int32, UInt64) +1 Variant(Int32, UInt64) +0 Variant(Int16, UInt64) +1 Variant(Int16, UInt64) +0 Variant(Int8, UInt64) +1 Variant(Int8, UInt64) diff --git a/tests/queries/0_stateless/03215_varian_as_common_type_integers.sql b/tests/queries/0_stateless/03215_varian_as_common_type_integers.sql new file mode 100644 index 00000000000..dcc69735534 --- /dev/null +++ b/tests/queries/0_stateless/03215_varian_as_common_type_integers.sql @@ -0,0 +1,8 @@ +set use_variant_as_common_type = 1; +set allow_experimental_variant_type = 1; + +SELECT if(number % 2, number::Int64, number::UInt64) as res, toTypeName(res) FROM numbers(2); +SELECT if(number % 2, number::Int32, number::UInt64) as res, toTypeName(res) FROM numbers(2); +SELECT if(number % 2, number::Int16, number::UInt64) as res, toTypeName(res) FROM numbers(2); +SELECT if(number % 2, number::Int8, number::UInt64) as res, toTypeName(res) FROM numbers(2); + From 2b369cccdd58902e9da3fd3947e5cb5759ba2881 Mon Sep 17 00:00:00 2001 From: avogar Date: Mon, 5 Aug 2024 09:35:56 +0000 Subject: [PATCH 220/227] Reduce table size in 03037_dynamic_merges_2_vertical_wide_merge_tree test --- ...3037_dynamic_merges_2_vertical_wide_merge_tree.reference | 6 +++--- .../03037_dynamic_merges_2_vertical_wide_merge_tree.sql | 6 +++--- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_wide_merge_tree.reference b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_wide_merge_tree.reference index afd392002e5..253d87de5f0 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_wide_merge_tree.reference +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_wide_merge_tree.reference @@ -1,3 +1,3 @@ -1000000 Array(UInt16) -1000000 String -1000000 UInt64 +200000 Array(UInt16) +200000 String +200000 UInt64 diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_wide_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_wide_merge_tree.sql index 104d6018e41..dd643f8dffd 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_wide_merge_tree.sql +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_wide_merge_tree.sql @@ -5,9 +5,9 @@ set allow_experimental_dynamic_type = 1; drop table if exists test; create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1, vertical_merge_algorithm_min_rows_to_activate=1, vertical_merge_algorithm_min_columns_to_activate=1, lock_acquire_timeout_for_background_operations=600; system stop merges test; -insert into test select number, number from numbers(1000000); -insert into test select number, 'str_' || toString(number) from numbers(1000000, 1000000); -insert into test select number, range(number % 10 + 1) from numbers(2000000, 1000000); +insert into test select number, number from numbers(200000); +insert into test select number, 'str_' || toString(number) from numbers(200000, 200000); +insert into test select number, range(number % 10 + 1) from numbers(400000, 200000); system start merges test; optimize table test final; select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); From cec8a5d52b83f0c1cdcaed833aec9bf79941b2a8 Mon Sep 17 00:00:00 2001 From: avogar Date: Mon, 5 Aug 2024 09:42:22 +0000 Subject: [PATCH 221/227] Reduce table size in similar tests --- ...dynamic_merges_2_horizontal_compact_merge_tree.reference | 6 +++--- ...03037_dynamic_merges_2_horizontal_compact_merge_tree.sql | 6 +++--- ...37_dynamic_merges_2_horizontal_wide_merge_tree.reference | 6 +++--- .../03037_dynamic_merges_2_horizontal_wide_merge_tree.sql | 6 +++--- ...7_dynamic_merges_2_vertical_compact_merge_tree.reference | 6 +++--- .../03037_dynamic_merges_2_vertical_compact_merge_tree.sql | 6 +++--- 6 files changed, 18 insertions(+), 18 deletions(-) diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_compact_merge_tree.reference b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_compact_merge_tree.reference index afd392002e5..253d87de5f0 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_compact_merge_tree.reference +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_compact_merge_tree.reference @@ -1,3 +1,3 @@ -1000000 Array(UInt16) -1000000 String -1000000 UInt64 +200000 Array(UInt16) +200000 String +200000 UInt64 diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_compact_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_compact_merge_tree.sql index e133ac3001f..fa64ed2f8fd 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_compact_merge_tree.sql +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_compact_merge_tree.sql @@ -5,9 +5,9 @@ set allow_experimental_dynamic_type = 1; drop table if exists test; create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000, lock_acquire_timeout_for_background_operations=600; system stop merges test; -insert into test select number, number from numbers(1000000); -insert into test select number, 'str_' || toString(number) from numbers(1000000, 1000000); -insert into test select number, range(number % 10 + 1) from numbers(2000000, 1000000); +insert into test select number, number from numbers(200000); +insert into test select number, 'str_' || toString(number) from numbers(200000, 200000); +insert into test select number, range(number % 10 + 1) from numbers(400000, 200000); system start merges test; optimize table test final; select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_wide_merge_tree.reference b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_wide_merge_tree.reference index afd392002e5..253d87de5f0 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_wide_merge_tree.reference +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_wide_merge_tree.reference @@ -1,3 +1,3 @@ -1000000 Array(UInt16) -1000000 String -1000000 UInt64 +200000 Array(UInt16) +200000 String +200000 UInt64 diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_wide_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_wide_merge_tree.sql index d527081b763..4b8a036f166 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_wide_merge_tree.sql +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_wide_merge_tree.sql @@ -5,9 +5,9 @@ set allow_experimental_dynamic_type = 1; drop table if exists test; create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1, lock_acquire_timeout_for_background_operations=600; system stop merges test; -insert into test select number, number from numbers(1000000); -insert into test select number, 'str_' || toString(number) from numbers(1000000, 1000000); -insert into test select number, range(number % 10 + 1) from numbers(2000000, 1000000); +insert into test select number, number from numbers(200000); +insert into test select number, 'str_' || toString(number) from numbers(200000, 200000); +insert into test select number, range(number % 10 + 1) from numbers(400000, 200000); system start merges test; optimize table test final; select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_compact_merge_tree.reference b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_compact_merge_tree.reference index afd392002e5..253d87de5f0 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_compact_merge_tree.reference +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_compact_merge_tree.reference @@ -1,3 +1,3 @@ -1000000 Array(UInt16) -1000000 String -1000000 UInt64 +200000 Array(UInt16) +200000 String +200000 UInt64 diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_compact_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_compact_merge_tree.sql index ebccfb77922..a4e67de76db 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_compact_merge_tree.sql +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_compact_merge_tree.sql @@ -5,9 +5,9 @@ set allow_experimental_dynamic_type = 1; drop table if exists test; create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000, vertical_merge_algorithm_min_rows_to_activate=1, vertical_merge_algorithm_min_columns_to_activate=1, lock_acquire_timeout_for_background_operations=600; system stop merges test; -insert into test select number, number from numbers(1000000); -insert into test select number, 'str_' || toString(number) from numbers(1000000, 1000000); -insert into test select number, range(number % 10 + 1) from numbers(2000000, 1000000); +insert into test select number, number from numbers(200000); +insert into test select number, 'str_' || toString(number) from numbers(200000, 200000); +insert into test select number, range(number % 10 + 1) from numbers(400000, 200000); system start merges test; optimize table test final; select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); From 0a7a67b8e0e05bc9476d4f9dd38747bf61b6bb8e Mon Sep 17 00:00:00 2001 From: avogar Date: Mon, 5 Aug 2024 09:57:13 +0000 Subject: [PATCH 222/227] Disable 03038_nested_dynamic_merges* under sanitizers because it's too slow --- ...sted_dynamic_merges_compact_horizontal.sql | 2 +- ...nested_dynamic_merges_compact_vertical.sql | 2 +- ...8_nested_dynamic_merges_small.reference.j2 | 84 +++++++++++++++++++ .../03038_nested_dynamic_merges_small.sql.j2 | 35 ++++++++ ..._nested_dynamic_merges_wide_horizontal.sql | 2 +- ...38_nested_dynamic_merges_wide_vertical.sql | 2 +- 6 files changed, 123 insertions(+), 4 deletions(-) create mode 100644 tests/queries/0_stateless/03038_nested_dynamic_merges_small.reference.j2 create mode 100644 tests/queries/0_stateless/03038_nested_dynamic_merges_small.sql.j2 diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_horizontal.sql b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_horizontal.sql index 1d5c63dcdf1..81888946681 100644 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_horizontal.sql +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_horizontal.sql @@ -1,4 +1,4 @@ --- Tags: long +-- Tags: long, no-tsan, no-msan, no-ubsan, no-asan set allow_experimental_variant_type = 1; set use_variant_as_common_type = 1; diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_vertical.sql b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_vertical.sql index 2bffe35c577..ba58ca471a2 100644 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_vertical.sql +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_vertical.sql @@ -1,4 +1,4 @@ --- Tags: long +-- Tags: long, no-tsan, no-msan, no-ubsan, no-asan set allow_experimental_variant_type = 1; set use_variant_as_common_type = 1; diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_small.reference.j2 b/tests/queries/0_stateless/03038_nested_dynamic_merges_small.reference.j2 new file mode 100644 index 00000000000..ae07c164074 --- /dev/null +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_small.reference.j2 @@ -0,0 +1,84 @@ +2 Tuple(a Dynamic(max_types=3)):Date +3 Tuple(a Dynamic(max_types=3)):Array(UInt8) +5 Tuple(a Dynamic(max_types=3)):String +5 Tuple(a Dynamic(max_types=3)):UInt64 +10 UInt64:None +3 Tuple(a Dynamic(max_types=3)):Array(UInt8) +5 Tuple(a Dynamic(max_types=3)):UInt64 +7 Tuple(a Dynamic(max_types=3)):String +10 UInt64:None +2 Tuple(a Dynamic(max_types=3)):DateTime +3 Tuple(a Dynamic(max_types=3)):Array(UInt8) +5 Tuple(a Dynamic(max_types=3)):UInt64 +7 Tuple(a Dynamic(max_types=3)):String +10 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) +10 UInt64:None +13 Tuple(a Dynamic(max_types=3)):None +5 Tuple(a Dynamic(max_types=3)):UInt64 +10 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) +10 UInt64:None +12 Tuple(a Dynamic(max_types=3)):String +13 Tuple(a Dynamic(max_types=3)):None +2 Tuple(a Dynamic(max_types=3)):Date +3 Tuple(a Dynamic(max_types=3)):Array(UInt8) +5 Tuple(a Dynamic(max_types=3)):String +5 Tuple(a Dynamic(max_types=3)):UInt64 +10 UInt64:None +3 Tuple(a Dynamic(max_types=3)):Array(UInt8) +5 Tuple(a Dynamic(max_types=3)):UInt64 +7 Tuple(a Dynamic(max_types=3)):String +10 UInt64:None +2 Tuple(a Dynamic(max_types=3)):DateTime +3 Tuple(a Dynamic(max_types=3)):Array(UInt8) +5 Tuple(a Dynamic(max_types=3)):UInt64 +7 Tuple(a Dynamic(max_types=3)):String +10 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) +10 UInt64:None +13 Tuple(a Dynamic(max_types=3)):None +5 Tuple(a Dynamic(max_types=3)):UInt64 +10 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) +10 UInt64:None +12 Tuple(a Dynamic(max_types=3)):String +13 Tuple(a Dynamic(max_types=3)):None +2 Tuple(a Dynamic(max_types=3)):Date +3 Tuple(a Dynamic(max_types=3)):Array(UInt8) +5 Tuple(a Dynamic(max_types=3)):String +5 Tuple(a Dynamic(max_types=3)):UInt64 +10 UInt64:None +3 Tuple(a Dynamic(max_types=3)):Array(UInt8) +5 Tuple(a Dynamic(max_types=3)):UInt64 +7 Tuple(a Dynamic(max_types=3)):String +10 UInt64:None +2 Tuple(a Dynamic(max_types=3)):DateTime +3 Tuple(a Dynamic(max_types=3)):Array(UInt8) +5 Tuple(a Dynamic(max_types=3)):UInt64 +7 Tuple(a Dynamic(max_types=3)):String +10 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) +10 UInt64:None +13 Tuple(a Dynamic(max_types=3)):None +5 Tuple(a Dynamic(max_types=3)):UInt64 +10 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) +10 UInt64:None +12 Tuple(a Dynamic(max_types=3)):String +13 Tuple(a Dynamic(max_types=3)):None +2 Tuple(a Dynamic(max_types=3)):Date +3 Tuple(a Dynamic(max_types=3)):Array(UInt8) +5 Tuple(a Dynamic(max_types=3)):String +5 Tuple(a Dynamic(max_types=3)):UInt64 +10 UInt64:None +3 Tuple(a Dynamic(max_types=3)):Array(UInt8) +5 Tuple(a Dynamic(max_types=3)):UInt64 +7 Tuple(a Dynamic(max_types=3)):String +10 UInt64:None +2 Tuple(a Dynamic(max_types=3)):DateTime +3 Tuple(a Dynamic(max_types=3)):Array(UInt8) +5 Tuple(a Dynamic(max_types=3)):UInt64 +7 Tuple(a Dynamic(max_types=3)):String +10 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) +10 UInt64:None +13 Tuple(a Dynamic(max_types=3)):None +5 Tuple(a Dynamic(max_types=3)):UInt64 +10 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) +10 UInt64:None +12 Tuple(a Dynamic(max_types=3)):String +13 Tuple(a Dynamic(max_types=3)):None diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_small.sql.j2 b/tests/queries/0_stateless/03038_nested_dynamic_merges_small.sql.j2 new file mode 100644 index 00000000000..7828c2af49c --- /dev/null +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_small.sql.j2 @@ -0,0 +1,35 @@ +set allow_experimental_variant_type = 1; +set use_variant_as_common_type = 1; +set allow_experimental_dynamic_type = 1; +set enable_named_columns_in_function_tuple = 0; + +drop table if exists test; + +{% for engine in ['MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000', + 'MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000, vertical_merge_algorithm_min_rows_to_activate=1, vertical_merge_algorithm_min_columns_to_activate=1', + 'MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1', + 'MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1, vertical_merge_algorithm_min_rows_to_activate=1, vertical_merge_algorithm_min_columns_to_activate=1'] -%} + +create table test (id UInt64, d Dynamic(max_types=3)) engine=MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1, vertical_merge_algorithm_min_rows_to_activate=1, vertical_merge_algorithm_min_columns_to_activate=1, lock_acquire_timeout_for_background_operations=600; + +system stop merges test; +insert into test select number, number from numbers(10); +insert into test select number, tuple(if(number % 2 == 0, number, 'str_' || toString(number)))::Tuple(a Dynamic(max_types=3)) from numbers(10); +insert into test select number, tuple(if(number % 3 == 0, toDate(number), range(number % 10)))::Tuple(a Dynamic(max_types=3)) from numbers(5); + +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; +system start merges test; +optimize table test final; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; + +insert into test select number, tuple(if(number % 3 == 0, toDateTime(number), NULL))::Tuple(a Dynamic(max_types=3)) from numbers(5); +insert into test select number, tuple(if(number % 2 == 0, tuple(number), NULL))::Tuple(a Dynamic(max_types=3)) from numbers(20); + +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; +system start merges test; +optimize table test final; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; + +drop table test; + +{% endfor -%} diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_horizontal.sql b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_horizontal.sql index fb686091ebb..a53c5b0b2a5 100644 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_horizontal.sql +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_horizontal.sql @@ -1,4 +1,4 @@ --- Tags: long +-- Tags: long, no-tsan, no-msan, no-ubsan, no-asan set allow_experimental_variant_type = 1; set use_variant_as_common_type = 1; diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_vertical.sql b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_vertical.sql index ed195452d56..4256b010ec0 100644 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_vertical.sql +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_vertical.sql @@ -1,4 +1,4 @@ --- Tags: long +-- Tags: long, no-tsan, no-msan, no-ubsan, no-asan set allow_experimental_variant_type = 1; set use_variant_as_common_type = 1; From 3802b1ed6c7174b0d95bf1c89d339187fe6dc69d Mon Sep 17 00:00:00 2001 From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com> Date: Mon, 5 Aug 2024 12:42:57 +0200 Subject: [PATCH 223/227] Update comment --- src/Interpreters/Cache/FileCache.cpp | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/src/Interpreters/Cache/FileCache.cpp b/src/Interpreters/Cache/FileCache.cpp index 4c17afb79be..aff4e48d01d 100644 --- a/src/Interpreters/Cache/FileCache.cpp +++ b/src/Interpreters/Cache/FileCache.cpp @@ -326,6 +326,8 @@ std::vector FileCache::splitRange(size_t offset, size_t size /// ^ ^ /// right offset aligned_right_offset /// [_________] <-- last cached file segment, e.g. we have uncovered suffix of the requested range + /// ^ + /// last_file_segment_right_offset /// [________________] /// size /// [____________________________________] @@ -335,8 +337,9 @@ std::vector FileCache::splitRange(size_t offset, size_t size /// and get something like this: /// /// [________________________] - /// ^ ^ - /// right_offset right_offset + max_file_segment_size + /// ^ ^ + /// | last_file_segment_right_offset + max_file_segment_size + /// last_file_segment_right_offset /// e.g. there is no need to create sub-segment for range (right_offset + max_file_segment_size, aligned_right_offset]. /// Because its left offset would be bigger than right_offset. /// Therefore, we set end_pos_non_included as offset+size, but remaining_size as aligned_size. From 6a0c0e7b1d4922b616d706820d9000ffe8040d63 Mon Sep 17 00:00:00 2001 From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com> Date: Mon, 5 Aug 2024 12:48:21 +0200 Subject: [PATCH 224/227] Update FileCache.cpp --- src/Interpreters/Cache/FileCache.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Interpreters/Cache/FileCache.cpp b/src/Interpreters/Cache/FileCache.cpp index aff4e48d01d..ed91e41db17 100644 --- a/src/Interpreters/Cache/FileCache.cpp +++ b/src/Interpreters/Cache/FileCache.cpp @@ -340,7 +340,7 @@ std::vector FileCache::splitRange(size_t offset, size_t size /// ^ ^ /// | last_file_segment_right_offset + max_file_segment_size /// last_file_segment_right_offset - /// e.g. there is no need to create sub-segment for range (right_offset + max_file_segment_size, aligned_right_offset]. + /// e.g. there is no need to create sub-segment for range (last_file_segment_right_offset + max_file_segment_size, aligned_right_offset]. /// Because its left offset would be bigger than right_offset. /// Therefore, we set end_pos_non_included as offset+size, but remaining_size as aligned_size. From 3244002cae58da99c8b088888376c0fda7f3f1f4 Mon Sep 17 00:00:00 2001 From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com> Date: Mon, 5 Aug 2024 13:22:10 +0200 Subject: [PATCH 225/227] Update FileCache.cpp --- src/Interpreters/Cache/FileCache.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Interpreters/Cache/FileCache.cpp b/src/Interpreters/Cache/FileCache.cpp index ed91e41db17..1a15efa7cf8 100644 --- a/src/Interpreters/Cache/FileCache.cpp +++ b/src/Interpreters/Cache/FileCache.cpp @@ -560,7 +560,7 @@ FileCache::getOrSet( FileSegment::Range initial_range(offset, offset + size - 1); /// result_range is initial range, which will be adjusted according to - /// 1. aligned offset, alighed_end_offset + /// 1. aligned_offset, aligned_end_offset /// 2. max_file_segments_limit FileSegment::Range result_range = initial_range; From c21b97672c86a18b693c8c60335271248c279fdd Mon Sep 17 00:00:00 2001 From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com> Date: Mon, 5 Aug 2024 14:36:54 +0200 Subject: [PATCH 226/227] Fix trailing whitespace --- src/Interpreters/Cache/FileCache.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Interpreters/Cache/FileCache.cpp b/src/Interpreters/Cache/FileCache.cpp index 1a15efa7cf8..217ae614c22 100644 --- a/src/Interpreters/Cache/FileCache.cpp +++ b/src/Interpreters/Cache/FileCache.cpp @@ -337,7 +337,7 @@ std::vector FileCache::splitRange(size_t offset, size_t size /// and get something like this: /// /// [________________________] - /// ^ ^ + /// ^ ^ /// | last_file_segment_right_offset + max_file_segment_size /// last_file_segment_right_offset /// e.g. there is no need to create sub-segment for range (last_file_segment_right_offset + max_file_segment_size, aligned_right_offset]. From e4c71aec4dfa40d7e9b75c626b0229a33c59f315 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Mon, 5 Aug 2024 16:12:25 +0000 Subject: [PATCH 227/227] Update version_date.tsv and changelogs after v24.5.5.78-stable --- docs/changelogs/v24.5.5.78-stable.md | 55 ++++++++++++++++++++++++++++ utils/list-versions/version_date.tsv | 1 + 2 files changed, 56 insertions(+) create mode 100644 docs/changelogs/v24.5.5.78-stable.md diff --git a/docs/changelogs/v24.5.5.78-stable.md b/docs/changelogs/v24.5.5.78-stable.md new file mode 100644 index 00000000000..415ea165101 --- /dev/null +++ b/docs/changelogs/v24.5.5.78-stable.md @@ -0,0 +1,55 @@ +--- +sidebar_position: 1 +sidebar_label: 2024 +--- + +# 2024 Changelog + +### ClickHouse release v24.5.5.78-stable (0138248cb62) FIXME as compared to v24.5.4.49-stable (63b760955a0) + +#### Improvement +* Backported in [#66768](https://github.com/ClickHouse/ClickHouse/issues/66768): Make allow_experimental_analyzer be controlled by the initiator for distributed queries. This ensures compatibility and correctness during operations in mixed version clusters. [#65777](https://github.com/ClickHouse/ClickHouse/pull/65777) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). + +#### Critical Bug Fix (crash, LOGICAL_ERROR, data loss, RBAC) +* Backported in [#66884](https://github.com/ClickHouse/ClickHouse/issues/66884): Fix unexpeced size of low cardinality column in function calls. [#65298](https://github.com/ClickHouse/ClickHouse/pull/65298) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#66691](https://github.com/ClickHouse/ClickHouse/issues/66691): Fix the VALID UNTIL clause in the user definition resetting after a restart. Closes [#66405](https://github.com/ClickHouse/ClickHouse/issues/66405). [#66409](https://github.com/ClickHouse/ClickHouse/pull/66409) ([Nikolay Degterinsky](https://github.com/evillique)). +* Backported in [#67814](https://github.com/ClickHouse/ClickHouse/issues/67814): Only relevant to the experimental Variant data type. Fix crash with Variant + AggregateFunction type. [#67122](https://github.com/ClickHouse/ClickHouse/pull/67122) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#67501](https://github.com/ClickHouse/ClickHouse/issues/67501): Fix crash in DistributedAsyncInsert when connection is empty. [#67219](https://github.com/ClickHouse/ClickHouse/pull/67219) ([Pablo Marcos](https://github.com/pamarcos)). +* Backported in [#67850](https://github.com/ClickHouse/ClickHouse/issues/67850): Fixes [#66026](https://github.com/ClickHouse/ClickHouse/issues/66026). Avoid unresolved table function arguments traversal in `ReplaceTableNodeToDummyVisitor`. [#67522](https://github.com/ClickHouse/ClickHouse/pull/67522) ([Dmitry Novik](https://github.com/novikd)). + +#### Bug Fix (user-visible misbehavior in an official stable release) +* Backported in [#65350](https://github.com/ClickHouse/ClickHouse/issues/65350): Fix possible abort on uncaught exception in ~WriteBufferFromFileDescriptor in StatusFile. [#64206](https://github.com/ClickHouse/ClickHouse/pull/64206) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#65621](https://github.com/ClickHouse/ClickHouse/issues/65621): Fix `Cannot find column` in distributed query with `ARRAY JOIN` by `Nested` column. Fixes [#64755](https://github.com/ClickHouse/ClickHouse/issues/64755). [#64801](https://github.com/ClickHouse/ClickHouse/pull/64801) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#65933](https://github.com/ClickHouse/ClickHouse/issues/65933): For queries that read from `PostgreSQL`, cancel the internal `PostgreSQL` query if the ClickHouse query is finished. Otherwise, `ClickHouse` query cannot be canceled until the internal `PostgreSQL` query is finished. [#65771](https://github.com/ClickHouse/ClickHouse/pull/65771) ([Maksim Kita](https://github.com/kitaisreal)). +* Backported in [#66301](https://github.com/ClickHouse/ClickHouse/issues/66301): Better handling of join conditions involving `IS NULL` checks (for example `ON (a = b AND (a IS NOT NULL) AND (b IS NOT NULL) ) OR ( (a IS NULL) AND (b IS NULL) )` is rewritten to `ON a <=> b`), fix incorrect optimization when condition other then `IS NULL` are present. [#65835](https://github.com/ClickHouse/ClickHouse/pull/65835) ([vdimir](https://github.com/vdimir)). +* Backported in [#66328](https://github.com/ClickHouse/ClickHouse/issues/66328): Add missing settings `input_format_csv_skip_first_lines/input_format_tsv_skip_first_lines/input_format_csv_try_infer_numbers_from_strings/input_format_csv_try_infer_strings_from_quoted_tuples` in schema inference cache because they can change the resulting schema. It prevents from incorrect result of schema inference with these settings changed. [#65980](https://github.com/ClickHouse/ClickHouse/pull/65980) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#66155](https://github.com/ClickHouse/ClickHouse/issues/66155): Fixed buffer overflow bug in `unbin`/`unhex` implementation. [#66106](https://github.com/ClickHouse/ClickHouse/pull/66106) ([Nikita Taranov](https://github.com/nickitat)). +* Backported in [#66454](https://github.com/ClickHouse/ClickHouse/issues/66454): Fixed a bug in ZooKeeper client: a session could get stuck in unusable state after receiving a hardware error from ZooKeeper. For example, this might happen due to "soft memory limit" in ClickHouse Keeper. [#66140](https://github.com/ClickHouse/ClickHouse/pull/66140) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Backported in [#66226](https://github.com/ClickHouse/ClickHouse/issues/66226): Fix issue in SumIfToCountIfVisitor and signed integers. [#66146](https://github.com/ClickHouse/ClickHouse/pull/66146) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#66680](https://github.com/ClickHouse/ClickHouse/issues/66680): Fix handling limit for `system.numbers_mt` when no index can be used. [#66231](https://github.com/ClickHouse/ClickHouse/pull/66231) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#66604](https://github.com/ClickHouse/ClickHouse/issues/66604): Fixed how the ClickHouse server detects the maximum number of usable CPU cores as specified by cgroups v2 if the server runs in a container such as Docker. In more detail, containers often run their process in the root cgroup which has an empty name. In that case, ClickHouse ignored the CPU limits set by cgroups v2. [#66237](https://github.com/ClickHouse/ClickHouse/pull/66237) ([filimonov](https://github.com/filimonov)). +* Backported in [#66360](https://github.com/ClickHouse/ClickHouse/issues/66360): Fix the `Not-ready set` error when a subquery with `IN` is used in the constraint. [#66261](https://github.com/ClickHouse/ClickHouse/pull/66261) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#66972](https://github.com/ClickHouse/ClickHouse/issues/66972): Fix `Column identifier is already registered` error with `group_by_use_nulls=true` and new analyzer. [#66400](https://github.com/ClickHouse/ClickHouse/pull/66400) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#66969](https://github.com/ClickHouse/ClickHouse/issues/66969): Fix `Cannot find column` error for queries with constant expression in `GROUP BY` key and new analyzer enabled. [#66433](https://github.com/ClickHouse/ClickHouse/pull/66433) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#66720](https://github.com/ClickHouse/ClickHouse/issues/66720): Correctly track memory for `Allocator::realloc`. [#66548](https://github.com/ClickHouse/ClickHouse/pull/66548) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#66951](https://github.com/ClickHouse/ClickHouse/issues/66951): Fix an invalid result for queries with `WINDOW`. This could happen when `PARTITION` columns have sparse serialization and window functions are executed in parallel. [#66579](https://github.com/ClickHouse/ClickHouse/pull/66579) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#66757](https://github.com/ClickHouse/ClickHouse/issues/66757): Fix `Unknown identifier` and `Column is not under aggregate function` errors for queries with the expression `(column IS NULL).` The bug was triggered by [#65088](https://github.com/ClickHouse/ClickHouse/issues/65088), with the disabled analyzer only. [#66654](https://github.com/ClickHouse/ClickHouse/pull/66654) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#66948](https://github.com/ClickHouse/ClickHouse/issues/66948): Fix `Method getResultType is not supported for QUERY query node` error when scalar subquery was used as the first argument of IN (with new analyzer). [#66655](https://github.com/ClickHouse/ClickHouse/pull/66655) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#67633](https://github.com/ClickHouse/ClickHouse/issues/67633): Fix for occasional deadlock in Context::getDDLWorker. [#66843](https://github.com/ClickHouse/ClickHouse/pull/66843) ([Alexander Gololobov](https://github.com/davenger)). +* Backported in [#67481](https://github.com/ClickHouse/ClickHouse/issues/67481): In rare cases ClickHouse could consider parts as broken because of some unexpected projections on disk. Now it's fixed. [#66898](https://github.com/ClickHouse/ClickHouse/pull/66898) ([alesapin](https://github.com/alesapin)). +* Backported in [#67197](https://github.com/ClickHouse/ClickHouse/issues/67197): TRUNCATE DATABASE used to stop replication as if it was a DROP DATABASE query, it's fixed. [#67129](https://github.com/ClickHouse/ClickHouse/pull/67129) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Backported in [#67379](https://github.com/ClickHouse/ClickHouse/issues/67379): Fix error `Cannot convert column because it is non constant in source stream but must be constant in result.` for a query that reads from the `Merge` table over the `Distriburted` table with one shard. [#67146](https://github.com/ClickHouse/ClickHouse/pull/67146) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#67576](https://github.com/ClickHouse/ClickHouse/issues/67576): Fix execution of nested short-circuit functions. [#67520](https://github.com/ClickHouse/ClickHouse/pull/67520) ([Kruglov Pavel](https://github.com/Avogar)). + +#### NOT FOR CHANGELOG / INSIGNIFICANT + +* Backported in [#66387](https://github.com/ClickHouse/ClickHouse/issues/66387): Disable broken cases from 02911_join_on_nullsafe_optimization. [#66310](https://github.com/ClickHouse/ClickHouse/pull/66310) ([vdimir](https://github.com/vdimir)). +* Backported in [#66426](https://github.com/ClickHouse/ClickHouse/issues/66426): Ignore subquery for IN in DDLLoadingDependencyVisitor. [#66395](https://github.com/ClickHouse/ClickHouse/pull/66395) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#66544](https://github.com/ClickHouse/ClickHouse/issues/66544): Add additional log masking in CI. [#66523](https://github.com/ClickHouse/ClickHouse/pull/66523) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#66859](https://github.com/ClickHouse/ClickHouse/issues/66859): Fix data race in S3::ClientCache. [#66644](https://github.com/ClickHouse/ClickHouse/pull/66644) ([Konstantin Morozov](https://github.com/k-morozov)). +* Backported in [#66875](https://github.com/ClickHouse/ClickHouse/issues/66875): Support one more case in JOIN ON ... IS NULL. [#66725](https://github.com/ClickHouse/ClickHouse/pull/66725) ([vdimir](https://github.com/vdimir)). +* Backported in [#67059](https://github.com/ClickHouse/ClickHouse/issues/67059): Increase asio pool size in case the server is tiny. [#66761](https://github.com/ClickHouse/ClickHouse/pull/66761) ([alesapin](https://github.com/alesapin)). +* Backported in [#66945](https://github.com/ClickHouse/ClickHouse/issues/66945): Small fix in realloc memory tracking. [#66820](https://github.com/ClickHouse/ClickHouse/pull/66820) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#67252](https://github.com/ClickHouse/ClickHouse/issues/67252): Followup [#66725](https://github.com/ClickHouse/ClickHouse/issues/66725). [#66869](https://github.com/ClickHouse/ClickHouse/pull/66869) ([vdimir](https://github.com/vdimir)). +* Backported in [#67412](https://github.com/ClickHouse/ClickHouse/issues/67412): CI: Fix build results for release branches. [#67402](https://github.com/ClickHouse/ClickHouse/pull/67402) ([Max K.](https://github.com/maxknv)). + diff --git a/utils/list-versions/version_date.tsv b/utils/list-versions/version_date.tsv index cb6b8f588da..75c10fa67b8 100644 --- a/utils/list-versions/version_date.tsv +++ b/utils/list-versions/version_date.tsv @@ -2,6 +2,7 @@ v24.7.2.13-stable 2024-08-01 v24.7.1.2915-stable 2024-07-30 v24.6.2.17-stable 2024-07-05 v24.6.1.4423-stable 2024-07-01 +v24.5.5.78-stable 2024-08-05 v24.5.4.49-stable 2024-07-01 v24.5.3.5-stable 2024-06-13 v24.5.2.34-stable 2024-06-13